Virtual Library

Background
Background: recent evidence shows a role for immunotherapy via checkpoint antibodies in a subgroup of patients with non-small cell lung cancer (NSCLC). We have recently published the complex interplay between tumors and the immune system[1]. The modulating effect of the tumor on the immune system, mainly immunosuppressive, is challenging. MDSC are described to play a major role in this tumor associated immune suppression by inhibiting cytotoxic Tcell function. The aim of the present study is to relate the number of MDSC to two well-known prognostic factors in NSCLC: stage of the disease and WHO performance status.

Methods
Methods: MDSC were determined in peripheral blood mononucleair cell fractions of 195 treatment-naive NSCLC patients (185 stage IV (aNSQ), 10 stage I-III(eNSQ), and 20 healthy controls (HC)). WHO performance status was determined by experienced investigators prior to the start of treatment. In this abstract the relation between clinical data and polynucleair MDSC are presented. Flowcytometric analysis was performed within 5 hours after the blood was drawn. MDSC were characterized as CD16low,CD11b+,CD14-,HLA-DR-,CD15+, and CD33+.

Results
Results: A large variation in the number of MDSC was observed in patients with NSCLC, also in the same stage of the disease and WHO performance status. In a subgroup of patients with eNSQ high MDSC levels were found, but also in patients with aNSQ low levels of MDSC were found. For the group as a whole, however, an increased number of MDSC was found in patients with aNSQ compared to HC and eNSQ (p<.001, in both comparisons). Comparison of the limited number of patients with of eNSQ and HC no statistical difference was found but there is a tendency for an increase in the number of MDSC per stage of NSCLC. The number of MDSC increased with worse WHO performance status ( p=0.05 between 0 and 2). Whether the number of MDSC has prognostic or predictive value to response to chemotherapy in all patients is subject of current research. Follow up data in all patients with response to chemotherapy, progression free survival and overall survival are at present collected.

Conclusion
Conclusion: Peripheral blood MDSC levels are increased in patients with aNSQ compared to eNSQ and HC. MDSC are increased in patients with worse WHO performance status. The number of MDSC shows large variation between patients with similar stage of the disease, showing the complex interplay between tumor and immune system. 1 Aerts JG, Hegmans JP, Cancer Research 2013

Background
Immune modulatory strategies directed against CTLA4 (Cytotoxic T-Lymphocyte Antigen 4), PD1(Programmed cell Death 1) and PD-L1 (Programmed cell Death 1 Ligand 1) either in monotherapy or in combination with chemotherapy and radiotherapy offer great promise as novel treatments against lung cancer. Its outcome and therapeutic ratio may depend on the expression of the target molecules in tumours and the surrounding non-malignant lung tissue. We here report the expression of CTLA4, PD1 and PD-L1 in primary resected NSCLC and in the surrounding non-malignant lung.

Methods
RNA sequencing was performed on tumor tissue and normal lung tissue from resection specimens of NSCLC patients. The association between CTLA4, PD1 and PD-L1 expression levels and histology, gender, age, smoking status (current smoker vs. smoking cessation of at least one year), COPD and CRP blood levels was calculated both in the primary tumour and the normal lung. Results are expressed as mean +/- SD and range. Means were compared using Wilcoxon’s signed rank test for the distributions were non-parametrical. P-values <0.05 were considered significant.

Results
Fourteen patients were studied, 11 males and 3 females, with a mean age of 64.6 +/- 10.0 years (41-79). All patients had a smoking history: 6 were current smokers, 8 former smokers. COPD status: 8 no COPD, 1 GOLD class I, 3 GOLD class II, 2 GOLD class III. Pre-operative CRP (mg/dL): 15.0 +/- 2.9 (0.20-54.7). Seven patients had squamous cell cancer, 7 adenocarcinoma. PD-L1 expression in the lung was 3833 +/- 787 (372-5072), PD-L1 in the tumour 2595 +/- 1657 (788-6689), with a tumour/lung ratio of 0.67 +/- 0.37 (0.21-1.32). PD-L1 squamous vs. adenocarcinoma was 1840 +/- 1012 vs. 3350 +/- 1896, p<0.001. PD-L1 in the lung of ex-smokers vs. current smokers: 4136 +/- 811 vs. 3430 ± 591, p<0.001. These associations were not found for CTLA4 and PD1 receptor: CTLA4 tumour squamous vs. adenocarcinoma: 100.9 +/- 46.3 vs. 95.3 +/- 73.6, p=0.86; CTLA4 lung smokers vs. ex-smokers: 78.4 +/- 41.3 vs. 107.0 +/- 75.1, p=0.75. PD1 receptor tumour squamous vs. adenocarcinoma: 591 +/- 244 vs. 689 +/- 263, p=0.31 PD1 receptor in the lung of smokers vs. ex-smokers: 599 +/- 278 vs. 695 +/-216, p=0.46. No associations between CTLA4, PD1 and PD-L1 in the tumour or the lung and gender, age, COPD and CRP blood levels were found.

Conclusion
RNA sequencing is a useful method to dissect relevant molecules in the tumour and the lungs. Our results show more PD-L1 expression in adenocarcinoma than in squamous cell cancer, more PD-L1 expression in the non-malignant lung of ex-smokers than in current smokers and reduced PD-L1 in tumours compared to adjacent lung tissue.

Background
Cancer metastasis is the main cause of cancer-related deaths. Metastatic cancer cells spread through blood vessels where they constantly interact with platelets and leukocytes, forming tumor microemboli and thereby protected from otherwise rapid elimination from host immune defense cells such as NK cells. Platelets are known to be prometastatic and pro-angiogenic. They release platelet-derived growth factors, chemokines, various adhesion molecules and coagulation factors that effectively promote cancer spreading and metastasis. We previously demonstrated that many metastatic cancer cells acquire immune-resistance by aberrant expression of KIRs on their surface. We showed that cancer cells with aberrant expression or with forced ectopic expression of KIRs are more resistant to NK killing than those with null or low KIR expression. Pre-blocking with anti-KIR antibodies effectively reverses their sensitivity to NK killing. Here we report that KIR-expressing cancer cells interact strongly with platelets leading to increase NK tolerance compared to cancer cells with null or low KIR expression. These data support a novel mechanism of immune-tolerance due to aberrant KIR expression on metastatic cancer cells.

Methods
DNA microarrays were performed on metastatic lung cancer cells re-derived from orthotopic human metastatic lung tumors in athymic nude rats. Aberrant expression of KIR genes was detected and verified with immunohistochemistry and flow cytometry. For ectopic expression, lung adenocarcinoma parental cells (H2122-GFP) were transfected with KIR2DL1 (LL454) and/or KIR3DL1 (LL456) plasmids. Stable transformants were enriched by cell sorting. Binding of these cancer cells with differential KIR expression with human platelets, pre-labeled with anti-CD41-APC antibodies, were analyzed by flow cytometry. NK killing of GFP-tagged cancer cells with differential KIR expressions in the presence and absence of human platelets was accessed with fluorescence intensity in a fluorescent plate reader.

Results
Using in vitro cytotoxic assays, we found that KIR expression or platelet coating on cancers clearly increased their resistance to NK cell killing when compared with parental cells. Interestingly, we found that platelet coating on those metastatic cancer cells with high aberrant KIR expression increased their IC50 values by 6 to 14 folds respectively when compared with parental cells, while platelet coating on those cells with forced KIR expression also increased their IC50 values by 9 to 12 folds respectively. Our observation shows that NK tolerance correlates positively with platelet coating and the levels of KIR expression on the cancer cells (with correlation coefficient = 0.9 to 0.98), and that the NK resistance is the highest when both KIR aberrant expression and platelet coating are present on the cancer cells.

Conclusion
We discovered a novel mechanism of immune resistance to NK cells due to aberrant expression of KIRs on cancer cells. Aberrant KIR expression on cancer cells enhanced their interaction with platelets leading to further increase in NK tolerance than those cells with null or low KIR expression. This observation suggests that anti-platelet antagonists and anti-KIR antibodies may have clinical potential for the treatment or prevention of metastatic and immune resistant cells.

Background
Malignant pleural mesothelioma (MPM) is an aggressive tumour linked to chronic inhalation of asbestos fibers, with poor prognosis and increasing incidence in industrialized countries. Currently available chemotherapeutic regimens achieve a median progression free and overall survival of 6 and 12 months respectively. Tumor necrosis factor-related apoptosis-inducing ligand (TRAIL) is a member of the TNF family, which induces cancer cell death through extrinsic apoptotic pathway, while sparing normal cells. The aim of this study was to investigate the antitumor activity of recombinant human Apo2L/TRAIL (Dulanermin, Amgen;Genentech) in combination with antifolate-based chemotherapy in MPM cell lines and in vivo preclinical mouse model.

Methods
In vitro apoptosis assay was performed using Annexin-V-Fluos staining kit (Roche) according to the manufacturer’s instructions. Epithelioid (ZL55) and sarcomatoid (ZL34) cell lines were treated with carboplatin plus pemetrexed (CP), dulanermin (D) or CPD and then Annexin V positive cells were detected by flow cytometry using a FACSCalibur apparatus and CellQuest software (BD Biosciences). p53 protein expression level was detected by western blot analysis using a specific antibody. TRAIL death receptors (DR4 and DR5) and decoy receptors (DcR1 and DcR2) expression levels were assessed by flow cytometric analysis. In vivo experiments were performed in 30 SCID male mice, implanted subcutaneously in the right flank with 2x10[6 ]ZL55 cells suspended in 0.1 ml volume of RPMI, aged 6 weeks. When tumor volume reached 50-150 mm[3], the mice were randomized in 4 treatment groups: 1) not treated (NT); 2) C (75 mg/Kg day 1) plus P(100 mg/Kg day 1); 3) D (60 mg/Kg days 1 to 3); 4) CPD. Tumor volumes were recorded every second day, and mice suppressed at the 21[th] day or when tumor volume reached 500 mm[3].

Results
We observed a significant increase of specific cell death in epithelioid and sarcomatoid cell lines treated with CPD compared to those receiving CP or D as single agent (p<0.001). We then observed p53 activation in both cell lines after chemotherapy, and a subsequent significant increase of DR4/5 expression levels (p<0.005) without upregulation of decoy receptors. We finally assessed antitumor activity of CP and/or D in a ZL55 mouse model. We observed a statistically significant reduction of tumor volume at every time point in the three treatment groups compared to not treated; moreover, tumor volume was significantly reduced in mice treated with CPD (mean volume 58 mm[3]) compared to CP (mean volume 175 mm[3]) or D (mean volume 109 mm[3]) as single agent at the 21th day (p= 0.005). Finally, no difference in tumor growth was observed between mice treated with D compared to CP.

Conclusion
CP sensitizes MPM cell lines to TRAIL-dependent apoptosis in vitro, probably through p53 activation and subsequent upregulation of death receptors. CPD significantly reduces tumor volume in epithelioid mesothelioma mouse model compared to chemotherapy alone or dulanermin as a single agent; furthermore antitumor activity of dulanermin was comparable to that reported with chemotherapy. In vivo experiments in a sarcomatoid mouse model are currently ongoing.

Background
Lung cancer is the leading cause of cancer-related mortality worldwide and there are more than one million new cases reported worldwide each year. Small cell lung cancer (SCLC) makes up 15% to 20% of all lung cancers with a 5-year survival rate of 5% to 10%. Bcl-2 is a central regulator of cell survival that is overexpressed in the majority of SCLC and contributes to both malignant transformation and therapeutic resistance. The emergence of BH3 mimetics that modulate Bcl-2 pathway by occupying the BH3 groove represents a rational approach for the treatment of this neoplasm. S1 is a BH3 mimetic that depends on Bax/Bak completely. The purpose of this work was to study the key factors that determine the sensitivity of SCLC cells to S1 and the mechanism underlying the resistance of BH3 mimetics.

Methods
Western bolt was used to evaluate the contribution of Bcl-2 family members to the cellular response of 11 SCLC cell lines to a BH3 mimetic S1. Viable cells were determined using MTS assay. To study the potential mechanism of resistance to S1, we derived resistant lines from initially sensitive H1688 cells. Quantitative PCR was performed to investigate Bcl-2 up-regulation. Gene silencing and MEK/ERK inhibitor PD98059 were used to demonstrate the involvement of ERK1/2 signaling in S1-induced Bcl-2 expression.

Results
Results showed relatively higher levels of Bcl-2 and phosphorylated Bcl-2 (pBcl-2) characterized naïve SCLC cell lines that were de novo resistant to S1. Likewise, a progressive increase in the relative levels of Bcl-2 and pBcl-2 characterized the increased acquired resistance of H1688 cells following chronic exposure to S1. Furthermore, acute treatment of S1 induced Bcl-2 expression and phosphorylation. We showed that BH3 mimetics including S1 and ABT-737 induced ER stress and then activated MEK/ERK pathway. The dual function of MEK/ERK pathway in defining BH3 mimetics was illustrated: ERK1/2 activation leaded to Bcl-2 transcriptionally up-regulation and sustain phosphorylation in naïve and acquired resistant SCLC cells. pBcl-2 played a key role in creating resistance of S1 and ABT-737 not only by sequestrating pro-apoptotic proteins, but also a positive feedback to promote ERK1/2 activation.Figure 1

Conclusion
These results provide significant novel insights into the molecular mechanisms for crosstalk between ER stress and endogenous apoptotic pathways in SCLC following BH3 mimetics treatment.

Background
Seven Sirtuin family members (SIRT1-7), comprising a family of NAD+-dependent protein deacetylases and ADP-ribosyltransferases, are key proteins that regulate multiple physiological processes. SIRT2 was recently reported to play an important role in carcinogenesis. However, its role in non-small cell lung cancer (NSCLC) has not yet been investigated.

Methods
In this study, we analyzed the expression pattern of SIRT2 in NSCLC tissues from clinical patients and in cell lines

Results
We found that SIRT2 was significantly down-regulated at both the mRNA and protein levels in tumor than non-tumor tissues or cells, which were corroborated by the NSCLC tissue microarray results. Overexpression of SIRT2 in A549 and H1299 cells caused cell proliferation inhibition, cell apoptosis induction and cell cycle arrest. Further analysis showed that SIRT2 overexpression increased the ROS (reactive oxygen species) production and P27 levels. Moreover, up-regulation of SIRT2 by Resveratrol in NSCLC cells increased the sensitivity to cisplatin treatment. .

Conclusion
Taken together, our results implied that down-regulation of SIRT2 was associated with NSCLC, and regulation of SIRT2 might be an important target for NSCLC therapy

Background
Erlotinib (ERL), a specific inhibitor of epidermal growth factor receptor (EGFR) tyrosine kinase, benefits patients with non-small cell lung cancer (NSCLC), especially if the cancer harbors EGFR active mutations (mtEGFR). However, those who initially respond eventually develop progressive disease through acquired resistance. Bevacizumab (BEV), a humanized anti-vascular endothelial cell growth factor monoclonal antibody, has been demonstrated to be effective in combination with standard chemotherapies for advanced NSCLC patients. We hypothesized that the combination of the two agents may be more effective because they work through different modes of action. In the present study, we examined the antitumor activity of BEV in combination with ERL in human NSCLC xenograft models harboring mtEGFR.

Methods
Mice (BALB-nu/nu) were subcutaneously inoculated with NSCLC cell lines harboring mtEGFR (exon19 deletion); HCC827 and B901L. BEV (5 mg/kg) was intraperitoneally administered once a week for 3 weeks, and ERL was orally given daily for 21 days at doses of 5 and 40 mg/kg for HCC827 and B901L, respectively. Antitumor activity was evaluated by tumor volume (TV; mm[3]) on day 22. In order to examine the prolonged antitumor effect of the combination of BEV with ERL, the B901L xenograft model was used. BEV (5 mg/kg) was intraperitoneally administered once a week for 13 weeks and ERL (60 mg/kg; maximum effective dose) was orally given daily for 91 days. The antitumor activity was evaluated on day 92, with an interim evaluation on day 22. The microvessel density (MVD) of tumor tissues was evaluated by CD31 immunohistochemistry of specimens obtained on day 2 from mice treated with BEV (5 mg/kg) and ERL (40 mg/kg). Statistical analysis was performed using the Wilcoxon test.

Results
In the HCC827 model, the TV (mean±SD) of control, BEV, ERL, and BEV+ERL was 1300±424, 686±84, 615±185, and 194±29, respectively. In the B901L model, the TV of control, BEV, ERL, and BEV+ERL was 1739±761, 819±293, 294±198, and 57±19, respectively. The antitumor activity of BEV+ERL was significantly better than that of BEV or ERL monotherapy (p≤0.05) in both models. In the prolonged treatment experiment using B901L, on day 22, BEV (1096±298) showed significantly higher (p≤0.05) tumor growth inhibition than control (2452±731), but the ERL (61±31) and BEV+ERL (53±30) showed even better (p≤0.05) tumor regression effects than BEV. However, during further treatment, tumor regrowth was observed in the ERL group, even though ERL was consecutively given, and the TV on day 92 (1105±1001) was significantly greater than that on day 22 (61±31). Tumor regrowth was also observed in the BEV+ERL group; however, it was much slower compared to the ERL group and there was no significant difference between TV on day 92 (79±70) and that on day 22 (53±30). The MVD was significantly decreased to the same level in the BEV and BEV+ERL groups.

Conclusion
Compared to ERL monotherapy, the combination of BEV and ERL demonstrated promising efficacy in mouse models of NSCLC harboring EGFR activating mutations. The encouraging preclinical results warrant further investigation in a clinical setting.

Background
Bevacizumab (BEV), a humanized anti-vascular endothelial cell growth factor monoclonal antibody, is used in combination with chemotherapy for patients with non-small cell lung cancer (NSCLC). The epidermal growth factor receptor tyrosine kinase inhibitors (EGFR-TKIs) benefit patients with NSCLC, especially when this cancer harbors EGFR mutations (mtEGFR); however, those who initially respond eventually develop resistance. To treat the patients with EGFR-TKI acquired resistance, a combination of BEV and docetaxel (DTX) may be a possible treatment modality based on different mechanisms of action. In the present study, we investigated the antitumor activity of BEV in combination with DTX in EGFR-TKI-resistant NSCLC xenograft models.

Methods
Mice (BALB-nu/nu) were subcutaneously inoculated with NSCLC cell lines; NCI-H1993 (MET amplification) and NCI-H1975 (mtEGFR/T790M). The treatment was started (day 1) after tumor growth was confirmed. BEV (5 mg/kg) was administered intraperitoneally on days 1, 8, 15 and DTX was administered intravenously on day 1 at a dose of 60 or 20 mg/kg as a maximum effective dose in NCI-H1993 or NCI-H1975 xenografts, respectively. The antitumor activity was evaluated by tumor volume (TV; mm[3]) on day 46 in the NCI-H1993 model and on day 50 (with an interim evaluation on day 19) in the NCI-H1975 model. Tumor cells in the proliferation phase were immunohistochemically evaluated using Ki-67 as an index (count/1000 tumor cells) on day 8 in both tumors. Statistical analysis was performed using the Wilcoxon test.

Results
In the NCI-H1993 model, the TV (mean±SD) of control, BEV, DTX, BEV+DTX on day 46 was 1630±492, 670±116, 204±85, 49±34, respectively. The antitumor activity of BEV+DTX was significantly better than that of DTX monotherapy (p≤0.05). In the NCI-H1975 model, TV of control, BEV, DTX, BEV+DTX was 4631±2384, 1581±572, 18±19, 16±10, respectively, on day 19. The BEV+DTX combination and DTX monotherapy showed similar effects on tumor regression at this point. However, after that, tumor regrowth was observed in the DTX group, and increased beyond the initial TV in 5 out of 6 mice, whereas no tumor regrowth was observed in BEV+DTX group at day 50. The TV of DTX and BEV+DTX groups on day 50 was 2631±2391 and <10 respectively. The number of Ki-67 positive cells (mean±SD) in tumor tissue taken from control, BEV, DTX, BEV+DTX on day 8 was, respectively, as follows: NCI-H1993: 672±50, 638±25, 559±38, 459±37; NCI-H1975: 533±28, 401±57, 381±49, 290±34. The number of Ki-67 positive cells in tumors treated with BEV+DTX was significantly lower than in those treated with DTX (p≤0.05) in both models, although there were no significant differences in TV between the two groups on day 8.

Conclusion
Combining BEV with DTX can decrease the tumor cells in a proliferation phase more effectively than DTX alone in the early stage of the therapy. This may provoke a stronger antitumor effect later on and delay tumor regrowth. These data are encouraging, and the combination of BEV+DTX may be a promising treatment modality for patients with NSCLC after acquiring a resistance to EGFR-TKIs.

Background
Bevacizumab (BEV) is a humanized monoclonal antibody against vascular endothelial cell growth factor (VEGF) that inhibits VEGF-mediated angiogenesis in many types of tumors. OS and PFS were significantly prolonged in patients with advanced non-small cell lung cancer (NSCLC) who received BEV maintenance after induction with BEV+chemotherapy (ECOG4599 study). The maintenance study AVAPERL demonstrated that, after induction with pemetrexed (PEM)+BEV+cisplatin therapy, BEV+PEM combination significantly prolonged PFS compared to BEV monotherapy. However, the comparative efficacy of BEV+PEM to PEM monotherapy is not clear. In the present study, we investigated the antitumor effect of the BEV+PEM combination compared to BEV or PEM monotherapy as continuation maintenance after BEV+PEM treatment in a human NSCLC xenograft model.

Methods
SCID mice were subcutaneously inoculated with human NSCLC cell line NCI-H2228. As an induction treatment, BEV (5 mg/kg; maximum effective dose) and PEM (400 mg/kg; maximum effective dose) were administered intraperitoneally on days 1, 8, and 15 after randomization by tumor volume (TV). On day 22, the mice treated with BEV+PEM were re-randomized and BEV and/or PEM were administered intraperitoneally weekly until day 85. The antitumor activity was evaluated by TV on day 85. Microvessel density (MVD) and thymidylate synthase (TS) expression in tumor tissues were evaluated using specimens taken on day 85. MVD was analyzed by CD31 immunohistochemistry and evaluated in terms of the ratio of stained area to observed area. TS expression was evaluated by Western blot analysis. Statistical analysis was performed using the Wilcoxon test.

Results
TV (mm[3]; mean±SD) of BEV+PEM group on days 1 and 22 was 335±72 and 329±59, respectively, whereas TV of control group on day 22 was 671±67, indicating that BEV+PEM completely inhibited tumor growth during induction treatment. TV of control, BEV, PEM, and BEV+PEM groups on day 85 was 2748±1334, 586±320, 883±271, and 348±48, respectively. Each treatment group showed a significantly higher antitumor activity (p < 0.001) compared to control. Specifically, BEV+PEM completely inhibited tumor growth throughout the experiment, and showed a significantly higher antitumor activity than BEV (p < 0.05) or PEM (p < 0.001) monotherapy. MVD (%) was significantly lower (p < 0.01) in BEV (1.40±0.36) and BEV+PEM (1.24±0.49) groups than in control (6.63±1.89) and PEM (4.62±0.71) groups, indicating that the decrease in MVD caused by BEV treatment was not affected by the combination with PEM. TS expression level, which is inversely correlated with the effect of PEM, was lower in BEV+PEM group than in PEM group.

Conclusion
Compared to PEM monotherapy, the combination of BEV with PEM showed better antitumor effect as a maintenance therapy in a xenograft model. Decreases in MVD and TS expression may contribute to the effect of combination therapy.

Background
Erlotinib, an epidermal growth factor receptor (EGFR) tyrosine kinase inhibitor, shows notable effects against non-small cell lung cancers (NSCLCs) harboring EGFR-activating mutations such as exon 19 deletion or L858R mutation. However, almost all patients eventually acquire resistance to erlotinib. In this study, we investigated whether the level of exposure to erlotinib affects incidence of acquired resistance in EGFR-mutated NSCLC cell lines and examined their resistant mechanisms.

Methods
Human NSCLC cells B901L and PC-9 harboring exon 19 deletion and II-18 cells with L858R mutation were continuously exposed to 0.5 or 5 µM of erlotinib in 96-well plates for several months. The upper dose was chosen because the recommended dose of erlotinib (150 mg) has been reported to achieve maximum plasma concentration (Cmax) of about 5 µM. Cell growth inhibition was determined by a crystal violet assay. EGFR mutation status was determined by melting curve analysis. MET copy number was determined by real-time PCR analysis. Phosphorylations of EGFR, HER2 and ERK were measured by Western blot analysis.

Results
Resistant B901L, PC-9 and II-18 cells were emerged in 24, 8 and 7 wells by exposing to 0.5 µM erlotinib, and in 13, 2 and 0 wells by exposing to 5 µM erlotinib, respectively. No alterations in EGFR mutation status including T790M mutation nor acquisition of MET amplification were detected in any resistant cells. In parent cells, treatment with erlotinib markedly inhibited phosphorylation of EGFR, HER2 and ERK. In the parent cells, treatment with erlotinib markedly inhibited EGFR phosphorylation but not HER2 and ERK phosphorylation, compared with the parent cells.

Conclusion
In present in vitro analysis, it was suggested that exposure to the recommended dose of erlotinib would avoid acquired resistance in NSCLC with exon19 deletion or L858R mutation of EGFR. Pathways activated by phosphorylated HER2 may affect acquired resistance to erlotinib.

Background
Despite rapid strategy improvments of advanced lung cancer , traditional chemotherapy or targeted therapies, especially EGFR-TKI (epidermal growth factor receptor) will emerge treatment resistant eventually. The main cause for treatment failure of advanced lung cancer is drug resistance, and it’s molecule mechanism could be key to improve theraputic effect. The transcription factors Oct4 and Nanog were reported to be involved in drug resistance of many malignant tumors, which are closely related with the prognosis of patients. These factors are expected to be potential drug targets. The present study is designed to explore potential mechanisms about stem cell transcription factors’ regulation of epithelial mesenchymal transition (EMT) invlolved in drug resistance of lung cancer.

Methods
MTT，immunohistochemistry, immunofluorescence, Western blot, RT-PCR experiments in vitro and in vivo were employed. Clinical samples (104 postoperation and 94 advance stage samples) were collected to assess the clinical prognostic value of stem transcription factors( Oct4 and Nanog), （CD133 and ABCG2）and epithelial mesenchymal transition (EMT) markers（E-Cadherin and N-Cadherin）with clinical features. Down-regulation of Nanog by RNA interference were used to find the change of chemosensitivity and EGFR-TKI sensitivity. Combined with the specific mechanisms in vitro and clinical samples to interpret the regulation of EMT involved in drug resistance.

Results
The immunocytochemistry experiments showed that expression pattern of A549 cell line are CD133 (+), ABCG2 (+ +), Oct4 (-) and Nanog (-); A549/DDP cell line are CD133 (+ + +), ABCG2 (+ +), Oct4 (+ +) and Nanog (+ + +). There are significant differences in expression of CD133, Oct4 and Nanog between the two cell lines. Cell immunofluorescence experiments showed that stem cell transcription factors expressed in cell nucleus of A549 cell line, and stem cell surface marker CD133 located on the cell membrane. RT-PCR and Western blot were used to the examination of A549 and A549/DDP cell lines for CD133, ABCG2 and Oct4, Nanog’ expression. The results showed that the expression of these markers in cisplatin resistance cell line. Examination by RT-PCR of the samples from 40 paired cases of lung cancer and adjacent tissue, showed that Oct4 and Nanog in lung cancer group was higher than those in paracancerous tissues, suggesting that Oct4 and Nanog might play an important role in lung cancer initiation and malignant growth process. The expression of these transcription factors in 104 samples from lung cancer patients, analysed by immunohistochemical method, showed the expression level of these factors were significantly higher in stage III and IV patients than in stage I and II patients. It also showed that the higher the Oct4 and Nanog expressed, the poorer prognosis were (54.3 vs 68.5 months, P=0.053) and (52.7vs.72.1months, P= 0.022). The patient with negative expression of CD133 or ABCG2 survived longer than those with positive expression (72.8 vs. 50.2 month, P = 0.021) and (76.5 vs.47.7 months, P=0.000). Patients with positive expression of E-Cadherin survived longer than those with negative markers (67.6 vs 54.1 months, P=0.035), While N-Cadherin+ patients survived shorter than N-Cadherin- patients (55.1 vs. 77.1 months, P=0.025). In overall survival between patients with and those without chemotherapy, there is no significant difference (63.8 vs. 60.6 months, P=0.851). But subgroup analysis revealed patients undergoing chemotherapy with positive expression of CD133 enjoyed significantly longer survival (53.8 vs. 32.3 months). A subgroup analysis also showed that patients with negative expression of Nanog with chemotherapy survived longer than those without chemotherapy (69.3 vs. 76.4 months), while patients with positive Nanog treated with chemotherapy could survive longer than those without chemotherapy (57.0 vs. 38.7 months). In the 74 patients received chemotherapy after progression, we found that many patients with PR are Nanog- (82.1% vs. 17.9%), while patients with Nanog+ tend to have progression diease (61.4% vs. 38.6%). K-M survival analysis indicated that patients with Nanog (-) N-Cadherin (-) CD133 (-) survived longer than those with Nanog (+) N-Cadherin (+) CD133 (+), Nanog (+) N-Cadherin (-) CD133 (-), Nanog (+) N-Cadherin (-) CD133 (+) and Nanog (-) N-Cadherin (-) CD1 33 (-) patients (101.9 vs. 60.0, 101.9 vs. 54.6, 101.9 vs. 38.2 months). Multivariate regression analysis showed that stage of lung cancer, CD133, N-Cadherin and Nanog were independent prognostic factors. Downregulation of Nanog could partially restore the cell sensitivity to cisplatin in cisplatin-resistant cell line under RNA interference. In the EGFR primary drug resistance experiment, we found that CD133 and ABCG2 were higherly expressed in H23 (EGFR primary resistance cell line) than PC-9 (EGFR sensitive cell line). Detection of CD133, ABCG2 and Oct4, Nanog expression by RT-PCR and western blot in PC-9 and H23 cell lines showed that these markers expressed significantly higher in H23; Using immunohistochemical method, we analysed EGFR, C-met and CD133, ABCG2 and Oct4，Nanog in 104 cases of postoperative samples and analysed the relationship between their expression and prognosis. The results showed that C-met expressed in patients in advanced stage (P=0.0860), adenocarcinoma(P=0.0355) and non-smoking patients. K-M analysis showed that C-met expressed in patients with wild type EGFR(P= 0.0436). We further tested stem cell factors in 94 cases of advanced stage who received ERFG-TKI and analysed relationship between drug efficacy and prognosis. The results suggest that Nanog tend to express in smokers (39.4%vs.14.8%, P=0.0071), and Oct4 in adenocarcinoma (38% vs.8.3%, p=0.0170). Patients with negative Oct4 expression have longer PFS (time to progression of disease) (11 vs.7.0 m, P=0.041), patients with negative expression of Nanog have significantly longer PFS than patients with higher expression (13.5 vs.3.8m, P=0.000). Patients with high E-cadherin expression have longer PFS than patients with lower expression (13.5m vs 6.8., P=0.015), Patients with lower N-cadherin expression have longer PFS than patients with high expression (12.3 vs.7.4m, P=0.020). Multiple regression analysis suggested that Nanog, ECOG score and RR are independent prognostic factor. RNA interference results showed that down-regulation of Nanog expression can increase sensitivity to EGFR-TKI in H23.

Conclusion
This study will reveal the mechanisms of drug resistance in lung cancer from a new point of view. To provide the evidence of stem cell factor and EMT related pathway as a new target for reverse the drug resistance in the future.

Background
Bronchoscopic ablation of lung cancer has to date been limited to carcinoma in-situ located in the central airway or, using high-power lasers, to palliation in advanced obstructive disease . Bronchoscopic ablation of peripheral lung cancer is still under development. We are developing a new technology platform for localization (by fluorescence) and enhanced photothermal therapy (PTT) of peripheral lung lesions, based on porphysomes, which are novel, multi-functional, all-organic porphyrin-lipid nanoparticles. Even without active targeting, porphysomes accumulate within tumor through the enhanced permeability and retention (EPR) effect. In parallel, we have developed a prototype fluorescent endoscope system that allows visualization of peripheral lesions by the porphyrin fluorescence. Using this combination of novel instrumentation and nanoparticles, endoscopic PTT of lung cancer is demonstrated in preclinical lung cancer animal models in vivo.

Methods
The in vivo porphysome biodistribution was evaluated in both orthotopic lung tumors (A549, H460, H520) in mice and in VX2 tumors implanted directly in rabbit lung. Porphysomes were administered intravenously at a dose of 20 mg/kg, and the tumor fluorescence was imaged in vivo daily for 3 days (excised lung in the mouse model and endoscopically in the rabbit model). Ex vivo VX2 tissue was illuminated using a 670 nm diode laser to determine the optimized treatment parameters for PTT. Subsequently, in vivo bronchoscopic visualization of VX2 fluorescence and PTT was demonstrated in the rabbit model. The extent of ablation was histologically evaluated by NADH metabolic activity staining.

Results
The highest tumor-to-normal lung fluorescence ratio was achieved at 48 h post-injection of porphysomes(Rabbit VX2: n=4). The same trend was confirmed in the 3 kinds of orthotopic lung cancer mouse Xenografts (n=8). The ex vivo study revealed that 250 mW and 10 min of 670 nm laser irradiation raised the tumor tissue temperature by more than 20[o]C, so that this setting was used also in vivo and achieved thermal coagulation zones within the tumor of up to 13 mm diameter. In comparison, laser treatment alone without porphysomes caused minimal ablation zones of < 1.5 mm diameter.

Conclusion
Systemically administrated porphysomes accumulated in the lung cancer tissue with a high enough concentration to enable marked photothermal coagulation. The combination of systemically-administrated porphysomes with endoscopic near-infrared laser irradiation is a promising strategy for minimally-invasive bronchoscopic interventional therapy for peripherally-located lung cancer.

Background
Lung adenocarcinomas with EGFR mutation initially show good responses to EGFR- tyrosine kinase inhibitors (TKIs). However, emergence of acquired resistance is almost inevitable. To analyze molecular mechanisms underlying acquired resistance, we established a cell line from a patient who acquired resistance to gefitinib/erlotinib.

Methods
A 64-year-old woman underwent a pulmonary resection for lung adenocarcinoma in February 2009 (pT2aN0M0, EGFR L858R mutation). In April 2010, pulmonary metastases, mediastinal lymph node metastases, and right pleural effusion were identified. Gefitinib was started and this led to a complete response. However, despite continuous treatment with gefitinib, pleural effusion and serum carcinoembryonic antigen (CEA) levels gradually increased. In December 2010, gefitinib was switched to erlotinib but the serum CEA level continued to increase, and erlotinib was stopped on March, 2011. Even though she received no treatment after erlotinib withdrawal, interestingly, the serum CEA level was decreased significantly (from 89.5ng/ml on March 2011 to 19.2ng/ml on April 2011), and erlotinib was re-started on April 2011. The serum CEA level increased after re-administration of erlotinib (55.7ng/ml on May 2011). Therefore the regimen was switched to cisplatin/pemetrexed and she responded to this combination chemotherapy (CEA 9.1ng/ml on July 2011). This clinical experience may be an actual case of “drug addiction phenomenon” that we and others have observed in preclinical acquired resistance models (Suda K, et al. Lung Cancer 2012; Das Thakur M, et al. Nature 2013). We established a cell line from her pleural effusion obtained on March 2011 in drug-free condition (designated as ACC-GR1 cells). We analyzed this cell line to evaluate the efficacy of erlotinib and dacomitinib, an irreversible EGFR-TKI. In addition, we examined phosphorylation status of 42 receptor tyrosine kinase (RTK) of ACC-GR1 cells using Human Phospho-RTK Array Kit (R&D Systems) with/without erlotinib or dacomitinib. We also examined PC9 lung adenocarcinoma cell line for phosphorylation status of RTKs with/without erlotinib.

Results
ACC-GR1 cells harbored the T790M mutation, in addition to the original L858R mutation in the EGFR gene. ACC-GR1 cells do not have amplification of MET proto-oncogene. IC~50~ values for erlotinib and dacomitinib were 2.9 uM and 0.05 uM, respectively. Phospho-RTK array analysis revealed marked activation of EGFR and MET, in addition, activation of ERBB2, ERBB3, RET, and AXL in a culture condition without EGFR-TKI. Treatment with 1 uM of erlotinib led to mild inhibition of EGFR and MET phosphorylation (71% and 58%, respectively, compared with control), and phosphorylation of other RTKs fell below detectable limits. Treatment with 1 uM of dacomitinib led to further inhibition of EGFR phosphorylation (35% compared with control). In PC9 cells, phosphorylation of EGFR and MET were also observed in drug-free condition, and remarkably inhibited by 1 uM erlotinib treatment (10% and 64%, respectively, compared with control).

Conclusion
A cell line model established from pleural effusion of a patient who acquired resistance to gefitinib/erlotinib harbored T790M mutation and responded to dacomitinib in vitro. The acquired resistant cells showed activation of several RTKs in drug free condition, and these are remarkably inhibited by EGFR-TKI treatment.

Background
Malignant mesothelioma is an aggressive cancer with a low response to current therapies and consequently a poor prognosis. There is an urgent need to identify novel and more effective treatments to improve survival and quality of life for patients with mesothelioma. While a number of novel therapeutic agents have recently been examined in clinical trials, to date none of these has resulted in changes to standard management. This is due in part to a lack of robustness and relevance of the pre-clinical models used to assess new treatments. Therefore validated and biologically relevant pre-clinical models that demonstrate the clinical behaviour and drug sensitivity of mesothelioma, as well as typical resistance mechanisms, are necessary to improve the discovery of new treatments. Here we describe the generation and evaluation of chemotherapy resistant pre-clinical models of mesothelioma derived from the previously utilized syngeneic II-45 rat mesothelioma model.

Methods
Cell lines resistant to the current standard of care agents: cisplatin, pemetrexed gemcitabine, vinorelbine, and cisplatin and pemetrexed in combination, were generated by 15 rounds of exposure to the respective agent. Normal rat mesothelial cells (4/4 RM.4), the parental II-45 and the 5 resulting chemo-resistant mesothelioma cell lines were characterised for resistance to these and other agents. Tumours arising from syngeneic pleural engraftment of these cell lines were assessed by size, morphology, immunohistochemical markers of human malignancy, chromosomal changes and expression of genes involved in drug resistance and metabolism. Engrafted rats were assessed for survival, and circulating haematological, cytokine and biomarker profiles were generated and compared.

Results
Five different II-45 cell lines with approximately 2-fold resistance to the chemotherapeutic agent they were repeatedly exposed to, were established (cisplatin, p < 0.05; pemetrexed, gemcitabine, vinorelbine, p < 0.001). Cross resistance to other classes of anti-cancer agents also developed indicating potential multi-drug resistant phenotypes. Tumours derived from both the parental and chemo-resistant cell lines were immunohistochemically indistinct from human mesothelioma with positive labeling for WT1, calretinin, HBME-1, cytokeratin and popoplanin and negative labeling for two carcinoma related markers TTF-1 and CD15. Homozygous deletion of p16[INK4A]/p14[ARF], and increased expression of several members of the ATP-binding cassette transporter superfamily and Androgen receptor (AR) were demonstrated, consistent with findings in human mesothelioma. Corresponding with biomarkers studies in human disease, increased levels of osteopontin (p < 0.01), mesothelin (p < 0.01), vascular endothelial growth factor (p < 0.001) and neutrophil to lymphocyte ratio were identified relative to control bloods. Further, the acquisition of chemo-resistance resulted in changes to tumour morphology with tumours ranging from essentially epithelioid in the gemcitabine resistant tumours to entirely sarcomatoid in the combination (cisplatin and pemetrexed) resistant tumours. Overall survival was also affected by the acquisition of resistance with rats with pemetrexed resistant tumours having decreased survival.

Conclusion
These models display many features corresponding with the human disease, and thus provide powerful and robust pre-clinical platforms for in vivo mesothelioma studies.

Background
Malignant pleural mesothelioma (MPM) is an aggressive asbestos-related malignancy characterized by frequent resistance to chemo- and radiotherapy. Signals induced by Fibroblast growth factors (FGF) and their high-affinity receptors (FGFR) have been identified as important drivers for malignant growth in several tumor entities including thoracic malignancies. We have recently demonstrated that FGFR signals stimulate growth and migration in MPM cell models, whereas inhibition of FGFR1 reduced tumor growth and had an antagonistic effect on malignant behavior of MPM cells. In several cell models of biphasic MPM, FGF2 induced phenotypical changes reminiscent of epithelial-mesenchymal-transition (EMT). In this study we analyzed these FGF-induced morphological and functional alterations and the associated signal transduction mechanisms in more detail.

Methods
Cells were stimulated with recombinant FGF2 and analyzed by microscopy and ImageJ software. The specific inhibitors PD166866, UO126, MK2206, LY294002 and SB431542 were used to block FGFR1, MEK, AKT, PI3K and TGFbeta receptors, respectively. Alterations in gene expression were determined via whole-genome expression arrays and further evaluated by immunofluorescence. Downstream signaling was investigated by immunoblotting.

Results
In M38K and SPC212 cells FGF2 induced morphological alterations that were characterized by a more spindle-shaped appearance and reduced contacts with adjacent cells. This correlated with increased cell migration. With respect to signal transduction, the effects of FGF2 could be blocked by inhibition of FGFR1 or MEK, whereas inhibition of AKT, PI3K or receptors of the TGFbeta/activin family had no effect. Expression arrays of both cell models indicated regulation of several matrix metalloproteinases (MMP1, MMP4), the integrin subunit ITGA6 and the two TGFbeta family-related proteins INHBB and Smad7. Since the phenotypical changes were similar to EMT, genes previously connected to EMT were analyzed. Whereas slug/SNAI2 and ZEB1 were increased, other mesenchymal genes such as vimentin or N-cadherin were already expressed at high levels in the untreated cells, likely due to the mesodermal origin of mesothelial cells. In M38K, E-cadherin was decreased whereas in the more “fibroblastoid-type” SPC212 E-cadherin was generally expressed at very low levels.

Conclusion
Our data suggest that FGF2 induces morphological changes that result in a more sarcomatoid cell morphology and expression of some markers connected to EMT. These effects depend on the MAPK pathway and are connected to more aggressive cell behavior, paralleling the higher aggressiveness and worse prognosis of the sarcomatoid and biphasic compared to the epithelioid histological subtype of MPM.

Background
Lung cancer is the most common cause of cancer death. Despite advances in treatment of other cancers, overall 5-year survival rates for advanced non-small cell lung cancer (NSCLC) are dismal. Expression of βIII-tubulin (encoded by TUBB3 gene) is associated with resistance to tubulin-binding agents in a range of tumor types including NSCLC[1]. We identified a multifactorial role for βIII-tubulin in chemosensitivity by showing that it can mediate response not only to tubulin-binding agents, but also DNA damaging agents in NSCLC[2, 3], validating βIII-tubulin as a therapeutic target in NSCLC. We hypothesized that in vivo targeting of the TUBB3 gene would sensitise NSCLC tumours to chemotherapy.

Methods
DNA-directed RNA interference (ddRNAi) is a potent gene silencing approach that can achieve prolonged suppression of target genes. We developed a triple cassette shRNA vector containing three unique targets against the TUBB3 gene (βIII~TRP~) and a control vector containing three non-targeting sequences (Cont~TRP~). Gene and protein analysis was achieved using RT-qPCR and western blotting respectively. An orthotopic mouse xenograft model of NSCLC was used to evaluate gene ddRNAi delivery, efficacy and chemo sensitisation. Mouse tumour growth was monitored using non-invasive xenogen imaging. Statistical analysis included Mann–Whitney test and Log rank (Mantel-Cox).

Results
Transient transfection of NSCLC H460 cells with the βIII~TRP~ vector resulted in >40% decrease of TUBB3 gene expression and a concomitant decrease in βIII-tubulin protein level compared to the Cont~TRP~ vector. This led to increased in vitro sensitivity to both paclitaxel and cisplatin. Using a clinically relevant orthotopic model of NSCLC we treated H460 tumour-bearing mice with the βIII~TRP~ vector encapsulated in a lipophilic delivery agent and this resulted in a potent decrease (71.6%) in TUBB3 expression in tumours compared to Cont~TRP~ vector (P=0.03). Median survival increased by 6 days in the Cont~TRP~~+ cisplatin~ versus Cont~TRP~ groups although this increase was not significant. In contrast, median survival increased by 21 days in the βIII~TRP + cisplatin~ versus βIII~TRP+ vehicle~ groups (P=0.003). Importantly, in vivo suppression of TUBB3 combined with cisplatin treatment led to a significant increase in overall survival compared to βIII~TRP + vehicle~ control mice (P=0.0174).

Conclusion
This study provides the first evidence that a vector-based gene suppression approach can potently silence TUBB3 expression and increase chemosensitivity in vivo, highlighting this as a promising treatment strategy for drug refractory NSCLC. References 1. Kavallaris M. Microtubules and resistance to tubulin-binding agents. Nat Rev Cancer. 2010;10:194-204. 2. McCarroll JA, Gan PP, Liu M, Kavallaris M. betaIII-tubulin is a multifunctional protein involved in drug sensitivity and tumorigenesis in non-small cell lung cancer. Cancer Res. 2010;70:4995-5003. 3. Gan PP, Pasquier E, Kavallaris M. Class III beta-tubulin mediates sensitivity to chemotherapeutic drugs in non small cell lung cancer. Cancer Res. 2007;67:9356-63.

Background
Lung cancer is the leading cause of cancer deaths worldwide. Despite advances and progresses in surgery, chemotherapy, and radiotherapy over the last decades, the death rate from lung cancer has remained largely unchanged, which is mainly due to metastatic disease and multi drug resistance. Because of the overall poor prognosis, new treatment strategies for lung cancer patients are urgently needed. The aim of this study was to investigate the interactions between pemetrexed and vinorelbine for human adenocarcinoma via various chemotherapy schedules.

Methods
HCC cells and cisplatin resistant HCC cells (HCC-res) were treated with different schedules of combination of cisplatin (Cis), vinorelbine (Vin), and pemetrexed (Pem) respectively, more specific as Cis, Vin, Pem, Cis+Vin, Cis+Pem, Vin+Pem, Vin→Pem, Pem→Vin, Cis+Vin+Pem, Cis→Pem→Vin, and Cis→Vin→Pem. Cell growth inhibition was determined by cell viability analysis. Cell apoptosis was analyzed via annexin V staining and FACS analysis. And cytoplasm Ca[2+] was measured with Ca[2+] indicator dye Fura-2 AM.

Results
Vin and Pem caused a strong dose-dependent cytotoxic effect in both HCC and HCC-res cells. The IC50 values of Vin against HCC and HCC-res cells were 10.34 ± 1.12 nM and 9.98 ±2.12 nM, respectively. The IC50 values of Pem against these cells were 110.77 ± 17.28 nM and 118.89 ±18.77 nM respectively. The application of different therapy schedules induced a significant time dependent cell growth inhibition on HCC naïve and cisplatin resistant cells. The therapy scheme of Cis→Pem→Vin showed the strongest inhibitory effect on both HCC and HCC-res cells. The application of different therapy schedules on HCC and HCC-res cells increased the percentage of cells undergoing apoptosis, except the application of Vin alone. In both HCC and HCC-res cells, Cis→Pem→Vin was found the most effective to induce apoptosis. The application of different therapy schedules on HCC and HCC-res cells increased [Ca[2+]]~c~. Only the application of Vin alone failed to increase [Ca[2+]]~c~ in HCC cells. The most elevated [Ca[2+]]~c ~was found in the cells treated with Cis→Pem→Vin in both HCC and HCC-res cells

Conclusion
We demonstrated that the sequential application of Cis, Vin and Pem has a synergistic effect in cell growth inhibition, apoptosis induction, and [Ca[2+]]~c~ elevation in HCC and HCC-res cells. The Ca[2+ ]overload could lead to apoptosis, which was related to the cell growth inhibitory effect of chemotherapeutics in lung cancer cells. It might cast a light to develop chemotherapy schedules for patients, and to overcome cisplatin resistance in lung cancer.

Background
Chromosome Region Maintenance 1 (CRM1) is a nuclear exporter which transports certain proteins from the nucleus to the cytoplasm, including tumor suppressor proteins (TSPs) and other modulators of proliferation. Overexpression of CRM1 correlates with cancer progression in several human cancers, suggesting that CRM1 could serve as a novel target for treatment of cancers. NSCLC is an aggressive carcinoma which is not yet curable. The aim of our study was to explore the therapeutic efficiency of novel drug-like CRM1 inhibitors in NSCLC in vitro and in vivo, and to investigate the cytotoxic mechanisms of CRM1 inhibitors in NSCLC cell lines with EGFR-TKI resistance mutation.

Methods
KPT-185 and KPT-276 are selective inhibitors of nuclear export (SINE) that block CRM1. Cell viability, apoptosis and cell cycle were evaluated in 6 NSCLC cell lines (H1975, H1650, A549, H2228, HCC827, H1650 Gefitinib Resistance (H1650GR)) after treated with KPT-185; expression level of CRM1 in NSCLC cell lines was detected after exposed to KPT-185; TSPs were detected by western blot to explore the possible mechanisms of KPT-185 inducing NSCLC cells growth inhibition and apoptosis. NOD-SCID mice bearing H1975 (epidermal growth factor receptor (EGFR) tyrosine kinase inhibitor (TKI) resistant) tumors were treated orally with KPT-276 (similar structure to KPT-185, but improved animal pharmacokinetics) to examine the efficacy and side-effects of KPT-276 in vivo.

Results
In 6 NSCLC cell lines, growth inhibition showed in a time- and dose-dependent way after treated with KPT-185. The EGFR TKI resistant cell lines H1975 and H150GR were sensitive to KPT-185. Cell apoptosis analysis showed that KPT-185 induced NSCLC cells apoptosis in a dose-dependent manner. Also, KPT-185 induced cell cycle arrest at the G1/S checkpoint in NSCLC cell lines. CRM1 protein expression of 6 NSCLC cell lines was down regulated when treated with KPT-185, which could be completely abolished by bortezomib. CRM1 inhibition by KPT-185 up-regulated the expression of proteins involved in apoptosis in NSCLC cell lines, and down-regulated the expression of EGFR and survivin. In the xenograft H1975 model, tumor growth was significantly inhibited in KPT-276 oral treatment group compared with vehicle control group and EGFR-TKI treatment group (P<0.01), and there was no significant loss in body weight or side-effects in KPT-276 treatment group.

Conclusion
SINE CRM1 inhibitors showed anti-tumor activity in NSCLC both in vitro and in vivo, especially in EGFR-TKI resistance cell lines, it could inhibit the growth of NSCLC, arrest cell cycle, reduce expression of CRM1 protein, and the anti-tumor activity of SINE is mainly through inducing cell apoptosis. SINE is a novel CRM1 inhibitor and a promising clinical candidate.

Background
This study was performed to validate a newly developed sentinel lymph node (SLN) targeting tracer, indocyanine green–neomannosyl human serum albumin (ICG:MSA), and a thoracoscopic version of the intraoperative color and fluorescence imaging system (ICFIS) for lung cancer SLN mapping.

Methods
5 μg/kg ICG concentrations of ICG alone or ICG:MSA were injected into rat thigh and the results were compared. The fluorescence signal-to-background ratio (SBR) of SLNs were recorded and evaluated over a 2-h period using ICFIS. In addition, a SLN biopsy by video-assisted thoracoscopic surgery (VATS) was performed using ICG:MSA in porcine lung by using thoracoscopic ICFIS.

Results
The newly developed ICG:MSA showed a significantly improved SBR compared to ICG alone throughout the trials. All SLNs were identified in both rat (10 SLNs in 10 rat thighs) and pig (10 SLNs in 10 porcine lungs) under in vivo conditions. All SLNs were dissected successfully using VATS with help of thoracoscopic ICFIS.

Conclusion
ICG:MSA is accumulated in the SLN by uptake and retention through the mannose-specific receptors on a macrophage. Thoracoscopic ICFIS successfully assisted SLN mapping despite low near-infrared (NIR) light transmission in the commercial thoracoscope. Based on the results of the thoracoscopic SLN mapping, we expect that the ICG:MSA and thoracoscopic ICFIS can be translated to clinical trials in the near future.

Background
Background: Small cell lung cancer (SCLC) has an overall 5-year survival rate of < 5% despite initial response rates to first line chemotherapy of 70-90%. MYC family amplification occurs in 30% of SCLC and MYC amplified tumors have been shown to be dependent on aurora kinase B for their survival. Aurora B kinase (AURKB) is a chromosomal passenger protein that plays a critical role in mitosis by regulating chromosome alignment, accurate segregation, and cytokinesis during the mitotic stages. We evaluated the effects of the selective aurora B kinase inhibitor AZD1152-HQPA (active drug) in a panel of 15-SCLC lines.

Methods
Methods: Nine lines had MYC-family amplification (MYC-5 lines, MYCL1-2 lines, MYCN-2 lines) and 6-lines had no MYC-family amplification. Growth inhibition by AZD1152-HQPA was assessed by tetrazolium based assays. Apoptosis was determined using the DNA binding dyes YOPRO and propidium iodine (PI) and analysis by flow cytomery. The induction of polyploidy was evaluated by flow cytometry following PI staining of the DNA. The effect of AZD1152-HQPA on anchorage independent growth was determined by soft agar colony forming assays. Finally, we evaluated the in vivo efficacy of AZD1152 (prodrug) on SCLC xenografts in nude mice.

Results
Results: AZD1152-HQPA GI50 values for the 4 most sensitive SCLC lines were 11-30 nM and < 25% of untreated control growth was observed at 60 nM. Five lines had GI50 values of 30-60 nM but at 600 nM AZD1152-HQPA cell viability remained at 49-55% of untreated control growth. In the remaining 6-lines cell viability at 600 nM was 60-93% of control growth. Growth inhibition marginally correlated with MYC amplification (p=0.044) and was strongly correlated with MYC + MYCL1 + MYCN amplification (p=0.011). Apoptosis (20-28%) was induced by 30 nM AZD1152-HQPA at 24-48 hours in the most sensitive cell lines. Marked increases in DNA ploidy (8N) was observed after 24 hour exposure to AZD1152-HQPA (30nM) in the most sensitive cell lines and at 48 hours in the 5 intermediate lines and in 2 of the resistant lines. In vivo AZD1152 (i.p. 100 mg/kg/M-F for two weeks) induced tumor regression in nude mice implanted with a sensitive cell line. AZD1152 at 50mg/kg/day inhibited tumor growth; however these tumors began growing following treatment cessation. A resistant line was also implanted into nude mice and AZD1152 at 50 and 100 mg/kg/day caused tumor regression. Polyploidy was induced in this cell line at 48 hours post 30 nM AZD1152-HQPA. Anchorage independent growth in this resistant line was also completely inhibited by AZD1152-HQPA 25 nM. The doses of AZD1152-HQPA used in this study are within the range reported to be clinically achievable.

Conclusion
Conclusions: AZD1152-HQPA growth inhibition significantly correlated with MYC-family amplification in SCLC lines but some SCLC lines without MYC-family amplification were also inhibited. Additional biomarkers are needed to identify SCLC patients most likely to respond to AZD1152.

Background
Neovascularization has a critical role in the growth and metastatic spread of tumors, and involves recruitment of circulating endothelial progenitor cells (EPCs) from bone marrow. In this study, we examined whether EPCs could promote tumor angiogenesis, and found that the tumor growth was enhanced by the administration of EPCs.

Methods
To test the hypothesis that genetically modified bone marrow-derived EPCs can be effective carriers of therapeutic agents to tumor sites, we conducted human interferon-beta (HuIFN- b) gene transfection of EPCs with a virus vector in vitro .

Results
When HuIFN-b was applied in the ex vivo culture of EPCs, HuIFN- β-transduced EPCs achieved efficient killing of the total population of SPC-A1 cells, indicating a bystander effect was elicited by HuIFN- b-transduced EPCs in vitro . When SCP-A1 cancer cells were coimplanted along with ex vivo cultivated EPCs subcutaneous injection in nude mice, the tumor growth was increased. However, the anti-tumor effect of interferon-beta (IFN- β) offset the tumor-progressive character of EPCs and the tumor growth, and the vascular density of tumor tissues increased by coimplanted EPCs were decreased upon IFN- b treatment. In addition, overall expression levels of vascular endothelial growth factor in tumor tissues were decreased upon IFN-b treatment.

Conclusion
Our results suggest that gene-transfected EPCs could be useful as a tumor-specific drug delivery system.

Background
Mutant Kras was observed in more than 20% of non-small cells lung cancers (NSCLC) and represented one of the most dominant oncogene during cancer progression. However, there were fewer effective inhibitors which specific targeting of mutant Kras in clinical trials. The immunotherapeutic effects of Kras DNA vaccine in lung cancer were not well understood.

Methods
In this study, we investigated anti-tumor efficacy of mutant Kras DNA vaccine in spontaneous mouse lung tumor model driven by Kras[G12D]. TetO-Kras[G12D] and Scgb1a1-rtTA bitransgenic mice were treated with doxycycline to induce lung tumor for 5 months. To evaluate therapeutic effects of Kras DNA vaccine, bitransgenic mice with lung tumor were treated with wild-type Kras or mutant Kras[G12D] plasmids by skin administration.

Results
In this model, we found that skin administration of mutant Kras DNA vaccine had better therapeutic effects compared to control group. Additionally, dominant negative Kras (N17) DNA vaccine significantly decreased tumor nodules on lung in transgenic mice expressed mutant Kras. Th-1 immune response was dramatically enhanced in vaccine-treated mice compared with in control mice.

Conclusion
Overall, our results indicated that Kras DNA vaccine produced an effective anti-tumor response. Furthermore, skin administration of Kras DNA vaccine is effective as a potential approach for lung cancer therapy.

Background
Complete surgical resection of solid malignancies is often not achieved due to growth of occult tumour cells, distant metastases, or the invasiveness of the tumour mass. Hence there is a need for adjuvant therapies that may be administered following surgery to target residual tumour. Using a murine mesothelioma model we have previously shown that combining gemcitabine with anti-CTLA4 is effective at generating an anti-tumour immune response, leading to tumour regression and improved survival. We sought to use this treatment regimen to improve the post-surgical outcome of debulking surgery.

Methods
The murine mesothelioma line generated in our lab, AB1-HA, was inoculated subcutaneously into the flanks of Balb/c mice. Established tumours (<50mm[2]) were debulked by surgical removal of 75% of the tumour mass. Treatment with gemcitabine and anti-CTLA4 were commenced on the day of surgery, and mice were monitored for residual tumour outgrowth and survival. Size-matched controls were treated in parallel.

Results
Dual administration of gemcitabine with anti-CTLA4 was effective at causing regression of remaining tumour and improving survival following partial debulking surgery, over either therapy alone. The outcome of this treatment when administered to debulked tumours was comparable to that achieved against smaller, size-matched tumours.

Conclusion
We have shown that larger tumours can be successfully partially debulked and the remaining tumour treated using a combination of gemcitabine and anti-CLTA4. Thus, combined chemo-immunotherapy is a feasible option for post-surgical treatment option to remove residual tumours.

Background
Asbestos-induced mesothelioma in MexTAg transgenic mice closely resembles the key features of human disease. This model is highly suited to testing new chemotherapies and cancer prevention strategies. The resultant mesothelioma responds to cytotoxic chemotherapy with efficacy of the same order of responsiveness as human mesothelioma. The transgene, large T Antigen is an oncogene encoded by Simian Virus 40, and the role of this viral oncogene in tumourigenesis has been widely studied. Here we investigate the gene expression differences between TAg tumours and wild type tumours and examine overall similarity to human mesothelioma at the molecular level.

Methods
not applicable

Results
We found TAg expressing mouse tumours were 90% identical to wild type mouse tumours. The key pathway that was affected by these gene differences was cell cycle. Gene set enrichment analysis showed that the TAg:wild type tumour differences were homolgous to the Rb null genotype in a TAg dose dependent manner. To address whether transgenic mouse tumours were a good representation of human mesothelioma when compared to wild type mouse tumours, we showed that of genes differentially expressed between i) TAg or ii) wild type mouse tumours and mouse mesothelial cells, there were the same number of gene similarities with the set of genes that differentiate between human mesothelioma and human mesothelial cells. Human mesotheliomas commonly have a deletion of the cdkN2 locus, encoding the tumour suppressor genes p16 and p15. We found that while wild type mouse tumours contained the p16 deletion, TAg tumours did not.

Conclusion
In summary, the asbestos-induced mesotheliomas in MexTAg mice are comparable to human mesothelioma at the molecular level. We hypothesize that TAg expressing mice develop tumours in a more uniform way than wild type mice following asbestos exposure and are not dependent on deletion of p16 for tumourigenesis.

Background
Background: As a result of the recent deep molecular profiling of NSCLC samples, FGF receptor inhibition has emerged as a promising strategy for targeting a sub-set of squamous NSCLC (Sq NSCLC) tumours that carry FGFR gene amplifications, mutations and fusions. AZD4547 is a potent, orally available and selective inhibitor of FGFR 1, 2 and 3 and is currently in phase II clinical development.

Methods
not applicable

Results
Results: In pre-clinical patient derived models of FGFR1 amplified Sq NSCLC, AZD4547 is able to induce dose dependent tumour growth inhibition and tumour regression and this is correlated to FGFR1 protein expression and inhibition of signalling pathways downstream of FGFR1. Since the FGFR2 mutations described in Sq NSCLC do not present as clear codon hot-spots we also investigated a sub-set of the FGFR2 mutations using inducible expression in non-transformed cells. We found these mutations to be constitutively active and capable of inducing 3D colony formation in non-transformed cells. Both FGFR signalling and 3D colony growth were inhibited potently by AZD4547 treatment. We have developed a number of biomarker assays that enable both patient selection and exploratory analysis of patient samples. We found that FGFR1 gene amplified samples are enriched for those expressing both FGFR1 mRNA and protein. AZD4547 is currently being tested as a monotherapy in Sq NSCLC patients whose tumours carry FGFR1 amplification and we will describe preliminary observations from this trial including a comprehensive molecular profile of the tumour from a patient who experienced a durable partial response following AZD4547 treatment.

Conclusion
Conclusion: In view of the strong and emerging platform of evidence that implicates dis-regulated FGFR signalling in Sq NSCLC and the early evidence of clinical activity, FGFR inhibition warrants continued clinical investigation in patients whose tumours carry FGFR genetic lesions including amplification, mutation and gene fusions.

Background
Resistance to apoptosis is a hallmark of cancer that can be reversed by several chemotherapy drugs and targeted therapies, including cisplatin and erlotinib. In vivo imaging of apoptosis would be of great interest to study molecular mechanisms of drug activity and resistance. The aim of this study is to investigate whether in vivo micro-imaging of apoptosis could be used for early assessment of treatment response in a murine xenograft model of human lung cancer.

Methods
A549 (EGFR wild type), H1650 (carrying EGFR exon 19 deletion that confers sensitivity to Erlotinib) and H1975 (carrying EGFR mutations L858R and T790M, resistant to Erlotinib) cell lines were used to induce subcutaneous tumors in Nude mice. In vivo micro-imaging of apoptosis was performed using fibered confocal fluorescence microscopy (FCFM) after intra-venous injection of a fluorogenic caspase 3 substrate. Tumors were treated by cisplatin (10mg/kg) (5 mice, A549 xenografts), erlotinib (25mg/kg) (6 mice, 2 A549, 2 H1650, 2 H1975), or vehicle (6 mice, 2 A549, 2 H1650, 2 H1975).

Results
In A549 xenografts treated by cisplatin, apoptosis was detected in vivo at 24h post-treatment. Fluorescence intensity ratio (FIR) was significantly higher than in untreated tumors (16.9+/-3.1 vs 4.8+/-2.1 respectively, p<0.001). 24h after treatment by erlotinib, FIR in H1650 erlotinib sensitive tumors was significantly higher than in A549 tumors, and than in H1975 erlotinib resistant tumors (12.2+/-2.4 vs 6.1+/-1.2 vs 1.9+/-0.7 respectively, p<0.01, Figures 1 and 2). Results were confirmed ex vivo on harvested tumor xenografts by TUNEL assay and immunohistochemistry for activated caspase 3, and on cell lines in vitro by flow cytometry and fluorescence microscopy. Figure 1 Figure 2

Conclusion
Early in vivo micro-imaging of apoptosis using FCFM makes it possible to differentiate sensitive from resistant tumors to Erlotinib in a murine xenograft model of human lung adenocarcinoma.

Background
In second line setting there is no standard treatment for patients with mesothelioma, since all agents tested in clinical trials failed to improve survival. An individual patient however, may actually benefit from second line treatment, considering the patient population is heterogeneous. Yet, there are no predictive markers to identify those patients likely to respond to a certain drug. We developed protocols for in vitro drug sensitivity testing with several cytotoxic agents using primary tumor cells derived from pleural fluid of our patients. Recently, we implemented a personalized second line treatment protocol. Here we will report the outcome of the pilot study.

Methods
Cells were isolated from pleural fluid, drawn from patients with mesothelioma for symptom relief. Diagnosis of each mesothelioma patient was confirmed by regular pathological staining. Cells were plated and incubated with a five point concentration range of 5 single drugs and 2 two-drug combinations for 72 hours. Cell viability was determined by a metabolic assay. Each concentration point was measured in triplo and a biological duplo experiment was performed to check reproducibility. The drugs used in this screen were all previously tested in clinical mesothelioma trials. Patients with pleural fluid that were fit and progressed after standard first line treatment were considered for second line chemotherapy. Choice of second line treatment was based on screening results.

Results
Thirty-nine mesothelioma patients had pleural fluid drawn for culture of primary tumor cells and subsequent drug sensitivity screening. Drug screens were successful in 22 patients (56%). Drug sensitivity profiles were available within two weeks after isolation of tumor cells, which is appropriate for clinical decision making. Agents tested for were cisplatin or carboplatin, pemetrexed, gemcitabine, vinorelbine, oxaliplatin, a combination of cisplatin and pemetrexed and of oxaliplatin and gemcitabine. Individual dose-response curves showed different sensitivity to the various cytotoxic agents. Ten patients were chemo-naive at the time of the drug sensitivity screen. Five of them received first line chemotherapy (cisplatin and pemetrexed). Two of them had progressive disease. Both demonstrated evident resistance to cisplatin and pemetrexed in their drug sensitivity profile. Four patients received second line treatment based on the drug sensitivity profile of their primary tumor cells. To date, treatment response is evaluated for two patients. Both patients received a combination of oxaliplatin and vinorelbine. The first patient had a clinical response. For the second patient, oxaliplatin/vinorelbine was the best option, although her in vitro profile suggested a rather resistant tumor. She was progressive and showed an unusual large amount of necrosis on repeated CT imaging. Patient three and four are currently treated with oxaliplatin and gemcitabine and gemcitabine monotherapy, respectively. Their treatment responses have to be awaited.

Conclusion
Personalized drug screening using primary tumor cells is feasible within a clinically relevant time period and may yield new treatment combinations with better responses.

Background
Lung cancer ranks among the most commonly occurring malignancies, and is the leading cause of cancer-related mortality worldwide. Chemotherapy for lung cancer can be made more specific to tumor cells, and less toxic to normal tissues, through the use of ligand-mediated drug delivery systems.

Methods
For the screening of targeting peptides, a phage-displayed peptide library underwent affinity selection with lung cancer cells. Targeting phage clones were selected by ELISA, immunofluorescence, flow cytometry and an in vivo homing assays. Targeting peptides can be used to develop ligand-mediated targeted therapy or targeting imaging probe. For preclinical study, we evaluated therapeutic efficacy of targeting liposomes using xenograft, orthotopic- and metastatic-tumor animal models

Results
In this study, we investigated the targeting mechanism of the ligand-mediated drug delivery system using lung cancer targeting peptides. Conjugation of lung cancer targeting peptides to liposomes enhanced the amount of drug delivered directly into NSCLC cells, through receptor-mediated endocytosis. Accumulation of peptide-conjugated liposomal doxorubicin (SP-LD) in tumor tissues was 11.2-fold higher than that of free doxorubicin, and the area under the concentration-time curve (AUC0-72 hours) was increased 159.2-fold. Furthermore, SP-LD enhanced therapeutic efficacy and increased the survival rate of tumor-bearing mice in syngenic, metastatic and orthotopic animal models.

Conclusion
The current study suggests that tumor-specific peptides may be used to create chemotherapies specifically targeting tumor cells in the treatment of NSCLC. The use of lung cancer targeting peptide-conjugated liposomes enhances pharmacokinetic properties, improves efficacy and safety profiles, and allows for controlled biodistribution and drug release.

Background
Lung cancer is the major cause of cancer death in the world while non small cell lung cancer (NSCLC) accounts approximately 85% of all lung cancer diagnosis. EGFR mutations, found in 10–30% of patients with NSCLC characterize a subpopulation with exquisite sensitivity to EGFR tyrosine kinase inhibitors (EGFR-TKIs). However, the clinical benefits of first-generation TKIs (like gefitinib or erlotinib) can be further improved because of the development of drug-acquired resistance within 10–14 months in patients who initially respond to the treatment. Therefore, there is a need to discover next generation medicines as EGFR-TKIs for NSCLC patients.

Methods
Our EGFR program was first started to screen 20,000 in-house compounds for EGFR (wild type, 32D- EGFRWild-Type) activity in EGFR-transfected 32D cells and further performed knowledge-based design. More than 300 analogues were synthesized in this series and US and ROC patents were granted in 2013 and PCT patent application is under examination. We had identified DBPR112 as a potent EGFR-TKI with oral in vivo activity in a mouse model for lung adenocarcinoma.

Results
DBPR112 showed IC50 of 2 nM in HCC827 cells and potent EGFRWild-Type (IC50: 10 nM) and EGFRL858R/T790M (IC50: 70 nM) kinase inhibition which are better than gefitinib and similar to that of BIBW2992 (Afatinib, developed by Boehringer Ingelheim) that is currently under phase III clinical trial. DBPR112 was orally (F = 49%) administered against the growth of human lung HCC827 tumors subcutaneously xenografted in nude mice. A dramatic reduction of the tumor size was noted in the tumors treated with DBPR112 without significant loss of body weights in the nude mice. In addition, the pharmacokinetics properties of DBPR112 are superior to those of Afatinib.

Conclusion
Considering the fact that EGFR kinase plays an important role in lung adenocarcinoma with drug-sensitive mutations in EGFR, our goal is to develop novel and potent EGFR kinase inhibitors for the treatment of lung cancer, either as a single agent or in combination with existing cancer treatment. DBPR112 is a highly potent small-molecule against various forms of EGFR kinase. The non-clinical profile of DBPR112 showed promising in vivo efficacy. We, therefore, propose to design a preclinical program to assess the developability of DBPR112 and conduct comprehensive pre-clinical studies. We hope to make DBPR112 an investigational new drug for the treatment of lung cancer in the near future.

Background
Patients with non-small cell lung cancer (NSCLC) who have activating epidermal growth factor receptor (EGFR) mutations derive remarkable benefit from treatment with EGFR-tyrosine kinase inhibitors (EGFR-TKIs). But drug-resistant problem appears sequentially and novel therapeutic strategies should be explored to overcome EGFR-TKI resistance. As the first EGFR-TKI in China and the third in world, icotinib shows good efficacy and tolerability in patients with advanced NSCLC. This study aims to establish icotinib resistant cell line-HCC827/IR and analyze it’s biological character for further study of EGFR-TKIs resistance.

Methods
HCC827 is a cell line with a deletion in the exon 19 of EGFR gene. HCC827 cells were exposed to increasing concentrations of icotinib (10nM to 20uM). Cells with the ability to grow in 20uM of icotinib were obtained 8 months after the initial drug exposure. The cell proliferation, viability, distribution of cell cycle, EGFR gene sequence (exon 18, 19, 20 and 21), EGFR-TKIs cross-resistance and the response to a histone deacetylases inhibitor (Chidamide, CS055) were evaluated after allowing the cells to grow in drug-free conditions for 2 months.

Results
Population doubling time (PDT) of HCC827/IR was not different from HCC827 (32.3±6.0h vs. 36.3±2.4h, P=0.198). In the cell cycle distributions of HCC827/IR, the cell number in G0/G1 phase were decreased (P=0.035), but the cell number in S and G2/M phase had no significant change compared with parent cells (P=0.388 and P=0.205, respectively). The resistance index (RI=HCC827IR~IC50~/ HCC827~IC50~) of HCC827/IR to icotinib was (1.98±0.15)×10[3]. And HCC827/IR cells also showed high resistance to the other two EGFR-TKIs (gefitinib and erlotinib), the RI of gefitinib and erlotinib was (2.36±0.082)×10[3 ]and (1.069±0.004)×10[3], respectively. But, the sequence of EGFR gene did not changed before and after resistance to EGFR-TKIs. In addition, HCC827/IR was sensitive to CS055 (IC~50~=254±32nM).

Conclusion
This study successfully established icotinib resistant NSCLC cell line-HCC827/IR. HCC827/IR cells had some different biological characters compared with parent cells and showed high cross-resistance to other EGFR-TKIs, but it was sensitive to CS055. The results could provide experimental reference for clinical application of TKIs and provide cell line model for further study of TKI-resistance.

Background
ROS1 is a receptor tyrosine kinase that regulates pathway signaling related to cell proliferation, growth and survival. Chromosomal rearrangements involving ROS1 have recently been identified as potential driver mutations in non-small-cell lung cancers (NSCLC) and preclinical and clinical studies indicate a significant response to crizotinib in tumor cells harboring this mutation. We developed an RT-PCR assay for all ROS1 fusion gene transcripts described in NSCLC to date which is suitable for use with FFPE tissue.

Methods
Primers and probes have been specifically designed to detect 14 known fusion transcripts from seven different fusion partners with the ROS1 gene using six multiplexed reactions. These fusion gene pairs include: CD74-ROS1, SLC34A2-ROS1, GOPC (FIG)-ROS1, SDC4-ROS1, EZR-ROS1, TPM3-ROS1, and LRIG3-ROS1. The 14 different ROS1 fusion products were represented by specifically designed and manufactured pCR 2.1-TOPO and pZErO 2.1 plasmids. After screening and identifying ROS1 rearrangements in clinical samples, the specific variants were further differentiated by gel electrophoresis and sequencing.

Results
All 14 ROS1 fusion genes could be detected using this RT-PCR assay. We screened 578 FFPE lung cancer samples that were negative for EGFR, KRAS and ALK mutations and detected 12 ROS1 rearrangements in this cohort (2.08%). All positive tumors were adenocarcinomas of the lung, 75% of the patients were female and median age was 53 years. The ROS1 rearrangements that were identified are: SLC34A2 (Exon 13)-ROS1 (Exon 32 and 34), CD74 (Exon 6)-ROS1 (Exon 34), EZR (Exon 10)-ROS1 (Exon 34) and SDC4 (Exon 2)-ROS1 (Exon 32). Messenger RNA expression of ROS1 was significantly higher in ROS1 positive samples than wild-type samples (median 5.2 vs.1.0; p<0.001). An inter-assay comparison of 31 samples containing four ROS1 positives using RT-PCR and FISH showed 100% concordance.

Conclusion
The clinical relevance of ROS1 testing in NSCLC and potentially other solid tumors as a predictive biomarker for sensitivity to crizotinib is rapidly increasing. We developed a multiplexed RT-PCR assay that enables the detection of ROS1 fusion genes using FFPE tissue and provide a rapid and cost effective screening tool.

Background
With existing therapeutic efforts, patients with lung cancer have a poor prognosis. Assessment of tumor size, lymph node status and presence of metastases is currently applied for determining prognosis and treatment modality, but predicted and real outcomes can vary significantly. Biomarkers with reliable prognostic significance are therefore of utmost importance but due to a lack of immediate correlation between levels of protein and their corresponding mRNA, a screen based on the kinase activity become a promising option to circumvent this limitation with the tremendeous advantage of focusing on therapeutically targetable enzymatic activities. Several protein tyrosine kinase inhibitors already clinically approved for the treatment of lung cancer are targeting some of the 400 types of DNA signatures described in silico in the human genome. The aim of this study is to clarify the following hypothesis: Is the in vitro multiplexed tyrosine phosphorylation of substrates a possible approach to molecularly classify the kinome of early stage lung adenocarcinoma biopsies and obtain a diagnostic /prognostic signature correlating with the survival of patients?

Methods
We have built a tumor bank of frozen malignant and non-neoplastic lung surgically obtained specimen and recorded all clinical interventions, follow up treatments and outcome for each of our patients. We incubated TNM stage 1 and 2 lung adenocarcinoma kinomes on PamChip®4 microarrays and followed the kinetics of the multiplexed tyrosine phosphorylation for 144 peptides substrates. Image quantification, quality control, statistical analysis and interpretation of data were performed with the BionavigatoR software.

Results
We screened 84 paired malignant TNM stage 1 and 2 lung adenocarcinoma and non-neoplastic lung biopsies for the multiplex tyrosine phosphorylation of substrates immobilized on a PamChip®4microarrays. Based on a 76-point ‘response-signature’ we obtained 73 % of correct prediction with a 10 fold cross validation PLS-DA analysis in TNM stage 1 lung adenocarcinoma biopsies. Moreover, we detected 26 peptides substrates significantly more inhibited in kinomes of long-term survivors than in kinomes of the short-term survivors.

Conclusion
In frozen biopsies of TNM stage I adenocarcinoma and with a PLS-DA analysis applied to a 76-point ‘response-signature’ we present the feasibility to discriminate between long-term and short-term survivors. Furthermore, the found differences in enzymatic activities in lung biopsies may result in the identification of new targets in future anti lung cancer therapy efforts.

Background
In this study, we attempted to detect EGFR and KRAS mutations in plasma/pleural effusion of patients with advanced NSCLC by pyrosequencing, to investigate whether plasma and pleural effusion DNA could be used as a substitute for tumor tissues for detecting gene mutations, and to explore the correlation of EGFR/KRAS mutations with the efficacy of the epithelial growth factor receptor-tyrosine kinase inhihitor (EGFR-TKI) as well as their correlation with survival in TKI-treated patients.

Methods
Blood, pleural effusion and tumor tissues were obtained from 146, 64 and 63 patients with advanced NSCLC, of whom there were 40 matched tissue and plasma samples, 24 matched tissue and pleural effusion samples. The exons 19, 20 and 21 of EGFR was amplified by mutant-enriched PCR using selective restriction enzyme digestion, and exon 2 of KRAS in plasma were amplified by nested-PCR. Then mutations were detected by pyrosequencing. The association between mutations and the patients’ survival was analyzed using Kaplan-Meier.

Results
EGFR mutations were detected in 34.38% tissues (22/64), 24.24% plasma samples (24/96) and 30.16% pleural effusion samples(19/63). KRAS mutations were detected in 4.69% tissues (3/64), 6.16% plasma samples (9/146) and 7.93% pleural effusion samples. No statistical significance was found in EGFR/KRAS mutations between plasma/pleural effusion and tumor tissues (p>0.05). The same EGFR genes were observed in plasma and matched tissue in 34 patients (consistency: 85%).The sensitivity of detecting EGFR mutations is 73.33% and the specificity is 92 %.There was only 1 KRAS mutation detected in the 40 tissues , but no mutation in the matched plasma (consistency: 97.5%). Same EGFR/KRAS genes were observed in 21, 23 pleural effusion and matched tissue respectively (consistency: 87.5%, 95,83%). The sensitivity of detection of EGFR/KRAS mutations is 77.78% and 66.67%. The specificity is 93.33 % and 100%. Significant correlation existed between EGFR mutations in tumor tissue/pleural effusion from patients with advanced NSCLC, and a smoking history, and histopathologic type (p<0.05). Among the 38 TKI-treated patients, the disease control rate (DCR) and objective response rate (ORR) were 90% and 60%, respectively, in patients with EGFR mutation in plasma, and 53.57% and 17.86%, respectively, in patients with wide-type EGFR (DCR, p=0.059; ORR, p=0.019).The DCR and ORR were 66.67% and 33.33%, respectively, in patients with wide-type KRAS in plasma, and 40% and 0%, respectively, in patients with KRAS mutation. Patients with EGFR activating mutations in plasma had a favoring median PFS of 10.5 months, significantly longer than the patients with wild-type EGFR (5.0 months) (p=0.228). The median PFS was 2.5 months for patients with KRAS mutation and 9 months for patients with wild-type KRAS, respectively (p=0.000).

Conclusion
A high consistency exists between EGFR/KRAS mutation detection in plasma/pleural effusion and tumor tissues. Plasma/pleural effusion could be used as a substitute for tumor tissues for detecting gene mutations. EGFR and KRAS mutations in plasma are highly associated with the treatment response and prognosis of TKI-treated patients. It is feasible to detect EGFR and KRAS mutations in plasma/pleural effusion by pyrosequencing. The detection sensibility of EGFR mutations can be increased by ME-PCR, facilitating choosing patients for TKI treatment.

Background
The elevated serum level of MUC1 (KL-6, CA15.3) in lung cancer patients is considered to be a marker for advanced adenocarcinoma and a prognostic factor for disease progression after treatment. The aim of this study was to evaluate the serum MUC1 levels in primary lung cancer patients in the Russian Federation, a large portion of which are patients with squamous cell lung cancer (SCC).

Methods
Serum samples were obtained from 346 patients at admission. Two hundred ninety three patients were diagnosed with a malignant lung tumor (MLT), i.e. SCC (n=136); adenocarcinoma, AC (n=89); adenosquamous cancer, ASqC (n=21); bronchioloalveolar carcinoma, BAC (n=12); large cell carcinoma, LCC (n=3); small cell cancer (n=4); anaplastic cancer (n=8); carcinoid (n=17); sarcoma (n=3). A non-malignant lung disease (NMLD) was confirmed histologically in 53 patients: benign tumors (n=10), pneumofibrosis (n=13), tuberculosis (n=9), a resolution stage of pneumonia (n=21). Anti-MUC1 ICO25 monoclonal antibody-based and validated inhibition enzyme-linked immunosorbent assay was used (normal range 9 - 38 U/ml, median 24 U/ml).

Results
An occurrence of the elevated serum MUC1/ICO25 levels (³40 U/ml) in patients with MLTs was significantly higher than in patients with NMLDs (151 [51.5%] vs. 7 [13.2%] cases, correspondingly; р<0.001), and it positively correlated with clinically aggressive tumor histology: AC, 61.0%; ASqC, 61.9%; SqCC, 44.9%; BAC, 25.0%; LCC and small cell cancer, 100%; anaplastic cancer, 50.0%; carcinoid, 23.5%; sarcoma, 0%. The significantly elevated serum MUC1/ICO25 levels (³60 U/ml) were found in 81 (27.6%) patients with MLTs and 1 (1.9%) patient with NMLDs. In patients with nonsquamous non-small cell lung cancers (AC, BAC, LCC) and ASqC the serum MUC1/ICO25 ³60 U/ml were found in 44.0% (55/125) of cases. The rate of these cases positively correlated with the advanced TNM stage and depended on the patients’ age: statistically significant difference between stages I-II and stages IIIB-IV was found in patients aged 65 years and older (15.8% [3/19] vs. 84.2% [16/19]; p<0.001), but not in patients aged less than 65 years (28.6% [8/28] vs. 48.6% [17/35]; p=0.127). In patients with SCC the overall rate of serum MUC1/ICO25 ³60 U/ml was 16.9% (23/136), and the reversed age-related dependence of those levels on the tumor stage was bserved. In patients aged less than 65 years the higher rate of the serum MUC1/ICO25 ³60 U/ml correlated with more advanced TNM stage: 0% (0/13), 13.8% (4/29),28.6% (6/21) and 52.9% (9/17) for stages I, II, IIIA and IIIB-IV, respectively (I-II vs. IIIB-IV, p<0.001). In patients with SCC aged 65 years or older the serum MUC1/ICO25 ³60 U/ml was found in 0% (0/19), 30.0% (3/10), 7.7% (1/13) and 0% (0/14) of cases for stages I, II, IIIA and IIIB-IV, respectively. In the entire group of patients the rate of the serum MUC1/ICO25 ³60 U/ml did not depend on the tumor histological grade, local tumor extension, location of distant metastases or comorbidities.

Conclusion
The abnormal MUC1/ICO25 serum levels correlate with unfavorable prognostic clinical characteristics of the lung cancer, and its potential clinical significance is not restricted to the cases of lung adenocarcinoma.

Background
Recent advances in molecular profiling of non-small cell lung cancer (NSCLC) have led to the replacement of platinum-based chemotherapy with targeted therapies for certain genetic subsets of NSCLC (ALK rearrangements, some EGFR activating mutations). It is also known that myriad pathways can drive resistance, the unfortunate norm for most patients. A greater understanding of the overlap across multiple biomarker subsets, including activating mutations, signal transduction pathways, and immune system markers, might aid in prognostic assessment, predictive biomarker development and the design of combination or sequential treatment regimens.

Methods
The prevalence and prognostic significance of nine biomarkers (TTF1, p63, EGFR mutation, KRAS mutation, MET immunohistochemistry [IHC], PDL1 IHC, PTEN IHC, NaPi2B IHC, ECDH IHC) across two independent sample sets (Set 1, n=561; Set 2, n=300) were tested. With the exception of ECDH, all assays were IVD or companion diagnostics. Set 1 was collected from patients who were eligible for surgery with curative intent from 2003–2005 at MD Anderson Cancer Center in the USA. Samples from Set 2 were part of a collaboration between the University of Colorado Cancer Center, USA and The Norwegian Radium Hospital, and contained surgically-resected NSCLC tissues collected from 2006–2011.

Results
The prevalence of each biomarker varied significantly by histology. For adenocarcinoma samples, the prevalence of each biomarker was: EGFR mutation (13%), KRAS mutation (29%), TTF1 IHC (83%), p63 IHC (7%), MET IHC (50%), PDL1 IHC (45%), PTEN loss IHC (11%), NaPi2B IHC (76%), EGFR IHC (FLEX cut-off, 11%). In squamous-cell carcinoma, the prevalence of each biomarker was: TTF1 IHC (2%), p63 IHC (87%), MET IHC (13%), PDL1 IHC (50%), PTEN loss IHC (13%), NaPi2B IHC (3%), EGFR IHC (FLEX cut-off, 40%). In addition, more than 67% of patients were positive for more than one biomarker and >33% were positive for at least three biomarkers. The diagnostic criteria for each biomarker and correlations with patient characteristics will be described in further detail. Figure 1. Biomarker Overlap in Adenocarcinoma in Set 1 (n=337) Figure 1

Conclusion
Collectively, these data suggest that the biomarker landscape in NSCLC is complex and will be increasingly dynamic as more experimental agents approach pivotal testing. Grant support: this study was partially funded by UT Lung Specialized Programs of Research Excellence grant (P50CA70907; IIW)

Background
Apoptotic pathways are controlled by activation or inhibition of the intracellular caspases. Inhibitor of apoptosis proteins (IAPs) are endogenous inhibitors of caspase activity. We aimed to investigate the relationship of the apoptosis regulators X-linked inhibitor of apoptosis (XIAP) and ubiquitin specific protease 8 (USP8) with clinical parameters, survival and response to chemotherapy in advanced stages of non-small cell lung cancer (NSCLC).

Methods
A total of 34 NSCLC patients (28 males, 6 females) and 44 healthy volunteers were included in the study. Most of the patients had squamous cell histology (62%), while others had adenocarcinoma (29%) and unclassified (8%) types. All patients had locally advanced stage IIIA-IIIB (50%) or metastatic (50%) disease. XIAP and USP8 levels were measured by ELISA method.

Results
The median serum XIAP levels were similar in patients and the controls (p=NS). USP8 level was higher in patients compared to the control group (p<0.0001). In univariate analysis, there was no significant relationship between XIAP and USP8 serum levels and age, sex, performance status, weight loss, stage of disease, histopatological type and response to chemotherapy. In the low and high XIAP and USP8 groups, there was no significant difference in progression-free survival (PFS) (p=0.432 and p=0.50, respectively) and overall survival (OS) (p=0.989 and p=0.90, respectively).

Conclusion
No relationship was found with serum XIAP and USP8 levels and clinical parameters, response to chemotherapy, PFS and OS in patients with advanced stages of NSCLC.

Background
The prognostic role of signal transducer and activator of transcription 3 (STAT3) and phosphorylated STAT3(p-STAT3) in non-small-cell lung cancer (NSCLC) remains controversial. To clarify its impact on survival, we performed a meta-analysis to quantitatively assess STAT3 and p-STAT3 expression on prognosis of NSCLC.

Methods
Published studies were identified using a systematic and thorough literature search.To be eligible, a study had to investigate STAT3 or p-STAT3 expression rates of NSCLC patients in different characteristics and survival data of STAT3 or p-STAT3 expression.

Results
A total of 17 retrospective trials were chosen for meta-analysis, including 1793 patients.We showed that the estimated pooled HR (0.67, 95%CI: 0.57-0.77) of 9 trials (STAT3: HR 0.71, 95%CI 0.38-1.04; p-STAT3: HR 0.67 ,95%CI 0.56-0.77) for NSCLC was statistically significant (P<0.0001), suggesting that high STAT3 or p-STAT3 expression is a strong predictor of poor prognosis among patients with NSCLC.For the risk factors, pooled analysis of patients with STAT3 positivity, demonstrated a statistically significant OR (3.82, 95%CI: 2.37-6.16) between poorly differentiated carcinoma and well-moderately, OR( 5.68, 95%CI: 3.16-10.21)between patients in stage III-IV and patients in stage I-II, and OR (3.41, 95%CI: 2.12-5.49 ) between patients with lymph node metastasis and patients without lymph node metastasis. However, pooled analysis of patients with p-STAT3 positivity, only demonstrated a statistically significant OR (4.51, 95%CI: 1.57-12.96) between poorly differentiated carcinoma and well-moderately(P<0.05).

Conclusion
High STAT3 or p-STAT3 expression is a strong predictor of poor prognosis among patients with NSCLC. The conclusion should be confirmed by large prospective studies with long-term follow-up.

Background
EGFR-activating mutations are predictive of high response rates and overall survival gains in patients (pts) with pulmonary adenocarcinoma, treated with EGFR- tyrosine-kinase inhibitors (EGFR-TKIs), as erlotinib. Mutations in the EGFR gene, especially in the exon 20 (T790M), are related to resistance to EGFR-TKIs. We investigated if a polymorphic DNA sequence in exon 20 (Q787Q, NCBI database 162093G>A, SNP ID: rs1050171) was associated with clinical outcomes in pulmonary adenocarcinomas, treated with erlotinib.

Methods
It is a prospective, observational study on all consecutively pts whose tumors were genotyped for EGFR-activating mutations. Tumor samples were formalin-fixed and paraffin-embedded. Tumor areas were selected and macrodissected, followed by whole DNA extraction and amplification by PCR. EGFR genotyping was performed through DNA sequencing (exons 18, 19, 20 and 21) by Sanger´s methodology. Pts with adenocarcinomas harbouring EGFR-activating mutations were treated with erlotinib.

Results
191 pts had tumor samples genotyped between Aug/2011 and Apr/2013. Median age was 64 y (17-90), 106 (56%) female. According to ethnicity, 154 pts were Caucasian (81%), 26 African-American (14%) and 11 Asian (6%). Seventy pts were classified as never-smokers (37%), 23 (12%) as light-smokers (≤ 10 p.y.) and 95 as current smokers (51%). EGFR activating mutations could be identified in 54 out of 191 samples (28%): 35 were exon 19 deletions (65%), 15 were L858R mutation in exon 21 (30%), and three were rare mutations (G719S and G719A in exon 18, and V774M in exon 20). Polymorphism Q787Q in EGFR gene (exon 20) was detected in 108 samples (56.5%). The polymorphic status did not correlate with gender (p=0.324), smoking status (p=0.810) or EGFR mutational status (p=0.238), but it was more frequently detected in Caucasian pts (p=0.0002). Considering all 191 studied pts, no difference in median overall survival was detected according to polymorphic status (19.6 mo. vs. 24.3 mo., HR 0.86; 95% CI 0.54-1.38, p=0.541). There was no difference in response rate to erlotinib according to the polymorphic status (p=0.248). In addition, no difference in median overall survival was detected according to polymorphic status among the 38 pts treated with erlotinib and presenting EGFR-activating mutations (not reached in both groups, HR 2.44; 95% CI 0.31-16.01, p=0.425).

Conclusion
Polymorphism Q787Q in EGFR gene (exon 20) was commonly detected in pulmonary adenocarcinomas (56.5%), being more frequent in Caucasian pts. The presence of polymorphic status was neither related to sensitivity to erlotinib, nor to survival outcomes in our pts.

Background
Identification of actionable driver mutations in non-small cell lung carcinoma (NSCLC) has become increasingly important for the prioritisation of targeted therapies. Mutational analysis of formalin-fixed paraffin embedded (FFPE) tissues presents several challenges including generally limited and fragmented DNA, the need to identify a range of biologically significant mutations and a pressing need for a fast turn around time at a cost-effective way. Our aim was to determine optimal methods for quantification of DNA for mutational analysis in NSCLC and to develop a new custom assay that could perform multigene mutational analysis on the limited quantity of DNA available in the small NSCLC samples frequently submitted for testing.

Methods
DNA was extracted from FFPE tissues including cytology specimens. Spectrophotometry quantification was compared with Qubit 2 Fluorometer measurements and the Sequenom SampleID assay for accurate and meaningful assessment of extracted DNA for diagnostic mutational profiling. We have previously established a diagnostic protocol for somatic mutation profiling in NSCLC using a commercial DNA mass spectrometry kit (Oncocarta v1.0) and compared it with a new custom kit “OncoFocus” developed in collaboration with Sequenom. These assays utilise target amplicons of small sizes for efficient amplification in fragmented DNA and simultaneously profile a range of actionable mutations in EGFR, KRAS, BRAF and NRAS. Preliminary verification of the “OncoFocus” assay was performed in 27 NSCLC samples, 3 lung cancer cell lines and 2 control genomic DNA samples.

Results
We found spectrophotometry significantly overestimated DNA quantity particularly at low concentrations. We also studied the correlation of DNA quantities with estimated copies of DNA templates as determined by SampleID. The results suggested that a minimum of 300 ng DNA is needed to achieve the required 300 – 500 amplifiable genomic copies per reaction for the OncoCarta analysis, which remains difficult to achieve for many diagnostic NSCLC samples. We developed a more focused diagnostic panel “OncoFocus” which could be performed reliably with less DNA but which includes key actionable mutations in 159 hotspots in EGFR (n=109), KRAS (n=17), BRAF (n=15) and NRAS (n=18) requiring only 150 ng of DNA. Somatic mutations were identified in 23 samples and 3 cell lines including EGFR (n=22), KRAS (n=6) and BRAF (n=1). No false positive results were observed in 4 FFPE and 2 control samples. The whole process from the receipt of FFPE samples to issuing a report can be completed within 5 working days and the “OncoFocus” panel has increased our capacity per chip (iPLEX II) from 15 to 31 samples. The “OncoFocus” panel also results in decreased per sample testing costs.

Conclusion
The Qubit fluorometer is a more reliable and accurate method to quantify DNA derived from FFPE for mutational analysis than spectrophotometry. We also conclude that DNA mass spectrometric analysis using a new custom “OncoFocus” panel is an effective and efficient test that simultaneously detects 159 mutational hotspots, in the generally lower quantity of DNA obtained from routine FFPE NSCLC samples.

Background
Patients’ EGFR mutation status prior to treatment impacts outcomes and, EGFR testing has been developed as a companion diagnostic; this relationship between therapeutic and diagnostic agents is known as personalized healthcare. Recently, it was reported that about half of patients may acquire resistance to EGFR-TKIs following therapy, mainly by appearance of EGFR mutations, such as T790M. Thus, it is important to assess EGFR mutation status before and during treatment to determine the most appropriate treatment regimens for patients. A number of PCR-based techniques are used in the assessment of EGFR mutations. In Japan, the “Scorpion-ARMS” therascreen® EGFR Rotor-Gene Q PCR Kitis (therascreen EGFR assay) the only available in vitro diagnostic (IVD) test. The cobas® EGFR Mutation Test (cobas EGFR assay) is the only FDA-approved kit for IVD testing in the US.In this study, we compared the performance of the cobas EGFR assay and the therascreen® EGFR assay using FFPE tissue specimens from NSCLC patients.

Methods
We extracted DNA from 149 FFPE tissues of NSCLC, according to the manufacturer’s instructions and performed a comparative study of cobas EGFR and therascreen EGFR methods.

Results
EGFR mutations were identified in 63 NSCLC specimens (42.3%) using the cobas EGFR assay and 61 samples (41.2%) using the therascreen EGFR assay. The concordance rate between the cobas EGFR assay and therascreen EGFR assays was 145/149 (97.3%). Only three discordants between these EGFR assays were observed. One T790M mutation in combination with an L858R mutation was identified by the cobas EGFR assay. No invalid assay results occurred with the cobas EGFR assay.

Conclusion
The cobas EGFR assay has two advantages. One is that the process is easily performed by stable methods. It takes only 8 hours, from tumor specimen to results generated using the semi-automated system. Thus, patients assessed using the cobas EGFR assay can begin the most appropriate treatment quickly. The other advantage is that only a very small amount of DNA (150 ng) is required to detect mutation status using the cobas EGFR assay. Our results show a high concordance rate (97.3%) of cobas EGFR with an existing IVD product, the therascreen EGFR assay. In this study, one double mutant, T790M in combination with L858R, was only identified by the cobas EGFR assay. The cobas EGFR assay appears to give the most accurate and appropriate results for NSCLC patients.

Background
The 5-year survival for stage IB non-small cell lung cancer (NSCLC) is only 55%, but the benefit of adjuvant chemotherapy in this setting remains equivocal. Numerous prognostic markers have been examined, but none to date have moved into clinical practice. There is an urgent need to identify novel molecular markers that can select high risk patients, who may potentially benefit from adjuvant chemotherapy.

Methods
We identified 471 consecutive patients with stage IB primary NSCLC according to the American Joint Commission on Cancer, (AJCC) 6[th] edition tumour-node-metastasis staging system, who underwent surgical resection between 1990 and 2008. Patients who received neoadjuvant or adjuvant treatments were excluded. Pathology reports were reviewed and pathologic characteristics were extracted. Expression of phosphorylated Akt (pAkt) in both cytoplasmic and nuclear locations was assessed by immunohistochemistry, and clinicopathologic factors were analyzed against 10-year overall survival using Kaplan-Meier and Cox proportional hazards model.

Results
455 (96.6%) cancers were adequate for pAkt immunohistochemical analysis. The prevalence of pAkt expression in the cytoplasm and nucleus of the cancers was 60.7% and 43.7% respectively. Patients, whose cancers expressed higher levels of pAkt in the cytoplasm, had a trend towards longer overall survival than those with lower levels of cytoplasmic pAkt (p=0.06). Conversely, patients whose cancers expressed higher levels of pAkt in the nucleus had a poorer prognosis than those with lower levels of nuclear pAkt expression (p=0.02). Combined low cytoplasmic/high nuclear expression of pAkt was an independent predictor of overall survival [HR=2.86 (95% CI:1.35-6.04); p=0.006] when modeled with age [HR= 1.05 (95% CI: 1.03-1.07); p<0.001], extent of operation [HR= 2.11 (95% CI: 1.48-3.01); p<0.001], visceral pleural invasion [HR=1.63 (95% CI: 1.24-2.15); p<0.001], gender, tumour size, histopathologic type and grade (p>0.05).

Conclusion
Levels of expression of pAkt in the cytoplasm and nucleus and visceral pleural invasion are independent prognostic factors that can help to select patients with high risk disease, who may potentially benefit from adjuvant chemotherapy.

Background
Activating EGFR mutations confer sensitivity to EGFR tyrosine kinase inhibitors (TKIs) in patients with NSCLC, but responses are transient, with delay in disease progression but no impact on survival. Concomitant genetic alterations could account for these incomplete clinical responses. Erlotinib-treated EGFR-mutant NSCLC patients harboring the EGFR T790M mutation had shorter PFS than those without the mutation (12 vs 18 months [m]). Low BIM levels were associated with gefitinib resistance in EGFR-mutant NSCLC.

Methods
The efficacy results of the EURTAC trial were updated at January 24, 2013. We have evaluated the frequency and potential impact of pretreatment EGFR T790M mutations and BIM mRNA expression in 95 patients with EGFR-mutant NSCLC included in the EURTAC trial.

Results
T790M mutations were detected in 65.26% of patients. PFS to erlotinib was 9.7 m for those with T790M mutations and 15.8 m for those without, while among patients receiving chemotherapy, it was 6 and 5.1 m, respectively (P<0.0001). BIM expression was successfully analyzed in 83 patients. PFS to erlotinib was 12.9 m for those with high BIM levels and 7.2 m for those with low/intermediate BIM levels, while among chemotherapy-treated patients, it was 5.8 and 5.5 m, respectively (P=0.0003). OS was 28.6 m for patients with high BIM expression and 22.1 m for those with low/intermediate BIM expression (P=0.0364). The multivariate analyses showed that erlotinib was a marker of longer PFS (HR, 0.35; P=0.0003), while high BIM expression was a marker of longer PFS (HR, 0.49; P=0.0122) and OS (HR, 0.53; P=0.0323).

Conclusion
BIM mRNA expression is a biomarker of PFS and OS in EGFR-mutant NSCLC. T790M mutations and BIM mRNA expression can potentially be used for designing combination therapeutic strategies for use in lieu of EGFR TKI monotherapy.

Background
In 3% of cases, lung cancer is diagnosed in patients younger than 45 years. The epidemiology, biology and clinical history of young lung cancer patients is generally different from the adult counterpart: the higher percentage of mutations has the potential to influence both tolerability and response to treatment with consequent impact on quality of life and survival. The biomolecular characterization of the disease in this subgroup will allow the design of clinical studies dedicated to young patients, that will lead to the identification of specific items that are not deducible from trials opened to the general adult population. In this study, Next-Generation Sequencing (NGS) technology has been applied to archival tissue samples to enhance tumor-specific genomic profile knowledge in this selected cohort of young patients (pts).

Methods
A retrospective analysis has been performed at the Thoracic Unit of San Luigi Hospital from January 2007 to March 2013, collecting 13 lung cancer-diagnosed pts (10 completely sequenced; in 3 cases the analysis is ongoing), aged between 15-39 years. Genomic DNA was extracted by microdissected formalin-fixed and paraffin embedded (FFPE) tumor samples of all pts and by lymphocytes of three healthy controls (ctrl). Amplicons NGS libraries for 46 oncogenes included in the Ion Torrent Cancer Panel were generated, following manufacture guidelines, and sequenced in Personal Genome Machine (PGM) Ion Torrent instrument. Variant Caller included in Torrent Suite Software was used to identify mutations.

Results
Twenty-two non-synonymous, 3 frameshifts, 3 stop-gain and 55 synonymous somatic sequence variations were found in 10 young adult patients (allelic frequency ≥ 10%). Excluding synonymous mutations, the most frequently altered genes in patients were TP53 (7 mutations; 25%), followed by EGFR and KDR (5 mutations; 18%), PI3K (3 mutations; 11%), KIT (7 mutations; 14%), FGFR3-ABL1-MET-ATM-RB1-SMO (1 mutation; 3.6%). Furthermore, 14 of these mutations are annotated in SIFT or in PolyPhen databases as “mutations that could damage affected protein”. Overall, we identified 28 mutations annotated in COSMIC database, among which the most frequent were COSM149673, COSM28026 and COSM6223-COSM22413 with 5,4 and 2 counts, respectively. We also found 93 SNPs in our cohort, including the most frequent rs7688609, rs1873778 and rs1050171 with 13 counts (10 pts; 3 ctrl) followed by rs1800861 (9 pts; 3ctrl); rs41115 (8 pts; 2ctrl); rs1870377 (4 pts; 1ctrl); rs3822214 (2 pts; 2ctrl); rs3729687 (2 pts; 1ctrl); rs2228230 (2 pts) and rs3730358-rs3135898 (1 pt; 1ctrl).

Conclusion
The development of new biological techniques, such as the next-generation sequencing, could allow to collect a wide number of mutations. From these preliminary results, some interesting data have been discovered concerning SNPs or mutations. This pivotal retrospective analysis is the basis for the ongoing prospective collection. A better definition of molecular-genetic pattern in this selected young population of patients could increase the knowledge about the lung cancer etiology and suggest age-related new trials design.

Background
It is reported that abundance of EGFR mutations is related with efficacy of EGFR TKI in advanced NSCLC patients with mutant EGFR. This study was designed to investigate influence of EGFR mutations and their abundance on efficacy of EGFR TKI by a quantitative method.

Methods
190 NSCLC treated with EGFR TKI and available tissue for EGFR mutations were enrolled into the study; 113 were FFPE specimen, and 62 were fresh tissue. EGFR mutation was detected with the kit of AmoyDx ARMS and percentage of mutant EGFR was tested with the method of an Allele Specific PCR with Competitive Blocker (ASB-PCR). In this assay, copies of EGFR mutants were calibrated by standard curve, and the mutation rates were estimated through normalizing by copies of a conserved sequence in EGFR exon2. The relationship between abundance of EGFR mutations and efficacy of EGFR TKI was analyzed.

Results
Of 190 samples, 15 were censored due to EGFR exon2 copies less than 100; finally, 175 enrolled into data analysis. Mutant percentage less than 0.1% was defined as wild-type, 0.1%~2% as low abundance, 2%~20% as moderate abundance, and more than 20% as high abundance. The mutant rate was 56.6% and 62.3% by using AmoyDx ARMS and ASB-PCR methods, respectively. The accordance rate of EGFR mutations was 89.7% by two methods. Of 175 samples, 20, 27 and 62 harbored low, moderate and high abundances of mutant EGFR, respectively; 66 were wild-type EGFR. Median progress free survival (mPFS) was 4.9 (95% CI, 3.4 to 6.4), 8.3 (95% CI, 3.3 to 13.3) and 16.0 months (95% CI, 10.4 to 21.7) in patients with low, moderate and high abundances of mutant EGFR (p =0.012). The mPFS of low abundance was not longer than that of those patients with wild-type EGFR (2.0 months, 95% CI, 0.2 to 4.1; p=0.261). Objective response rate (ORR) was 67.7%, 44.4%, 25.0% and 19.7%, and disease control rate (DCR) was 90.3%, 81.5%, 55.0% and 45.5% in patients with high, moderate, low abundance and wild-type EGFR, respectively (p<0.001). But ORR and DCR were no difference between low abundance and WT groups (p=0.754; p=0.610, respectively).

Conclusion
The abundance of EGFR mutations could affect the efficacy of EGFR-TKI, and quantitation of mutant EGFR could better predict for efficacy of EGFR TKI in advanced NSCLC.

Background
The search for novel biomarkers to define more successful and individual treatment approaches represent an important challenge for those involved in the care for patients with malignant pleural mesothelioma (MPM). In this exploratory study, we have systematically investigated the proteins present in plasma of MPM patients and correlated their levels with disease outcomes.

Methods
Plasma samples from twelve MPM patients (6 ‘short-’ and 6 ‘long-term’ survivors from parallel phase II studies investigating thalidomide) were used for proteomic analyses. Our series included samples from 9 patients with epithelial MPM and 3 patients with biphasic MPM. Plasma samples were immuno-depleted of the 14 most abundant proteins prior to labelling for isobaric tag for relative and absolute quantitation (iTRAQ) analysis using mass spectrometry. The most promising candidates and mesothelin were chosen for selected reaction monitoring mass spectroscopy (SRM-MS) quantification and enzyme-linked immunosorbent assay (ELISA) validation. Statistical analyses using T-Test of peak areas were used to identify proteins that were differentially expressed between the short- and long-term survivor groups.

Results
Median survival of short- and long-term survivors (1.2 and 38.3 months, respectively) differed significantly (p = 0.001). This was also the case for the neutrophil-to-lymphocyte ratio (NLR) that was significantly higher in the group of short-term survivors (p=0.03). Other baseline characteristics did not reveal major differences between the short- and long-term survivors. The total number of proteins identified was 226 (1% false discovery rate) in iTRAQ. A number of those were found to be differentially expressed between short- and long-term survivors (≥1.2-fold change; p≤0.05) by iTRAQ: selenoprotein P; tetranectin; insulin-like growth factor-binding protein 2 (IBP2); osteonectin (SPARC); platelet basic protein (CXCL7); and attractin. Mesothelin was assessed to validate the proteomic methodology: SRM-MS quantification was highly correlated with the MESOMARK ELISA values with a Pearson correlation of 0.82 (p=0.001). SRM-MS quantification revealed that the concentrations of attractin (p=0.02), tetranectin (p=0.003) and selenoprotein P (p=0.001) were higher in long-term survivors. In contrast, there was a trend for an increase in the concentration of SPARC (p=0.32), IBP2 (p=0.12) and CXCL7 (p=0.19) to be correlated with shorter survival. Furthermore, quantification by ELISA demonstrated an association between long survival and low concentration of SPARC (p=0.07) as well as high tetranectin (p=0.13).

Conclusion
We have demonstrated the feasibility of using the iTRAQ and SRM-MS proteomic techniques to investigate potential prognostic protein markers in plasma of MPM patients. Potential prognostic biomarkers worthy of further studies include SPARC and tetranectin and we plan to validate these in a larger clinical cohort.

Background
Small cell lung is a rapidly proliferating disease. Because of its high proliferation index, cancer cells rely on glycolysis, rather than oxidative phosphorylation, for ATP generation. Furthermore SCLC tumours often contain regions of hypoxia which switches tumour cells to a glycolytic phenotype. Increased glycolysis leads to increased lactate production which is effluxed from the cell in order to prevent reduced intracellular pH or inhibition of metabolic pathways via the monocarboxylate transporter (MCT) proteins MCT1 and MCT4. Inhibition of these transporters has been proposed as a method of selectively targeting highly glycolytic cancer cells. AZD3965 is an orally bioavailable MCT1 specific inhibitor currently under evaluation in phase I clinical trials. In an in vitro model of SCLC we have recently shown that in hypoxic conditions resistance to AZD3965 is associated with increased MCT4 levels (Polanski et al. submitted). Aim: To examined the expression and clinical significance of MCT1 and MCT4 in SCLC.

Methods
Archival SCLC biopsy specimens and clinical data from 78 patients presenting to the University Hospital of South Manchester and the Christie Hospital between 1994 and 2005 were analyzed. Nine representative cores from the tumour specimens were used to generate three TMAs. Sections were then stained for the markers MCT1, MCT4 and for the hypoxic marker CAIX. Staining was evaluated by two independent scorers and extent and intensity of the staining were estimated. A combined score for each case was calculated as the mean product of extent and intensity for all the cores in a case. The association between MCT1, MCT4 or CAIX and known prognostic factors was evaluated by Fisher’s exact test. Kaplan-Meier analysis was used to assess overall survival rates and Log Rank test was used for the comparison of the survival distributions.

Results
The proportion of tumours with any expression of MCT1, MCT4 or CAIX was 99%, 99%, 90% respectively. Higher levels of expression (intensity x extent) were observed for MCT1 (median=8.17) compared to MCT4 (median=2.21; p<0.001). A positive correlation was observed for CAIX expression and MCT4 expression. Tumours with CAIX expression, high MCT1 expression (>median) and low MCT4 expression ( 10 x 109/L, platelets < 150 x 109/L, Na < 135 mmol/L; LDH > 550 IU/L). However, high MCT1 expression score was associated with worse survival (14 vs. 32 months; p=0.019). Neither MCT4 nor CAIX expression was prognostic. Of the known prognostic factors assessed, extensive stage was significantly associated with shorter overall survival (6 vs. 27 months; p<0.001).

Conclusion
MCT1 and MCT4 are often expressed in SCLC and 21% cases in this series express a pattern associated with potential sensitivity to MCT1 inhibition. Higher expression of MCT1 is an adverse prognostic factor in univariate analysis reinforcing further evaluation of MCT1 inhibition in this disease.

Background
Malignant Pleural Mesothelioma (MPM) is an aggressive disease, often diagnosed at an advanced stage. It is characterized by a long latency period and prior asbestos exposure. Currently accurate diagnosis of MPM is difficult due to the lack of sensitive biomarkers, and despite minor improvements in treatment, median survival rates rarely exceed 12 months. Accumulating evidence suggests that aberrant expression of long non-coding RNAs (lncRNAs) plays an important functional role in cancer biology. LncRNAs are a class of recently discovered non-protein coding RNAs >200 nucleotides in length with a role in regulating transcription. The aims of this study were to characterize the expression and function of these lncRNAs in MPM.

Methods
To identify novel lncRNAs involved in MPM, microarray profiling was performed on five cell lines - the immortalized normal mesothelial cell line (MeT-5A) and four MPM lines (two epithelioid H28 and H226 and two biphasic MM05 and MSTO) using Invitrogen’s NCode lncRNA microarrays. These allow simultaneous assessment of mRNA and lncRNA content. High priority candidate lncRNAs were selected on the basis of statistical (P<0.05) and biological (>3-fold difference) significance. Expression of high priority candidates were technically validated using RT-qPCR, and biologically validated in three independent test sets. Pathway analyses were performed to interrogate the relationship between lncRNA and mRNA expression. Cell proliferation and colony formation assays were used to investigate lncRNA function.

Results
Microarray profiling and real-time qPCR validation identified 9 lncRNA candidates with significant differential expression in MPM compared with normal mesothelial cells Validation in three independent test sets by RT-qPCR analysis demonstrated consistent up-regulation of four of these lncRNAs. Receiver Operating Curve analysis showed that two of these candidates were able to separate benign pleura and MPM with high sensitivity and specificity. In addition, high expression of AK054908 was associated with nodal metastases with lower levels of AK130275 and AF268386 observed in patients receiving induction chemotherapy. Cases with higher EF177379 levels also demonstrated a trend to improved survival. The majority of mRNAs co-expressed with candidate lncRNAs were associated with cellular and metabolic processes including cell cycle, cell death and apoptosis. In functional studies, siRNA knockdown of AK130275 showed suppression of cell growth and colony formation in MPM cells with moderate changes observed following knockdown of EF177379.

Conclusion
To our knowledge this is the first systematic study of lncRNA expression profiles in MPM. We have found that lncRNA expression profiles can distinguish malignant mesothelium from normal pleural tissue, and that some lncRNAs are associated with nodal metastasis and long term survival. We also demonstrate that lncRNAs have potential prognostic and diagnostic utility with functional roles in regulating cell growth. Further work is required to evaluate whether these lncRNAs are capable of differentiating mesothelioma from lung cancer and benign asbestos-related diseases, and to reveal their specific functions in MPM pathogenesis.

Background
There has been a focus on the use of thymidylate synthase (TS) as a biomarker for the long term administration of metabolic antagonists for lung cancer. TS mRNA levels have been measured by real time RT-PCR in tumors collected using a micro-dissection method; however, this method has not been applicable for general use and its clinical use has been problematic. Therefore, this study investigated the correlation of TS mRNA expression level and a commonly used semiquantitative immunohistochemical method.

Methods
Resected lung specimens were collected from 47 patients with Stage IA (T1b) and IIB primary non-small-cell lung cancer who had undergone lobectomy between 2006 and 2012 at our Hospital. Levels of mRNA expression were measured using the Danenberg Tumor Profile method while TS immunostaining with anti-TS mouse IgG MoAB was graded into 4 levels.

Results
Of 47 patients, 24 patients were male and 23 patients were female. Thirty six people had adenocarcinoma while 11 patients exhibited squamous cell carcinoma. The number of patients at each IA, IB, IIA and IIB stage were 20, 20, 5, and 2, respectively. TS mRNA expression levels of these patients were between 0.69 and 12.68 (a median value of 2.47). Immunostaining indicated that Grade 0, 1, 2, and 3 were observed in 3, 17, 15 and 12, respectively. A moderate positive correlation was observed at a correlation coefficient of r=0.4880 between mRNA expression levels and immunostain levels (A). Moreover, a significant correlation between mRNA expression levels and immunostain levels was observed when mRNA expression levels were divided into two groups, one with higher than median values and the other with lower than median values, where Grade 0 and 1 were assigned as negatively stained and Grade 2 and 3 were assigned as positively stained (B).Figure 1

Conclusion
Immunostaining is a useful method for measuring TS level when TS is being considered as an appropriate biomarker for postoperative adjuvant chemotherapy with 5-FU delivertives.

Background
The diagnosis and treatment of non-small cell lung cancer (NSCLC) was revolutionised with the discovery of epidermal growth factor receptor (EGFR) activating mutations and the corresponding treatment of mutation-positive cancers with EGFR tyrosine kinase inhibitors (EGFR-TKI, i.e. gefitinib, erlotinib). While EGFR-TKI’s are highly effective in this setting, patients invariably relapse due to acquired resistance to the targeted therapy. In approximately 50% of the cases, resistance to gefitinib and erlotinib is associated with a T790M mutation that renders the EGFR tumor molecule unresponsive to these inhibitors. A number of 2[nd] and 3[rd] generation inhibitors are currently in development to address the primary T790M resistance mechanisms (afatinib, AZD9291, Clovis 1686). Identifying the relevant patient population for such drugs may rely on diagnostic analysis of re-biopsy material to determine the mechanism of acquired resistance. However, such invasive surgical procedures may present a significant challenge in a proportion of patients with EGFR TKI-resistant disease. Availability of a reliable, minimally invasive method to assess the EGFR mutation status of NSCLC patients could have major clinical benefits for patients and support access to novel targeted therapies. Recently, several highly sensitive technologies have been described for detection of EGFR mutations in the small amount of tumor DNA that is shed into the circulating plasma (ctDNA). ctDNA has potential as a minimally-invasive alternative to surgical biopsies and as such, holds promise for disease diagnosis and early assessment of disease progression.

Methods
We undertook a comprehensive cross-technology comparison of three technology platforms for detection of T790M mutation in ctDNA extracted from patient plasma: 1) ARMS-based detection using the Roche cobas® EGFR mutation detection kit; 2) digital droplet PCR using the BioRad ddPCR instrument (by MolecularMD); and 3) bead-based digital PCR using the Inostics BEAMing technology. In total, 140 frozen plasma samples (approximately 80 EGFR mutation +ve and 60 EGFR mutation -ve), obtained from EGFR-TKI resistant patients enrolled on two AstraZeneca clinical studies were used to assess mutation prevalence, false negative, and false positive rates. Each individual patient plasma sample was split and evaluated across multiple platforms. ctDNA was extracted and tested by operators blinded to the EGFR tumor mutation result for 3 common mutations (Exon 19 deletion E746_A750delELREA, L858R and T790M). The EGFR mutations status of patient-matched tumor biopsies was available for all plasma samples to allow for ctDNA-tumor concordance testing.

Results
Overall, concordance of EGFR mutation status was high between all 3 methods. In addition, the false positive rate in the known EGFR mutation -ve cases was low for all 3 methods. However, differences were seen in the rate of false negative results (assay sensitivity) between methods, with digital PCR showing increased assay sensitivity compared with the ARMS based method. A comprehensive comparative analysis for all 3 technologies will be presented.

Conclusion
Digital droplet and BEAMing PCR platforms both provide sensitive detection of EGFR mutations in ctDNA isolated from circulating plasma. Importantly, both platforms also provide results in relation to wild-type EGFR molecules, allowing for quantification EGFR mutation load

Background
Thymidylate synthase (TS) is a potential predictive marker for efficacy of treatment with pemetrexed. The current study aimed at investigating whether TS expression changes during non-pemetrexed chemotherapy of non-small cell lung cancer (NSCLC) thus making rebiopsy necessary for deciding on pemetrexed second line treatment.

Methods
TS immunohistochemístry was performed on biopsies and available resection specimens from 65 NSCLC patients stage T1-3N0-2 treated with preoperative carboplatin and paclitaxel (NAC-group) and 53 NSCLC patients stage T1-4N0-1 treated with surgery without preceding chemotherapy (OP-group) served as controls. The diagnostic biopsies and subsequent resection samples were compared in order to evaluate for concordance in TS expression in groups with and without preoperative chemotherapy.

Results
No statistically significant change in TS expression was observed between diagnostic biopsies and subsequent surgical resections of primary tumors in either the OP-group (p=0.186) or the NAC-group (p=0.542). Primary tumors were discordant between diagnostic biopsies and resection specimens when TS expression was dichotomized into high (H-score>150) and low (H-score ≤150), in 45% and 33% in the OP-group and NAC-group, respectively (p=0. 288).

Conclusion
The discordance observed between paired serial samples likely reflects intratumoral heterogeneity of TS expression and highlights the need of sufficient representative material for TS expression analysis if this biomarker is to be used for treatment selection.. TS expression in primary tumors remained unchanged and new biopsies for deciding on 2nd line pemetrexed do not seem warranted based on the current results.

Background
In a recently published study our group proposes a possible predictive impact of immunohistochemically detected RRM1 on vinorelbine efficacy in advanced NSCLC. This thesis was based on results from a randomized phase III trial comparing triplet-chemotherapy (paclitaxel, cisplatin, gemcitabine) to standard doublet-therapy (cisplatin, vinorelbine). We found that increased expression of RRM1 was associated with significantly decreased progression-free survival (PFS) and overall survival (OS) only in the patients receiving cisplatin-vinorelbine therapy. These findings were suprising since the overexpression of RRM1 is conventionally associated with resistance towards the chemotherapeutic agent gemcitabine, which is a potent inhibitor of ribonucleotide reductase (RNR). RNR is an essential enzyme for DNA synthesis that converts ribonucleoside di-phosphates into deoxyribonucleoside di-phosphates. The enzyme consists of a large sub-unit (RRM1) and a small sub-unit (RRM2). Vinorelbine is a spindle-poison and resistance towards this agent has never been associated with RRM1 over-expression. It has however been shown that vinorelbine can reduce the repair of radiotherapy-induced DNA damage in small-cell lung-cancer cell-lines, pointing to a possible interaction between vinorelbine and the DNA repair system. This study aimed at further exploring the possible predictive value of immunohistochemically detected RRM1 in patients receiving vinorelbine therapy. For this purpose we chose a cohort of malignant pleural mesothelioma patients treated with cisplatin and vinorelbine in a phase II trial.

Methods
Fifty-four consecutive patients with MPM, were enrolled between February 2003 and September 2006 into a phase II trial with cisplatin and vinorelbine. The formalin-fixed paraffin-embedded bioptic tumor specimens from these MPM patients were retrospectively evaluated for RRM1 expression by immunohistochemistry (IHC) using an H-score. The cut-off point was chosen as the upper quartile value of the H-scores to separate positive (H-score ≥upper quartile) from negative (H-score

Results
Sixty-six patients had enough tumor tissue for IHC. The upper quartile H-score was 3 yielding 15 positive and 34 (69%) negative tumors in the cisplatin-vinorelbine treated group. In the carboplatin-pemetrexed treated group there were 6 positive and 11 (64%) negative tumors. There was a significant overall survival advantage only in the RRM1-negative patients treated with cisplatin and vinorelbine (log rank p= 0.002). This group had a two-year survival rate of 47% opposed to 13 % for the RRM1-positive tumors treated with this combination. In the carboplatin-pemetrexed-treated group there were no differences in overall survival according to marker status, with two-year survival rates of 0% and 9% in the RRM1-positive and –negative group respectively.

Conclusion
Patients with RRM1-negative tumors treated with cisplatin-vinorelbine combination therapy had a prolonged overall survival. There was no survival advantage in the RRM1-negative group when patients were treated with carboplatin and pemetrexed. Our study suggest a possible role of RRM1 in predicting efficacy of cisplatin-vinorelbine combination chemotherapy ,also in MPM, which supports the similar finding from our group in NSCLC patients. Thus, RRM1 may be a biomarker for vinorelbine though the results require further validation.

Background
Variety of molecular methods for somatic mutations detection in NSCLC demonstrate discrepancies in sensitivity, cost and performance. Presence of activating mutation in TK domain of EGFR gene qualifies NSCLC patients to targeted therapy with TK inhibitors. Sanger sequencing is still a 'gold standard' for EGFR point mutation analysis and FISH is the primary method for detecting copy number of specific genes in tissue sections. Alternative methods like qRT-PCR, CISH or MLPA are more and more commonly used to detect EGFR point mutations or MET or HER2 amplification, yet results on their accuracy remain inconclusive.

Methods
We compared EGFR mutation analysis performed on 278 adenocarcinoma ( NSCLC FFPE and cytology) samples using Real-Time PCR technology with modified taqman probes to detect 29 EGFR mutations in exon 18, 19, 20 and 21 and Multiplex Ligation-dependent Probe Amplification (MLPA) to detect the most common deletion in exon 19, substitutions in exon 21 (L858R), 20 (T790M) and 18 (G719X), and insertion in exon 20 (S768I). Secondly, we evaluated MET and HER2 amplification in all 278 samples by MLPA only.

Results
EGFR mutation analysis demonstrated identical results in 97.5% cases between qRT-PCR and MLPA; 2.5% of discrepancy came from limit detection between those two methods (1% Limit of Detection for qRT-PCR; and 10-15% for MLPA). The fluorescent signals to cross the threshold for those 2.5% cases were detected in late cycles and positively correlated with small percentage of tumour cells in the sample. Both methods are more sensitive than traditional Sanger Sequencing which limit of detection is in a range 15%-25%. Additional MET and HER2 amplification analysis indicated that 10% of NSCLC samples carried MET and 1.8% - HER2 amplification.

Conclusion
Both Real-Time PCR and MLPA are accurate tools to detect point mutations or chromosomal aberration, respectively - as an alternative method to time consuming Sanger sequencing and FISH analysis for NSCLC biomarker detection. As MET amplification is known negative prognostic factor, in particular for patient treated with TK inhibitors, we suggest to use both technologies in NSCLC diagnostics: qRT-PCR for EGFR somatic mutation analysis and MLPA to detect MET amplification.

Background
Genomic profiling of SqCC has identified somatic alterations in NRF2 or its negative regulators (NFE2L2 mutations/amplifications, KEAP1 or CUL3 mutations/deletions) in ~1/3 of tumors. These alterations result in activation of the NRF2 transcriptional program, but the clinical significance of this pathway in lung SqCC patients is unknown. We hypothesize that a gene expression signature that reflects somatic NRF2-activating alterations may be identified and correlated with NRF2 protein over-expression. Furthermore, such gene expression or its immunohistochemical correlates may have prognostic significance and/or may be predictive of adjuvant chemotherapy benefit in early stage resectable lung SqCC patients.

Methods
Logistic regression (LR) and SAM analysis were applied independently to 104 SqCC cases from The Cancer Genome Atlas (TCGA) that had both microarray gene expression and mutation data to identify genes associated with NRF2 pathway mutational status. Overlapping genes were used to define the signature, which was then tested in 3 independent SqCC microarray datasets to evaluate its prognostic value. Correlation of the signature with NRF2 and KEAP1 mutations and with NRF2 and KEAP1 immunoreactive protein expression by immunohistochemistry (IHC) was evaluated. We also tested the gene expression signature as a potential predictor of adjuvant chemotherapy benefit in a subset of NCIC JBR.10 adjuvant chemotherapy trial patients with microarray data.

Results
A 28-gene signature that distinguished SqCC with or without aberration of the NRF2 pathway genes (NFE2L2/KEAP1/CUL3) in the TCGA dataset was identified. This gene signature that putatively represents NRF2 pathway activation status separates consistently SqCC into 2 groups in independent datasets. Both NRF2/KEAP1 mutation and NRF2 protein expression by IHC were significantly correlated with the NRF2 pathway activation signature (p<0.001 for each comparison). KEAP1 protein expression was not associated with the gene expression signature. No prognostic effect of the activated signature was observed in three independent datasets. In the JBR.10 patient cohort, a trend toward improved survival with adjuvant chemotherapy was observed in patients with the NRF2 “wild type” signature (HR 0.32, 95%CI 0.065-1.6 p=0.16), but not in patients with the “activated” signature (HR 2.28, 95%CI 0.24–22, p=0.48; interaction p=0.15).

Conclusion
A gene expression signature based on mutational activation of the NRF2 pathway may be predictive of benefit from adjuvant cisplatin/vinorelbine in SqCC. Patients with NRF2 pathway activating somatic alterations may have reduced benefit from this therapy. NRF2 immunohistochemistry could potentially be useful to identify NRF2-activated lung SqCC patients who may not benefit from adjuvant chemotherapy but this requires further validation.

Background
Oncogenic mutations of KRAS play important roles in cancer progression and resistance to chemotherapy in non-small cell lung cancer (NSCLC). Ras-related proteins, RALA and RALB, are members of the RAS superfamily of small GTPases and act as downstream effectors of KRAS. Therefore RAL proteins may be involved in cancer cell survival, invasion and metastasis. In this study, authors investigated the prognostic role of RALA and RALB overexpression in non-small cell lung cancer (NSCLC) patients with KRAS mutations.

Methods
Expression status of RALA and RALB in the tumor tissue of NSCLC patients whose tumor harbors KRAS mutations were evaluated by immunohistochemistry(n=12). In addition, we investigated 11 metastatic lung tumor tissue samples from KRAS mutant colorectal cancer patients. Predictive factors for survival were calculated using Kaplan-Meier estimator and Cox proportional hazards model.

Results
There was no statistically significant relation between RALA/RALB overexpression and KRAS mutation subtypes. However, RALB was frequently overexpressed in the tumor of advanced stageof NSCLC with KRAS mutation(p=0.015, χ[2]-test). Furthermore, the overall survival of the patients with RALB overexpression was significantly shorter than that of the patients without RALB overexpression (2.7 vs. 7.3 months, P<0.001; Mann-Whitney U test). In metastatic lung tumor tissues from KRAS mutant colorectal cancer patients, RALB protein overexpression showed only a trend toward shortened disease free survival.

Conclusion
RALB overexpression might be related to rapid disease progression and poor prognosis in NSCLC patients with KRAS mutations.

Background
The chemokine CXCL16 and its receptor CXCR6 are expressed on a variety of immune cells, and have been shown to influence angiogenesis. Hence, these markers are potentially interesting markers in NSCLC treatment. The expression of CXCR6 and CXCL16 has been examined in a variety of human cancers; however no studies have investigated their influence on prognosis in non-small cell lung cancer (NSCLC). We therefore wanted to explore their prognostic significance in NSCLC, in addition to examining associations with our previously investigated immunologic and angiogenic markers.

Methods
Resected tumor tissue from consecutive unselected stage I-IIIA NSCLC patients (1990-2005) were collected in two different cohorts; Nordland Central Hospital, NCH (n = 176), and University Hospital of North Norway, UNN (n = 159). Tissue microarrays were constructed from duplicate cores of neoplastic epithelial and stromal areas of each specimen. Immunohistochemistry was used to evaluate the expression of CXCR6 and CXCL16. Correlation analyses with well-known adaptive immunological (CD4, CD8, CD20), innate immunological (CD68, CD56, CD1A) and angiogenic (vascular endothelial growth factors and receptors, platelet derived growth factors and receptors and fibroblast growth factor-2 and receptor-1) markers were done. Disease-specific survival was used as endpoint, and survival analyses were performed in each cohort separately and in the combined total population.

Results
In univariate analysis, ↑stromal cell CXCL16 expression was a significant positive prognostic factor in the combined patient cohort (P = 0.016), supporting the similar trends in each cohort separately (NCH, P = 0.045; UNN, P = 0.137). Interestingly, the combinations (Figure 1) increased (↑stromal and ↑cancer cell), mixed (↑↓ and ↓↑) and reduced (↓stromal ↓cancer cell) CXCL16 expression patterns showed 5-year survival rates of 71%, 58% and 48%, respectively (P = 0.016). CXCR6 was expressed in cancer cells, but did not show any significant prognostic impact. There were no strong correlations between CXCR6/CXCL16 and immunological or angiogenic markers. In the multivariate analysis, combined ↓cancer and ↓stromal cell CXCL16 expression was an independent negative prognostic factor in the total cohort when compared to ↑stromal and ↑cancer cell expression (hazard ratio: 2.41; 95% confidence interval: 1.14 – 5.12, P = 0.022). Figure 1

Conclusion
We have shown that combined ↑cancer and ↑stromal cell CXCL16 expression is an independent positive prognostic factor in NSCLC. Further studies are warranted to elucidate the biological mechanism underlying this finding.

Background
Cancer-Testis antigens (CTAs) are immunogenic molecules exclusively expressed in normal testes but with aberrant expression in up to 30% of NSCLC. NY-ESO-1, is a CTA whose expression is associated with poorer survival but improved response to chemotherapy.[1] The promoter regions for many CTA genes contain CpG islands which are predominantly methylated in most tissues. In cancer, hypomethylation of CTA gene promoters hsvr been shown to be determining factor for the frequent re-expression of these genes. Herein we sought to correlate NY-ESO-1 promoter methylation (PM) with protein expression in lung cancers and test whether methylation status can be used as a predictive and prognostic marker in NSCLC.

Methods
We reviewed the clinicopathological data for patients with pathological N2 NSCLC treated with surgical resection followed by either observation or adjuvant chemotherapy, if treated after 2004. Genomic DNA from formalin fixed paraffin embedded (FFPE) samples were purified and bisulfite converted according to the manufacturer’s (QIAGEN) instructions. Mutational status was determined in genomic DNA using Sequenom's Oncocarta panel v1.0. Tumour samples were cut and stained with NY-ESO-1 antibody (E978). Previously described protocols for methylation-specific NY-ESO-1 primers[2] and beta-actin[3] control probes were used. Using real-time quantitative methylation-specific polymerase chain reaction (qMS-PCR), the level of NY-ESO-1 methylation was determined and expressed as the ratio of methylated to unmethylated molecules (M/UM). We used a cut-off fold-difference of 5 (Log2, M/UM) to separate tumour samples into hypomethylated and hypermethylated groups. Survival estimates were obtained using the Kaplan-Meier method. Comparison across groups was performed using the Chi-squared or Fisher's test.

Results
NY-ESO-1 PM MS-PCR was successful in 92/100 (92%) of specimens. NY-ESO-1 was hypermethylated in 86% (79/92) of samples. Age, smoking history, gender, histology and oncogenic mutations were not associated with presence of NY-ESO-1 PM. NY-ESO-1 hypomethylation was inversely associated with protein expression (p<0.0001). In multivariate analysis, hypomethylation of the NY-ESO-1 promoter was an independent poorer prognostic marker in patients not treated with chemotherapy (HR 2.97, 95%CI 1.06-8.29, p=0.038), after adjusting for age, gender, stage IIIA/B and histology. Conversely, in patients treated with chemotherapy there were no differences in survival associated with NY-ESO-1 PM, suggesting hypomethylated tumours remain chemosensitive.

Conclusion
NY-ESO-1 hypomethylation was an independent adverse prognostic marker in patients not treated with chemotherapy. Given their poorer outcome, association with chemosensitivity and known immunogenicity, CTAs may make attractive targets for immunotherapy, demethylating agents and immune checkpoint inhibition. References 1.John et al. PLOS One, In press. 2.James et al. Oncogene 2006. 3.Eads et al. Cancer Res 2001. .

Background
Intra-tumor heterogeneity can confound mutational status in oncodriver genes and challenge targeted cancer therapy strategies. Ultra-deep sequencing can detect low-frequency and expanded clonal mutations in primary tumors to better inform treatment decisions

Methods
KRAS coding exons in 61 treatment-naïve colorectal cancer (CRC) tumors were sequenced, along with KRAS, EGFR, ALK, and MET in lung tumors from 3 Chinese non-small cell lung cancer (NSCLC) patients

Results
Forty-two percent of CRC patients (28/61) harbored mutations in the KRAS active domain, 12 patients harbored >1 mutation, and eleven patients (18%), of which 5 had frequencies <10%, were not detected by Sanger sequencing. Low frequency KRAS active (G12R) and EGFR kinase domain mutations (G719A) were identified in one NSCLC patient. Multiple low frequency mutations in KRAS, EGFR, and MET and ALK gene copy number increases were found in a second NSCLC patient; a third NSCLC patient had EML4-ALK fusion with ALK, EGFR, and MET mutations. Multiple low frequency mutations occurred within individual gene in all three patients.

Conclusion
A complex pattern of intrinsic low frequency driver mutations in well-known tumor oncogenes may exist prior to treatment, potentially resulting in resistance to targeted therapies. Ultra-deep sequencing can characterize intra-tumor heterogeneity and identify such mutations which could ultimately impact treatment decisions.

Background
In a previous report of EGFR WT advanced NSCLC patients treated with first-line platinum-based chemotherapy we observed a worse clinical outcome for KRAS-mutants compared with KRAS WT patients (Metro et al. ESMO 2012). Here, we assessed whether this phenomenon could be due to different levels of ERCC1 expression.

Methods
From a prospectively maintained database of EGFR WT advanced NSCLC patients diagnosed at a single Institution between January 2006 and November 2012, we identified the individuals who had a known KRAS mutation status and tissue available for assessment of ERCC1 mRNA expression. Total RNA was isolated from paraffin-embedded tumor specimens using RNeasy Mini kit and automatically purified by QiaCube instrument (Qiagen). Quantification of mRNA expression levels of ERCC1 was analyzed by real-time one-step RT-PCR using QuantiFast technology by RotorGeneQ instrument (Qiagen), and the results were compared considering β-actin as the internal reference gene.

Results
One hundred and eleven patients were evaluable, 60 of which were KRAS-mutants. Among KRAS-mutants, the rate of codon 12/13/61 mutations were 80%/13.3%/6.7% respectively. Baseline patients characteristics were as follows: median age was 62 years (35-84), 36.9% were male, 63.9% were stage IV, 78.3% were PS 0 or 1, 87.3% were ever-smokers, and 71.1% had received a first-line platinum-based chemotherapy. More ever-smokers were present in the KRAS-mutant subgroup compared with WTs (90% versus 76.5%, respectively, P = 0.08). ERCC1 average scores ranged from 0.1 to 26.7, the values being not normally distributed (Kolmogorov-Smirnov test, P<0.0001). Median and mean overall ERCC1 values for all patients were 1.3 and 2.2 [standard deviation (SD) 3.4], respectively. There was no statistically significant difference in terms of ERCC1 median values betwen KRAS-mutants and KRAS WTs (1.4 vs. 1.3, respectively, P = 0.27). Nevertheless, mean ERCC1 expression levels were found to be significantly higher in KRAS-mutants compared with KRAS WTs [2.9 (SD 4.5) vs. 1.4 (SD 0.8), respectively, P = 0.02]. This finding was due to 7 KRAS-mutant patients (ERCC1 high) coming out with ERCC1 levels higher than 5.0, thus notably incresing mean ERCC1 values. In the group of patients treated with first-line platinum-based chemotherapy (n = 79), median progression-free survival was 1.9 months for KRAS-mutant, ERCC1 high patients (n = 6), 5.1 months for KRAS-mutant, ERCC1 low patients (n = 38), and 7.1 months for KRAS WT patients (n = 35) (P = 0.003).

Conclusion
KRAS-mutant NSCLCs may express higher levels of ERCC1 compared with KRAS WTs, which could translate into poor sensitivity to first-line platinum-based chemotherapy. Combination strategies of platinum-based chemotherapy plus KRAS-targeting agents may represent an appealing upfront strategy for KRAS-mutants advanced NSCLCs, particularly in presence of concomitant expression of high ERCC1 levels.

Background
Metabolomics is an approach that is increasingly used in research of different types of cancer as it reflects the most proximate deviations in metabolism of cells and tissues. It has been shown before that untargeted metabolomics can effectively distinguish diseased tissue samples from healthy ones and diseased individuals from healthy ones regarding the general metabolite pattern of the samples.

Methods
In this study lung cancer tissue and macroscopically non-cancer tissue were obtained from lung cancer patients (n=45) and analysed using liquid chromatography combined with tandem mass spectrometry to acquire full spectra of metabolites between mass-to-charge ratio (m/z) 50-1000 extracted in hydrophobic and hydrophilic phase in positive and negative ionisation mode. Principal component analysis (PCA), sparse partial least squares discriminant analysis (SPLS-DA), paired t-test were used for data analysis. Fragmentation and comparison of fragment spectra to public and in-house databases were used for metabolite identification.

Results
PCA and SPLS-DA indicated the existence of systematic differences in between the two types of tissue based on the data matrix extracted from mass-spectra. All together 496 signals (13% of all measured signals) were found to have significant (p<0.001) alteration by t-test between cancer and non-cancer spectra. Eleven most prominent of them were selected for fragmentation. Their mass-to-charge ratios (m/z) were in hydrophilic phase as follows: +86, -147, +168, -171, -201, and in hydrophobic phase +735, +736, -774, -775, -888 and -889. Signals with m/z of -171, +735, +736, -774 and -775 were found more abundant in non-cancer tissue and the signals with m/z of +86, +168, -147, -201, -888 and -889 in the cancer tissue compared to non-cancer tissue. From hydrophobic phase signals +735 and +736 were identified as unspecified species of phosphatidylcholines. M/z -888 was identified as phosphatidylinositol 18:0-20:3 (stearic acid and Mead acid as lipid acid residues) having on average two-fold stronger signal in the spectra of cancer tissue compared to non-cancer tissue. The phosphatidylinositols are, in addition to structural purposes in membranes, known to be used as a depot for PUFA-s, such as arachidonic acid (AA; 20:4). As the increase of free Mead acid in blood has been linked to the deficiency of essential fatty acids (as AA) in food, we suggest our finding indicates that lung cancer cells have significantly increased demand for inflammatory mediators synthesized from AA.

Conclusion
Cancer tissue is clearly distinguishable from non-cancer tissue with mass spectrometry. From all low-molecular weight metabolites 13% are differing in cancerous and non-cancer tissue. Most striking changes were assigned to phospholipids and in particular to Mead acid containing phosphatidylinositols, which are upregulated in cancer tissues by 100%. The results imply differential regulation of polyunsaturated fatty acids in lung cancer, which improves our knowledge of molecular mechanisms of cancer and opens new ways for accurate prognostic markers and personalized approach to postoperative treatment.

Background
Radiation-induced lung damage (RILD) is a dose-limiting toxicity of lung radiotherapy. Individual sensitivity can be measured by changes in Hounsfield Units over time (delta HU) on CT scans (De Ruysscher et al. Acta Oncol 2013). This endpoint is specific for lung damage and does not correlate with dyspnoea, which is multi-factorial. In this study, we investigated the association between density changes over time and SNPs aiming at finding individual sensitivity for RILD.

Methods
Delta HU/Gy and delta HU/Gy x MLD (Mean Lung Dose), the latter to take into account a volume factor for RILD, were correlated with 314 SNPs related to fibrosis and inflammation. The outcome variables were square root transformed because both were not normally distributed. Univariate ANOVA analyses were performed for comparisons of means. P-values of less than 0.01 were considered to be significant.

Results
Eighty-nine lung cancer patients were studied, 63 men and 26 females. Twenty patients were treated with radiotherapy alone, 31 with sequential chemo-RT and 38 with concurrent chemo-RT. Twenty percent of the patients developed grade 2 or more clinical dyspnoea after treatment. Three SNPs were significantly correlated with delta HU/Gy: rs2252070 (p=0.006, MMP13), rs2230588 (p=0.009, JAK1) and rs12901071 (p=0.009, SMAD3) [Table 1A]. For delta HU/Gy x MLD, significant associations were found for rs3819122 (p=0.008, SMAD4), rs2230529 (p=0.009, ITGB2) and rs2230588 (p=0.009, JAK1) [Table 1B]. Figure 1

Conclusion
Quantification of CT density changes due to radiotherapy, measured as HU changes over time as a specific and quantitative endpoint for RILD correlates with specific SNPs in genes involved in signal transduction of cytokines (SMAD3/4, JAK1), in the extracellular matrix (MMP13) and in cell adhesion (ITGB2). External validation will follow.

Background
Mutational profiling for targetable oncogenic drivers is systematically recommended in the pretherapeutic workup of metastatic lung adenocarcinoma. Pathological and molecular diagnoses may be performed on specimens from metastatic lesions, when the primary pulmonary tumor is not accessible, either because of proximal location increasing the risk of transthoracic procedures, or when the material collected is insufficient in size. The feasibility of performing molecular diagnoses on small specimens from bone metastases has been questioned over time. To prospectively study patients with bone metastases from lung cancer, we set up a multidisciplinary group at the Hospices Civils de Lyon, including rheumatologists, pulmonologists, oncologists, interventional radiologists, pathologists and molecular biologists.

Methods
POUMOS was a prospective observational study aiming at evaluating the feasibility of routine percutaneous biopsy of synchronous bone metastases from lung adenocarcinoma, to perform pathological diagnosis and mutational profiling on the bone lesion. Results were correlated with that obtained on specimens from the primary tumor, if available. Technically, bone metastasis specimens, usually multiple, were sent fresh for immediate formalin-fixation, and, if possible, snap-freezing. Decalcification of bone was performed using EDTA for a better preservation of cell morphology and DNA. Mutational profiling of EGFR, KRAS, HER2, BRAF, and PIK3CA, as well as testing for ALK rearrangements, was conducted as recommended by the French National Cancer Institute (INCa) for all adenocarcinoma cases. DNA extraction was performed after laser microdissection of cancer cells; genotyping was based on direct sequencing and/or SNaPshot. Complete description of the process will be presented at the meeting.

Results
Starting April 2011, 46 patients with lung adenocarcinoma and synchronous bone metastasis were enrolled. No grade 3-4 adverse effects were reported after the bone biopsy. Mutational profiling was obtained in 45 (98%) cases; one specimen did not provide with sufficiently good quality DNA for the analysis. EGFR mutation was observed in 6 (13%) patients, KRAS mutation in 14 (30%) patients, HER2, BRAF and PIK3CA mutations in 1 (2%) patient each. Updated results, especially correlations between the mutational profiles of primary lung tumors and bone metastases, will be reported at the meeting.

Conclusion
Our data demonstrate the feasibility of percutaneous biopsy of synchronous bone metastasis from lung adenocarcinoma to conduct mutational profiling for common oncogenic alterations. The establishment of multidisciplinary teams to ensure the coordination between clinicians, radiologists and pathologists, makes routine pathological and molecular diagnosis on bone metastasis specimens a fast, reliable and safe procedure.

Background
KRAS mutations are common driver mutations in 30% of non-small cell lung cancer. However, treatment for lung cancer patients with KRAS mutations remains unestablished. Therefore, to screen other targetable receptor tyrosine kinases (RTKs) in lung cancers with KRAS mutations, we performed phospho-RTK array analyses using 6 KRAS mutated lung cancer cell lines.

Methods
Sixteen NSCLC cell lines were analyzed using Human-Phospho-RTK Array Kit (R&D Systems, MN) and relative phosphorylation levels of 42 RTKs were examined. Phosphorylation levels of RTKs were quantified relative to the average of positive controls using image analysis software JustTLC (Sweday, Lund, Sweden). We also examined growth inhibitory effect of small interfering RNA (siRNA) and erlotinib in KRAS mutant lung cancer cell lines. In addition, we developed erlotinib-resistant cell lines from erlotinib-sensitive H358 cells by chronic drug exposure for 3 months.

Results
In the phospho-RTK array analysis, phosphorylation levels of EGFR were lowest in the cell lines with KRAS mutation (Table1). Three of six cell lines with KRAS mutation showed IGF-1R phosphorylation and the difference was statistically significant compared with 0 of ten cell lines without KRAS mutation (P=0.03, Fisher's Exact test). KRAS siRNA did not induce apoptosis in 5 cell lines with KRAS mutation but mild apoptosis in H358 cells. Moreover, only H358 showed mild sensitivity to erlotinib (IC~50~ 120nmol/L). Analysis for erlotinib-resistant H358 cells identified increased phosphorylation of IGF-1R compared with H358 parent cells (18.0 times).Figure 1

Conclusion
These results suggest IGF-1R may be a candidate molecular target in lung cancers with KRAS mutation.

Background
Epidermal growth factor receptor (EGFR) gene variation is one common driver mutation in lung cancer. In lung adenocarcinoma, EGFR mutations are the powerful predictors to the efficacy of EGFR-TKIs. However, the phenomenon seemed to be inconsistent in lung squamous carcinoma. The present study aims to investigate the sensitivity of lung squamous cell carcinoma (SCC) harboring EGFR mutation to EGFR-TKIs, and explore the likely molecular mechanisms through constructing EGFR mutant lung SCC cell lines and analyzing the results transcriptomics.

Methods
This study retrospectively enrolled 91 patients with advanced lung SCC who received EGFR-TKIs in Beijing University Cancer Hospital. Denaturing high-performance liquid chromatography (DHPLC) and immunohistochemistry (IHC) were used for the detection of EGFR mutations and protein expressions of P63 and TTF-1, respectively. Stable cell lines were constructed through lipofectamine and confirmed by western-blot, soft agar cloning formation and tumoregeneity in nude mice. Cell titer Glo was used for cell number counting. RNASeq was used for the analysis of transcriptomics and related signaling pathways were screened in KEGG website. SPSS (17.0 version) was used for statistics.

Results
The frequency of EGFR mutation was 16.2% (64/396) in lung SCC. In lung cancer patients with EGFR mutation, lung SCC patients accounted for 7% (64/863), commonly in male (71.9%), smoker (59.4%) and EGFR exon 19DEL (64.1%). In total 91 patients who received EGFR-TKIs, the objective response rate (ORR) and the disease control rate (DCR) were 11.0% (10/91) and 56.0% (51/91) respectively and the progression-free survival (PFS) and the overall survival (OS) were 2.9 months and 16.3 months. In terms of 60 patients who provided EGFR mutated status, 29 patients harbored EGFR mutation and the other 31 without EGFR mutation. The PFS of the subgroup with EGFR mutation after EGFR-TKIs showed prolonged trend compared with that of the subgroup without EGFR mutation, however, the statistics failed to amount to significant difference (2.4 months vs. 1.8 months, p=0.071). No statistical differences were observed in PFS ORR, DCR and OS between the two subgroups. IHC results demonstrated that most of lung SCC patients harboring EGFR mutation showed P63 positive and TTF-1 negative (15/18). We then constructed lung SCC cell lines with EGFR mutation (Beas-2B/EGFR 19DEL cell line and Beas-2B/EGFR 21L858 cell line), which showed resistant to erlotinib with the IC50 of 3.43±0.56μM和11.9±1.18μM. After the treatment of erlotinib, the mainly up-regulated signaling pathways included TGFβ pathway and hedgehog pathway, and RNA levels of some mesenchymal genes were also up-regulated.

Conclusion
Lung SCCs with EGFR mutation were not uncommon, who showed insensitivity and even resistance to EGFR-TKIs. The likely molecular mechanisms included TGFβ related epithelial-mesenchymal transition (EMT) and tumor stem cell like differentiation, and the upregulation of bone morphogenetic protein 4 (BMP4).

Background
Acquired resistance in NSCLC with activating EGFR mutations has been defined by Jackman et al. The molecular mechanisms have been studied in small series of patients: in about 50% of cases, T790M mutation confers acquired resistance, in 25% c-Met overexpression detected by IHC. Other as of yet unknown mechanisms might also play a role. Here we present the acquired resistance mechanisms in a cohort of 12 patients (out of 33 pts.) with activating EGFR mutations treated with TKI and identified at our center (see Halbfass et al.). Treatment methods and outcome are also described

Methods
159 consecutive patients (s. abstract Halbfass et al.) were molecularly studied for EGFR mutations in exons 18-21. EGFR mutation analysis was performed by Sanger Sequencing or by hybridization based COBAS methodology after microdissection of tumor tissue to ascertain a high percentage of tumor tissue. c-Met IHC was performed automatized by standard procedure (BondMax, Menarini). ALK rearrangement was studied using a break apart FISH-Probe (Abbott). Remission was measured by RECIST 1.1 criteria.

Results
Of 159 patients tested, 33 had an EGFR mutation, of which 3 had primary resistance mutations (T790M 1 case, Exon 20 insertions 2 cases). Of these 30 pts., 20 had acquired resistance and 12 were rebiopsied. 8 were not rebiopsied for various reasons, the most common being progress in the CNS. Of the 12 rebiopsied pts, c-Met was successfully studied in 8, T790M in 12 and ALK in 6. In 5/8 pts., c-Met was amplified, in 4/12 pts. the T790M mutation was found and in 1/6 pts., an ALK rearrangement was observed. 4 pts. received afatinib, 2 pts. chemotherapy, 2 pts. afatinib and cetuximab, 1 pt. afatinib and crizotinib and 2 pts. other treatments. Detailed treatment results will be presented at the meeting.

Conclusion
The understanding of resistance mechanisms in acquired TKI resistance is of academic interest and might be important for subsequent treatment options. Therefore, rebiopsy at progress while on TKI therapy should be discussed with patients. As targeted treatment options are available for c-Met, ALK as well as T790M, resistance mechanisms potentially may guide therapeutic strategies in EGFR mutated patients with acquired resistance.

Background
The interactions between radiation dose, toxicity and systemic responses are poorly understood during radiotherapy of patients with NSCLC). The purpose of this study was to monitor DNA damage using the gamma-H2AX assay in tissues inside and outside irradiated volumes of patients with NSCLC.

Methods
This prospective ethics approved study assessed 12 patients receiving radiotherapy was planned to 60 Gy in 30 fractions over 6 weeks. Six patients receive concurrent platinum doublet chemotherapy (chemoRT), and six patients had radiotherapy alone. The number, distribution and kinetics of gamma-H2AX foci were compared with the irradiated volume. Lymphocytes and eyebrow hairs were processed for gamma-H2AX staining and microscopy at each of 5 time-points; baseline, 1 and 24 hours post-first fraction, 4 weeks into radiotherapy, and 3 months after treatment completion.

Results
The mean irradiated target volume was 384 cm[3] (range 87-1137 cm[3]). We observed the presence of a small subpopulation of lymphocytes with multiple (>5) gamma-H2AX foci at 1-hour post-first fraction. There was no difference in this subpopulation between patients receiving chemoradiotherapy or radiotherapy alone at baseline (p=0.26) nor at 1-hour (p=0.24) There was a strong correlation between the size of this subpopulation and irradiated volume (r=0.84, p<0.01), indicating direct radiation exposure. This suggests potential utility of the gamma-H2AX assay as a human in-vivo biodosimeter. This subpopulation was not detected at 24 hours due to DNA damage repair. A trend for reduction of this subpopulation and the average number of foci in lymphocytes analysed at 4 weeks of radiotherapy suggests an impaired radiation response after multiple radiotherapy fractions. By contrast, the mean number of observed hair follicle gamma-H2AX foci was not different from baseline to 1-hour post first-fraction (p=0.42), elevated at 24 hrs (p=0.10) and 4 weeks (p=<0.01) but was not different from baseline at 3 months (p=0.31). The scattered dose at the eyebrow was recorded at <0.01 Gy per fraction and was insufficient to directly induce the observed gamma-H2AX signal Figure 1 Figure 1 - a brisk DNA damage response in out-of-field eyebrow hair 24-hours post radiation (gamma-H2AX foci in green)

Conclusion
The Gamma-H2AX assay on peripheral blood at 1-hour may be a useful as a human in-vivo biodosimeter. To our knowledge, this study is the first report of abscopal DNA damage response in hair follicles associated with radiotherapy in cancer patients. Further validation of our findings on a larger patient cohort is warranted.

Background
Rash, liver dysfunction, diarrhea and pneumonitis are known as adverse events of epidermal growth factor receptor tyrosine kinase inhibitors (EGFR-TKIs). Transforming growth factor-β1 (TGF-β1) is a cytokine and acts as an anti-proliferative factor in normal epithelial cells and epithelial-mesenchymal transition (EMT) with invasion and metastasis in cancer cells. TGF-β1 mediated Smad activation caused EMT, and activation of the EGFR. TGF-β1-mediated EGFR activation was abolished by EGFR suppression or extracellular EGF depletion pathway. TGF-β1 genotypes are associated with serum level of TGF-β1. It has been hypothesized that TGF-β1 genotypes may be implicated in clinical outcomes of EGFR-TKIs.

Methods
Patients were identified through a query of patient information for subjects enrolled in the Medical Information System in Osaka City University Hospital between January 1999 and February 2012. The associations between three genetic polymorphisms of TGF-β1 (C-509T, T869C, and G915C) and adverse events of erlotinib and gefitinib have been studied. Genomic DNA was extracted from peripheral blood or formalin-fixed and paraffin-embedded tissues. TGF-β1 genotypes were determined using RT-PCR method. The primers were designed to amplify the target fragments of TGF-β1 rs1800469, rs1800471, and rs1982073, respectively. In order to identify the risk factors for the adverse events, gender, age, stage and three genetic polymorphisms of TGF-β1 were selected and estimated for their potential confounding effects on rash, diarrhea, and liver dysfunction by multivariate analysis. Unconditional logistic regressions were used to compute the odds ratios (ORs) and their 95% confidence intervals (CIs). All analyses were two-sided, and p values of less than 0.05 were considered statistically significant. This study was approved by the ethics committee of Osaka City University (approval number, 1700).

Results
A total of 255 patients received gefitinib, and 75 patients received erlotinib. In the gefitinib group, the rates of rash, diarrhea, and liver dysfunction of grade 1 or more were 66.7%, 26.0% and 48.5%, respectively. In the erlotinib treatment group, the rates of rash diarrhea, and liver dysfunction of grade 1 or more were 84.1%, 43.5% and 33.3%, respectively. The C-509T consisted of TT in 26.6%, CT or CC in 73.3% of patients (n=289), respectively. The T869C consisted of CC in 26.1%, TC or TT in 73.9% of patients (n=272), respectively. The G915C consisted of GG in 100% of (n=313) patients. TT of the C-509T and CC of the T869C were significantly associated diarrhea of grade 1 or more in erlotinib group (OR 0.21, 95% confidence interval [CI] 0.054-0.72, p < 0.001, and OR 0.21, 95% CI 0.05-0.73, p = 0.014, respectively). No associations were observed between TGF-β1 genotypes and any adverse events in gefitinib group.

Conclusion
Minor alleles of TT of the C-509T and CC of the T869C were associated with erlotinib induced diarrhea. These alleles are generally associated with high level of TGF-β1 in serum. The increase level of TGF-β1 may be a risk factor for mucosal damage of the gastrointestinal tract in patients treated with erlotinib.

Background
Non-small cell lung cancer (NSCLC) is the most common type of lung cancer. Although some progress has been made in the development of its treatment, most patients are diagnosed at advanced stage and have a short overall survival rate. Just as specific genes have to be analyzed before selecting a targeted therapy, microRNAs (miRNAs) are emerging as biomarkers for NSCLC diagnosis and prognosis. miRNAs are small oligonucleotides that regulate target mRNAs in the post-transcriptional step driving the expression of some genes involved in the tumorigenesis of NSCLC. Due to their resistance to nuclease digestion, miRNAs can be detected both in lung tissue samples and in body fluids as circulating factors. Our aim was to identify specific non-invasive cancer biomarkers involved in NSCLC tumorigenesis regulation.

Methods
33 NSCLC patients, 19 male and 14 female, were enrolled. The mean age was 63.3 and 66% were smokers. 88% of patients had advanced stage (IIIb-IV) tumors of which 23 were adenocarcinoma and 10 squamous cell carcinoma. We collected serum before chemotherapy and, when available, tissue samples from biopsy. MiRNA expression profiling of 33 serum and 10 tissue paired samples and 10 serum samples from normal volunteers were investigated by Affymetrix GeneChip miRNA Array. meV software was used for statistical analysis, and target genes identified by deregulated miRNAs were analyzed through the miRWalk database.

Results
Statistical analysis revealed 47 miRNAs differentially expressed in NSCLC serum samples compared to control serum (p<0.05) and 29 miRNAs deregulated in NSCLC tissue samples compared to their normal counterparts. 326 miRNAs resulted differentially expressed between the serum and the tissue of NSCLC patients. Among these deregulated miRNAs, the Venn diagram comparing tumor/normal serum, tumor/normal tissue and tumor serum/tissue samples showed that the serum of NSCLC patients was characterized by 22 miRNAs, while 10 miRNAs identified the tumor tissue of NSCLC patients. Only one miRNA, miR-133a, was detected both in NSCLC serum samples and in tissue ones. Focusing on miRNAs involved in NSCLC pathogenesis, of 22 miRNAs miR-486-5p had a lower mean expression ratio (MER) in tumor vs normal serum (8.70±3.28 vs 10.78±1.35), as did let7b which had a MER of 5.28±2.63 in tumor serum vs 7.14±1.8 in normal serum samples. Of 10 miRNAs, miR-200c had the lowest MER in tumor vs normal tissue (1.25±0.10 vs 2.37±0.09 respectively) as did miR-29c* which had a MER of 2.18±0.08 in tumor tissue samples vs 2.34±0.11 in normal tissues. miR-133a had a comparable MER in tumor serum vs tumor tissue (2.41±0.17 vs 2.46±0.17 respectively). These miRNAs are able to target PIK3CA and PTEN genes, according to the miRWalk database, which are involved in the EGFR-related pathway.

Conclusion
Our results highlighted specific miRNA expression profiles, both in the serum and in the tissue of NSCLC patients, able to identify circulating miRNAs that could be used as non-invasive biomarkers for early diagnosis, or to predict prognosis in NSCLC patients thus improving personalized therapy. These data are preliminary to a prospective clinical validation in a multicentric trial.

Background
Epidermal growth factor receptor (EGFR) is widely distributed in human epithelial cell membrane, including esophageal squamous cell carcinoma (ESCC). The purpose of this study was to evaluate the relationship between malignant biological behavior and EGFR status in ESCC.

Methods
We investigated tumor specimens from 56 patients with surgically resected ESCC from 2004 through 2008. Immunohistochemistry (IHC) was performed to analyze the expression of EGFR. Fluorescence in situ hybridization (FISH) was performed to assess the EGFR gene amplification. EGFR mutations in exons 19-21 were detected by pyrosequencing technology. A chi-square test or Fisher exact test for independence was used to examine the correlation among the status of EGFR protein, gene and the several clinicopathological factors. Overall survival and disease-free survival were constructed using the Kaplan-Meier method, and the log-rank test was used to evaluate the statistical significance of differences. Multivariate analysis was performed using the Cox proportional hazard method.

Results
EGFR was overexpressed in 30 of 56 ESCC (53.6%) and was correlated with tumor differentiation (p=0.047) (Figure 1.). EGFR gene amplification was found in 13(23.2%) cases and was correlated with the presence of lymph node metastasis (p=0.001) and higher pathological tumor-node-metastasis (pTNM) stage (p=0.042). There was no EGFR mutation in the clinical samples of 56 patients with ESCC. In univariate analysis, there was no correlation between the prognosis and EGFR protein expression, but EGFR gene amplification was a significant predictor of better prognosis (p=0.031). The multivariate analysis revealed that EGFR gene amplification was significantly correlated with better disease-free survival (p=0.037) and overall survival (OS) (p=0.026). However, patients with EGFR gene amplification did more likely receive aggressive treatment in clinical practice.Figure 1

Conclusion
EGFR protein overexpression and gene amplification in ESCC were correlated with the malignant biological behavior, including tumor differentiation and lymph node metastasis. Further studies should be conducted to analyze the prognostic value of EGFR protein overexpression and gene amplification in patients with ESCC.

Background
Trametinib is a reversible and highly selective allosteric inhibitor of MEK1/MEK2. In a 2:1 randomized, open-label phase 2 study, trametinib showed an overall response rate of 12% but did not show improvement in progression-free survival (PFS) over docetaxel as second-line treatment in patients with KRAS-mutant NSCLC. Median overall survival (OS) was 8 months in the trametinib arm and has not been reached in the docetaxel arm (ASCO 2013 abstract #8029). Several CAF markers were identified as prognostic and predictive of clinical benefit in patients with renal cell carcinoma (Tran, Lancet 2012), and of tumor shrinkage in patients with NSCLC (Nikolinakos, Cancer Research 2010), after receiving targeted therapy.

Methods
Of 134 patients randomized, plasma samples (n = 116 baseline [113 KRAS-mutant], n = 89 day 22) from patients who consented for this optional study were analyzed for 38 CAFs using SearchLight multiplex assays in a CLIA-certified laboratory. Baseline CAF levels were tested for association with PFS using proportional hazards regression within and between arms. Correlation between baseline CAFs and baseline tumor burden was assessed using the Spearman rank correlation test. Change from baseline was assessed using Wilcoxon signed rank tests. P < .01 was considered significant; for treatment arm by CAF-level interaction, P < 0.05 was considered significant.

Results
Lower baseline levels of IL-6 and OPN and higher baseline levels of TRAIL were associated with longer PFS in patients treated with trametinib but not those treated with docetaxel. Interaction between CAFs and trametinib or docetaxel treatment was significant for IGFBP-1, MMP-9, E-selectin, and VEGF. There was a positive correlation between IL-6 levels and baseline tumor burden. At day 22, decreases (trametinib: ANG-2, IGF-1, IGFBP-3, IL-10, IL-2R, TIMP-1; docetaxel: IL-6, MIP-1A, MIP-1B, SDF-1) and increases (trametinib: IGFBP-1, IL-6, MMP-2, PIGF, VEGF) in CAF levels were observed. Additional results (pairwise correlation, CAF levels and OS, change from baseline CAF levels and PFS and OS) will be reported.

Conclusion
Circulating baseline CAF levels may be predictive of PFS for patients treated with trametinib or docetaxel. CAF levels are modulated by each treatment. The impact of circulating baseline CAFs on PFS warrants additional investigation in well-designed clinical trials. Plasma CAF profiling may aid in the prognostic evaluation of patients and determine potential therapeutic response to trametinib treatment in patients with NSCLC.

Background
Lung cancer is the leading cause of death among malignant tumors worldwide therefore it is important to develop novel diagnostic techniques.

Methods
This is a prospective case control study including two groups of patients: Group I: healthy volunteers as control .Group II: Patients with advanced non-small cell lung cancer (NSCLC). Plasma VEGF 165 levels were measured at baseline by ELISA before first line Gemcitabine-Cisplatin regimen. High VEGF 165 cutoff taken was >703 pg/ml .Primary end point was comparison of plasma VEGF 165 levels in cases and controls. Secondary endpoint was correlation between High VEGF 165 and clinical response (CR), Progression free survival (PFS) and overall survival (OS) in advanced NSCLC patients.

Results
35 patients with advanced NSCLC and 34 age and sex matched normal subjects as control were included and followed up during the period from 10/2009 to 10/2012 with median follow up of 10.5 months. Median plasma VEGF 165 level was 707 pg./ml in cases versus 48 pg./ml in controls, (p < 0.001).No significant correlation was found between plasma VEGF 165 level and clinical response(p < 0.5).No significant correlation was found between plasma VEGF 165 level and Median PFS and OS (p=1 and 0.7 respectively.

Conclusion
Plasma VEGF 165 is a potential diagnostic marker in advanced NSCLS.

Background
Although a high level of thymidylate synthase (TS) expression in malignant tumours has been suggested to be related to a reduced sensitivity to the antifolate drug pemetrexed, no direct evidence for such an association has been demonstrated in routine clinical samples from patients treated with this drug. The purpose of this study was to evaluate the impact of quantitative TS expression in tumor cells as predictor of the efficacy in patients with advanced non-small cell lung cancer, small cell lung cancer (SCLC) and mesothelioma treated with pemetrexed in our institution.

Methods
54 patients were included in this study: 40 stage IV NSCLC (26 adenocarcinomas, 11 large cell, and 3 squamous cell carcinoma), 3 SCLC and 11 mesothelioma. 21 patients received platins-pemetrexed as first line NSCLC, 20 pemetrexed in monotherapy as second and further lines and 3 carboplatin-pemetrexed fo extensive disease SCLC. Total RNA was isolated by RNeasy FFPE procedure (Qiagen). The expression of TS was analyzed by RT-qPCR using appropriate mRNA specific primers and probes in LightCycler 480II platform at 45 cycles. TS levels was calibrated to expression in normal tissue.

Results
From 54 cases, TS expression data were available in 32 cases, detecting overexpression in 23 (71.8%) and low expression in 9 (28.2%) patients. The response rate for patients with low TS expression was 0.63 compared with 0.15 in patients with overexpression (p=0.015). A significant benefit in time to progression was observed in patients with low expression (median TTP 12 vs. 2 months respectively, p= 0.002), whereas did not impact on overall survival (median OS 20 vs. 19 months respectively, p= 0.595).

Conclusion
TS overexpression in tumor cells correlated with a reduced response to pemetrexed-containing chemotherapy and might be used as a predictive biomarker in advanced lung and mesothelioma cancer patients.

Background
The EGFR mutated status becomes a very important factor for NSCLC patients considering of the treatment, but the mechanism of the mutation is still unknown.Our study aimed to detect the correlations among EGFR mutations and polymorphisms of EGFR and CYP1A1 genes and their associations with clinical outcome of NSCLC patients treated with EGFR-TKI.

Methods
We evaluated the EGFR mutations, the genotypes for EGFR Intron1 (CA) n, R497K and CYP1A1 *2A, *2C polymorphisms in 70 Chinese patients with NSCLC. Genetic polymorphisms were correlated to EGFR mutations. As to subgroup of 36 patients who accepted the EGFR-TKI treatment and had systemic 5 years follow up data, the associations among the somatic EGFR mutations, the genomic polymorphisms of EGFR and CYP1A1 and clinical outcome of the EGFR-TKI were analyzed.

Results
The data show that EGFR Intron1 (CA) n and CYP1A1*2A, *2C polymorphisms were correlated with EGFR mutations (P=0.006, P=0.001, and P=0.008, respectively) and all the three polymorphisms were also associated with EGFR 19 exon delection (P=0.007, P=0.033, and P=0.006, respectively); whereas the multivariate analysis demonstrated that only CYP1A1*2A polymorphism was associated with EGFR somatic mutations (P=0.021). For 36 patients treated with EGFR-TKI, the EGFR mutation and CYP1A1*2A polymorphism showed correlation with clinical response of EGFR-TKI(P=0.001, and P=0.011, respectively); However, the multivariate analysis confirmed that the EGFR mutation is still the most effective predictive factor (P=0.006) ; Either the log-rank test and Cox regression analysis demonstrated that the CYP1A1*2A polymorphism is independent prognostic factor for patients’ overall survival treated with EGFR-TKI( P=0.000 for both statistical analysis).

Conclusion
The results demonstrate that the CYP1A1*2A polymorphism is correlated with EGFR somatic mutation; for advanced NSCLC patients with EGFR-TKI therapy, the EGFR mutation status is still most effective predictor for clinical response of EGFR-TKI, whereas the CYP1A1*2A polymorphism is an independent prognostic factor. The inner mechanisms deserve thorough study.

Background
A large number of studies have evaluated the impact of TP53 mutation on the prognosis of patients with non-small cell lung cancer (NSCLC); however, the results of these studies are still controversial. Recently, considerable intratumor heterogeneity for genetic alterations has been demonstrated in various human cancers, including lung cancer.

Methods
In the present study, we evaluated TP53 mutations in NSCLCs by direct sequencing and observed remarkable variation in the values of the relative intensity (RI, the height of the peak of mutated allele/the height of the peak of non-mutated allele) of TP53 mutation. We also examined whether the RI values were associated with intratumor heterogeneity of TP53 mutation. In addition, we evaluated the relationship between TP53 mutation and survival outcome.

Results
The patients with TP53 mutation did not have significantly worse survival compared to those without the mutation. However, when tumors with TP53 mutation were categorized into two groups, those with a low and those with a high RI, the latter group had significantly worse survival compared to those with wild-type TP53 (adjusted hazard ratio = 2.58, 95% confidence interval = 1.21-5.48, P = 0.01), whereas the former group did not.

Conclusion
These results suggest that intratumor genetic heterogeneity may be an important factor in determining the role of TP53 mutation on the prognosis of NSCLC patients.

Background
Met is a receptor tyrosine kinase that is encoded by c-Met a proto-oncogne. Abnormal c-met expression has been described in several solid tumors most notably in non small cell lung carcinoma and gastric carcinoma where tumor growth and survival is driven by met signaling. Furthermore, met is also a resistance pathway for EGFR tyrosine kinase inhibitors. Met-targeted therapies are currently being tested in clinical trials and met IHC is being used as one of the methods to evaluate met expression. Malignant mesothelioma is an aggressive tumor with limited treatment options and short survival. Met-based therapies could be considered in mesotheliomas with deregulated c-Met expression.

Methods
The study cohort included 39 malignant mesothelioma (Age 15-91 years; Sex: 11 female and 28 male; Histologic type: 1 sarcomatoid, 7 biphasic and 21 epithelioid, primary location: 8 peritoneal and 31 pleural). Immunohistochemistry was performed using an anti-total c-Met rabbit monoclonal antibody (clone SP-44, Ventana Medical Systems, Tucson, AZ, USA) using 4 micron sections of large biopsies or resection specimens and an automated platform (Ventana Benchmark Ultra, Ventana Medical Systems, Tucson, AZ, USA). The staining was evaluated for intensity (none, week, moderate, strong) and extent (percent of tumor staining) and a score was assigned to each tumor on a scale from 0 to 3+ (0: no staining, 1+: any staining in less than 50% of tumor, 2+: moderate to strong staining in >50% of tumor, 3+: strong staining in >50% of tumor). Tumors with and IHC score of 3+ and 2+ were considered positive.

Results
Twenty (20) tumors were scored negative and 19 tumors positive. Strong (3+) staining was seen in 3 epithelioid mesotheliomas (all pleural), 2+ staining was seen in 1 biphasic and 15 epithelioid mesotheliomas, while the remaining 6 biphasic, 1 sarcomatoid and 13 epithelioid mesotheliomas were negative (1+: 19; 0: 1).

Conclusion
c-Met positivity (3+ or 2+) is seen in the majority of epithelioid mesothelomas (18 of 21) and in the minority of biphasic mesotheliomas (1 of 7), while the single sarcomatoid mesothelioma was negative. Since the majority of malignant mesotheliomas are of the epithelioid subtype we expect the majority of malignant mesotheliomas be considered met positive as determined by immunohistochemistry and potentially amenable to met-targeted therapy. Correlation with met amplification and activating c-met mutations needs to be investigated to better understand the mechanisms of c-met over expression in malignant mesotheliomas.

Background
The discovery of various genetic alterations that underlie lung cancer has opened-up a new era in the development of specifically targeted therapies employing specific alteration-dependent inhibitors. In the vast majority of NSCLC patients genetic alterations occur in EGFR (20-25%), BRAF (3%), and ALK (3-10%), seemingly in a predominant mutual exclusive manner. Hence, patient stratification for EGFR, KRAS, BRAF, and ALK alterations is becoming common practice among oncologists. The mutational status of NSCLC patients is considered dynamic and can change in due course of the disease and selection in response to therapy. Longitudinal monitoring of the mutational status of NSCLC patients is of crucial importance to tailor targeted therapy by adapting treatment regimens according to mutational status. However, technical challenges associated with serial tumor biopsy constitute a major challenge in longitudinal molecular monitoring and targeted treatment of NSCLC patients. Blood-based assays may overcome such limitations and allow for frequent assessment of the mutational status.

Methods
Here, tumor tissue, platelets, and/or plasma of NSCLC patients were collected and analyzed for the presence of translocated ALK using FISH, IHC, and/or RT-PCR. Monitoring of EML4-ALK in platelets and plasma was performed by RT-PCR on EML4-ALK positive patients treated with crizotinib and correlated to the clinical response as measured by serial radiographic CT.

Results
From a multicenter cohort of NSCLC patients we identified EML4-ALK positive patients by FISH, IHC, and RT-PCR, for blood platelet and/or plasma isolation. In blood platelets of ALK positive NSCLC patients (n=24) and ALK negative control subjects (n=54) we demonstrated an EML4-ALK test sensitivity of 70-80% and specificity of 100%. We detected EML4-ALK in plasma of ALK positive NSCLC patients (n=22), however with inferior sensitivity of 20-30%. In addition, longitudinal EML4-ALK monitoring in blood platelets of an ALK positive NSCLC patient correlated with the crizotinib response.

Conclusion
We demonstrate here that blood platelets of NSCLC patients are a biosource for the detection of the EML4-ALK translocation and may proof useful for longitudinal monitoring of ALK inhibitors in NSCLC patients, with superior outcome over plasma.

Background
Epidermal growth factor receptor (EGFR) antagonists are therapeutic agents that can be effective in colorectal cancer (CRC) treatment. It has been shown that 40% of CRC tumors have activating KRAS exon 2 codon 12 and 13 mutations and that these mutations are associated with a poor response to EGFR antagonists. Recent studies have shown that mutations in KRAS exons 3 and 4 as well as NRAS exons 2, 3, and 4 also predict poor response to EGFR agonists such panitumumab. Sensitive detection (down to 1%) with Sanger sequencing of such diagnostic biomarkers is necessary to determine the presence or emergence of drug resistant tumor cell populations. Locked Nucleic Acid (LNA) containing oligonucleotides (oligos) have been used in microRNA (sample preparation), RNA (in situ hybridization) and DNA (SNP detection using allele-specific PCR) analysis applications. Incorporation of LNA into oligos has the advantage of increasing the melting temperature of the LNA/complementary template duplex after hybridization as compared to duplexes using standard oligos.

Methods
To improve the mutation detection limits of Sanger sequencing, an LNA-based approach has been developed to cycle sequence the mutant allele selectively in the presence of the wild-type allele. During cycle sequencing, an additional annealing step is added to hybridize the LNA-containing oligo (BLOCker-oligo) to the template DNA. A denaturing step is then performed at a temperature at which the BLOCker-oligo remains annealed to the wild-type sequence while the LNA oligo denatures from the mutant sequence. The sequencing primer then anneals to the mutant sequence and is subsequently extended. Both forward and reverse strand BLOCker oligos can be developed enabling sensitive enrichment for bidirectional sequencing.

Results
To show applicability of this methodology in CRC samples, the limit of detection in both the forward and reverse sequencing directions for multiple codon 12 and codon 13 KRAS exon 2 mutations with and without the addition of the KRAS exon 2 wild-type specific BLOCker-oligo will be demonstrated. To date, the increase in sensitivity is ~10 fold. In addition, a series of FFPE samples will be analyzed for mutations in NRAS and KRAS exons 2 – 4 with and without BLOCker-oligos to show the increase sensitivity of this approach.

Conclusion
BLOCker sequencing is an efficient methodology for the detection of any mutation present in a sample using standard laboratory equipment and Sanger sequencing. The assays being developed at Transgenomic will be offered as CE IVD kits for the detection of all TKI-response associated mutations in KRAS and NRAS Exons 2, 3 and 4.

Background
Malignant pleural mesothelioma (MPM) is an aggressive tumor in the pleural lining of the lung often resulting from prior exposure to asbestos. Median overall survival following frontline chemotherapy with pemetrexed/cisplatin is approximately 12 months. There is no established second line treatment. New therapeutic modalities are urgently needed to improve the prognosis of MPM patients. Approximately 50% of MPM patients exhibit homozygous disruption of the NF2 tumor suppressor gene by mutation and/or deletion resulting in lack of expression of functional merlin protein. VS-6063 is a potent, orally active, inhibitor of focal adhesion kinase (FAK).

Methods
Proliferation of drug-treated mesothelioma cell lines in 3-dimensional (3D) Matrigel culture was assessed by MTS, and orthotopic MPM xenograft growth in the lungs was measured in vehicle- vs. VS-6063-treated SCID mice. Since absence of merlin expression can theoretically result from several mechanisms including NF2 mutation and chromosome 22 abnormalities, we assessed NF2 gene deletion by FISH and merlin protein levels by IHC in the same human mesothelioma tumor samples.

Results
Among a panel of mesothelioma cell lines in 3D culture, MPM lines lacking expression of merlin protein were found to be especially sensitive to the selective FAK inhibitor VS-6063. In contrast, MPM cell lines with wild-type merlin were less sensitive with EC~50~ values greater than 1 μM. Accordingly, oral dosing of VS-6063 induced significant tumor growth inhibition in a merlin-negative mesothelioma model pre-implanted in the lungs of mice. To enable the planned stratification of MPM patients by merlin status, an immunohistochemistry (IHC) assay has been optimized to quantify merlin protein levels. A merlin IHC H-score below the defined cutoff was associated with loss of at least one copy of chromosome 22, indicating that chromosomal deletion is an important mechanism of merlin loss in mesothelioma patients.

Conclusion
These data support the clinical development of VS-6063 for treatment of malignant pleural mesothelioma patients stratified by merlin/NF2 status. VS-6063 is being studied in a registration-directed mesothelioma trial.

Background
Erlotinib (E) is currently indicated as first line therapy in patients with EGFR mutations, and as second and third line treatment for patients with advanced metastatic NSCLC. In Cyprus comprehensive EGFR testing started in January 2011. In this study we reviewed all the patients that received E between 2007-2010, without having undergone EGFR mutation testing, looking at predictive factors of response to E. In view of evidence suggesting that EGFR mutations are found in predominantly TTF1 +ve tumours, TTF1 status as well as clinical factors (i.e. adenocarcinoma histology, female sex and non smoking status) were assessed as potential predictive factors of response to E.

Methods
100 consecutive patients are included. 3 patients received this as 1[st] line, 49 patients as 2[nd] line and 48 patients as 3rd line and beyond. Previous treatments included gemcitabine platinum doublet as first line, and either pemetrexed or docetaxel as 2[nd] line therapy. Patients had regular Chest x-rays (every 2-3 cycles) and CT scans (every 3-4 cycles) and were assessed clinically on a monthly basis. Survival outcomes for both progression free survival (PFS) and overall survival (OS) were calculated with Kaplan Meier. Subset analyses using smoking status, sex, histology type and TTF1 and finally Cox regression was undertaken.

Results
40 female and 60 male were treated with E. Median age is 66 years (range 32-79). 45 patients were WHO performance status (PS) 0-1, 47 PS 2-3. For 8 patients PS was not recorded. Median progression-free survival (PFS) for all patients was 95 days (95% CI 68-118). Median overall survival (OS) from starting E was 144 days (95% CI 103-185). Toxicity was predominantly grade 0-2. Ten (10) patients had a partial response and 34 patients had stable disease. Subset analyses were undertaken based on smoking status, sex, histology type and TTF1. Univariate analysis for PFS using KM plots showed a statistically significant difference for sex, smoking and TTF1. On Cox regression only gender and TTF1 were statistically significant (0.007 and 0.006). There was a striking difference in median OS between patients with TTF1-ve tumours (33 days) and TTF1 +ve tumours (149 days). See table underneath.

	Median PFS (days)	95% CI (days)	Log Rank	Median OS (days)	95% CI (days)	Log Rank
All patients	95	68-118	-------	144	103-185	-------
Female	165	46-284	0.0137	236	45-427	0.1470
Male	80	40-120		104	60-148
< 10 Pack Yrs	203	149-257	0.0009	291	193-389	0.0034
> 10 Pack Yrs	54	15-93		84	53-115
TTF1 – ve	28	11-45	0.0002	33	28-38	0.0001
TTF1 +ve	109	34-184		149	35-263

Conclusion
TTF1 is a better predictor of benefit to E than histology, sex and smoking status. The very low median PFS and OS for patients with TTF1 –ve tumours would suggest that such patients derive no benefit from E, hence TTF1-ve status acts as a negative predictor of benefit to E.

Background
Recent advances in next generation sequencing provide us with the ability to simultaneously analyze both mutation status and copy number changes across a large number of cancer related genes. Here we discuss our experience in developing, validating and applying a 47 gene panel across our lung tumor patient population in a clinical laboratory.

Methods
Genomic DNA was extracted from 20 Formalin Fixed Paraffin embedded (FFPE) tissue samples from lung tumors for validation and over 80 patient samples upon clinical implementation. Extracted DNA was analyzed for quality and amplifiable quantity with suitable performance criteria established during validation. Between 10 and 250 ng of total genomic DNA was then subjected to the Truseq Amplicon (Illumina) assay for enrichment of the target regions across 47 different genes. Libraries were then sequenced on the Miseq (Illumina) to an average depth of coverage between 2000 and 4000x with 186 basepair, paired end reads. Data was then analyzed using analysis pipelines composed of various in house and open source tools, to detect insertions, deletions, amplifications and single nucleotide variants to an allele frequency of 5%.

Results
Across the clinical sample set, 67% of samples were found to have at least one disease associated mutation, 8% had only an unclear variant, 13% were ‘normal’, while 12% had DNA that was not adequate for testing. The average number of mutations seen in samples with disease associated changes was 1.6, with a range from 1 to 4. Half of the samples analyzed were found to have Tp53 mutations, followed by EGFR changes seen in 11%. While only 16 patients had EGFR changes, nearly half (7) of these had co-mutations, including changes at positions 709, 719, 768 and 790. Many of the EGFR changes were complicated insertion/deletions found in exons 19 and 20, which were missed by standard bio-informatic algorithms and only captured by in house tools. KRAS changes were seen in 21% of patients followed by PTEN, MET, STK11 and APC mutations seen in at least 2 patients. Other known somatic mutations were also identified in ATM, BRAF, CTNNB1, FBXW7, FLT3, GNAS, HRAS, NRAS, PIK3CA and ALK in at least one patient. This included the hotspot mutation F1174L in ALK, which is seen with high frequency in neuroblastoma.

Conclusion
The mutation status of many clinically relevant genes can be reliably detected in FFPE samples using a single molecular assay followed by high throughput sequencing. Using this approach known somatic mutations seen frequently in other tumor types can be readily identified in lung tumors, and highlights the future benefits of tumor profiling.

Background
The use of biomarkers for selecting patients with non-small cell lung cancer (NSCLC) for treatment with epidermal growth factor receptor (EGFR) tyrosine kinase inhibitors (TKIs) is essential for achieving a satisfactory therapeutic outcome. EGFR mutations have been found to be predictive of response to EGFR-TKIs. The aim of this study was to explore whether biomarkers which can be identified in plasma, such as EGFR mutations, circulating free DNA, and levels of expressed cytokines are predictive for response to EGFR-TKIs and patient survival time.

Methods
Formalin-fixed and paraffin-embedded biopsies from tumor tissues and paired plasma samples were collected from 134 patients with advanced NSCLC, EGFR mutations in both types of specimens were assessed by an ARMS/Scorpion assay using real-time PCR. Expression levels of transforming growth factor-alpha and beta one (TGF-α and TGF-β1) were assessed using an enzyme-linked immunosorbent assay (ELISA). Concentrations of circulating free DNA were detected in both NSCLC patients and healthy subjects by a colorimetric assay using ultraviolet spectrometry. The clinical significance of EGFR mutations, levels of cytokines, and circulating free DNA was assessed in advanced NSCLC patients treated with EGFR-TKIs.

Results
EGFR somatic mutations were detected in the tumors from 68 of 134 (50.7%) advanced NSCLC patients, and EGFR mutations were detected in the plasma samples from 17 (12.7%) NSCLC patients. Also, the concentrations of circulating free DNA were higher in NSCLC patients than in healthy subjects (P<0.01). EGFR-TKI treatment produced significant effects on progression-free survival (PFS) and overall survival (OS) that were related to the presence of EGFR mutations detected in the tumor tissues (P<0.01).Patients with high levels of TGF-β1 showed shorter OS and worse response to EGFR-TKI treatment than patients with low TGF-β1 levels (P<0.01); however, patients with different expression levels of TGF-α showed no difference in either PFS or OS (P>0.05). Multivariate analysis showed that younger age, adenocarcinoma, never smoking and EGFR somatic mutation were associated with a longer PFS time, and adenocarcinoma, never smoking, low performance status (PS) score, EGFR somatic mutation and low levels of TGF-β1 were associated with greater OS (P<0.05).

Conclusion
Plasma levels of TGF-β1 may be a marker for predicting response to EGFR-TKIs and survival time in NSCLC patients, and levels of circulating free DNA could be a biomarker for differentiating between NSCLC patients and healthy individuals.

Background
Long term health related quality of life (HRQoL) among patients operated for non-small cell lung cancer (NSCLC) has not been extensively studied, as most studies has end-point within 24 months following surgery. Endpoints ike progression-free survival and overall survival, postoperative HRQoL could be more important to the patient.

Methods
A total of 586 patients were operated for NSCLC in the Department of Cardiothoracic surgery of the Helsinki University Hospital between January 2000 and June 2009. Two validated quality of life "questionnaires, the 15D and the EORTC QLQ-C30 with its lung cancer specific module QLQ-LC13, were sent to patients alive in June 2011. Results of the 15D were compared with those of an age- and gender-standardized general population. Patient and treatment features predicting higher or lower long-term HRQoL were identified.

Results
Of the 276 patients who were sent the questionnaires, 230 (83.33%) answered. The median follow-up time was 4.85 years. When compared with the general population (picture), our patient group had significantly lower scores in the 15D total score, representing the total HRQoL of the patients, and in the dimensions Mobility, Breathing, Usual activities, Depression, Distress and Vitality. The patient group scored significantly higher in the dimensions Hearing and Mental function. Features predicting lower long-term HRQoL were comorbidity, measured with the Charlson comorbidity index (CCI), post-operative complications and pre-operative FEV~1~% 70% of the predicted value. Adjuvant-therapy was observed to predict higher long-term HRQoL.Figure 1

Conclusion
NSCLC patients suffer from permanently reduced long-term HRQoL compared to the age- and gender-matched normal population. Factors associated with reduced HRQoL were presence of comorbidity, postoperative complications and reduced FEV1-status preoperatively. The occurrence of complications was also associated with a significantly reduced survival rate. Adjuvant-therapy was associated with a higher HRQoL. Age and gender were not associated with significant differences in the total 15D-score or the QLQ-C30 global health score.

Background
Systematic mediastinal lymphadenectomy is still essential for an adequate postoperative staging of NSCLC. We tried to investigate the controversial role of sentinel node biopsy (SNB) in early stage non small cell lung cancer (NSCLC) surgery

Methods
A total of 52 patients with clinical T1N0MO NSCLC underwent SN navigation lobectomy using Tc-99 labeled tin colloid followed by systematic mediastinal lymphadenectomy (SML) in two years time period (2010-2012). Mapping of the mediastinal lymph nodes by their number and station followed by histopathological evaluation was performed. Patients data were statistically analyzed.

Results
Intraoperative SN was identified in 45 (87%) of these patients with 92% of accuracy. We found lobe specific skip nodal metastases in 5 (10%) patients resulting in upstaging. The incidence of ML metastases seemed to be more often in adenocarcinoma patients (p<0.05), but skip nodal metastases showed higher rate in squamous cell carcinoma patients. Intraoperative frozen section was not confirmed accurate for detecting micrometastases in two (4%) patients. Operative time was prolonged for 10 (8-25) minutes showing no difference in complication rate.

Conclusion
Procedure showed absolute safety and high accuracy. Our results indicated that SN identification could reduce mediastinal lymph node dissection in early stage NSCLC. Further clinical studies should be carried out in order to prove that minimally invasive surgical procedures could be curative for T1N0MO NSCLC.

Background
The extent of mediastinal lymph node assessment during surgery for non-small cell cancer remains controversial. Different techniques are used, ranging from simple visual inspection of the unopened mediastinum to an extended bilateral lymph node dissection. Furthermore, there are different terms to define these techniques: Sampling is the removal of one or more lymph nodes guided by preoperative finding. Systematic (full) nodal dissection is the removal of all mediastinal tissue containing the lymph nodes systematically within anatomical landmarks.

Methods
A Medline search was conducted to identify articles in English, addressing the role of mediastinal lymph nodes resection in the treatment of NSCLC

Results
Opinions favoring full lymphatic dissection include complete resection, improved nodal staging and better local control due to resection of undetected micrometastasis. Arguments against routine full lymphatic dissection are increased morbidity, increased operative time and lack of evidence of improved survival. For complete resection of non-small cell lung cancer a systematic nodal dissection is recommended for many authors, as the standard approach during surgery: it ascertains both adequate nodal staging and completeness of resection.

Conclusion
Whether extending the lymph node dissection influences survival or recurrence rate remains to be determined. There are valuable arguments in favor of not only an improved local control but also an improved long-term survival. However, the impact of lymph node dissection in long-term survival should be further assessed by large-scale multicenter randomized trials.

Background
The use of resections lesser than lobectomy as definitive management of a stage I non-small cell lung carcinoma (NSCLC) is a topic that creates controversy in the global medical community. To describe the current conclusions concerning the relative indications of each type of resection in the surgical treatment of stage I NSCLC, as well as the international results from their application concerning the local recurrence, disease-free survival, and five-year survival rates.

Methods
Thirty four prospective and retrospective studies registered in PubMed and Scopus electronic databases during the last twenty five years were reviewed. Bibliographies and handsearching of journals were used to identify trials. Studies’ authors, citations, objectives, and results were extracted. No meta-analysis was used. Validation of results was discussed.

Results
Segmentectomies were superior to wedge resections in terms of local recurrence and cancer-related survival rates. Sublobar resections were superior to lobectomy concerning preservation of pulmonary parenchyma. It was recommended that high-risk patients undergo segmentectomy. Lobectomies were superior to segmentectomies only for tumors >2 cm (T2bN0M0) as regarding disease-free and overall 5-year survival. There was no significant difference for tumors <2 cm in most studies. Free surgical margins were crucial for local control rates. Systematic lymphadenectomy was mandatory regardless of type of resection. In cases of pure bronchoalveolar carcinoma, segmentectomy was recommended. Shorter hospital stay was achieved with sublobar resections.

Conclusion
The choice of type of resection for T1aN0M0 tumors should rely on specific patient and tumor characteristics. Patient age and tumor size are the most important factors. Further prospective randomized trials are needed to determine minimal resections in early lung cancer patients.

Background
Endobronchial ultrasound-guided transbronchial needle aspiration (EBUS) has been gradually application in clinic, but its value in staging of lung cancer and dignosing of mediastinal mass need to be evaluated. We intends to study comparatively the clinical value of EBUS and mediastinoscopy, and, want to provide a basis for the rational choice applications.

Methods
Between July 2009 and December 2012, 361 patients accepted mediastinoscopy, including 199 cases of lung cancer and 162 of mediastinal mass. During the same period, 348 patients accepted EBUS , including 216 cases of lung cancer and 132 of mediastinal mass. All of the biopsy samples were tested in the pathological department of Shanghai chest hospital. Analyzing the data of two groups and comparing the clinical value of mediastinoscopy and EBUS both on preoperative staging of lung cancer and diagnosing of mediastinal mass.

Results
With the pathological results as a gold diagnosis standard, the accuracy, sensitivity and specificity of mediastinoscopy diagnostic efficacy are 98.28%, 98.03% and 100%, respectively on lung cancer, and, 98.11%, 97.62% and 100%, respectively on mediastinal mass. For EBUS, the accuracy, sensitivity and specificity of diagnostic efficacy are 95.69%, 94.74% and 100%, respectively on lung cancer, and, 82.64%, 77.42% and 100%, respectively on mediastinal mass. Two techniques have not significant difference on diagnosing and staging of lung cancer (P>0.05), but have statistically difference on mediastinal mass (P<0.05), especially on sarcoidosis (P=0.0109), tuberculosis(P=0.0135) and lymphoma(P=0.0036).

Conclusion
Mediastinoscopy and EBUS have a similar clinical value in diagnosing and staging on lung cancer, but mediastinoscopy is superior to EBUS in diagnosing on mediastinal mass.

Background
To determine factors predicting early and long term mortality and complications in patients who underwent a thoracotomy because of primary lung cancer.

Methods
Data of patients who underwent a thoracotomy in the Catharina Hospital Eindhoven between January 1995 and January 2011 have been collected retrospectively from the medical files. Early mortality was defined as mortality <30 days after surgery. Last date of follow up was January 1, 2013. Patients were divided in three groups according to date of surgery (1: 1995-1999, 2: 2000-2004 and 3: 2005-2010). Complications were divided into major and minor complications. Predicting factors were assessed with uni- and multivariate logistic regression analysis. For long-term mortality and survival predicting factors were assessed using the Cox proportional hazards model.

Results
In total 501 patients underwent a thoracotomy due to primary lung cancer. Overall 30 day mortality (early mortality) was 5.6% (n=28). Multivariate analysis showed that age over 70 (p=0.002), pneumonectomy (p=0.008) and a pre-operative VO2max of <15 ml/kg/min (p=0.02) were significant predictors of early mortality. With respect to long term survival, 308 (61%) patients had died at the end of the follow-up period. Median survival time was 44 months, with an overall 5- and 10- year survival of 45% and 27%, respectively. The 5-year survival for stage I, II and III-IV was 61%, 46% and 16%, respectively (p<0.0001, log rank test). Furthermore Cox regression analysis showed that stage (stage I, stage II compared to stage III-IV), FEV1% ≤70%, a history of cerebrovascular disease (CVD) and surgery in an earlier time period (1, 2 compared to 3) were significant predictors of long term mortality. Finally, the only significant predictor for major postoperative complications was a history of COPD (OR 2.32; 95% CI 1.38-3.91).

Conclusion
In this cohort, mortality and complication rates in lung cancer patients after thoracotomy were in line with literature. Significant predictors of early mortality were age, pneumonectomy and pre-operative VO2max. Significant predictors of long term mortality were disease stage, FEV1%, a history of CVD and surgery in an earlier time period.

Background
The incidence of lung cancer in the older population is increasing. In case of resectable primary lung cancer, surgery remains the best treatment for cure, independent of age. However, the prevalence of co-morbidities among elderly lung cancer patients is significantly higher. A presumed fear of increased postoperative morbidity and mortality in the elderly patients has resulted in the delivery of sub-optimal cancer surgery. We believe all elderly cancer patients should be offered optimal treatment depending on their functional status not on their age; a major step in ensuring this would be to establish the appropriate surgical management protocol for elderly cancer patients. Thus it is important to determine whether the elderly would indeed benefit from the same management standards as their younger counterparts. This study aimed to determine the suitable operative procedures for elderly patients with primary lung cancer.

Methods
Between January 2006 and December 2012, 98 patients aged over 75 years with primary lung cancer received lobectomies in our hospital. We divided the patients into group A (75-79 years old) and group B (over80 years old) and analyzed their relapse-free survival and postoperative complication incidence.

Results
The patients in group A (46 men and 17 women) had a mean age of 76.8 years, 36 patients had adenocarcinomas, 24 had squamous cell carcinomas, and 3 had other tumors. A mediastinal lymph node dissection (MLND) was performed in 52 patients but not in 11 patients. The patients’ pathological stages were 1A (21), 1B (13), 2A (5), 2B (12), 3A (11), and 4 (1). The 3-year relapse-free survival was 66.3 % (MLND-positive, 62.6 %; MLND-negative, 90.9 %). Postoperative complications occurred in 30.8 % of the MLND-positive patients and 18.2 % of the MLND-negative patients. The patients in group B (26 men and 9 women) had a mean age of 82.4 years, 17 patients had adenocarcinomas, 12 had squamous cell carcinomas, and 6 had other tumors. An MLND was performed in 24 patients but not in 11 patients. The patients’ pathological stages were 1A (12), 1B (10), 2A (5), 2B (3), 3A (1), and 3B (4). The 3-year relapse-free survival was 73.3 % (MLND-positive, 75.7 %; MLND-negative, 68.2 %). Postoperative complications occurred in 16.7 % of the MLND-positive patients and 9.1 % of the MLND-negative patients.

Conclusion
This study showed a survival benefit in elderly lung cancer patients who underwent lobectomies. In clinically well-documented early nodal stage disease (N0 patients and N1 patients with limited hilar disease), an MLND did not have a therapeutic effect and thus may not be necessary. Omitting MLNDs in elderly lung cancer patients also provided fewer postoperative complications. In this study, we proved that elderly patients with resectable lung cancer who received lobectomies without MLNDs could achieve long-term survival and be a safe procedure.

Background
The term "ground glass opacity (GGO)" on high-resolution computed tomography (HRCT) is defined as “hazy increased attenuation in the lung that does not obliterate the bronchial and vascular margins” by Fleischner Society. The identification of small lung nodules of GGO on HRCT often implies lung cancer, especially well differentiated adenocarcinoma or atypical adenomatous hyperplasia; however, there is no objective definition of GGO, such as the computed tomography number.

Methods
A single institutional retrospective study. To access the correlation between radiological and pathological diagnosis of the patients with small pure GGO on HRCT. Thirty-nine consecutive surgically resected patients with pure GGO less than 30 mm detected by HRCT between July 2008 and March 2013 in our department were retrospectively examined. The median follow-up of these patients was 28.7 (1.9 - 92.7) months.

Results
The median age of the patients was 64 (range 42-82) years old, 19 patients were male and 20 were female. The median size of major axis of lung nodules was 11 (range 5-25) mm, and 29 (74.4%) were less than 15 mm and 10 were between 15 and 30 mm in diameter. Twenty-eight (71.8%) patients had a single nodule, whereas 11 patients had multiple nodules. Six of the 39 patients had a previous history of malignancy (three lung cancers and three other cancers). During the follow-up period, 22 patients had nodules that were stable in size or appearance, and five patients had nodules that either became enlarged or in which the opacity increased, as determined by HRCT. The other twelve patients were operated based on the findings of their first HRCT, basically by the attending surgeons’ decision. Partial resection was performed in seven patients, segmentectomy in 11 patients and lobectomy was performed in 21 patients. Histologically, thirty-seven patients had adenocarcinoma, one had small cell carcinoma and one had a benign tumor. Among the 37 patients with adenocarcinoma, 14 were adenocarcinoma in situ, five tumors were minimally invasive and 18 were invasive according to the IASLC/ATS/ERS classification. There was no postoperative recurrence during the follow-up period.

Conclusion
Even if the small pulmonary nodules present as pure GGO, they may still be adenocarcinoma with an invasive nature. The timing of surgery should be considered carefully so that a chance to achieve a cure of such patients is not missed.

Background
Superior sulcus tumors (SST) comprise a subgroup of non-small-cell lung cancers that arise near the pulmonary apex or superior sulcus. They generally invade the chest wall and brachial plexus, and occasionally the subclavian vessels. Induction chemoradiation therapy followed by surgery is the recommended treatment for SST. However, surgical approaches for SST remain controversial, partly because of their infrequent use. Several approaches to resecting these tumors have been described, depending on the precise localization and involvement of the surrounding organs. These include posterolateral thoracotomy, hemi-clamshell and transmanubrial osteomuscular-sparing approaches. It is necessary to establish the appropriate multimodality therapy for SST, including the optimal surgical approaches.

Methods
We retrospectively analyzed the clinical courses of patients with SST treated with surgery at our institution. A total of 2765 patients with non-small-cell lung cancer were treated surgically at Osaka City General Hospital, Japan, from January 1995 to December 2012. Among these, 34 patients with SST were investigated in this study.

Results
The mean age of the patients was 62 years (range, 42–90 years). There were 32 men and two women. Seventeen patients had squamous cell carcinoma, 12 had adenocarcinoma, and five patients had tumors of other histological types. There were 21 patients with stage 2B, 10 with stage 3A, and three patients with stage 3B disease. Two patients received induction chemotherapy, and 22 patients received induction chemoradiotherapy. Posterolateral thoracotomy was performed in 11 patients and anterior thoracotomy (hemi-clamshell, transmanubrial osteomuscular-sparing approaches) in 22 patients. A combination of anterior and lateral thoracotomies was applied in one patient. Pulmonary lobectomy was performed in 25 patients, segmentectomy in one patient, and pulmonary partial resection in nine patients. The resected surrounding organs, other than the chest wall, were the subclavian artery in two patients, the superior vena cava in two, and the aortic arch and vertebral body in two patients each. The median follow-up period was 16 months (range, 3–154 months). Postoperative 1-, 3- and 5-year survival rates were 72%, 46%, and 34%, respectively. Investigation of clinicopathological factors with potential impacts on postoperative outcome identified pathological nodal extension as the only significant factor indicating poor prognosis (p < 0.01). Tumor markers, surgical approach, type of pulmonary resection, and type of resected surrounding organ had no effect on postoperative outcome. No viable tumor was observed in seven of 22 patients treated with induction chemoradiotherapy, and the postoperative 5-year survival rate in these seven patients was 86%. Recurrent disease was observed in 17 patients during the postoperative follow-up period. Local recurrence was observed in five patients and recurrence in distant organs was observed in 13 patients.

Conclusion
Patients with node-positive SST have a poor prognosis, and surgical indications should be investigated fully in these patients. Induction chemoradiotherapy is necessary to treat SST. The major sites of recurrence are in distant organs, and the type of pulmonary resection does not affect postoperative outcome. Partial resection may be an acceptable option in patients with no detectable viable tumor after induction chemoradiotherapy.

Background
Recently, multiple primary lung cancer is inreasing with progress of imaging diagnostic technology and improvement in treatment results. We assesed the selection of the type of pulmonary resection, operative morbidity, mortality and the outcome of our cases which enforced surgical treatment for multiple primary lung cancer.

Methods
Of 840 patients who underwent operation for primary lung cancer between January 2001 and May 2013.We consider 57(6.7%) to have a multiple primary lung cancer. We use the criteria of Martini and Melamed as a diagnosis of multiple primary lung cancer.

Results
The group comprised 27 men and 30 women. Median age was 66.7 years (48-80). It occured synchronous in 21 cases and metachronous in 36 cases. Cell type combination was adenocarcinoma-adenocarcinoma in 37 cases, adenocarcinoma-squamous cell carcinoma in 10 cases, squamous cell carcinoma-squamous cell carcinoma in 6 cases, adenocarcinoma-large cell carcinoma in 3 cases and large cell carcinoma-large cell carcinoma in 1 case. Operative procedures was lobectomy-lobectomy in 13 cases, lobectomy-segmentectomy in 12 cases, lobectomy-edge resectionin 19 cases, edge resection-edge resection in 5 cases and lobectomy/edge resection-radio therapy in 3 cases. Pathological stage for the first cancer was IA/IB/IIA/IIB/III = 35/14/4/0/4. The 5 year survival rate after first surgery was 81.7%, and second surgery was 72.2%.

Conclusion
With this result, we consider that surgical resection may prove beneficial in cases which postoperative good survival can be expected by the aggressive surgical approach, despite it is difficult to distinguish multiple primary lung cancers and metastatic cancers preoperatively.

Background
Postoperative acute respiratory distress syndrome (ARDS) is a recognized complication of pulmonary resection. ARDS following lung resection has a miserable prognosis, with overall hospital mortality rates over 25%. Previous studies demonstrated that there were risk factors of ARDS after pulmonary resection including age, chronic obstructive pulmonary disease (COPD), interstitial pneumonia, concurrent cardiac disease, prior therapy, remaining lung perfusion, duration of operation, increased blood loss and so on. Neutrophils and neutrophil elastase (NE) are believed to play a key role in the endothelial injury and increased vascular permeability characteristic of ARDS. Sivelestat sodium hydrate is a selective NE inhibitor and has been shown to improve respiratory status in cases of ARDS. It has not been well known whether or not NE inhibitors are beneficial for prevention of ARDS after lung resection.

Methods
We conducted a pilot study to investigate the efficacy of perioperative administration of sivelestat sodium hydrate to prevent postoperative ARDS in 34 non-small cell lung cancer (NSCLC) patients who had the various preoperative risk factors of the incidence of ARDS after pulmonary resection in Chiba University between 2009 and 2011. They received sivelestat sodium hydrate (5mg/kg/day) intravenously for 7 days starting at the beginning of operation.

Results
The patient demographics were as follows: median age, 68 years of age (range 47 to 83 years), male/female ratio, 31/3, clinical stage I/II/III, 9/6/19. The histology was adenocarcinoma (n =19), squamous cell carcinoma (n=10) and others (n =5). Risk factors of ARDS included induction chemotherapy (n=3), induction chemoradiotherapy (n=17), interstitial pneumonia (n=10), COPD (n=3) and medical history of ARDS (n=1). All 34 patients underwent complete resection. The operations included 2 partial resections, 31 lobectomies, and 1 pneumonectomy. Of the 31 patients who received lobectomy, bronchial or arterial plasty was performed in 9 patients. The postoperative mortality rate was 2.9%. One patient died of heart failure on the nineth postoperative day. There was no incidence of ARDS after pulmonary resection in all patients.

Conclusion
Perioperative administration of sivelestat sodium hydrate can be beneficial to prevent postoperative ARDS in NSCLC patients. Prospective studies are required.

Background
In 2005, we reported the study of intentional limited resection for small peripheral lung cancer based on intraoperative pathologic exploration. At that time, only 14 patients with a small adenocarcinoma showing ground-glass opacity (GGO) had undergone limited resection. After that, we have continued limited resection and follow-up. The median follow-up time from the operation has reached 80 months, so we analyze the long-term results of this procedure.

Methods
Between 1996 and 2013, we enrolled 56 patients in this study. Entry criteria were: 1) cT1aN0M0 peripheral adenocarcinoma, 2) High resolution computed tomography (HRCT) findings suspected of having a Noguchi type A or B adenocarcinoma, and 3) pulmonary function adequate to permit lobectomy. When the tumor consisted of GGO only or GGO with a solid component that accounted for less than 50% of the surface area on HRCT, the tumor was suspected to be Noguchi type A or B adenocarcinoma. Wedge resection or segmentectomy was performed, and was followed by an intraoperative pathologic exploration. After confirming the diagnosis of Noguchi type A or B by intraoperative pathologic exploration, operation was completed. No systematic lymph node dissection or sampling was performed. If the lesion was not Noguchi type A or B, extended segmentectomy or lobectomy with systematic lymph node dissection was performed instead.

Results
Between 1996 and 2013, we enrolled 56 patients in this study. Limited resection was performed in all patients, wedge resection in 52, and segmentectomy in 4. Intraoperative pathologic exploration revealed that the lesion was not Noguchi type A or B in 11 patients. In these 11patients, we underwent extended segmentectomy in 2 and lobectomy in 9 with systematic lymph node dissection. Intentional limited resection was completed in 45 patients. Of these, 7 had Noguchi type A tumors, and 38 had Noguchi Type B tumors based on intraoperative pathologic exploration. Postoperative pathologic examination revealed 10 patients with Noguchi type A, 31 patients with Noguchi type B, and 4 patients with Noguchi type C. We recommended reoperation to 4 patients with Noguchi type C, but all refused reoperation and has been carefully followed at 3-month intervals. There was no postoperative and 30-day mortality or in-hospital mortality. There were no morbidities. All patients but one are alive without recurrence of lung cancer at the time of writing. Only one patient died of malignant lymphoma without recurrence. The follow-up periods have ranged from 6 to 195 months, and median follow-up period is 80 months. The overall 5- and 10-year survival rates were 96% and 96%, respectively. The 5- and 10-year recurrence –free proportions were 100% and 100%, respectively.

Conclusion
When patients are carefully selected by preoperative HRCT and intraoperative pathologic exploration, limited resection can be an acceptable option for the treatment of T1aN0M0 adenocarcinoma showing GGO lesion.

Background
N1 disease is a subset of non-small cell lung cancer (NSCLC) with a different prognosis than other subsets. Although the NCCN guideline recommends adjuvant chemotherapy alone in completely resected N1 disease, locoregional failure, which could have been prevented by postoperative radiotherapy (PORT), may not be uncommon. Using PORT with modern techniques has resulted in significantly higher rates of local control, disease-free survival, and overall survival. Therefore, this study aimed to evaluate the actuarial rates of locoregional failure in patients with pathologic N1 NSCLC and to identify the risk factors associated with an increased risk of locoregional failure, which could have been potentially prevented by PORT after complete resection.

Methods
Between 2003 and 2010, we enrolled 136 patients who underwent complete resection with pathologically confirmed N1 disease through the prospective lung cancer database of Seoul National University Bundang Hospital. Patients who underwent neoadjuvant therapy, adjuvant radiotherapy, or operative mortality were excluded. Multiple factors potentially related to outcomes including patient-related factors, surgery-related factors, and pathologic factors were extensively evaluated. Locoregional failure, which could have been potentially prevented by PORT, was defined as recurrence at either a bronchial stump, or a resected margin of the lung, hilum, and mediastinum. Other failures were ipsilateral lung recurrence, pleural seeding, and metastasis of distant organs. Univariate analysis by a log rank test and multivariate analysis by the Cox proportional hazards model were performed to identify risk factors independently associated with a higher risk of locoregional failure.

Results
The median follow-up duration was 45 months (6-114) and recurrence developed in 54 (40%) patients. The actuarial 5-year rates of disease-free survival and overall survival were 56% and 66%, respectively. From the perspective of first site recurrence, 23 (17%) locoregional failures, which could have been potentially prevented by PORT, included recurrence in the mediastinum in 10, bronchial stump in 5, regional lymph nodes in 4, and mediastinum + others in 4. The 31 (23%) other failures included a distant organ in 17, ipsilateral lung in 12, and pleural seeding in 2. The median survival time from locoregional failure and other failures was 41 and 57 months, respectively; however, there was no significant difference. Risk factors of locoregional failure were squamous cell carcinoma, number of involved node (>1), pathologic stage (IIIA), interlobar node involvement, more than 2 node stations of involvement, and a lymph node ratio greater than 10% by univariate analysis. Pathologic stage (HR=4.768, 95% CI=1.641-13.859, p=0.01), interlobar node involvement (HR=2.783, 95% CI=1.057-7.327, p=0.04), and squamous cell carcinoma (HR=2.449, 95% CI=0.929-6.454, p=0.07) were independent risk factors by multivariate analysis.

Conclusion
Locoregional failure was more common than expected, and pathologic stage, interlobar node involvement, and cell type were independent risk factors for locoregional failure after complete resection of N1 NSCLC. A prospective clinical trial may be necessary to evaluate the effectiveness of adjuvant radiotherapy in patients with these risk factors.

Background
Malignant pleural effusion or pleural seeding is detected in advanced non-small cell lung cancer (NSCLC) patients, and they are generally associated with poor prognosis. Systemic chemotherapy is the mainstay modality in these patients. However, it is not enough to improve the survival. Intrapleural perfusion hyperthermic chemotherapy (IPHC) provides direct effect to pleural seeding cancer cells. This study attempted to evaluate the efficacy and safety of IPHC.

Methods
From 2003 to 2012, 41 patients who underwent IPHC for malignant pleural effusion or pleural seeding for NSCLC in our institute. The IPHC was performed with cisplatin (dose:150-200mg/m[2]) for 90 minutes after resection of primary tumor. Efficacy was determined by computed tomography and Positron Emission Tomographic (PET) standardized uptake value (SUV) postoperatively.

Results
The IPHC group consisted of 25 males and 16 females. The mean age was 60.98±9.80 year ranging from 33 to 78. Preoperative pleural mean SUV was 1.46±1.88 (R: 0-6) and postoperative pleural mean SUV was 1.58±1.72 (R: 0-5) (p= 0.933). Sixteen patients were received adjuvant systemic chemotherapy. The 2-year and 5-year survival rate were 82.1% and 44.9%, respectively. Major post-IPHC complications are acute renal insufficiency (n=4, 9.76%) and arrhythmia (n=2, 4.88%). There were no difference in sex, age, adjuvant systemic chemotherapy, tumor size, nodal statues between the patients who survived more than 2 years and less than 2 years. There was no difference in SUV of preoperative main mass and pleura between the patients who survived more than 2 years and less than 2 years. However, the SUV of postoperative pleura in the patients who survived more than 2 years (SUV: 2.92±1.98) was less than that of the patients who survived less than 2 years (SUV: 1.16±1.45) (p=0.031).

Conclusion
IPHC would be safety procedure for malignant pleural effusion or pleural seeding. IPHC may provide better survival compared with the systemic chemotherapy only in the highly selected patients. Low post-IPHC SUV uptake would be provide longer survivor.

Background
In Japan, an aging society, the number of surgical operations performed in the elderly has been increasing. In this study, to investigate whether the surgical resection of non-small cell lung cancer benefits extremely elder patients (85-year-old or over), we analyzed the clinical outcome of our patients.

Methods
Sixteen consecutive patients aged 85 years or older who underwent surgical resection of primary non-small cell lung cancer in our hospital from May 2002 to September 2012 were enrolled in this study. The patients’ operative procedure, respiratory function before operation, histological type, tumor size, clinical stage, comorbidity, surgery-related complications, prognosis, and recurrence were retrospectively reviewed.

Results
There were 11 males and 5 females. Their mean age was 86.6 years (range: 85-93 years). The mean follow-up period after operation was 1290 days (range: 249-4029 days). Operative procedures include lobectomy (N=4), segmentectomy (N=6), and wedge resection (N=6). Among them, thoracoscopic surgeries were performed in 14 patients. Patients treated by segmentectomy had poorer pulmonary function than others in terms of the forced expiratory volume in 1 second (FEV1.0) (mean±SD: 1.43±0.40 vs 1.94±0.39 L, p<0.05) and vital capacity (VC) (mean±SD: 2.02±0.51 vs 2.55±0.53 L, p=0.06). Histological type include adenocarcinoma (N=11) and squamous cell carcinoma (N=5). The tumor size ranged 11 mm to 48 mm. All 16 patients were negative for lymph node metastasis. Pathological stage was IA in 11 patients, IB in 4 patients, and IIB in 1 patient. Seven patients had comorbidity (e.g. COPD, post-bypass grafting for coronary aretery, and others). Seven patients were smokers. Mortality rate was 0% and morbidity rate was 25% (e.g. pulmonary fistula treated with pleurodesis in 1 patient, atrial fibrillation in 2 patients, heart failure in 1 patient, and delirium in 1 patient). Fourteen patients are alive, whereas 2 patients died from other diseases than lung cancer. The 2 patients died after 1950 days and 1344 days after operation, respectively. No patients had recurrence during the follow-up period.

Conclusion
No patients died from surgery or post-surgical complications. Our study indicates that the surgical resection of non-small cell lung cancer benefits patients, even with extremely elder age (85-year-old or over). For our institution’s principal indications for pulmonary surgery, patients with elder age (75-years-old or older) have to (i) be able to undergo pulmonary physiotherapy, (ii) have good performance status (0-1 in ECOG scale), and (iii) be free from psychological disorder such as dementia. Additionally, in cases with extremely older age (85-year-old or over), we consider patients with clinical stage I or patients without lymph node metastasis are good candidates. Lobectomy can be performed in patients with better pulmonary function, whereas patients with worse pulmonary function should undergo segmentectomy. Wedge resection is selected only if the lesion is judged to be completely resectable. We consider these inclusion criteria and the manner of selecting operative procedures will provide a good prognosis, even though the patient is 85 years or older.

Background
Induction chemoradiotherapy (CRT) followed by surgery (iCRT) is one of treatment strategies for locally advanced non-small cell lung cancers (NSCLCs). We have previously reported its feasibility and good clinical outcome with approximately 60% of 5-year overall survival rate. .Perioperative nutritional status,is considered as one of important factors for improved clinical outcome after surgery and other treatments. Here, we investigated the perioperative nutritional status in NSCLC patients treated by iCRT (CRT group) to evaluate the influence of nutritional variables on clinical outcome by comparing that in NSCLC patients with simple pulmonary resection without CRT (non-CRT group) .

Methods
Thirty-three consecutive patients with locally adcanced NSCLC who underwent iCRT from January 1, 2009, until December 31, 2011 at our institute were included in this study. The regimen of CRT was two cycles of docetaxel (40 mg/m[2]) plus cisplatin (40 mg/m[2]) with concurrent radiotherapy (46 gray) and the surgery was performed within 6 weeks of completing induction CRT. We compared nutrition-related factors and clinical outcome in 33 iCRT patients with those in 58 consecutive NSCLC patients who underwent lobectomy during January 1 to December 31, 2011 at out institute. .As for blood nutritional factors, total lymphocyte count (TLC), albumin (Alb), total cholesterol (T-cho), choline esterase (ChE), were examined. The prognostic nutritional index (PNI) was also calculated by Alb and TLC. Each nutrition-relatd factors were examined 1) before CRT, 2) before surgery and 3) one month after surgery.

Results
Median age of CRT group (61 years old) was significantly younger than that of non-CRT group (69 years old). Twenty-one males and 12 females and 44 males and 14 females were enrolled in CRT and non-CRT groups, respectively. Before any treatment, no significant difference was observed in body mass index and any blood nutritional factors in both groups. After induction CRT, TLC was significantly decreased, and additionally, Alb, T-cho, and ChE were significantly decreased after surgery comparing with those before surgery (after CRT). As for preoperative status in both groups, TLC, Alb and PNI were significantly lower in CRT group than in non-CRT group. Regard with surgery, extended surgery, operating time, and blood loss was significantly heavier in CRT group than in non-CRT group. Perioperative mortality rate was 0% in both groups and the frequency of post-operative complication was similar in both groups (51% and 41% in CRT and non-CRT groups, respectively). The length of hospital stay after surgery was significantly longer in CRT group (median 23 days) than in non-CRT group (median 14 days). Among CRT group, patients with loiw PNI index could not administrate adjuvant chemotherapy.

Conclusion
Perioperative nutritional status, especially TLC, is suppressed after CRT and moreover after surgery. Suppression of nutritional status continued one month after surgery with induction CRT and severe suppression of nutritional status disturbs further treatment such as adjuvant chemotherapy.

Background
Patients with cerebro- and cardio-vascular comorbidities (CCVC) who undergo surgery represent a high-risk group and require careful perioperative management. In the present study, we aimed to retrospectively analyze the postoperative complications (POC) of patients with CCVC who had undergone pulmonary resection for lung cancer. Patients with cerebro- and cardio-vascular comorbidities (CCVC) who undergo surgery represent a high-risk group and require careful perioperative management. In the present study, we aimed to retrospectively analyze the postoperative complications (POC) of patients with CCVC who had undergone pulmonary resection for lung cancer.

Methods
Among 288 patients who underwent pulmonary resection at our institution from January 2009 to December 2011, we examined the records of 51 patients with CCVC (17.7%) to identify the risk factors for developing POC. Among the analyzed patients, we noted the presence of 34 POC, including tachyarrhythmia in 9, prolonged pulmonary fistula in 9, pyothorax in 2, cerebral infarction in 2, requirement of long-term oxygen therapy in 2, interstitial pneumonia in 2, delirium in 2, and other POC in 4. Several patients had multiple POC.

Results
We examined 43 male patients (84.3%); the median age was 72 years and the median preoperative forced expired volume in 1s (FEV~1~) was 2200 mL (range, 1120–3420). The patients with CCVC included 12 with cerebral infarction, 2 with transient cerebral ischemic attacks, 2 with cerebral hemorrhage, 1 with subarachnoid hemorrhage, 4 with cerebral aneurysm, 10 with arrhythmia, 17 with ischemic heart disease, 1 with valvular heart disease, 8 with aortic aneurysm/dissection, 11 with peripheral arterial disease, and 1 with a left atrial myxoma; several of these patients had multiple CCVC. Moreover, 2 patients underwent pneumonectomy, 37 underwent lobectomy, 3 underwent segmentectomy, and 9 underwent wedge resection. Postoperative morbidity rates were 21.4% in cerebrovascular comorbidity patients (p = 0.015), 53.5% in the cardiovascular comorbidity patients (p < 0.0001), 71.4% in CCVC patients (p = 0.0028), and 12.3% in patients without CCVC. No operative or in-hospital mortality was noted. Gender, age, smoking status, and smoking index were not found to be significantly related to the incidence of POC. However, patients with an FEV~1~ < 2200 mL were found to be significantly more likely to develop POC (p = 0.036).

Conclusion
We noted that patients with CCVC and low FEV~1 ~were more likely to develop POC.

Background
Multidisciplinary team (MDT) meetings have developed to enhance disease-specific interdisciplinary communication towards accelerating the initiation of appropriate therapy and improving both patient care and survival. They have long been a mainstay in the management of patients with cancer. This study aims to compare resection rates, surgical cure (based on post-resection pathological status) and co-morbidity status in consecutive lung cancer cases (stages IA-IIB, IASLC TNM staging 6[th] or 7[th] ed.), as presented at our institution’s Lung Cancer MDT meeting, compared with a similar contemporaneous control group not presented at the MDT meeting.

Methods
Data for this study was drawn from the Clinical Cancer Registry Project (ClinCR) database in NSW, which captures all cases of lung cancer managed in the public health system as well as databases used by the Lung Cancer MDT and Cardiothoracic Surgery. Cases of non-small cell lung cancer (NSCLC) appropriate for curative surgery based on pre-operative stage IA-IIB were analysed for MDT presentation, resection rate, surgical cure rate and for pre-operative Charlson co-morbidity score.

Results
Data were analysed from the time period 2006-2011. During this time 755 cases of NSCLC were diagnosed, 377 were presented at MDT (49.9%) and 378 were not (50.1%). A total of 138 cases of early stage disease (stages IA-IIB) were identified, 90 of which were presented at MDT (65.2%) and 48 were not (34.8%). Significantly more cases of early stage disease presented at MDT, 33/90 (36.7%) had resection than cases of early stage disease not presented at MDT 13/48 (27.1%) (p=0.015). Overall resection rate in early stage disease (MDT and non-MDT) was 46/138 (33.3%). In resected cases the surgical cure rate (according to surgical margins) did not differ significantly between the two groups with 30/33 (90.9%) in the MDT group and 12/13 (92.3%) in the non-MDT group achieving surgical cure (p=0.75). For cases of early stage NSCLC presented at MDT the odds ratio for surgical resection was 2.5. Further analysis of Charlson co-morbidity score will clarify associated co-morbidities, which may help explain this difference.

Conclusion
Surgical resection rates are statistically higher in early stage NSCLC cases presented at MDT than in contemporaneous cases not presented at MDT in our study population. Surgical cure rates for resected patients appear high in both groups. Further data on co-morbidities and identifiable referral patterns, including patients managed outside the public hospital system, may clarify reasons for the disparity in resection rates and enable identification of potentially reversible barriers to treatment.

Background
Although large-cell neuroendocrine carcinoma (LCNEC) was categorized as a variant of large cell carcinoma on the WHO histologic classification of lung carcinomas, the clinical and biological features of LCNEC resemble those of small cell lung carcinoma. Therefore, there is no consensus on the treatment strategy for LCNEC, and an indication of surgical treatment for LCNEC is still controversial. Even though preoperative accurate diagnosis of LCNEC is difficult, the aim of this study was investigating patients with pulmonary LCNEC in whom better postoperative outcome is expected.

Methods
We retrospectively reviewed patients with pulmonary LCNEC on permanent pathologic diagnosis who underwent pulmonary resection at the 3 institutions between 1999 and 2011. We reviewed the medical records of each patient for demographic, clinical, and pathologic data including age, sex, smoking status, preoperative serum CEA, radiologic tumor size, c-stage, surgical procedure, extent of lymphadenectomy, p-stage, lymph node metastasis, visceral pleural invasion, lymphatic permeation, vascular invasion, and adjuvant chemotherapy. Disease-free survival (DFS) was calculated using the Kaplan-Meier method, and factors associated with DFS were analyzed with the log-rank test.

Results
Of the 18 patients eligible for this study, 14 were male and 4 were female. The median age was 74 years (range, 53 to 85). According to the current TNM classification, 12 patients had c-stage I disease, 4 had c-stage II disease, and 2 had c-stage IIIA disease. The majority of patients (13 patients, 72%) underwent lobectomy, 1 underwent pneumonectomy, 1 underwent bilobectomy, and 3 underwent wedge resection. On pathologic diagnosis, 8 patients had p-stage I disease, 5 had p-stage II disease, and 5 had p-stage IIIA disease. Following surgical treatment, cisplatin-based adjuvant chemotherapy was applied for 3 patients. The 1-year and 2-year DFS were 39% and 39%, respectively, with the median follow-up period of 9 months (range, 2 to 80). During the follow-up period, 10 patients (56%) developed recurrence, and the recurrence was identified within the first year post-resection in all the 10 patients. By the log-rank test, smoking status (non- or former, vs. current) and surgical procedure (lobectomy or greater, vs. limited resection) were identified as significant factors associated with DFS.Figure 1

Conclusion
Of patients with pulmonary LCNEC undergoing surgical treatment, a long-term prognosis might be expected if no recurrence is identified within the first year post-resection. If diagnosis of LCNEC is preoperatively obtained, surgical treatment is recommended for patients without current smoking status, and lobectomy or greater resection should be the surgical procedure of first choice.

Background
Although positron emission tomography/computed tomography (PET/CT) seems to be able to provide accurate information of lymph nodes status in non-small cell lung cancer (NSCLC) patients, we sometimes experience surgical cases with unexpected lymph nodes metastasis. The objective of this study is to demonstrate clinicopathological characteristics of patients with unexpected pathological (p) N1 and N2 NSCLC.

Methods
All patients with lung cancer underwent enhanced CT of the chest and PET/CT preoperatively for evaluating of lymph node status. Mediastinoscopy or EBUS-TBNA was not routinely performed. Unexpected pN1 and pN2 diseases were defined as surgical cases which were proved to have hilar or mediastinal lymph nodes metastasis postoperatively in spite of negative 18-fluoro-2-deoxy-D-glucose (FDG) uptake in hilar or mediastinal lymph nodes on preoperative PET/CT. We retrospectively reviewed clinical features of these patients and analyze predictive factors for these unexpected diseases.

Results
Between January 2007 and December 2012, 533 patients with lung cancer underwent　surgical resection at our institution. Among them, we reviewed 182 patients who had available preoperative PET/CT data and underwent a curative-intent operation with systematic or selective lymph node dissection. One hundred fifty-one patients (83%) had lung cancer with expected lymph node status, whereas, 31 patients (17%) were found to have lung cancer with unexpected lymph nodes metastasis, consisting of 12 (39%) unexpected pN1 and 19 (61%) unexpected pN2 diseases. There were 16 men and 15 women with a median age of 67 years. Seventeen patients were current- or past- smokers, and 14 were never-smokers. Tumor size ranged from 12 to 52 mm, and pathological T factor was pT1a in 3, pT1b in 5, pT2a in 19, pT2b in 3, and pT3 in 1. Histological type of the primary tumor were adenocarcinoma in 28 (90%), squamous cell carcinoma in 2, and large cell carcinoma in 1. Among 28 adenocarcinomas, the most common predominant subtype was papillary (20 patients; 71%), followed by acinar (5 patients), solid with mucin, micropapillary, and invasive mucinous (one patient, respectively). Of these patients, 7 patients had micropapillary component in varying proportions in their tumors. EGFR gene mutation status was available in 22 patients, and of these, 12 patients (55%) had tumors with EGFR gene mutation, consisting of exon 19 deletion in 5, exon 21 point mutation in 5, and other minor mutation in 1. In univariate analysis, tumor size (> 3cm), pleural, and lymphatic invasion were significant predictive factors for unexpected pN1 and pN2 diseases. Only in 151 adenocarcinomas, papillary predomiant tumor and having micropapillary component were significant predictive factors for unexpected lymph node metastasis.

Conclusion
We should take care that false-negative rate is relatively high on preoperative PET/CT for lymph node status in NSCLC. Histological findings of the primary tumor are often important because they can provide predictive information for lymph nodes status even if there is no FDG uptake in regional lymph nodes on preoperative PET/CT. We hope new accurate imaging modality which can reflect tumor histology and lymph nodes micrometastasis.

Background
The average age of the general population is increasing in the worldwide. Therefore, there is also an increasing number of elderly patients presenting with potentially-resectable lung cancer. We retrospectively reviewed the outcomes of octogenarians or over who underwent pulmonary resection for primary non-small cell lung cancer (NSCLC) to identify the independent factor of overall survival.

Methods
We conducted a retrospective single-institution review of patients aged 80 years or over who underwent pulmonary resection for primary NSCLC from 1990 to 2012 at Nagasaki university hospital. The various clinicopathological data, including gender, histological type, body mass index, comorbidity, clinical stage, surgical procedure, extent of lymph node dissection, and pathological stage were analyzed.

Results
119 octogenarians or over underwent pulmonary resection. The median age was 82 years (range 80-92 years). Of the total patient number, 56 (47.1%) had respiratory and 44 (33.6%) had cardiovascular comorbidity diagnosed preoperatively. The clinical stage was I in 97 (81.5%) patients, II in 13, III in 6, IV in 3. Operations included 82 (68.9%) lobectomies, 2 (1.7%) bilobectomies, 15 (12.6%) segmentectomies, and 19 (16.0%) partial resections. Only 31 (26.1%) were performed mediastinal lymph node dissection. The pathological stage was I in 79 (81.5 down to 66.4 %) patients, II in 16 (13.4%), III in 21 (17.7%), IV in 3 (2.5%). 26 (21.8%) patients presented with postoperative respiratory complications, and 11 (9.2%) were cardiovascular, and the operative mortality was 1 (0.8%). The 5-year survival rates were 46.0% for all patients, 60.8% for stage I patients. The disease specific 5-year survival rates were 60.1% for all patients, 79.5% for stage I patients, respectively. In univariate analysis, female (p<0.04), clinical stage (p<0.002), and pathological stage (p<0.000) was independent and cardiovascular comorbidity was marginally (p<0.05) factor for overall survival. In multivariate analysis, only advanced pathologic stage (stage II, more) was independent predictor of overall survival [p<0.0001, Hazard ratio: 3.17, 95% confidence interval 1.76-5.73]. Figure 1

Conclusion
Surgical treatment for selected patients aged 80 years or over or with primary NSCLC can be performed safely with low morbidity and mortality in this study. We recommend that limited operation might be the best surgical treatment, especially for stage I NSCLC. In the future, establishment of accurate clinical staging as well as early detection for lung cancer, and the appropriate treatment for advanced stage NSCLC for aged people should be studied for the upcoming aged society.

Background
In order to clarify clinical implication and therapeutic strategies managing minute malignant pleural effusion unexpectedly found at thoracotomy, we analyzed diagnostic results of cytodiagnosis of pleural effusion found at thoracotomy and prognosis of the patients treated in our institute retrospectively.

Methods
From 2004 to 2013, 502 non-small cell lung cancer patients were surgically treated in our institute. At the time of thoracotomy pleural effusion cytology (PEC) was performed in the cases that small amount of subclinical, minute pleural effusion was found unexpectedly. In some cases pleural lavage cytology (PLC) was performed simultaneously.

Results
In 254 patients (50.6%) out of 502, minute pleural effusion was detected and PEC was performed. In 25 patients (9.8%), malignant pleural effusion was demonstrated. PLC was performed in 191 cases, and positive results were obtained in 16 cases (8.3%). In these 16 cases, PEC was also performed in 12 cases and positive results were obtained in 11 cases. And in 4 cases no pleural effusion was found but PLC was positive for malignancy. Post operative 5 yrs survival of malignant pleural effusion group was 43.7%. In p-N0 cases, post operative survival of PEC positive cases was tend to poorer than surgically treated PEC negative cases (p=0.13), but in p-N1, N2 cases, post operative survival was almost equal between these groups (p=0.66).

Conclusion
Cytodiagnosis of minute pleural effusion found at thoracotomy was performed in 51% of surgical cases of NSCLC. Subclinical malignant pleural effusion found at thoracotomy may be a prognostic factor in clinical stage one case. PLC at the time of thoracotomy may become a complementary examination of PEC.

Background
An extended sleeve lobectomy is a useful procedure so as to spare the lung parenchyma. However, the resection of the bronchus can cause an increment in the tension at the site of the anastomosis and mismatches in the size of the bronchial orifices. Induction chemoradiotherapy (CRT) followed by surgery is a therapeutic option for locally advanced non-small cell lung cancer (NSCLC). Induction CRT, especially radiotherapy, has a negative effect on bronchial healing in the bronchial stump or anastomosis in a pulmonary resection.

Methods
The medical records were reviewed for nine NSCLC patients who underwent extended sleeve lobectomy after CRT between December 2007 and January 2013. Disease stage was evaluated with imaging analyses, including enhanced chest computed tomography (CT) scan, brain magnetic resonance imaging, positron emission tomography-CT scan and bronchoscopy. Induction CRT was performed for eight cases using cisplatin and docetaxel with concurrent thoracic radiation. For one patient who had synchronous laryngeal cancer, 5-fluorouracil and nedaplatin were used as chemotherapy. The radiation dose was 46 or 40 Gy using a conventional fractionation (2 Gy/day). Patients without progressive disease or good general condition underwent surgery. The bronchial anastomosis was basically wrapped with an omental pedicled flap or pericardial fat pad with prophylactic intent. The pre- and postoperative first-second forced expiratory volume was measured. The overall survival (OS) and the disease-free survival (DFS) were calculated from the date of initialing induction CRT until the date of death or the last follow-up for OS and until confirmed death of any cause or recurrence at local or distant site for DFS. The survival curve was calculated by the Kaplan-Meier method.

Results
The median patient age was 60 years (range, 50 to 73 years). There were seven men and two women. The histological subtype was squamous cell carcinoma in six patients and adenocarcinoma in three patients. Five patients had clinical stage (c-stage) IIIA, two patients had c-stage IIIB, and two patients had c-stage IIB. The radiation dose was 46 Gy in seven patients and 40 Gy in two patients. An extended sleeve lobectomy was performed for the left lingular division and the lower lobe in four patients, the right upper lobe and trachea in one patient, the right upper lobe, carina and trachea in one patient, the right middle and lower lobe in one patient, the right upper and middle lobe and the carina in one patient, and the right upper lobe and superior segment of the lower lobe in one patient. While no postoperative 90-day deaths occurred in this series, one case developed a bronchopleural fistula on postoperative day (POD) 25 and one case developed a bronchovascular fistula on POD 163. No cases of local recurrence occurred. The first-second forced expiratory volume before surgery was 2.52 ± 0.58 L (mean ± standard deviation), while that after surgery was 1.80 ± 0.66 L. The 2-year overall survival and disease-free survival rates were 63.5% and 47.6%, respectively.

Conclusion
Our experience suggests that an extended sleeve lobectomy after induction CRT is feasible, but careful patient selection and perioperative management is mandatory.

Background
Oschner and DeBakey in 1940, reviewed the world literature of 191 esophageal resections with a 72% mortality rate. Later on with advent of better preoperative, intaoperative and postoperative management advances, the risk of mortality had decreased to less than 10%. Various radical resections were described both in resection of primary (Eg: Transhiatal esophagectomy(THE), transthoracic esophagectomy (TTE),enbloc esophagectomy,etc..,) and lymph node dissection (two field, three field , etc.,). The role of radical surgery is still controversial and its becomes necessary to decide which patient can tolerate a radical procedure in the preoperative setting itself.

Methods
We retrospectively analyzed surgical and medical records of 154 patients of carcinoma esophagus operated in a single unit at our center to assess the predictive factors for postoperative morbidity for the period 2006-2012.

Results

Table 1: Demographic profile of the Patients: Characteristics Number of patients ( Total N = 154) Age Mean Range 55.857 ± 10.3989 20-79 years Sex Male Female 92 62 Histopathological Examination Squamous cell carcinoma Adenocarcinoma Adenosquamous carcinoma 124 28 2 Site of lesion Middle third esophagus Lower third thoracic esophagus and Gastroesophageal tumors 36 118 Duration of symptoms ( Mean ± S.D) Preoperative albumin (gm/dl)( Mean ± S.D) Comorbidities Neoadjuvant treatment 2.481± 1.7684 (Range 1 -12) 3.938 ± 0.3684 (Range 3-4.9) 25 ( COPD-12, Diabetes Mellitus-10, Hypertension-8, Ischemic heart disease-3; Hypothyroidism-3, Previous history of pulmonary Koch’s-3) 10 (6 patients - NACTRT, 4 patients- NACT) Surgery performed Transhiatal esophagectomy Transthoracic esophagectomy 140 14 Complications Pulmonary complications Anastamotic leak Mortality Abdominal infection Chyle leak 18 11 12 ( 7 had pulmonary complication also) 3( All 3 had pulmonary complication ) 2 1( also had pulmonary complication)

NACTRT- Neoadjuvant chemoradiotherapy, NACT-Neoadjuvant chemotherapy Of 154 patients who underwent esophagectomy at our center, 140 patients underwent THE and 14 underwent TTE. Mean duration of dysphagia was 2.4 months. One hundred and twenty four patients had squamous cell carcinoma and 30 patients had other histological types. Twenty five patients had comorbidities as described in Table.1. Eighteen patients developed postoperative complication of which three died - one case due to massive pulmonary embolism and two cases due topulmonary infection and septicaemia). Twelve patients had anastamotic leak (7.7%), all were managed conservatively. When multivariate regression analysis was performed for the predicted risk factors and development of complication, preop albumin ( less than 4gm/dl) and histological type (non squamous cell carcinoma) were associated significantly with increased post operative complications (Table.2). Abnormal pulmonary function tests though showed increased risk of complications it didn’t attain statistical significance. Table 2: Multivariate logistic regression analysis of predictive factors for early postoperative complications:

Independent variables Coefficient Standard Error t P Albumin (Less than 4 gm/dl) -0.1505 0.07519 -2.001 0.0472*(significant) Comorbidity (Yes) 0.1012 0.06716 1.507 0.1339 Duration of dysphagia (Absolute dysphagia) 0.004364 0.01576 0.277 0.7823 Histopathological examination ( Non squamous cell carcinoma) -0.1877 0.07332 -2.559 0.0115*(Significant) Neoadjuvant treatment ( Yes) 0.1043 0.1147 0.910 0.3645 Site of lesion ( Middle 3rd) 0.06701 0.06641 1.009 0.3146 Pulmonary function tests (Abnormal) 0.04156 0.02232 1.862 0.0646

Conclusion
We achieved a 2% mortality rate during the study period. Preoperative serum albumin and histological subtype were associated with increased postoperative mortality. Neoadjuvant therapy was not associated with increased complication rate.

Background
A new lung adenocarcinoma classification is being proposed by the International Association for the Study of Lung Cancer, American Thoracic Society, and European Respiratory Society (IASLC/ATS/ERS), which was based on histomorphologic subtype and had recently been validated in a North American series of 514 stage I lung adenocarcinomas. However, its distribution of patient number was biased, especially adenocarcinoma in situ (AIS), minimally invasive adenocarcinoma (MIA), and non-mucinous lepidic predominant groups, compared with in Japan. We evaluated an appropriateness of new classification in a series in Japan and whether the classification could be useful for selecting limited cases undergoing sublobar resection.

Methods
We retrospectively reviewed clinical records of all patients operated on for non-small cell lung cancer from 1997 to 2011 (n=825). 292 patients (35.4%) had pathological stage IA adenocarcinoma of the lung classified according to the Union for International Cancer Control/American Joint Committee on Cancer 7th Edition. Some pathologists, blinded to patient outcome, performed histopathologic subtyping according to the proposed new IASLC/ATS/ERS classification. Statistical analyses were made including Kaplan–Meier and Cox regression analyses.

Results
There were 160 females (54.8%) and 132 males (45.2%) with a median age of 67 years (29–84 years) and 212 pT1a and 80 pT1b patients (tumor size: 16.7±7.1 [2-30] mm). Three overall prognostic groups were identified: low grade: AIS (n=103, 35.3%) and MIA (n=24, 8.2%) had 97.1% and 95.1% of disease-free survival at 5 years (DFS, median follow-up was 54 months); intermediate grade: non-mucinous lepidic predominant (n=61, 20.9%), acinar predominant (n=38, 13.0%), and papillary predominant (n=48, 16.4%), with 80.1%, 86.5%, and 62.4% of DFS; and high grade: invasive mucinous adenocarcinoma (n=5, 1.7%), solid predominant (n=12, 4.1%) and micropapillary predominant (n=1, 0.3%), with 88.9% of DFS. DFS in low grade was significant better than in other two grades (P<0.005), however, there was no significant difference between in intermediate and high grade groups due to lower DFS in papillary predominant or insufficient patient number in high grade group. Preoperative imaging examinations such as consolidation/tumor (C/T) ratio on high resolution CT and maximum standardized uptake value (SUVmax) by FDG-PET were correlated with histopathologic grade according to new classification (P<0.01). Moreover, sublobar resection was undergone for 144 cases (49.3%), more cases had been identified small tumor, low C/T ration, low SUVmax, and low grade subtypes, and DFS in sublobar resection was 96.4% which was significant better than in lobectomy (78.6%, P=0.0002).

Conclusion
In this study, we can evaluate with enough number of patients classified to AIS, MIA, or non-mucinous lepidic predominant according to the new IASLC/ATS/ERS classification. Most of subtypes correlated with DFS, except of papillary predominant and subtypes in high grade clinical aggressiveness, which may need more clinical investigation. Patients in low grade subtypes who underwent sublobar resection had better DFS, which can be predicted using tumor size and preoperative imaging examinations such as C/T ratio and SUVmax. So, the new classification has advantages for better selection of limited cases undergoing sublobar resection as a curative surgery.

Background
Postoperative complications are common after major lung resections for cancer. Although there have been important advances in operative technique and perioperative care, lung resection for cancer is still a major surgery with higher incidence of complications in centres with a low case volume. We undertook a retrospective analysis of a prospectively maintained surgical database to ascertain whether the grade of the operating surgeon influenced the occurrence of postoperative complications and mortality in a tertiary level teaching cancer centre.

Methods
Data from a prospectively maintained database (Aug 2004 to May 2013) was analysed and the following parameters were retrieved: age, sex, type of surgery, grade of surgeon (Consultant vs trainee), postoperative major complications including pulmonary complications, air leak, bronchopleural fistula and mortality. ICU and hospital stay were also compared between surgeries performed by consultants and trainees. Categorical variables were analysed by the Chi square or Fischer’s exact test and numerical variables using the unpaired Student t test. P values less than 0.05 were considered statistically significant.

Results
A total of 654 patients (494 male, 160 female; mean age 54.4, range 14 to 83 years) underwent lung resection for primary lung tumors in the study period. The overall major morbidity and postoperative mortality was 10.6% and 1.7% respectively. Consultant thoracic surgeons performed 532 surgeries while trainees performed 122 procedures. Major morbidity was 11.2% and 8.2% (p=0.336) and postoperative mortality was 1.9% and 0.8% (p=0.628) when lung resection was performed by consultants and trainees respectively. The incidence of pulmonary complications (10.2% vs 9.9%, p=0.916) and bronchopleural fistula (2.2% vs 1.8%, p=0.749), median ICU (both 0 days) and hospital stay (both 5 days) were also similar in the two groups.

Conclusion
Early postoperative outcomes after major lung resection for primary lung tumors are independent of the level of training of the operating surgeon. These results are qualified by the fact that operations performed by trainees were closely supervised and assisted by experienced consultant thoracic surgeons in a teaching tertiary level cancer centre.

Background
The treatment of choice for early-stage lung cancer is surgical resection. Over the years, there has been an increase in minimally-invasive surgical (MIS) options for lung resection, including video-assisted thoracoscopic (VATS) lobectomy and robotic-assisted thoracoscopic (RATS) lobectomy. These MIS approaches have reformed lobectomies, decreased overall morbidity as well as decreased post-operative pain and hospital length of stay (LOS) when compared to traditional thoracotomy lobectomies. In spite of the recent rapid surgical expansion of MIS lobectomy, little is known about the learning curve of RATS lobectomy. In order to move forward in this new surgical technologic era while enhancing patient safety, determination of this learning curve is crucial.

Methods
We retrospectively analyzed the perioperative outcomes of 236 consecutive patients who underwent robotic lobectomy at our institution between September 2010 and June 2013. Patients were grouped chronologically into four quartiles, with 59 patients in each quartile. A comparison was performed between quartiles with respect to operative times, intraoperative estimated blood loss (EBL), hospital LOS, and in-hospital mortality. Statistical analysis was undertaken using analysis of variance (ANOVA), linear regressions, and t-tests. Significance was set at p-value <0.05.

Results
A total of 236 patients with a mean age of 67 ± 10 years underwent RATS lobectomy between September 2010 and June 2013. Each of the four quartiles had overall conversion-to-thoracotomy rates of ≤10% and emergency conversion rates of ≤5%. Although intraoperative EBL and hospital LOS were not significantly different among the quartiles, both EBL and hospital LOS showed a decreasing trend over each chronologic quartile. Operative times increased during the 2nd quartile, but also showed a decreasing trend with subsequent quartiles. There was an overall pathologic upstaging in 39% of the RATS lobectomy patients. Lastly, the in-hospital mortality rate was 5.1% for the 1st quartile and 0% for each subsequent quartile.

Conclusion
In contrast to the steep learning curves associated with other MIS procedures, RATS lobectomy may result in only a modest learning curve for surgeons with extensive VATS lobectomy experience, as is suggested from the unchanging and low hospital LOS, intraoperative EBL, and conversion to open lobectomy rates across all quartiles. Additionally, the increase in operative time in the 2nd quartile is most likely associated to the extension of the RATS approach to more difficult lobectomy cases after relative comfort with RATS lobectomy has been achieved. The high percentage of overall pathologic upstaging is likely due to more complete mediastinal lymph node dissection than would be achieved with either deliberate lymph node sampling only or else inadequate lymph node dissection due to limitations in visualization, instrumentation, or technical experience with VATS lobectomy. The decrease in mortality rate from the 1st to subsequent quartiles can be attributed to increased proficiency in RATS lobectomy. Thus, a bi-phasic learning curve is demonstrated by increased operative times after establishment of comfort with RATS lobectomy after initial success and subsequent extension of RATS to more complicated patients (1[st] phase), while operative times, EBL, hospital LOS, and mortality decrease with improved patient selection and more RATS experience (2nd phase).

Background
Lung cancer remains by far the biggest cause of cancer deaths worldwide, exceeding the next four cancers combined. Meanwhile, health systems struggle with ballooning costs and an unclear relationship between cost and outcomes. Within this context, there exists a great need for reliable, goal-oriented, and cost-effective care. ProvenCare[®] Lung Cancer is a national, multi-institutional care improvement collaborative that identifies and prescribes best practice elements of care in order to improve outcomes of pulmonary resection for lung cancer.

Methods
Incorporating principles of reliability science, the ProvenCare[® ]group established a protocol to promote adherence to and completion of best practice elements of care for patients undergoing pulmonary resection for lung cancer. Thirty-eight care elements span the continuum from preoperative assessment through intraoperative and inpatient postoperative care, and on to post-discharge follow-up. An original 6 thoracic surgical teams, subsequently expanded to twelve institutions, prospectively measured adherence to the prescribed elements of care in an “all or none” measure of success (adherence to all 38 elements is required for “compliance”). These data are to be compared both in a pre- and post-ProvenCare[®] involvement within participating sites, as well as to contemporaneous data within the Society of Thoracic Surgeons (STS) General Thoracic Database, to compare operative morbidity and mortality between the two groups. Figure 1

Results
To date, nearly 900 patients have been enrolled at the twelve Collaborative sites. Certain elements have proven more difficult than others to accomplish reliably, namely adherence to postoperative pain control and pulmonary toilet regimens. Even so, the redesign of the complex care required by the lung resection patient to embed evidence-based best practices into everyday patient flow, has resulted in high adherence to the prescribed elements of care, ranging from 92 to 100%. A comparison of the ProvenCare[®] study group to the STS Database control will be undertaken once an additional 200 patients are enrolled.

Conclusion
Quality and reliability represent frontiers of cancer care, and have the potential to improve patient outcomes as much as blockbuster medications or innovative radiation or surgical techniques, and perhaps more cost-effectively. We present an example of a national, collaborative, multidisciplinary model of reliable care that represents the first attempt to test patient care outcomes in this context.

Background
The incidence of a second primary lung cancer has been reported with 1-2% per patient year. Still, relatively few data has been published about this selected group of patients with sometimes conflicting results. Most data has been included low patient numbers that derived either from multiple institutions or from a long time period that may have rendered conclusions difficult caused by varying diagnostic procedures and therapeutic developments. Moreover, data regarding clinical characteristics is lacking for patients with first primary lung cancer who might be at risk for developing second primary lung cancer

Methods
From January 1999 to December 2008, 65 patients with second primary lung cancer, classified by the criteria proposed by Martini and Melamed, were treated at our Institution. We had 34 patients with a synchronous tumour and 31 with metachronous. As second treatment, we performed 11 lobectomies and 54 segmentectomies and widewedge resections. Histology was adenocarcinoma in 58, squamous in 4, adenosquamous in 8, large cells in 2 and pleomorphic in 1.

Results
Overall 5-year survival from second surgery was 69%; overall operative mortality was 0.65% (1 patient). Regarding the interval of surgery, the second operation had performed later than 2 years group showed a better 5-year survival than within 2 years group (80.6% and 69.2%, respectively, P = 0.008). Compared with lobectomies, segmentectomy showed a no significantly changes in survival rate(69 and 60%, respectively, P >0.051).

Conclusion
From our experience, lobectomy should still be considered as the treatment of choice in the management of second primary lung cancer, but sublobar resection remains a valid option in high-risk patients with limited pulmonary function.

Background
The clinical impact of mediastinal lymph node dissection (MLND) during pulmonary metastasectomy remains controversial. Especially the prognostic contribution of MLND on the prevention of tumor recurrence in patients with no evidence of mediastinal lymph node metastasis has not been clearly defined. We aimed to clarify the role of MLND during pulmonary metastasectomy in this population.

Methods
We retrospectively reviewed 632 patients who underwent pulmonary metastasectomy from January 2006 to December 2010 in Asan Medical Center. Among them, two hundred nine patients were identified to meet the following criteria and comprised the current study population: the presence of preoperative computed tomography (CT) and positron emission tomography (PET) within 2 months before pulmonary metastasectomy, definite control of the primary tumor, and no evidence of mediastinal lymph node metastasis. Of 209 patients, sixty-seven patients underwent MLND during pulmonary metastasectomy (MLND group), whereas 142 patients underwent pulmonary resection only (non-MLND group). Between-group recurrence-free survival was compared, and risk factors for tumor recurrence were evaluated. The data on tumor recurrence were obtained through a median follow-up duration of 42 months (range 2-83 months).

Results
The study population was composed of 119 male and 90 female, and the age at the first pulmonary metastasectomy ranged from 13 to 82 years (median, 56 years). Primary tumor pathologies included colorectal cancer (n=104, 49.8%), hepatobiliary cancer (n=38, 18.2%), kidney cancer (n=17, 8.1%), sarcoma (n=14, 6.7%), and the others (n=36, 17.2%). Disease-free interval from initial primary tumor treatment to the first metastasis ranged from 1 to 94 months (median, 25 months). Overall 5 year recurrence-free survival rate was 30.1%. There was no difference in recurrence rates between the MLND group and the non-MLND group (5 year recurrence-free survival: 30.0% vs. 24.5%, p=0.927). On multivariable analysis, primary tumor histopathology (p<0.001), disease-free interval (p=0.016), and the number of nodules (p<0.001) emerged as significant and independent prognostic factors for recurrence. After adjustment by these three significant variables, mediastinal lymph node dissection did not affect recurrence-free survival (hazard ratio, 0.924; 95% confidence interval, 0.641-1.333; p=0.672).

Conclusion
Tumor recurrence after pulmonary metastasectomy was affected by the histopathology of the primary tumor, disease-free interval, and the number of metastatic nodules. However, the role of mediastinal lymph node dissection as a part of pulmonary metastasectomy is obscure in patients with no evidence of mediastinal lymph node metastasis based on CT and PET.

Background
Lung cancer generally presents in an advanced stage and tuberculosis(TB) is a common disease in the subcontinent. This study describes locally advanced lung cancer showing low nodal involvement by cancer and concomitant TB.

Methods
Retrospective analysis of lung cancer database of Department of Surgical Oncology , BRA IRCH , AIIMS (2012 -2013 ) was performed and 28 cases were identified who underwent surgeries for lung mass. Only primary lung cancers were included. The clinical features , histopathology and management of these patients were analyzed.

Results
A total of 1293 cancer patients were operated between 2012 and 2013 in the department of surgical oncology. Out of which 28 patients were diagnosed to have lung mass with an incidence of 2.1%. Lung cancer was common in fifth decade. Predominant in males [M:F - 3.6:1]. Equal incidence of left and right lung. Five had old Koch's disease. Eighteen were smokers and majority were diagnosed to have lung cancer by computed tomography guided tissue diagnosis. Bronchoscopy detected 9 central tumors. All underwent R 0 resection except one case which was unresectable. Majority were in stage III [ 18 cases]but only 2 patients had nodal invovlement. Chest wall was resected in 4 patients with an average of three ribs resected. Final histopathology showed majority of adenocarcinoma [ 12 cases] followed by squamous cell carcinoma [ 8 cases]. Five patients showed features of concomitant TB.

Conclusion
Due to high prevalence of TB in this subcontinent , nodal staging in pre-operative imaging assessment might be fallacious. Imaging and positron emission tomography results should be interpreted more cautiously while making surgical decisions regarding operability.

Background
This meta-analysis aims to compare the perioperative outcomes of video-assisted thoracic surgery (VATS) versus open thoracotomy for propensity score-matched patients with early-stage non-small cell lung cancer (NSCLC).

Methods
Four relevant studies with propensity score-matched patients were identified from six electronic databases. Endpoints included perioperative mortality and morbidity, individual postoperative complications and duration of hospitalization.

Results
Results indicate that all-cause perioperative mortality was similar between VATS and open thoracotomy. However, patients who underwent VATS were found to have significantly fewer overall complications, and significantly lower rates of prolonged air leak, pneumonia, atrial arrhythmias and renal failure. In addition, patients who underwent VATS had a significantly shorter length of hospitalization compared to patients who underwent open thoracotomy. Figure 1 Figure 2

Conclusion
In view of a paucity of high level clinical evidence in the form of large, well-designed randomized controlled trials, propensity score matching may provide the highest level of evidence to compare VATS with open thoracotomy for patients with NSCLC. The present meta-analysis demonstrated superior perioperative outcomes for patients who underwent VATS, including overall complication rates and duration of hospitalization.

Background
To explore the clinical application value of complete video-assisted thoracoscopic(cVATS) lobectomy in the mini-invasive treatment of lung cancer.

Methods
90 patients with non-small cell lung cancer(NSCLC) who had undergone lobectomy were reviewed. According to surgical approach, complete video-assisted thoracoscopic lobectomy group (cVATS,n=47) and video-assisted mini-thoracotomy group(VAMT,n=43) were studied. Numbers of dissected lymph nodes,operation duration,volumes of intraoperative bleeding, duration of postoperative catheter drainage, length of postoperative hospital stay, incidence rates of postoperative complications, postoperative pain scores of patients were compared between the two groups retrospectively.

Results
There were no significant differences in numbers of dissected lymph nodes, operation duration, bleeding during operation, incidence rates of postoperative complication between the two groups(P>0.05).Duration of postoperative catheter drainage and length of postoperative hospital stay of patients in cVATS group were shorter than those in VAMT group(P<0.05).Pain scores of patients in cVATS group were lower than those at the same time in VAMT group(P<0.05).

Conclusion
Complete video-assisted thoracoscopic lobectomy is safe and effective surgical strategy for lung cancer patients with advantage of rapid recovery.

Background
Optimal pathologic nodal staging of NSCLC requires MLN dissection (MLND) or systematic sampling (SS). In our prior audit of a citywide database, 45% of resections were claimed by surgeons as MLND, none of which met pathology criteria, only 9% of all resections met SS criteria, 50% of all resections had random sampling (RS) and 42% had no sampling (NS) of MLN. An independent surgeon audit suggested that 29% of operation notes described a MLND, but 26% were RS and 45% NS. The concordance rate for MLND or SS between surgeon claims and pathology report audit was only 11%. We examined the impact of corrective intervention with a pre-labeled lymph node specimen collection kit and a checklist on the verifiable quality of MLN examination in a repeat audit of surgeon claims.

Methods
Prospective cohort study of NSCLC resections performed with the kit at 4 Memphis, TN hospitals from 11/2010 - 01/2013. Surgeons, operating room and pathology staff received training on the value of rigorous MLN examination and proper kit use. Surgeons marked the stations harvested on a checklist during the operation. Resections were classified into 4 pre-defined groups based on MLN stations marked on the checklist (surgeon claims), and the pathology report: MLND, SS (both by ACOSOG Z0030 trial criteria), RS (>0 MLN present, but MLND/SS criteria not met), NS (0 MLN present). Audited operation notes were categorized by surgeons from two independent academic cancer centers into one of the 4 MLN examination groups. The primary endpoints were the verifiable rate of MLND + SS and the concordance rate between observers.

Results
N = 161; 51% female, median Charlson comorbidity score 2 (IQR 1-3), 58% right-side resections. Clinico-demographic characteristics were similar between patients in each MLN category. Surgeons claimed MLND in 49%, SS in 9% of cases; vs 76% and 14% in the independent surgeon audit. The kappa score between independent surgeons was 0.44 ('moderate agreement'). Figure 1

Conclusion
The verifiable MLND+SS rate increased from 9% in the previous pathology audit to 83%; and from 29% in the previous independent surgeon audit to 89%. Concordance between operating surgeon claims and the pathology report increased from 11% to 83%. The improved lymph node yield and verifiable quality of MLN mapping indicates that implementing a corrective intervention with a pre-labelled specimen collection kit and checklist improves surgical MLN collection practice, fosters better communication with pathologists and improves the quality of pathologic nodal staging of NSCLC.

Background
The optimal surgical approach for early stage NSCLC continues to be debated. Nodal upstaging could be a surrogate measure for the quality of surgery and may help define superiority of a particular approach. However, nodal upstaging is only one measure of oncologic equivalence and may not translate into cancer recurrence or survival. Additionally, recent publications focusing on oncologic equivalence have compared approaches from different time periods. This study compares nodal upstaging, recurrence rates, disease-free and overall survival between matched groups of open and MIS (VATS/Robotic) lobectomy performed concurrently in a single time period.

Methods
We retrospectively compared patients undergoing lobectomy via thoracotomy to MIS for primary, clinical stage I/II NSCLC from 01/04-05/11. Patients were matched for age, gender, comorbidities, PFTs and clinical TNM status.

Results
Two hundred and fourteen patients were evaluated with 107 in each group. Preoperatively the MIS group had more T1a tumors (Table). The rate of nodal upstaging was significantly higher in the open group compared to the MIS group [N0 to N1: 12% (13) vs 4% (4), p=0.02; N0 to N2: 7% (7) vs 1% (1), p=0.03]. At median follow-up of 38 and 33 months respectively, recurrence rates for open vs MIS were similar: local 4% (4) vs 2% (2), regional 8% (9) vs 3% (3) and distant 13% (14) vs 12% (13), p=0.23. Disease-free survival was 74% (79) and 83% (88) for open and MIS groups respectively at 36 and 28 months, p=0.14. Overall 2 year survival was 89% (95) for the open group and 91% (97) for the MIS group, p=0.65.

Characteristics of open versus MIS groups

	OPEN (N=107)	MIS (N=107)	p-Value
Age (median)	69	68	0.574
Female	63 (59%)	71 (66%)	0.260
Smoker	82 (77%)	84 (79%)	0.340
# Comorbidities (median)	1	1	0.589
FEV1% (median)	83	85	0.835
DLCO/VA% (median)	79	83	0.700
Investigations
PET Scan	93 (87%)	103 (96%)	0.264
Mediastinoscopy	78 (73%)	85 (79%)	0.264
Clinical Stage			0.150
IA	73 (68%)	82 (77%)
IB	33 (31%)	24 (22%)
IIA	1 (1%)	1 (1%)
Pathologic Stage			<0.001
IA	38 (36%)	68 (63%)
IB	41 (38%)	32 (30%)
IIA	16 (15%)	3 (3%)
IIB	2 (2%)	1 (1%)
IIIA	9 (8%)	3 (3%)
IV	1 (1%)	0
Pathologic T size (cm) (median)	2.8	2.1	0.003
Pathologic N0 status	87 (81%)	102 (95%)	0.001

Conclusion
In the same time period, nodal upstaging after open lobectomy for early stage NSCLC was significantly higher compared to MIS. However, there were no differences in local, regional or distant recurrence rates. Disease-free and overall survival was equivalent at median follow up of 38 months despite the difference in upstaging rates.

Background
After pulmonary resection, the immediate postoperative observed forced expiratory volume at 1 second (FEV1) is expected to be more correlated with the incidence of postoperative complications than the postoperative predicted FEV1. However, the clinical factors that affect the decrease in postoperative FEV1 have not been reported in detail.

Methods
Seventy patients who underwent lobectomy or segmentectomy for lung cancer were prospectively enrolled in this study. Of these patients, 25 who performed all pulmonary function tests in the preoperative period and in the early (at 3 or 4 days), middle (3 weeks later), and late (> 3 months later) postoperative periods were analyzed. At each postoperative period, the ratio of the postoperative observed FEV1 to the postoperative predicted FEV1 (po/ppFEV1) was evaluated according to the clinical factors.

Results
The mean po/ppFEV1 in the early / middle / late postoperative periods in all cases was 0.83 / 1.02 / 1.11. In the early period, the mean po/ppFEV1 of the resected upper / middle/ lower lobes was 0.74 / 0.97 / 0.88. The early poFEV1 after upper lobe resection was significantly lower than the ppFEV1 after middle or lower lobe resection (P=0.019). In the middle and late periods, no significant relationships were found between po/ppFEV1 and the resected lobe. No other clinical factors showed significant relationships with po/ppFEV1 in any postoperative period. The postoperative forced volume capacity (FVC) did not show a significant relationship with any clinical factor.

Conclusion
Resection of the upper lobe leads to a decrease of the FEV1 in the early postoperative period that is lower than the postoperative predicted FEV1. Therefore, in the early postoperative period after upper lobe resection, careful management is needed in order to avoid pulmonary complications.

Background
Lobectomy is considered the gold standard curative treatment for early lung cancer. Its indication does not only rely only on the extent of the disease but also on postoperative lung capacity. With the new adenocarcinoma classification for lung cancer peripheric lesions smaller than 2cm might be treated with anatomic segmentectomy.

Methods
We performed a retrospective analysis of 29 patients submitted to anatomic segmentectomy by VATS technique in the period of 05/07/2009 to 31/05/2013 in Salvador, Bahia, Brazil due to peripheral lung lesion suspected of malignancy. A total of 30 procedures were performed by two surgeons (Ugalde PA, Pereira ST). The purpose of this study is to analyze the indications, length of stay, surgical complications and final pathology.

Results
Of the 29 patients 17 (58,62%) were female, the age ranged from 13 to 89 years with an average of 53 years (± 20.49). 53,3% of cases final pathology revealed neoplastic disease, 10 adenocarcinomas in situ, 4 metastatic diseases, 1 carcinoid tumor, and 1 spindle cell tumor proliferation. The postoperative hospitalization ranged from 1 to 6 days with an average of 2.73 days (± 1.44), the timing of chest drain varied 1 to 6 days with an average of 2.80 days (± 1.52). Two patients had to be submitted to bronchoscopy due to postoperative atelectasis.

Conclusion
Anatomic segmentectomy although controversial in the treatment of primary early stage lung cancer, in our series when performed through VATS was safe and effective with minimum rate complication. The advantages of a minimally invasive procedure resulted in a shorter hospital stay and recovery time, enabling patient early return to their activities.

Background
We have previously reported favorable outcomes in elderly patients treated with surgery for NSCLC. In this study we have evaluated all elderly patients that were diagnosed with NSCLC in Iceland and specifically studied those that were not operated on.

Methods
A whole nation study has been conducted that included all patients diagnosed with NSCLC in Iceland between 1991 and 2010 according to the Icelandic Cancer Registry. Data on staging, functional status and survival was obtained from medical records. All tumors were staged clinically (cTNM) and the reasons for exclusion from surgery were registered for elderly patients with cTNM stages IA – IIIA.

Results
Of 2263 patients with NSCLC, 735 (32.5%) were defined as elderly. Resection rate for the elderly was 15% compared to 26% for younger patients (p<0.001). Out of 627 elderly patients 55% had localized/regional disease that was considered potentially resectable, and of that group 42% were not operated on. The most common reasons for exclusion from surgery were poor pulmonary function (40%) ECOG-performance status (15%), central tumor location (14%), underlying heart disease (11%) or that the patient rejected treatment (9%).

Conclusion
Elderly patients with potentially resectable NSCLC are frequently excluded from surgery due to co-morbid conditions. Their favorable 30-day and long-term survival may reflect bias in selection of lower risk patients.

Background
Lung cancer is the leading cause of cancer deaths in Brunei Darussalam for the past 5 years. Curative surgical resection in early stage non-small cell lung cancer (NSCLC) have been shown to improve survival. This study aimed to assess the 5-yr survival of patients undergoing surgical resection of NSCLC in Brunei Darussalam.

Methods
From 2000 to 2013, 64 patients underwent surgical resection of NSCLC at the RIPAS Hospital in Brunei Darussalam. Demographic and Clinical data of these 64 patients were retrospectively retrieved from the clinical notes. All deaths and date of death were obtained and cross-check with the National birth and death registry at the Imigration Department. Data were analysed using SPSS statistical software and 5-yr Kaplan-Meier survival curves were derived. Predictors of 5-yr survival were analysed using Cox regression analysis.

Results
Mean age of the 64 patients was 60.6 ± 12.2 (27.4 – 80.0 years) with male to female ratio of 39:25. Racial distribution consisted of 82.9% (53/64) Malay, 15.6% (10/64) Chinese and 1 foreign national. Histological types of NSCLC consisted of 57.8% (37/64) adenocarcinoma, 12.5% (8/64) Squamous cell carcinoma, 4.7% (3/64) large cell carcinoma, 18.8% (12/64) bronchioalveolar carcinoma and 6.3% of other origin. Overall 5-yr Kaplan-Meier survival curves according to stage of disease were 67.9% for Stage 1A, 51.1% for Stage 1B, 30.0% for Stage 2B, 29.0% for Stage 3A, 20.8% for Stage 3B and 33.3% for Stage 4. 5-yr disease-free survival were 59.6% for stage 1A, 43.5% for stage 1B, 20.8% for Stage 3B and 33.3% for stage 4. 5-yr survival for Stage 1A was better in patients who had systematic mediastinal lymphnodes dissection carried out as routine procedure (72.5%). Only Stage of NSCLC at the time of surgery was a significant independent predictor of 5-yr survival (p=0.01).

Conclusion
Overall 5-yr survival of patients with treated early stage 1A NSCLC is good and comparable to the surrounding region. This is further improve if routine SMLD were performed. Stage of NSCLC at time of diagnosis was the only significant independent predictor of 5-yr survival.

Background
Accurate estimation of poFEV1 remains challenging. Underestimation excludes patients with early stage lung cancer from a potential curative resection, overestimation gives more postoperative complications. The anatomic segment method (ASM) and the perfusion scintigraphy (QS) are standard methods, although both underestimate actual poFEV1 (Holvoet 2011). FRI, describing flow characteristics in the lungs, estimates poFEV1 after virtual resection (De Backer 2010). We compared the accuracy of FRI with ASM and QS in the estimation of poFEV1 in patients planned for resection of lung cancer.

Methods
23 consecutive patients underwent pre- and postoperative FEV1 measurements by spirometry, ASM, QS and FRI. ASM-poFEV1 was estimated by the product of preop FEV1 with (1-number of resected segments /19). QS-images were obtained following intravenous injection of 175 MBq of [99m]Tc-MAA, using geometric means method to determine left and right sided contribution. QS-poFEV1 was the product of preop FEV1 x (1- number of resected segments/total number of segments in operated lung x perfusion in this lung). Pneumotach controlled CT scan at maximum expiration (RV) and inspiration (TLC) allowed to segment lobar volumes and calculated regional expansion (EXP). Resistances (iRaw before and after virtual resection) were acquired through computational fluid dynamics. FRI-poFEV1 equation: Figure 1

Results
Data of 14 patients are available: 8 male, median age 63 y (51-73), median preop FEV1 2490 ml (1660-4070), 10 lobectomy, 2 bilobectomy, 2 pneumonectomy. Median actual poFEV1 was 2100 ml (1210-3210). The Pearson correlation coefficient (R[2]) paired t-test and root mean square error (RMSE) between actual and predicted poFEV1 are shown below. FRI-poFEV1 underestimated fewer cases with respect to the actual- 7% variation corrected-poFEV1 (Oostveen,2013 in press) than the other methods.

Method	Predicted median poFEV1 (ml) (range)	Underestimated cases (n)	R[2]	Slope	RMSE (ml)	P (paired t-test)
ASM	1660 (1240-2970)	11/14	0.77	1.06	436	<0.001
QS	1980 (1320-3040)	7/14	0.84	1.10	309	0.007
FRI	2050 (1450-3440)	4/14	0.86	0.98	213	0.76

Conclusion
FRI seems a superior predictor of actual poFEV1 in resected lung cancer pts than either conventional method, allowing for 20% more functionally operable patients. Confirmation in a larger serie is ongoing.

Background
Indwelling pleural catheters (IPCs) have a role in the management of pleural effusions. The TIME-2 trial demonstrated equivalence in dyspneoa relief for first time pleurodesis. This single centre study aimed to compare experience of patients receiving talc pleurodesis versus IPC.

Methods
A retrospective review of all patients undergoing IPC insertion or talc pleurodesis within a single Trust between October 2007 and September 2012. We had a policy of selective IPC insertion for trapped lung or recurrent pleural effusion, with talc pleurodesis the procedure of choice for expansile lungs. We examined resource utilisation including pre-operative intervention, length of stay (LOS), re-accumulation and re-intervention.

Results
130 patients were identified. 61 (47%) patients underwent talc pleurodesis; 69 had an IPC inserted. 13.1% of talc patients and 59.4 % in the IPC group had received a previous pleural intervention (p<0.001). 23.0% of the talc and 29.0% of the IPC group received their procedure on an urgent basis (p=0.44 ). Significantly more patients underwent a general anaesthetic in the talc group (IPC 26 (37.7%), talc 57 (93.4%) p<0.001). Patients treated with IPC had a significantly shorter post-operative stay than those treated with talc (IPC median 2 (range 2-46) days; talc 5 (0-36), p<0.001). Significantly fewer patients experienced re-accumulation following IPC than talc pleurodesis at 30 days (8 (11.6%) vs 19 (31.3%) p=0.006), and overall (12 (17%) vs 27(44%) p<0.001). There were no differences in post-procedure mortality (IPC 3 (4.35%), talc 1 (1.64%) p=0.372); effusion requiring re-admission to hospital (IPC 5(7.25%), talc 7 (11.5%) p=0.406, or re-intervention rates (IPC 6 (8.7%), talc 7 (11.5%) p=0.60).

Conclusion
Despite being used in patients with more complicated pleural effusion, IPC placement was associated with a significantly shorter post-operative length of stay and fewer cases of effusion re-accumulation. IPC placement should be considered for the treatment of pleural effusion.

Background
Plasma fibrinogen levels have been shown to correlate with outcomes in various extra-thoracic malignancies. In patients with NSCLC, positive associations have been shown between fibrinogen levels and tumour pathology, but the clinical correlates are unclear. We aimed to examine whether pre-operative fibrinogen levels are a prognostic factor in patients undergoing surgical resection for suspected NSCLC.

Methods
All NSCLC patients undergoing surgery between 29/8/2007 and 30/3/11 were included. Pre-operative plasma fibrinogen levels were measured and correlated with clinicopathological factors, pathological TNM stage and survival. Survival analysis was performed on 17/06/13.

Results
722 patients underwent surgery for suspected NSCLC. In 519 (71.9%) patients (54.5% males, median age 68.5 (range 37.8 - 90.8) years), pTNM stage and preoperative fibrinogen level were available. Median fibrinogen level was 4.1 (range 1.7 - 10.2) g/dL. 330 (63.6%) of patients had fibrinogen level > reference range (2-4g/dL). Fibrinogen correlated with tumour size (p<0.001) and pTNM stage (p<0.001), but not with nodal stage, histological grade or cell type. At the time of analysis, 309 (59.5%) patients were alive. Fibrinogen > 4g/dl (p=0.01), pTNM stage (p<0.001), Nstage (p=0.001) and tumour size (p=0.003) were univariate prognostic factors. In Cox multivariate analysis, fibrinogen level (p=0.02), pTNM stage (p<0.001), age (p<0.001) and gender (p=0.023) were independent predictors of prognosis.

	Fibrinogen <4g/dL		Fibrinogen >4g/dL		p
	n	Median Survival	n	Median Survival
StageI	125	Not reached	176	63.0	0.011
Stage II	45	55.4	97	Not reached	0.677
Stage III	19	42.3	57	34.5	0.396

Conclusion
Fibrinogen is associated with tumour size and pTNM stage. Whilst survival data are not yet mature, pre-operative fibrinogen > 4 g/dl may be a novel independent prognostic factor following surgical resection of NSCLC. Further work is required to determine the clinical implications of high fibrinogen levels, and to investigate the underlying mechanisms.

Background
Female gender has been associated with worse outcomes in cardiovascular surgery, including vein bypass for limb salvage and coronary artery bypass grafting. Women have also been found to prefer to suffer arthritis pain rather than risk orthopedic surgery and to delay surgery to await better technology and to avoid disrupting caregiving roles for spouses and other dependents. We investigated the effect of gender on perioperative outcomes after robotic-assisted lobectomy.

Methods
We retrospectively analyzed 180 consecutive patients who underwent robotic-assisted lobectomy by one surgeon between September 2010 and February 2013. Intraoperative estimated blood loss (EBL), operative times (skin incision to skin closure), conversion to open lobectomy, chest tube days, hospital length of stay (LOS), and in-hospital mortality were analyzed. All clinically significant perioperative complications were noted, including minor complications, such as wound infection and anemia, to more serious major complications, such as empyema and DVT/PE. Comparison of perioperative outcomes between men and women was significant at p-value <0.05.

Results
Of 180 total patients, there were 90 men (mean age 68yr; range 37-86yr) and 90 women (mean age 68yr; range 29-85yr; p=0.19). Skin-to-skin operative times were 191+12 min for men and 174+12 min for women. Men had median (+SEM) EBL of 235+60mL compared to 150+48mL for women (p=0.79). Intraoperative complication rates were 7/90 (8%) in men and 10/90 (11%) in women (p=0.45). The most common intraoperative complication in men was bleeding not requiring conversion in 2/90 (2%), compared to pulmonary artery (PA) bleeding in 5/90 (6%) of women. The overall conversion rate to open lobectomy was 14/90 (16%) in women versus 6/90 (7%) in men (p=0.06); although the emergent conversion rate was 5/90 (6%) in women versus 1/90 (1%) in men. The most common reasons for conversion to open lobectomy in women was PA bleeding in 5/90 (6%) and dense hilar pleural and/or tumor adhesions in 5/90 (6%); while the latter was the most common reason for conversion in men, occurring in 3/90 (3%). A minor and/or major postoperative complication occurred in 48/90 (53%) of men, compared to 39/90 (43%) in women (p=0.18). The most common postoperative complications in men were prolonged air leak 20/90 (22%), atrial fibrillation 14/90 (16%), pneumonia 11/90 (12%), and mucus plugs requiring intervention 7/90 (8%), while the most common in women were prolonged air leak 14/90 (16%; p=0.25), pneumonia 12/90 (13%; p=0.82), atrial fibrillation 8/90 (9%; p=0.12), and mucus plugs requiring intervention 7/90 (8%; p=1.00). Women had 4.0+0.5 chest tube days (median+SEM) and 5.0+0.5 hospital days, compared to 4.0+1.1 chest tube days and 5.5+0.6 hospital days for men (p=0.14 and p=0.44, respectively). In-hospital mortality was 4/90 (4%) in men compared to 1/90 (1%) in women (p=0.17).

Conclusion
While women had a slightly higher conversion rate to open lobectomy than men, female gender was not associated with increased intraoperative or postoperative complications nor with increased hospital LOS or in-hospital mortality. Our study suggests that robotic-assisted pulmonary lobectomy is feasible and safe in women.

Background
Adjuvant chemotherapy is not required for pathological stage (p-stage) IA non-small cell lung cancer (NSCLC). And the prognosis of p-stage IA NSCLC is comparatively good. In the Japanese Joint Committee for Lung Cancer Registration, the 5-year survival rate for p-stage IA NSCLC was reported with 86.8% [J Thorac Oncol 2011; 6: 1229-35]. But, there are some cases to cause recurrences more than expected among them. The purpose of this study was to evaluate the risk factors for recurrence in p-stage IA NSCLC patients who had undergone complete resection.

Methods
A total of 409 primary NSCLC patients underwent pulmonary resections between January 2003 and December 2012. Among them, 145 patients with p-stage IA NSCLC completely excised tumors in surgical procedures of segmentectomy or more were evaluated. All patients were divided into two groups : the group with recurrence (group R, n=15) and the group with no-recurrence (group NR, n=130). We compared the both groups regard to the patient-related factors (age, sex, smoking habits, previous history, consultation motivation, timing of definitive diagnosis), the tumor-related factors (tumor markers, histologic type and differentiation, tumor size, lymphatic permeation, vascular invasion, EGFR gene mutation) and the treatment-related factors (extent of resection and nodal dissection, thoracoscopic surgery or open thoracotomy, skill levels of operators, operative duration, amount of bleeding, postoperative hospital stay, adjuvant chemotherapy). The postoperative 5-year survival rate was calculated with the Kaplan-Meier method. The differences of significance between the two groups were compared using the Chi-square test. P values less than 0.05 were considered statistically significant.

Results
The 5-year survival rate was 45.0% in the group R and 92.3% in the group NR (p<0.001). In the patient-related factors, the smoker/non-smoker ratio was 7/8 in the group R and 95/35 in the group NR (p<0.05). The previous history of hypertension/no-hypertension ratio was 9/6 in the group R and 44/86 in the group NR (p<0.05) and that of diabetes/non-diabetes ratio was 4/11 in the group R and 10/120 in the group NR (p<0.05). There were no significant differences in other patient-related risk factors. In the tumor-related factors, the pathological lymphatic permeation ly1/ly0 ratio was 3/6 in the group R and 3/76 in the group NR (p<0.001). There were no significant differences in other tumor-associated risk factors. In the treatment-related factors, the enforcement of adjuvant chemotherapy/nothing ratio was 2/13 in the group R and 3/127 in the group NR (p<0.05). There were no significant differences in other treatment-related risk factors.

Conclusion
A history of hypertension or diabetes, lymphatic permeation and adjuvant chemotherapy were risk factors for recurrence, respectively. And aggressive enforcement of adjuvant chemotherapy is desirable in p-stage IA NSCLC patients both with pathologically lymphatic permeation and with a history of diabetes or hypertension.

Background
Due to a recommendation from The Danish Board of Health, the department for Heart, - Lung, - and Vascular Surgery on Odense University Hospital, Denmark has developed a specialized program of rehabilitation for lung cancer patients who have recently undergone lung surgery.

Methods
Purpose The patient is offered a specialized and precisely targeted physical lung rehabilitation program supplemented with nurse intervention with the goal of an improved quality of life, functionality and capability. Method The project is conducted as a qualitative study e.g. with questionnaires about life quality and smoking habits. Spirometry, walking test after the Borg 10 scale, and also VAS-score related to pain, are conducted. A preoperative FEV1 is also measured. The physical rehabilitation begins three weeks post surgery and takes place twice a week throughout the total length of the program of 4-10 weeks. A specially trained physiotherapist is responsible for tests and physical exercise. The physical rehabilitation is accompanied by patient interviews conducted by a specially trained nurse and also by supervision from other health professionals. The areas of focus of the nurse intervention are dyspnea, coughing, nausea, loss of appetite, pain, fatigue, smoking habits, and other psychological symptoms, and consultation with specially trained nurse is offered 3, 6, 9, and 12 months postoperative. At the same time the patient is offered a CT scan. Due to ethical standards no control group of patients was established in this study.

Results
The nurse intervention has shown a clear positive effect on the patients smoking habits. The results on the effect on dyspnea, coughing, nausea, loss of appetite, pain, fatigue, smoking habits and psychological symptoms will be presented on the poster. Data show improvement in FEV1, and significant improvement in 6-minute walk test after finishing the physical rehabilitation program.

Conclusion
In the beginning of the program the patients are very motivated. Statements from patients and relatives show that they choose rehabilitation offers which are available in the Health Care System with the anticipation of a future positive influence on their prognoses. The patients are expressing that the tests which are a part of the physical rehabilitation program have a huge effect on their motivation. Data shows that physical and emotional rehabilitation has a very positive effect on the patients’ life quality, functionality, and capability. According to statements from patients the interdisciplinary intervention of physiotherapist, nurse and doctor in this project is very valuable. Professional experience has shown that the collection of data from surgical lung cancer patients is difficult due to the fact that a lot of patients disappear from the program of rehabilitation. A large group of the patients transfer to oncological treatment, aborts the rehabilitation offer or dies. This influences the results of the study and therefore also the conclusion.

Background
Malignant airway obstruction is associated with severe respiratory distress and poor outcome. Airway stents have been used to relieve malignant airway obstruction and may provide significant palliation of symptoms, but quality of life data after airway stenting is lacking. We report the Mayo Clinic Arizona experience with the Boston Scientific Ultra Flex stent in the management of malignant airway obstruction.

Methods
A single center study evaluating quality of life before and after placement of the Boston Scientific Ultra Flex stent for malignant airway obstruction. Quality of life data (SF-36 questionnaires) before and after stent placement were obtained prospectively. Charts were reviewed retrospectively. Patients were enrolled from 2007-2011 at Mayo Clinic Arizona. The study was approved by the Mayo Foundation IRB. Informed consent was obtained and documented in the clinical record. SF-36 questionnaires were completed by the patients in the outpatient clinic before (baseline) and after stent placement.

Results
From 2007 to 2011, 19 patients underwent placement of Boston Scientific Ultraflex non covered tracheobronchial stents for airway obstruction. 3 of the 19 patients underwent stent placement for benign etiologies and were excluded from the analysis. Of the remaining 16 patients, 8 were male and 8 were female. The median age was 64.5 years. Non-small cell lung cancer was the most common underlying diagnosis (11 patients). The remaining patients had head and neck cancer (2 patients), esophageal cancer (1 patient), metastatic Hurthle cell thyroid cancer (1 patient) and metastatic renal cell carcinoma (1 patient). The most common presenting symptom of airway obstruction was dyspnea. Data from SF36 forms were available in 8 of 16 patients. There was a significant improvement in the physical component score (PCS) (10.4±10.1, P=0.02), and a non-significant improvement in the mental component score (MCS) (1.4±6.2, P=0.54) after stent placement.

Conclusion
Placement of airway stents for malignant bronchial obstruction resulted in significant palliation of symptoms and improvement in patient quality of life.

Background
Gene therapy has shown promise as a novel treatment for pleural mesothelioma. This treatment requires intrapleural placement of a catheter, a routine procedure if a patient has an effusion. After surgery or pleurodesis, however, the pleural space will be fused and access can be difficult. Previously these patients were excluded from gene therapy. The objective of this study was to develop a minimally invasive technique that would allow placement of a catheter into a fused pleural space.

Methods
Over the past 50 months 34 patients with fused chest cavities were enrolled in a gene therapy clinical trial. Utilizing a videoscopic saphenous vein-harvesting device, a new surgical technique was developed to safely create a space in the chest cavity for biopsy of tumor, creation of space to accommodate vector instillation and placement of a tunneled catheter. Often the procedure required tunneling directly between the lung and the chest wall, where there was no visible disease, in order to access a pleural nodule that was detectable on CT scan. In every case this procedure was performed through a single 15 mm incision (fig 1). Figure 1 Figure 1 a) The videoscopic saphenous vein tunneling device b) The device in use – not transillumination in the chest approximately 10cm distal to the incision c) Closure of the single video port with the catheter emerging from a proximally tunneled site

Results
All 34 patients had successful catheter placement, with 27 as outpatients. The only surgical complication was a transient air leak in 1 patient requiring overnight admission. 5 patients were admitted for urinary retention and 1 for resumption of anticoagulation. One patient had a local wound infection at the catheter site requiring readmission and 2 more were treated as outpatients, all clearing with antibiotics. All patients received gene therapy.

Conclusion
This technique represents a safe, effective and minimally invasive way to access a fused pleural space. This technique has proven useful in extending eligibility for pleural mesothelioma gene therapy, but could be used to safely access a fused pleural space for any purpose. This would include obtaining a biopsy in a patient who had undergone pleurodesis without an established diagnosis or placement of a catheter for any type of intrapleural therapeutic.

Background
To analyze causes, predictors and consequences of conversions to open surgery in VATS lobectomy.

Methods
Retrospective analysis of a prospectively maintained database.

Results
Since February 2009, 297 patients with NSCLC were scheduled for VATS anatomical resections. Conversion to open surgery was necessary in 22 patients (7.4%). Reasons for conversion were bleeding in 10, oncologic reason in 8 and technical considerations in 4 patients (adhesions after pleuritis or radiotherapy for other tumors: 3, limited space: 1). In univariate analysis, conversion rate was significantly higher in patients after neoadjuvant therapy (21% vs 4.3%, p=0.002). There was a statistical trend for higher conversion rate in patients during the first half of the series (9.5% vs 3.4%, p=0.062) and larger tumor size (T1 vs >T1, 4.8% vs 12.9%, p=0.098). The conversion rate was not influenced by age of the patient, nodal stage (pN0 vs pN+), body mass index, COPD, FEV1, or benign disease. Conversion did not translate into higher overall postoperative complication rates (33.3% vs 29.5%), longer chest drain duration (median 5 vs 5 days), or mortality (0% vs 2%). However, length of hospital stay was significantly longer in the conversion group (median 10 vs 6 days, p=0.0066).

Conclusion
Neoadjuvant therapy is an independent risk factor for conversion to thoracotomy in this VATS lobectomy series. Patients after neoadjuvant therapy should be selected carefully for a VATS approach. Conversion to thoracotomy did not increase the postoperative complication or mortality rate, but significantly increased the length of stay.

Background
It was reported that in chronic obstructive pulmonary disease (COPD) patients, tiotropium improves lung function. However diagnosis of COPD is often made during evaluation of patients with lung cancer for surgical intervention and the efficacy of tiotropium for these patients is unclear. Thus a prospective study is needed to investigate it.

Methods
A prospective study was conducted on patients undergoing pulmonary resection for lung cancer with COPD (ratio of forced expiratory volume in 1 second (FEV~1~)/ forced vital capacity (FVC) less than 70%) between July 2011 and January 2012. Patients with a known history of asthma, chronic respiratory disease other than COPD were excluded. Primary endpoint was evaluating the incidence of postoperative complication. Secondary endpoints were improvement of pulmonary function tests after more than 1-week treatment using tiotropium preoperatively.

Results
Of 168 lung cancer patients for six months, 21 (12.5%) patients with COPD were enrolled. Pulmonary complications (prolonged air leak; 4 (19.0%), sputum retention; 2 (9.5%), hypoxia needing transient home oxygen therapy; 2 (9.5%)) were observed in seven (33.3%), although there was no critical complication such as acute respiratory failure and no side-effect related tiotropium. Treatment of tiotropium resulted in a significant improvement of FVC (pre-FVC 2.96±0.70 vs post-FVC 3.18±0.58; p=0.005) and FEV1 (pre-FEV~1~ 1.78±0.44 vs post FEV~1~ 1.91±0.18; p=0.0003), but there was no significant difference between pre-RV (residual volume) / TLC (total lung capacity)% and post-RV/TLC% ( 108.8±20.1 vs 102.4±16.1; p=0.237).

Conclusion
In this prospective study, we were safely able to use tiotropium without critical complication and it improved FVC and FEV~1~ in patients with COPD. But it did not improve RV/TLC% statistically and there remains doubt about efficacy of titropium. We thought that there is a problem of compliance in inhalation drug and need to reveal the population in which tiotropium was effective, and then we should perform a prospective randomized control trial.

Background
Significance of blood immune cell circuit for start of phase transition (PT) “norm---early lung cancer” (LC) was investigated.

Methods
In trial (1987-2012) consecutive cases after radical surgery (R0, bi/lobectomies=48, N2-lymphadenectomies=48; squamous=21, adenocarcinoma=25, large cell=2; G1=16, G2=21, G3=11), monitored 48 early LC patients (LCP) (age=59±6.5 years, m=40, f=8; T1AN0M0=48, tumor size=1.7±0.3 cm, 5-year survival=100%) and 120 healthy donors (m=69, f=51) were reviewed. Variables selected for study were input levels of immunity blood parameters. The percentage, absolute count and total population number (per human organism) of T, B, CD4, CD8, CD16, CD1, CDw26, monocytes, CD4+2H, CD8+VV, leukocytes, lymphocytes, monocytes, eosinophils, stick and segmented neutrophils were estimated. The laboratory blood studies also included input levels of NST (tests of oxygen dependent metabolism of neutrophils spontaneous and stimulated by Staphylococcus aureus or by Streptococcus pyogenes), index of stimulation of leukocytes by Staphylococcus aureus or Streptococcus pyogenes, index of thymus function, phagocytic number, phagocyte index, index of complete phagocytosis. Differences between groups were evaluated using discriminant analysis, clustering, structural equation modeling, Monte Carlo, bootstrap simulation and neural networks computing.

Results
It was revealed that start of PT “norm---early lung cancer” significantly depended on T-, B-, CD16-, CD8- cell circuit, neutrophils, monocyte circuit (P=0.000-0.027). Neural networks computing, genetic algorithm selection and bootstrap simulation revealed relationships of PT “norm---early lung cancer” and neutrophils (rank=1), CD16 (rank=2), monocytes (3), lymphocytes (4), CD4 (5), CD8 (6), B-cells (7), T-cells (8), eosinophils (9). Correct detection of start of PT “norm---early lung cancer” was 100% by neural networks computing (error=0.000; urea under ROC curve=1.0).

Conclusion
Start of phase transition “norm---early lung cancer” significantly depended on blood immune cell circuit.

Background
We hypothesized that modern postoperative radiotherapy (PORT) could decrease local recurrence (LR) and improve overall survival (OS) in patients with stage IIIA-N2 non-small-cell lung cancer (NSCLC).

Methods
To investigate the effect of modern PORT on LR and OS, we identified published phase III trials for PORT and stratified them according to use or non-use of linear accelerators. We modelled the potential benefit of modern PORT in stage IIIA-N2 NSCLC treated with induction chemotherapy and resection.

Results
Of the PORT phase III studies, eleven trials (2387 patients) were included for OS analysis and eight (1677 patients) for LR. PORT decreased LR, whether given with cobalt, cobalt and linear accelerators, or with linear accelerators only. An increase in OS was only seen when PORT was given with linear accelerators, along with the most significant effect on LR (relative risk for LR and OS 0.31 (p=0.01) and 0.76 (p=0.02) for PORT vs. controls, respectively). Four trials (357 patients) were suitable to assess LR rates after surgery and/or induction chemotherapy in stage III NSCLC. LR as first relapse was 30 % (105/357) after 5 years. In the modelling part, PORT with linear accelerators was estimated to reduce LR rates to 10 % as first relapse and to increase the absolute 5-year OS by 13 %.

Conclusion
This modelling study generates the hypothesis that modern PORT may increase both LR and OS in stage IIIA-N2 NSCLC even in patients being treated with induction chemotherapy and surgery.

Background
Whether IFRT could replace ENI or not has been a controversial topic for years because of rare data from large sample sizes, prospective, randomized studies were available. This study is to evaluate the locoregional failure and its impact on survival by prospectively comparing involved field radiotherapy (IFRT) with elective nodal irradiation (ENI) for locally advanced non-small cell lung cancer with concurrent chemoradiotherapy.

Methods
Patients were randomized into IFRT or ENI arm，treated with 2 cycles of carboplatin combined with paclitaxel. Those without distant metastasis continued to receive chemoradiotherapy. The target volumes for IFRT included primary tumor, ipsilateral hilum and positive mediastinal lymph nodes, while that for ENI included the primary lesion, ipsilateral hilum, bilateral mediastinal lymph node drainage areas and bilateral supraclavicular fossa. The radiation dose was prescribed as high as possible if the restrictions could be met: percent volume of bilateral lung receiving ≥ 20Gy (V20) was ≤35% and the maximum dose to spinal cord was ≤50Gy.

Results
99 consecutive patients were assigned (45 IFRT vs. 54 ENI), with more patients in IFRT iiradiated with ＞60Gy than those in ENI (48.9% vs. 25.9%, P=0.018). the local failure rates were 34.1% and 30.0%（P=0.673）, Of which the isolated ENF was 0.0% and 2.0%, respectively (P=0.363). The median survival time was 27.8 months (95% CI, 18.0-37.5 months) and 16.7 months (95% CI, 15.0-18.4 months); the 1-, 2- and 3-year local tumor progression-free survival rates were 78.1%、72.6%、62.9% and 85.5%、61.2%、56.1% （P=0.895）, respectively; the 1-, 2- and 3-year overall survival rates were 80.0%、53.3%、36.6% and 70.4%、34.9%、30.3% （P=0.08）, respectively.

Conclusion
Preliminary results indicated that IFRT did not increase locoregional failure related to ENF. With IFRT, higher radiation dose could be administered compared with ENI and it is expected to improve survival. Further investigation is warranted.

Background
Stereotactic body radiotherapy (SBRT) is an effective treatment for early stage inoperable non-small cell lung cancer (NSCLC), with loco-regional control of 80-90%. However, the median overall survival of these patients is limited. We evaluate the impact of co-morbidities on patient survival and whether a subset of patients who may not benefit from SBRT can be identified.

Methods
Patients treated on a prospective protocol at a single cancer center with SBRT for T1-T2N0 NSCLC from Oct 2004-May 2012 were evaluated. The most common doses delivered were 48Gy/4fr and 54Gy/3fr. The presence of significant medical co-morbidities including cardiac disease, COPD, cerebro-vascular disease, diabetes, previous pneumonectomy and oxygen dependence were recorded at baseline. Patient, tumor, and treatment data as well as outcomes were prospectively collected. Log rank tests were performed for survival analysis and chi squared tests used to analyze deaths within 1 year from radiotherapy treatment (D<1y). Cancer specific deaths (CSD) were defined as any death following a recurrence of the previously treated NSCLC.

Results
There were 279 patients identified, 134 female (48%) and 145 male (52%). The median age was 76 years (range 48-93). The performance status was ECOG 0 in 87 patients (31%), ECOG 1 in 127 patients (46%), ECOG 2 in 53 patients (19%) and ECOG 3 in 9 patients (3%). There were 212 (76%) with T1 tumors, the remainder (24%) T2 tumors. The median follow up was 1.3 years. At last follow up, 111 patients (40%) had died, including 42 (15%) patients with D<1y. Of all deaths, 25 (22.5%) were CSD, the remainder from other causes. There were 222 patients (80%) identified as having a significant co-morbidity, collectively these conditions did not influence deaths from any cause (DAC) or CSD. The presence of cardiac disease (N=67) led to an increased risk of DAC (HR 4.1, p = 0.04) but not CSD (HR 1.2, p=0.28). These results were more pronounced for D<1y, patients with cardiac disease having increased D<1y, (HR 7.34, p=0.007), but not CSD<1y, (HR 2.9, p=0.09). Other co-morbidities were not correlated of survival. ECOG status was correlated with both DAC (HR 15.1, p=0.005) and CSD (HR 9.3, p=0.05).

Conclusion
The presence of respiratory and vascular co-morbidities should not necessarily preclude a patient from receiving SBRT. ECOG status and prognosis from a cardiac point of view may be associated with poorer overall survival at 1 year and should be considered when assessing a patient’s suitability for SBRT.

Background
There has been recent interest in dose escalation in LA-NSCLC, with the aim to improve both loco-regional control and overall survival. Attempts to dose escalate CT-defined volumes for radiotherapy (RT) for LA-NSCLC have been limited due to organ at risk (OAR) toxicity. We investigated the potential for adaptive dose-escalation to PET-defined volumes, using 4DPET/CT scans acquired prior to and during a course of radical chemo/RT (CRT).

Methods
This single institution study prospectively enrolled patients with NSCLC receiving CRT to a dose ≥60Gy, delivered in daily 2Gy treatments. 4DPET/CT scans were acquired prior to (week 0) and at weeks 2 and 4 during RT. RT was delivered using the intensity modulated RT (IMRT) plan developed from the week 0 scans. Three alternative dose escalated IMRT plans were developed offline based on the week 0, 2 and 4 scans. The PET avid primary (PET-T) and nodal disease (PET-N) volumes were auto-contoured using the 50%SUV~max~ metric. PET-T and PET-N were dose escalated to as high as possible while respecting OAR constraints and ensuring coverage of the clinical plan PTV. The D95% and D~max~ of the PET-T and PET-N were calculated and compared between week 0-2-4.

Results
Thirty-two patients were recruited, with 27 completing all scans. Sixteen patients were stage IIIA (60%), 9 were IIIB (33%) and 2 were IIA (7%). Eight patients (30%) had been prescribed a clinical dose of 60 Gy, 17 (63%) had 66 Gy, 1 patient 70Gy and 1 patient 74Gy. 25 patients (93%) were boosted successfully above the clinical plan doses at week 0; this reduced to 23 (85%) at week 2 and 20 (74%) at week 4. For all weeks combined, the D95 for PET-T was higher than that delivered to clinical PTV by a median of 16.2 Gy (4.2-37.4Gy). The D95 for PET-N exceeded that delivered to clinical PTV by 13.4Gy (6.8-29.7Gy). The median D95% to the PET-T at week 0, 2 and 4 were 74.4 Gy, 75.3Gy and 74.1Gy respectively. The median D~max~ to PET-T at week 0, 2 and 4 were 85.9Gy, 83.8Gy and 81.2Gy. The median D95% to PET-N at week 0, 2 and 4 was 74.3Gy, 71.0Gy and 69.5Gy. The median D~max~ to PET-N at week 0, 2 and 4 were 82.7Gy, 82.5Gy and 78.9Gy.

Conclusion
Using 4DPET/CT derived volumes, it is feasible to dose escalate a majority of patients, either at the onset or during RT. Though the PET-T was able to be escalated to higher doses than PET-N, nodal disease can still be boosted to significant doses. More patients were able to be dose escalated at the onset of RT; however mid-RT dose escalation allows the additional potential for adaptation.

Background
To investigate the efficacy of single dose 8-Gy palliative chest re-irradiation (PCRI) in metastatic non-small cell lung cancer (M-NSCLC) patients with painful recurrences in previously irradiated thoracic region.

Methods
Clinical data of 63 M-NSCLC patients, who received single dose 8-Gy PCRT due to painful thoracic recurrences from February 2007 to June 2010 were retrospectively analyzed. All patients had previously received upfront definitive 60-66 Gy concurrent chemoradiotherapy (C-CRT), and 52 of them had also received salvage chemotherapy. Primary endpoint was change in visual analogue score (VAS), and secondary endpoints were time to lowest VAS record and duration of pain control.

Results
Treatment was well tolerated with only 3 (4.8%) grade III radiation-induced pneumonitis. For all patients median, 1-, 2-year survival were 9.2 months, 28.4%, and 12.3%. Median pre-PCRI and minimum achievable post-PCRI VAS values were 6.7 (range: 5-8) and 3.4 (range: 0-8), and the decline in VAS values was statistically significant (p<0.001). Objective response defined as reduction of at least 2 points in VAS value was achieved in 54 (85.7%) patients. Median time to lowest VAS and duration of pain control were; 27 (95% CI: 21 - 33) days and 7.1 (95% CI: 6.3 - 7.9) months, respectively.

Conclusion
Single dose 8-Gy PCRI is safe and highly efficient in palliating moderate to severe pain in locoregionally recurrent M-NSCLC patients, who have previously received upfront definitive C-CRT

Background
Aim of this study was to compare surgical resection and radiosurgical treatment modalities directed to brain metastasis (BM) in terms of survival outcomes in newly diagnosed non-small cell lung cancer (NSCLC) patients with synchronous single brain metastasis (SSBM).

Methods
Medical records of medically fit NSCLC patients with SSBM treated at our department were retrospectively evaluated. A total of 64 patients were staged with PET-CT besides conventional staging tools. TRT to a total dose of 66 Gy in 2 Gy fx was delivered with 2 cycles of concomitant cisplatin –based chemotherapy (CT) following surgery+30 Gy whole-brain RT (WBRT) (n:33) or 30 Gy WBRT+stereotactic radiosurgery (SRS) (n:31) for their BM.

Results
Pretreatment patient characteristics were given in Table. Whole study population could complete planned treatment to their BM and thoracic primaries. At median 20.7 months (7.1-58.7) of follow-up, Median overall (OS), neurological progression-free (NPFS), locoregional progression-free (LRPFS) and progression-free survival (PFS) were 23.4, 22.8, 14.9, ve 9.4 months, respectively. There was no statistically significant survival difference between two groups (p>0.05 for each survival parameter) (Figure 1). On univariate analyses, patients with ECOG performance status of 0-1 (p<0.001), younger than median 50 years (p=0.031) and with no weight loss (p=0.001) revealed superior OS. On multivariate analyses performed with these factor, all factors except age retained their independent prognostic value (p<0.05 for each parameter). Table 1. Pretreatment patient characteristics Figure 1Figure 2

Conclusion
Results of this study has shown that, in line with the literature, surgical or radiosurgical treatment modalities directed to SSBM in NSCLC patients could not demonstrate any superiority over each other. Therefore, management of BM should be tailored according to the facilities available, and patient or disease-specific conditions. Moreover, although larger prospective data are needed, survival outcome in such patients similar to locally advanced patients indicates the potentially curative role of definitive CRT.

Background
There are no randomized trials comparing CT versus FDG-PET/CT based radiotherapy planning for lung cancer or any other disease site.Based on phase II studies, a convincing body of data has emerged within the last 10 years incorporating the use of FDG-PET scans for radiotherapy planning in lung cancer.Published data comparing changes in volume measured with FDG-PET/CT to CT alone indicates that the magnitude of treatment volume changes with incorporation of PET in radiotherapy planning for lung cancer varies from 27% - 100%. However, volumetric data only provides information on changes in size and does not account for potential changes in position and shape of the target, thereby affecting variability of the GTV in NSCLC. In this study we describe influence of FDG-PET/CT or CT alone for the primary and mediastinal nodal disease in radiation planning for stage III NSCLC in relation to changes in volume, position and overlap of the GTV. We also report the interobserver variability between radiation oncologists for FDG-PET/CT and CT alone-derived GTV. In addition to volumetric measurements, we have used a vector displacement method for three-dimensional (3D) positional analysis. We have further evaluated the overlap of the primary and nodal GTV with Dice Similarity Coefficient (DSC) method

Methods
Patients (n=29) underwent Three Dimensional Conformal Radiotherapy (3DCRT) planning by three different radiation oncologists. Simultaneous co-registered CT and FDG-PET/CT were obtained in the same treatment planning position. Gross Tumor Volume (GTV) for lung tumor and mediastinal lymphadenopathy was contoured and compared for changes in volume and position. Interobserver variability was determined using three-dimensional analysis with vector displacement and the Dice Similarity Coefficient (DSC). Concordance for the number of lymph nodes contoured was performed.

Results
Mean GTV for lung tumor with FDG-PET/CT and CT alone was 62.0 cm[3] and 74.64 cm[3], respectively (p=0.0005), resulting in a 17% reduction by FDG-PET/CT. Mean GTV for mediastinal lymphadenopathy was 15.72 cm[3] and 19.02 cm[3] (p=0.084), equalling a 17% reduction GTV for FDG-PET/CT. Mean vector displacement of lung tumor was 2.0 mm with FDG-PET/CT versus 7.1 mm with CT alone (p = 0.0016), equating to a 3.6 fold reduction in interobserver variability of position. Mean vector displacement of the mediastinal lymphadenopathy was 1.53 mm with FDG-PET versus 10.2 mm for CT alone (p= 0.0005), resulting in a 6.7 fold reduction in interobserver variability. Median Dice Similarity Coefficient (DSC) for the primary GTV contours was 0.87 for FDG-PET/CT and 0.74 for CT alone. For the nodal GTV DSC were 0.79 and 0.59, respectively. Physician agreement on the number of lymph nodes contoured was 15/29 on CT and 27/29 patients for FDG-PET/CT. Only two of the three physicians agreed on the number of lymph nodes contoured for CT alone in 12/29 versus only 2/29 patients for FDG-PET/CT (p=0.0018).

Conclusion
FDG-PET/CT significantly reduces mean lung tumor and mediastinal nodal GTV, is more precise for size and position in defining target volumes, and reduces interobserver variability. There was greater agreement for the number of lymph nodes contoured on FDG-PET/CT compared to CT alone.

Background
To evaluate the clinical outcomes following SABR to early-stage NSCLC and pulmonary metastases (Mets) from NSCLC, including potential predictive factors and a focus on chest wall (CW) toxicities.

Methods
From our prospective lung SBRT database, we identified 240 NSCLC lung tumors in 209 patients treated with SABR between 2008 and 2011. 228 were primary NSCLC and 12 were NSCLC Mets. The institutional policy was to deliver 48-52 Gy in 4 fractions (fx) for peripheral tumors and 50 Gy in 5 fx for central tumors. Local control (LC) was defined as the absence of recurrence within or ≤1 cm beyond the planning target volume (PTV). Lobar LC was defined as absence of recurrence within the same lobe. All tumors were categorized as adjacent to the CW (PTV touching the CW) or not, to determine if this localisation can predict for rib fracture. Age, comorbidities, lobe of the target lesion, stage, tumor size, maximal tumor standardized uptake value (SUV), total dose and various dose-volume histogram metrics were also evaluated for their predictive significance.

Results
The median follow-up was 20.8 and 18.1 months (range, 0.2–52.1 and 15.3–46.2) for primary NSCLC and Mets, respectively. 168/240 tumors (70%) were biopsy-proven NSCLC. Of the 228 primary NSCLC, the majority were stage I (192/228; T1a=136, T1b=21, T2a=35). Among all primary tumors, the 2-year LC was 94.8%. LC did not differ between biopsy-proven or presumed tumor, and did not differ between primary NSCLC and NSCLC Mets. For the whole cohort, only the presence of a respiratory comorbidity predicted for better local control (p=0.021) on multivariate analysis. In stage I NSCLC, the 2-year LC, lobar LC, regional control, distant control (DC) and overall survival (OS) were 95.5%, 91.3%, 89.3%, 74.7% and 77.9%, respectively. The 2-year OS rates for non-stage I primary NSCLC and Mets from lung were 69.2% and 48.6%, respectively. The 2-year DC rates in all primary NSCLC and NSCLC Mets were 73.6% and 20.8%, respectively. Among primary NSCLC, upper lobe tumors predicted for a better DC compared to lower lobe tumors (p=0.009). Of the 240 lung tumors, 132 (55%) were adjacent to the CW and had a significant greater risk of rib fracture. The 2-year risk of fracture was 29.6% versus 8.1%, for tumors adjacent and not adjacent to the CW, respectively (p<0.001). The median time to fracture was 15.9 months. 51% of rib fractures were symptomatic. There was a suggestion that a higher conformity index (ratio of the 95% isodose volume to the PTV) predicted for a higher risk of rib fracture (p=0.067).

Conclusion
The excellent LC rate post-SABR seems similar in primary NSCLC and in NSCLC Mets, and having a respiratory comorbidity predicted favourably for LC. Patients with lower lobe tumors had higher risk of distant relapse. Patients with tumors adjacent to the CW are at significantly greater risk of rib fracture post-SABR and should be well informed prior to treatment.

Background
In February 2012 a specific GPA scale for assessing the prognosis of BM in NSCLC was published (JCO, 2012; 30(4): 419-426). Retrospective series have showed longer survival for patients with cerebral metastases and EGFR sensitizing mutations (m) compared with those without them.

Methods
Charts from EGFRm NSCLC p with BM were reviewed. We classified p with the lung-GPA prognosis scale at the time of diagnosis of BM and compared the expected GPA overall survival (GPA-OS) with the Observed OS (OS) for each group.

Results
24p were diagnosed from January 2007 to December 2012. 15 p were treated with whole-brain radiotherapy (WBRT), total dose 30Gy (2p GPA 0-1; 10p GPA 1.5-2; 3p GPA 2.5-3), 1 p with stereotactic radiosurgery (SRS), mean dose 18Gy (GPA 3.5-4), 1p with SRS and WBRT (GPA 2.5-3) and 2p with surgery (S) and WBRT (1p GPA 2.5-3; 1p GPA 3.5-4). 5p receive no treatment (3p GPA 0-1; 2p GPA 1.5-2). Outcomes were [Expected median (m) GPA-OS vs Observed m OS]:

GPA	Expected m GPA-OS(months)	Observed m OS (months)	Number of p
0-1	3.4	<1	5
1.5-2	4.7	6	12
2.5-3	8.8	34	5
3.5-4	14.8	Not reached (100% OS at 40 months)	2

Two p did not receive any EGFR tyrosine kinase inhibitor (TKI). 5p were receiving an EGFR TKI when they were diagnosed with BM and 12p were treated with EGFR TKI after the diagnosis of BM. Sixteen p have died. Further details of p characteristics and treatment received will be presented at the meeting.

Conclusion
Observed m OS of p with EGFR m-NSCLC and brain mets was much longer than expected by GPA-OS. The GPA subsets might predict prognosis in patients with mutated tumors as well.

Background
Lung cancer remains a challenging site to treat in radiotherapy. For tumour delineation CT combined with FDG PET is the current gold standard for delineating lung cancer volumes. MRI has been utilized in a limited number of cases such as superior sulcus tumours. However its use has been limited due to reduced MRI signal as a result of low proton density in lung, inhomogeneity of the magnetic field in lung and cardiac and respiratory motion. As technology improves, the utility of MRI in imaging lung cancer has become plausible. The aim of this literature review was to identify evidence to support the use of MRI in lung cancer imaging for radiotherapy.

Methods
Pubmed and Google Scholar were used to identify literature on MRI in Lung cancer using the keywords Lung Cancer, MRI, MR, thoracic imaging and radiotherapy. Articles were limited to human and phantom subjects. A total of 22 articles were identified between 1995 and 2013 and reviewed to assess the potential of MRI for lung radiotherapy imaging.

Results
Eighteen studies were performed on 1.5T and two on 3T magnets with magnet strength not specified in two studies. Gradient echo sequence and TrueFISP were most common sequences utilised. Thirteen articles examined tumour motion and nine concentrated on tumour volume analysis. One study illustrated minimal geometric distortion and inter-cycle reproducibility of tumour volume on free breathing MRI sequence. The possibility of defining tumour internal target volume for radiotherapy treatment was demonstrated in three studies Two further studies demonstrating variability of lung tumour motion based on location and variability of motion between tumour and non-tumour bearing lung. There was conflicting evidence on pulmonary node detection based on results from two studies. Four comparative studies with FDG PET and functional MRI sequences demonstrated diffusion weighted image (DWI) had higher specificity for lesion detection in the presence of inflammation. Correlation was seen between SUV and DWI value. Two studies demonstrated the potential for radiotherapy planning based on biological function by avoiding functional lung tissue.

Conclusion
This literature review indicates that improvement in MRI technology has overcome some of the initial limitations of utilising MRI for lung cancer imaging. MRI can not only provide the morphological information required for identifying lung cancer based on anatomical sequences but also functional information to identify both tumour activity and surrounding healthy or pathological structures. There is potential for MRI to be utilised in the clinical setting for defining tumour volumes for radiotherapy planning and in evaluating tumour response during and after treatment. However further research is required to determine protocols with defined standard sequences specific for lung cancer imaging.

Background
Multiple primary lung cancers (MPLC) are not an uncommon clinical presentation, with an incidence in the surgical literature of 1-8%. ESMO guidelines state that synchronously detected lesions should be treated as multiple primary tumors, and a curative approach for both lesions has been associated with improved survival in the surgical series. However, many patients with MLCP are elderly and have multiple co-morbidities, which can render them unfit to undergo surgery for both lesions. We analyzed clinical outcomes in such patients who were treated with SABR.

Methods
SABR was performed in 62 patients diagnosed with MPLC at the VUmc from 2003 – 2012. Staging included a mandatory FDG-PET scan, and all patients were discussed in a multi-disciplinary tumor board. A pathological diagnosis was available for both lesions in 3%, and for one lesion in 48%. Invasive nodal staging was performed in 13% of patients. SABR was used as a single modality for both lesions (n=56), or in combination with surgery for the second lesion (n=6). SABR was delivered to a total dose of 54-60 Gray (Gy) in 3-8 fractions, depending on tumor size and location. Clinical outcome, including survival, patterns of relapse and toxicity (CTC v4.0) was evaluated. A sub-analysis was performed for ipsilateral and bilateral lung lesions.

Results
Median overall survival was 31 months, with an actuarial survival of 56% at 2 years. Overall lesion local control rate was 84% at 2 years. Local control correlated significantly with number of fractions (p=0.013) and lesion location (p=0.004) on univariable Cox regression analysis. Lesion control at 2 years for bilateral lesions was 92% versus 74% for ipsilateral lesions (p=0.009). Regional failures at two years were observed in 13% (n=6) of all patients, and in 0% versus 31% in patients with respectively bilateral and ipsilateral lesions. Of the patients who developed a subsequent regional recurrence, only one had undergone EUS/EBUS prior to treatment, and others did not as pre-treatment FDG-PET-scans showed no nodal uptake. Distant failures were observed in 27% of all patients, at two years. No grade ≥3 early toxicity was observed. Late grade 3 toxicity was reported in 3 patients (5%), consisting of pneumonitis (n=1), rib fracture (n=1) and chest wall pain (n=1). No grades 4-5 late toxicity was reported.

Conclusion
Curative treatment of MPLC using SABR, either alone or combined with surgery, can lead to long-term survival with limited toxicity. The disappointing local control rates observed after SABR for ipsilateral double lesions merits further investigation. The higher rate of nodal recurrences in patients presenting with multiple ipsilateral lesions suggests that systematic nodal staging may be appropriate in such cases.

Background
As part of a feasibility study for a Hadron Therapy Centre in Belgium, an economic evaluation was performed to assess the potential cost-effectiveness of proton radiotherapy (PT) delivered concurrently with chemotherapy for locally-advanced non-small cell lung cancer (LA-NSCLC), compared to concurrent chemoradiotherapy employing photon therapy, either 3D-conformal (3D-CRT) or intensity-modulated (IMRT) radiotherapy.

Methods
A Markov decision-analytic model was developed using Microsoft Excel 2007 software. The model was defined for a time horizon of 10 years, allowing patients to transition between 5 health states (treatment, controlled disease, loco-regional progression, distant progression and death) using transition cycles of 3 months. Transition probabilities were derived from photon and proton literature on LA-NSCLC and from nationally available data. Results were to be expressed in cost per (quality adjusted) life years (LY and QALY). The occurrence of grade 3 toxicity or higher in terms of radiation pneumonitis, radiation esophagitis/dysphagia and pulmonary radiation fibrosis was accounted for in the calculation of QALYs. Treatment costs of the standard 3D-CRT and IMRT treatments were obtained from an Activity-Based Costing (ABC) exercise in Belgian radiotherapy centers (KCE report 198). Similarly, the cost of PT was calculated using ABC in different technical (proton-only vs. combined proton and carbon-ion center) and financing (private vs. public) scenarios. Toxicity and follow-up costs were based on literature evidence but adapted to the Belgian context.

Results
The base case analysis used the scenario of a publicly financed combined center. The survival curves generated by the model demonstrated it accurately predicts survival of published literature and of the Belgian Cancer Registry. Compared to 3D-CRT res. IMRT, PT generates 0.837 res. 0.664 extra LYs and 0.549 res. 0.452 extra QALYs. When combined with the higher cost (18,875€ res. 14,257€), this translates into an incremental cost-effectiveness ratio (ICER) of 22,543€/LY for PT compared to 3D-CRT and of 21,489€/LY compared to IMRT. Expressed in cost per QALY, the ICERs amount to 34,396€/QALY and 31,541€/QALY respectively. Assessing the effect of different technical scenarios and/or financing methods, the ICER ranges between 21,489€ to 53,685€/LY and 31,541€ to 78,873€/QALY, with the highest figures found for a combined center with private financing. One-way sensitivity analyses reveal that the results are most sensitive to the effect of proton therapy on disease control, loco-regionally as well as at distance, and to the quality of life pre-progression.

Conclusion
Based on a public financing scenario for a combined center, PT delivered concurrently with chemotherapy is found borderline cost-effective in the Belgian health care context, compared to the best available photon radiotherapy alternatives. These results are however highly sensitive to the cost of PT (hence the financing scenario) and the expected clinical advantage of PT, both in terms of improved survival and decreased long-term toxicity impacting on quality of life. Apart from clinical appropriateness and budgetary possibilities, such results support decision-making on the feasibility and desirability of introducing hadron therapy in Belgium.

Background
Concomitant chemo-radiotherapy (CRT) has been shown to be superior to single modality radiotherapy (RT) and to neoadjuvant chemotherapy (NeoCT) followed by RT. In this retrospective study we report the influence of age on survival in two groups of patients with locally advanced radiotherapy treated from 1995 to June 2012: 1) NoCRT-group: RT +/- NeoCT without CRT and 2) CRT-group: NeoCT followed by CRT.

Methods
Data for 446 patients that completed 3-D conformal RT or IMRT in planned doses of 60-66 Gy at 2 Gy/F without elective nodal irradiation was obtained. CRT was used for 114 patients with the regimens: weekly Docetaxel (N=43), Carboplatin-Vinorelbine (N=70) and Cisplatin-Vinorelbine (N=1). The NoCRT group consisted of 332 patients

Results
Figure 1 The median and 5 year survival was 17.2 months and 15% for NoCRT, and 21.4 months and 29% for CRT (p<0.02). The data was analyzed in the age groups ≤65 and >65 years. For NoCRT, no differences in survival was found (17.2 vs. 17.1 months), but in the CRT group significant longer survival was found in the younger age group, 29.2 vs. 20.0 months in the older group (p=0.03). CRT was the only significant factor among patients ≤65, while CRT was an insignificant factor among patients >65. The patients treated with CRT had significantly lower frequency of local relapse (33%) compared with patients treated without (46%, p = 0.016), while the rate of distant metastases was unaffected by the use of CRT. Patients ≤65 years had a low complication death rate (1-2%) independent of CRT, while the complication death rate among patients >65 years was 6-8%, also independent of CRT.

Conclusion
Patients ≤65 years had a very positive effect of CRT while CRT did not influence survival in older patients. The reason for this is unknown

Background
In this study we evaluated the feasibility and outcome of stereotactic body radiotherapy (SBRT) for lung tumors treated with helical tomotherapy.

Methods
Between December 2007 and January 2013, 26 Patients with a total of 49 lung tumors were treated with helical tomotherapy at a median 75 Gy of 10 fractions over 2 weeks. Thirteen lung lesions of primary lung tumors (group 1) and 36 of recurrent or metastatic lung tumors (group 2) were analyzed. Three patients received re-SBRT due to local recurrence after first SBRT were included. (Total 29 cases were analyzed). Radiation pneumonitis was graded according to the Common Terminology Criteria for Adverse Events V 4.0.

Results
The patients’ age was median 72 years (range, 47-82). Median follow up was 16 months (range, 2-57 months). Total 13 cases (45%) showed progressive disease after SBRT, and one of them was distant metastasis. Of 12 thoracic recurrences, in-field recurrences were noted in 5 patients and out-field recurrences in 7 patients. In group 1 (13 cases), there were 3 recurrences with 1 in-field thoracic recurrence, 1 out-field thoracic recurrence and 1 distant metastasis. Grade≥2 radiation pneumonitis was noted in 8 patients (30%). One patient died due to radiation pneumonitis at 2 months follow up after second SBRT.

Conclusion
The preliminary findings of our study suggest SBRT with helical tomotherapy is feasible for lung tumor treatment. But further studies with more patients and longer follow-up duration are required.

Background
Radiotherapy (RT) is commonly used for palliation of symptoms in patients with lung cancer. However, the time before onset of relief may be several weeks. It is therefore of interest to study how many patients who die within few weeks after the initiation RT since these patients are not likely to benefit from the treatment.

Methods
This is a retrospective review of the overall survival rate of 293 patients with lung cancer who received palliative RT in 2010 at our institution. If patients had received more than one course of RT, only data from the last course was included in the study.

Results
The planned fractionation (F) schedules were 8Gy/1F(N=34), 10Gy/1F(N=34), 20Gy/4-5F(N=16), 20Gy/10F(N=2), 25Gy/5F(N=44), and 30Gy/10F(N=167). The median survival was 11.9 weeks. The two-week mortality rate 10.2% and the four-week mortality rate was 22.2%.

	N	2 week mortality	4 week mortality	Median survival
8-10 Gy/1F	64	15 (23%)	26 (41%)	5.6 w
20-25Gy/4-5F	60	8 (13%)	20 (33%)	6.0 w
20-30Gy/10F	169	7 (4%)	19 (11%)	21.7 w
Total	293	30 (10%)	65 (22%)	11.9 w

Among the patients planned for 4-5F, only one patients (2%) did not receive the planned number of fractions, while 13 patients (8%) of the patients planned for 10F did not receive the planned number of fractions. In a logistic regression analysis using 2-week mortality as endpoint, only planned number of fractions <10F and bone metastases as indication of RT were significant factors for poor survival, while gender, CNS metastases, age, performance status and histology were not. The results are strikingly similar to Gupta et al. Radiother. Oncol. 2012;103 suppl.1(PO-0744).

Conclusion
Although 10% of the patients receiving palliative RT die within 2 weeks from the start of RT, the results indicate that prolonged treatment regimens were reserved for patients with longer life expectation. Nearly 1 of 4 of patients receiving 1F died within two weeks after start of RT indicating that the RT was futile in these patients.The radiation oncologists seem to expect a palliative effect of RT for bone metastases even in patients with short life expectations.

Background
Radiotherapy treatments of locally advanced NSCLC patients are associated with poor local control and low overall survival rates. Local recurrence often occurs within the initial primary tumour. Therefore, higher doses to locally advanced NSCLC tumours with conventional fractionation are required in order to increase the local control. However, current tumour dose levels are limited by the toxicity levels of the surrounding normal tissue, e.g. healthy lung tissue and mediastinal region. This study presents the concept of clinically applicable inhomogeneous dose plans that distribute higher doses in the tumour without compromising the expected lung toxicity, not only for isolated lung tumours, but also for patients with involved lymph nodes.

Methods
Twenty consecutive NSCLC patients previously treated with conventional radiotherapy treatment with a prescribed dose of 66Gy/33F were included in this planning study. The patients were staged T1b-T4 N0-N3 with a median tumour size of 56.7 cm[3] (range 3.2-399.2 cm[3]). Highly modulated IMRT plans were used for treatment with standard dose coverage of 95%-107% of the prescribed dose. For each patient, a new dose-escalated treatment plan was created with the same mean lung dose as obtained in the standard plan using an inhomogeneous dose distribution with minimum dose coverage of ≥95% of the prescribed dose. The maximum tumour dose was only limited by conservative tolerance levels of the surrounding healthy tissue: maximum doses of 45 to spinal canal, 66 Gy to oesophagus, and 74 Gy to less than 1 cm[3] within the mediastinal region. Two different tumour control probability (TCP) models were used to evaluate the homogeneous and inhomogeneous treatment plans. Furthermore, in order to estimate the reliability of the calculated doses in comparison to actual delivered doses, the treatment plans were re-calculated in 4D by accounting for uncertainties arisen from respiration, delineation, setup, and baseline shift.

Results
Dose escalation was possible for all patients. TCP values increased approximately 15 and 10 percentage points based on calculations related to the GTV and CTV, respectively, from an initial level of 18%. This increase in expected local control was obtained without increasing the mean lung dose. However, small increases in maximum doses to the mediastinum were observed: 2.5 Gy for aorta, 4.4 Gy for the connective tissue, 1.6 Gy for the heart, and 2.6 Gy for trachea and bronchi. The increase in TCP predictions from 4D calculations differed only slightly from the corresponding 3D calculation.

Conclusion
Increased target doses and TCP values using dose-escalated inhomogeneous dose distributions could be achieved for all patients, regardless of lymph node involvement, tumour stage, location, and size. These new treatment plans have the potential to increase the local tumour control with 10-15 percentage points without increasing the mean lung dose, which is often regarded as a measure for the expected lung toxicity. The conservative dose constraints used for the organs at risk ensures clinical applicable treatment plans that can be implemented today. Based on these promising results, a national randomised phase III study is under development.

Background
The aim of this study is to determine the interobserver variability between thoracic surgeons and radiation oncologists in defining and delineating target volume for postoperative radiotherapy for patients with non small cell lung cancer (NSCLC). These patients were offered postoperative RT due to microscopic non-radical operation.

Methods
Between 2002 and 2010, 48 NSCLC patients were offered postoperative radiotherapy at Department of Oncology, Odense University Hospital due to microscopic non-radical operation. Two thoracic surgeons (S1 and S2) and two clinical oncologists (O1 and O2) were retrospectively asked to delineate clinical target volume (CTV) on the 3D CT scanning used for radiotherapy planning in each patient. Instruction were given to include the non-radical site on basis of surgical-, pathological-, and radiological reports. The delineation was done independently by each physician. There are no local guidelines for delineation postoperative volume after non radical microscopic operation. The spatial volume discrepancy between the different physicians was the end-point.

Results
The mean volumes were very different between the physicians: S1 20.5 (SD 17.4) cm[3], S2 21.5 (28.1) cm[3], O1 14.4 (20.5) cm[3], and O2 32.9 (SD 35.6) cm[3]. A large spatial volume discrepancy between the different physicians was observed as well. Mean discrepancies were O1-O2 18.5 (SD 25.3) cm[3 ](p<0.0001), O1-S1 6.1 (SD 21.4) cm[3] (p=0.06), S1-S2 1.3 (SD 22.9) cm[3] (p=0.7). Mean Soerensen-Dice index O1-O2 0.27 (SD 0.19), O1-S1 0.27 (SD 0.16), S1-S2 0.29 (SD 0.18). There was a reasonable overlap in 25 patients (52%) between all 4 physicians. In another 9 cases (19%) 3 physicians had a reasonable overlap and one “outlier”. Figure 1 illustrates one case with reasonable overlap, and another case with one “outlier”.

Conclusion
Figure 1 Several conclusions can be drawn from this study. At first it is very challenging to define postoperative volume, when there is no gross tumor volume as is the case when RT it is due to microscopic non-radical operation. Secondly, it is important to have guidelines in order to define the approximate size of the CTV, this in context of the large difference in the size of the volume between the oncologists. Thirdly, surgeon and oncologist should delineate the postoperative target volume in collaboration to ensure the correct conclusion is drawn on the basis of operation and pathological reports, in order to hit the right target and reduce irradiated volume.

Background
Patient (pt) safety is of utmost concern to radiation oncologists. Pemetrexed (Pem) is an effective and well-tolerated treatment for advanced nonsquamous NSCLC. The safety of palliative radiation (XRT) during Pem treatment was studied in this subset of pts in the PARAMOUNT trial.

Methods
In PARAMOUNT, a randomized, double-blind study, 939 pts received 4 cycles of induction Pem (500 mg/m[2]) + cisplatin (Cis) (75 mg/m[2]) on day 1 every 21 days. Patients without progressive disease (PD) and with an Eastern Cooperative Oncology Group performance status (ECOG PS) of 0/1 (n=539) were then randomized (2:1) to maintenance Pem (500 mg/m[2], day 1) + best supportive care (BSC) (Arm A) or placebo + BSC (Arm B) until PD. Best supportive care (BSC) was defined as treatment without a specific antineoplastic regimen and included palliative XRT to extrathoracic structures. Safety was assessed via the incidence of adverse events (AEs) by maximum grade (Gr; CTCAE, v3).

Results
The 55 pts who received palliative XRT to extrathoracic structures during treatment had stage IV nonsquamous NSCLC. The majority of pts were male (58%), with an ECOG PS of 1 (75%). Patients’ median age was 61 yrs (range, 32-74) yrs, with 13% of pts ≥70 yrs. The most common location irradiated was bone (43/55 pts). Non-bone locations were: lymph node (3), mediastinum (2), chest (2), and adrenal gland, intraocular, lung, brain, and abdomen (1 each). Forty-five pts received XRT during Pem+Cis induction, 3 of whom also received XRT during maintenance. Seven pts (Arm A) and 6 pts (Arm B) received palliative XRT during maintenance. Total XRT doses ranged from 8-66 Gy. The time interval between day 1 of last chemotherapy cycle and the start of palliative XRT ranged from 0-28 days. Of 55 pts, 12 (22%) had ≥1 AE(s) during XRT considered possibly related to Pem and/or XRT (Table 1). All pts except 1 experienced the AE during induction. The most common AE was Gr 2 anemia. Three pts had Gr 3/4 anemia. Five pts had nonhematologic toxicities. One pt in Arm B, who received a total dose of 20 Gy in the hip during maintenance treatment, had pneumonitis. No AEs were reported for pts who received palliative XRT during Pem maintenance treatment.

Table 1: AEs during palliative XRT or within 2 weeks after the end of the last fraction in both phases of the PARAMOUNT trial.
Pts receiving palliative XRT (N=55)
Patients with AEs during induction and/or maintenance (n=12, 22%)
Toxicity	Gr 1, n (%)	Gr 2, n (%)	Gr 3-4, n (%)
Hematologic
Hemoglobin	1 (1.8)	4 (7.3)	3 (5.5)
Leukocytes	0	2 (3.6)	1 (1.8)
Platelets	0	1 (1.8)	0
Nonhematologic
Rash/dermatitis	1 (1.8)	1 (1.8)	0
Rash/desquamation	1 (1.8)	1 (1.8)	0
Pneumonitis	0	0	1 (1.8)*
*Pneumonitis was the only event reported for a pt during the maintenance phase. The pt was assigned to placebo.

Conclusion
Conclusions: In PARAMOUNT, palliative XRT is well tolerated and can be safely administered at low and high doses during Pem+Cis chemotherapy or Pem monotherapy to pts with advanced nonsquamous NSCLC.

Background
Traditionally, the same radiotherapy dose is prescribed to the primary tumor (PT) and involved lymph nodes (LN) for locally advanced NSCLC patients, independent of volume and metabolic activity of each lesion. On the other hand, the PT and lesions with a high [18]F-FDG uptake are often believed to have a higher recurrence rate. The purpose of this study was therefore to find prognostic parameters for the lesion control probability in locally advanced NSCLC.

Methods
A total of 279 patients treated between 2007 and 2011 with 24×2.75Gy IMRT and concurrent daily low dose cisplatin were included in this retrospective analysis. Patients had a staging FDG-PET scan within 6 weeks prior the start of treatment. For follow-up, CT thorax was performed 4-6 weeks after treatment and at 3-monthly intervals chest x-ray or CT scans were made up to 2 years after CCRT. Medical records of the patients were retrospectively reviewed. The PT and/or LN progression were classified based on follow-up records and scans by two physicians. The volume of each separate lesion on the planning CT and their SUVmax from the pre-RT staging PET scan were then tested as prognostic factors for disease progression using Cox proportional hazard model with patient as random effect. Data were analyzed using R, package “coxme”.

Results
A total of 291 PTs (8 patients had no PT, 252 had 1 PT, 18 had 2 PTs, 1 had 3 PTs) and 627 LNs (59 patients without LN, 57 had 1 LN, 57 had 2 LNs and 106 had >=3 LNs involved) were analyzed. The majority (92%) of the patients had TNM stage III and 86% patients had ≥N2. At a median follow-up of 30 months (95%CI 27-35) and median OS of 25 months (95% CI 21-29), 47 PTs had progression and 14 LNs had progression, 38% patients had systemic relapse. The log(Vol), SUVmax and lesion type (TP vs. LN) were each significant (p<0.001) as prognostic factor for progression in the univariate cox regression. In the multivariate analysis, log(Vol) remains as a significant predictor (p<0.001) with a trend toward significance for lesion type (p=0.07) (Figure 1). Figure 1

Conclusion
Our regimen of 66 Gy in 24fx results in excellent local-regional control in locally advanced NSCLC patients. The PT yields a significantly higher risk of progression than LNs. A larger lesion volume and higher SUVmax were associated with an increased risk of progression.

Background
Introduction: The treatment of locally advanced lung cancer with radiation therapy has traditionally utilised multiple beam conformal plans. On occasions, the curative intent is compromised due to excessive normal tissue dose, especially to healthy lung tissue. This necessitates a reduction in total delivered dose to a more palliative regimen. With the advent of Volumetric Modulated Arc Therapy (VMAT) at Nepean Cancer Care Centre (NCCC), the question arises of whether VMAT plans can be utilised in these circumstances in order to deliver curative tumour dose, whilst still delivering acceptable normal tissue doses. This study shows a planning comparison of the VMAT and traditional 3D Conformal Radiotherapy (3DCRT) planning techniques. Hypothesis: It is expected that VMAT plans will produce equivalent target volume coverage, whilst resulting in lower toxicities to surrounding healthy tissue, when compared to 3DCRT plans.

Methods
Method: This study was conducted retrospectively on three patients who had previously been treated with 3DCRT for stage IIIA or IIIB Non-Small Cell Lung Cancer at NCCC, and had undesirably high healthy lung doses when prescribed 60Gray (Gy) in 30 fractions. The original treatment volumes (planning target volume (PTV), clinical target volume/gross tumour volume) and total prescribed dose for each planning method was left unchanged for this study. The original 3DCRT plans utilised, on average, 4 static beams with 8mm field margins. Angles, wedges, weighting and energy were selected as a best fit to treat the target volume and avoid spinal cord irradiation and healthy lung tissue. Each VMAT replan comprised of one 352° arc, with 89 control points, delivered over a maximum of 90 seconds, with all radiotherapy dose delivered through this dynamic arc. This study compared the two planning methods based on PTV coverage, Conformity Index (CI), V20 (volume of normal lung receiving 20Gy) <30% and V30 (volume of normal lung receiving 30Gy) <20% of the combined healthy lung volume and mean lung dose.

Results
Results: Improved coverage of the PTV by 60Gy is achieved with VMAT, with an average of 92.5%, whilst 3DCRT plans achieved an average of 77.7% 60Gy PTV coverage. VMAT combined lung doses were reduced when compared to 3DCRT, with an average V20 of 29.9% for VMAT plans compared to 34.1% for conformal plans, and V30 of 18.4% for VMAT plans, compared to 27% for 3DCRT plans. Combined mean lung dose was also lower for VMAT plans, with an average reduction of 2Gy, with average mean lung dose of 18.01Gy for VMAT plans compared to 20.09Gy for 3DCRT plans. The CI of the D95% was improved in all three cases, with an average 100% CI of 1.01 for VMAT plans, and an average CI of 1.2 for 3DCRT plans.

Conclusion
Conclusion: VMAT is a viable planning method to achieve radical treatment intent to 60Gy in 30 fractions, adequate PTV coverage by D95% and improved healthy lung doses for patients with stage IIIA and IIIB NSCLC.

Background
Curative radiotherapy for non-small cell lung cancer (NSCLC) is usually delivered over a time period of 5-7 weeks. Tumour regression has been observed over the course of curative radiotherapy. Currently, the entire radiotherapy treatment is delivered based on the original tumour volume. An evolving approach is adaptive radiotherapy planning whereby the radiotherapy plan is modified during a course of treatment to account for tumour and patient changes. This has the potential to reduce the size of radiotherapy fields, allowing dose escalation and normal tissue sparing.

Methods
A cohort of 20 consecutive NSCLC patients receiving a curative course of radiotherapy and who had weekly kV cone beam computer tomography (CBCT) images was identified. The gross tumour volume (GTV) and anatomical reference points were delineated on all CBCT scans. Volume and positional changes were recorded and analyzed.

Results
Six patients with non-small cell lung cancer who received curative radiotherapy were assessed so far. Five patients had sufficient image quality for contouring. There was a mean percent decrease of 24.8% by fraction 16 and 37.2% by fraction 26. Average tumour migration was 0.4cm. Progressive anatomical changes were more prominent where there was tumour associated with atelectasis. Updated results for the whole cohort will be presented.

Conclusion
Tumour regression was observed in all patients and a proportion showed significant reduction. Adaptive planning was a feasible option in selected patients.

Background
Intensity-modulated radiation therapy (IMRT) is not covered by national health insurance in Korea until yet. This study is to retrospectively evaluate the clinical outcomes following IMRT in the patients with inoperable non-small cell lung cancer (NSCLC).

Methods
From May 2010 to November 2012, 43 patients with newly diagnosed, pathologically confirmed NSCLC, who seemed to be at excessive risk of pulmonary toxicity if treated with conventional radiation therapy (RT) techniques based on their disease extent and pulmonary function status, received IMRT. The median age was 58 (35~80) years. Clinical stages were IIIA in 7 (16.3%) and IIIB in 36 (83.7%) patients, where 26 patients (65.1%) had supraclavicular nodal metastases. Thirty-six (83.7%) received concurrent chemotherapy during IMRT. The most common chemotherapy regimen was weekly docetaxel plus cisplatin (N=18), followed by weekly paclitaxel plus cisplatin (N=11). Simulation with 4-dimensional CT was done in 27 patients (62.8%). RT was delivered with 6-MV photon beams using step-and-shoot IMRT method. The median RT dose was 66 Gy in 33 fractions. The median clinical target volume was 357.5 (89.3~881.2) cm[3], and elective irradiation to the uninvolved lymphatics was not added. Normal tissue constraints were as follow: maximum spinal cord dose was <46 Gy; relative lung volumes receiving 20 Gy/5 Gy were <35%/65%; and mean lung dose was <20 Gy. Early toxicities including treatment-related pneumonitis (TRP) and esophagitis were graded using the CTCAE version 4.0. In-field locoregional control (LRC), progression-free survival (PFS), and overall survival (OS) were estimated by the Kaplan-Meier method.

Results
At median follow-up of 11.6 (2.3~39.6) months, 30 patients (69.8%) experienced disease progression: distant metastases in 23 patients (53.5%); and locoregional relapse in 13 patients (30.2%) (Figure). Among 13 patients who experienced locoregional relapse, ten patients (23.3%) had in-field or marginal failure, while three (7.0%) had recurrence at initially uninvolved lymphatic regions. Grade 3 TRP and esophagitis occurred in one (2.3%) and ten (23.3%) patients. The one-year LRC, PFS, and OS rates were 75.0%, 33.7%, and 81.7%, respectively. Figure 1

Conclusion
The early experience of IMRT for the patients with advanced NSCLC, who are poor candidates for conventional RT techniques, seems favorable with respect to locoregional control and toxicity. Further studies will be directed to address the issues on the elective lymphatic irradiation extent, radiation dose escalation, long-term clinical outcomes, and comparison with conventional RT techniques. Authors hope to develop optimal clinical indications of IMRT that can be reimbursed by the national insurance system in Korea.

Background
Applying Intensity-Modulated Radiotherapy (IMRT) techniques to patients with stage III NSCLC allows treating large tumour volumes and bulky mediastinal lymph nodes while respecting the usual normal tissue constraints to the lungs and organs at risks. IMRT techniques lead to irradiating a large lung volume with low doses (lung bath). The volume of lung receiving doses of 5 Gy or more (V5) may estimate the toxicity related to low dose irradiation. In the current study, an analysis was performed to test whether the introduction of IMRT was associated with increased incidence of severe radiation pneumonitis (RP) compared to three-dimensional conformal radiotherapy (3D-CRT) and whether introducing a new dose constraints to the V5 (V5 = 60%) would reduce the incidence and/or severity of radiation pneumonitis.

Methods
The control group included 105 patients with pathologically confirmed inoperable stage III NSCLC receiving radical radiotherapy (60-66Gy in 2Gy/fraction) between 2007 and 2009, at our department using 3D-CRT. The Study group included ninety consecutive matching patients receiving the same radical dose of radiotherapy using IMRT in the period January 2011 to June 2012. The first 35 patients (group I) were treated with standard dose constrains on MLD (maximum 20 Gy) and V20 (Maximum 40%). In group II including 55 patients, a new dose constraint to V5 was introduced (maximum 60%). All patients were seen weekly under radiotherapy, 6 weeks after, and then with 3 months intervals. Radiation Pneumonitis was graded using CTC 3.3. The clinical and dosimetric parameters related to RP were analysed using SPSS.

Results
IMRT was delivered using 4 to 8 beam arrangements. The incidence of grade 3 or more RP in the control group reciving 3D conformal radiotherapy was 16.3% (2% lethal). This was increased to 33% (14% lethal) in group I. Introducing a new dose constrain for the volume receiving low dose (V5) reduced the incidence of severe radiation pneumonitis to less than 10% (0% lethal) in group II. Neither the Mean Lung Dose nor the volume receiving 20 Gy (V20) values were significantly different in the 3 groups.

Conclusion
Irradiating large lung volumes with low radiation doses of 5 Gy or more is associated with higher incidence of severe pneumonitis that is potentially lethal despite respecting the V20 and MLD constraints. Using IMRT in patients with large tumours especially if receiving concurrent chemotherapy should be performed with caution. The cut level of V5 of 60% was applied in our department and was found feasible.

Background
Respiratory induced tumour motion is one of several challenges encountered when delivering radical radiotherapy to lung cancer patients. In recent years, four-dimensional computed tomography (4DCT) has improved our ability to accurately define lung tumour motion during breathing. Using 4DCT images, our study aims to compare the magnitude of lung tumour motion due to respiration, amongst different anatomical lobes and pulmonary zones. This may help guide personalised radiotherapy margins for patients with lung cancer.

Methods
This is a retrospective study of 100 consecutive patients from the Peter MacCallum Cancer Centre treated with curative intent radiotherapy for lung cancer. All 4DCT scans accessible from patients scanned between December 2009 and May 2013 were included. Images were analysed using the MIM v5.6 software. Tumour volumes were delineated by a single observer and propagated to include all 10 phases of the respiratory cycle. Movements were tracked in the superior-inferior (SI), anterior-posterior (AP) and medio-lateral (ML) directions by changes in the gross tumour volume centroid coordinates. Tumour motion characteristics were correlated with anatomical lobe, pulmonary zone, tumour volume, histopathology, spirometry and T-stage. Tumours with chest wall or mediastinal invasion were excluded. Statistical analyses were performed using Prism v6.0.

Results
Preliminary data from 82 patients showed the greatest mean movement in the SI direction among lower lobe tumours compared to those located in the upper lobes [Left lower, 8.0mm, n = 13, vs. Left upper, 1.3mm, n = 24] [Right lower, 6.4mm, n = 19, vs Right upper, 1.9mm, n = 28], p < 0.01. In all lobes, mean movements were similar in the AP [1.6mm, Right lower; 2.1mm, Right middle; 1.8mm, Right upper; 2.3mm, Left lower; 1.6mm, Left upper] and lateral directions [0.9mm, Right lower; 2.4mm, Right middle; 1.2mm, Right upper; 1.5mm, Left lower; 1.2mm, Left upper]. 35 patients were staged as T1, 30 as T2 and 14 as T3. Mean lung tumour motion decreased with increasing T stage in the SI direction [3.9mm, T1; 3.7mm, T2; 3.5mm, T3], however this was not statistically significant. Assessment of the association between tumour motion and spirometry findings is ongoing. Figure 1

Conclusion
The degree of lung tumour motion varies widely according to its position within the lung. The largest differences in tumour motion was between the upper and lower lobes in the SI direction. Analysis of all 100 patient datasets is ongoing.

Background
To analyze the effects of hypofractionated-simultaneous integrated boost-intensity modulated radiation therapy (Hypo-SIB-IMRT) on medically inoperable patients with stage I non–small-cell lung cancer (NSCLC).

Methods
65 qualified patients from December 2003 to November 2012 at three centers in China were included, with a median follow-up time of 39 months. Hypo-SIB-IMRT was delivered in 15 fractions with iGTV 75Gy, CTV 60Gy, and PTV 45Gy in 3 weeks. The slow computer tomography (CT) scan, conventional planning CT scan with active breath control (ABC), [18]fluorodeoxyglucose-position emission tomography ([18]FDG-PET) scan, or four dimensional CT scan (4D CT) were employed to do simulation. During the treatment, electrical port imaging device (EPID) or cone beam CT (CBCT) were performed 2-3 times per week to verify the reproducibility of the targets. All IMRT plans were optimized with Pinnacle or Eclipse systems using heterogeneity correction.

Results
The 1-, 3- and 5-year overall survivals (OS) were 100%, 87% and 58%, respectively with a median survival of 66.5 months. The 1-, 3- and 5-year progression free survivals (PFS) were 98%, 83% and 74%, while cancer-specific survivals (CSS) were 100%, 92% and 73%, respectively. The distant metastasis-free survivals (DMFS) were 98%, 85% and 76%, meanwhile local control (LC) were 100%, 93%, and 91%, respectively. There were no significant differences in OS, PFS, CSS, DMFS and LC compared with stage (I~A~ vs. I~B~) or gross tumor volume (GTV＞20 cm[3] vs.≤20 cm[3]). However, patients with T~1~ tumors (stage I~A~) were inclined to better survivals than those with T~2a~ (stage I~B~) ones, and local relapses were more frequent for larger GTVs. 18.5% (12/65) had grade 1or 2 radiation pneumonitis (RP), and only 1.5% (1/65) grade 1 esophagitis. Of 12 patients with RP, 10.8% (7/65) developed grade 1 radiation pulmonary fibrosis (RPF).

Conclusion
Due to the favorable long-term OS, PFS, CSS, DMFS, LC in addition to the minimal toxicities for medically inoperable stage I NSCLC patients, Hypo-SIB-IMRT presented in this prospective study may be an option to stereotactic body radiotherapy.

Background
With the high radiation dose used in stereotactic ablative body radiotherapy (SABR), it is vital to accurately delineate radiosensitive normal tissues in the planning process such that the dose distributions can be optimised to avoid these regions. This can become a time consuming process. It is also well understood that there is variation in how these volumes are delineated and this has influence on the plan and follow-up data. This work aims to determine if the use of automatic atlas segmentation can help to quickly and accurately delineate CT volumes for SABR planning - saving time, reducing complexity and improving the uniformity of process between patients.

Methods
From a retrospective sample of 5 patients delineated by an oncologist, each patient’s CT scan and contours was used as an atlas for a deformable image registration to map volumes to the other patients. A range of quantitative metrics (based on both overlap and distance) were used to evaluate the agreement between the expert contours and the atlas contours for each.

Results
The Dice similarity co-efficient showed overlap varied depending on the structure – lungs and heart performed well with a mean of 90% (range 71 – 99%) while the airway structures (trachea, oesophagus, proximal bronchi) performed poorer, with a mean of 52% (range 17 – 99%). This indicates while some structures are well suited to atlas segmentation, some are not. The mean Hausdorff distance, indicated that contours requiring editing after applying the atlas, had mean Hausdorff distances on the order of 1 mm. Typically, atlases that did not perform acceptably had a mean Hausdorff distance > 3 mm indicating it was better to start again with manual delineation. From the sample of 5 atlases, none performed significantly better than the others across all cases.

Conclusion
The use of atlases could potentially encourage consistency in delineation and could reduce the laborious task of delineation for lung SBRT treatment planning, however editing of contours is still required. Larger structures performed well, however they could also be easily delineated with threshold-based automatic segmentation. Further investigation to develop a bank of representative atlases may improve performance.

Background
Few studies are dedicated to elderly patients with unresectable stage IIIA/B. We used a Standardized Geriatric Assessment (SGA) to select fit elderly patients and assess if this population can benefit from standard of care, namely concurrent CRT.

Methods
The aim of this multicentric phase II opened-study was to assess CRT in patients older 70 years with locally advanced NSCLC, evaluated as “fit” according to SGA.CRT associated oral vinorelbine (30 mg/m²/week) and IV cisplatinum (30 mg/m²/week) during 6 weeks concurrently with radiotherapy (66 Gy, 33 fractions, 6,5 weeks). Main inclusion criterias were : PS ≤ 1, weight loss < 10%, creatinine clearance ³ 50 ml/mn abreviated, VEMS ³ 40%, PaO2 ³ 60 mm Hg, KCO ³ 60% and patient classified as fit according to SGA. The principal end-point was early treatment tolerance (number of patients with adverse event grade ³ 3 (except nausea and vomiting) or grade 4 for hematologic toxicity and asthenia. Secondary end-points were RECIST response 4 weeks after treatment, quality of life, tolerance, progression-free survival and overall survival. Using a Simon's optimal plan in 2 steps, the total number of patients to be included was 59 with an intermediate analysis after 19 patients. Toxicities and serious adverse events were monitored by an independent peer committee.

Results
Interim analysis was done after 23 inclusions in 19 evaluable patients: males 84% , mean age 74.6 (70 to 83) years, 3 patients didn’t end the treatment (1 disease progression, 1 cons-indication for radiotherapy, 1 patient choice). Four patients had adverse event ≥ 3 (except nausea and vomiting) or grade 4 hematologic toxicity and asthenia. Treatment efficacy was: 1 RC, 10 RP, 5 SD, 1 PD. Two patients were not evaluable (1 early death, 1 patient’s refusal to further treatment). The independent peer committee judged that toxicities were acceptable and consistent with what was expected. Study is ongoing with 44 enroled patients currently.

Conclusion
The interim analysis of a phase II study of CRT in fit elderly patients with no resecable locally advanced NSCLC assessed by SGA showed an acceptable toxicity. Results will be upgraded for the congress.

Background
Mediastinal lymph node involvement (MLNI) is the strongest prognostic factor of resected locally advanced non-small cell lung cancers (LA-NSCLC). The 7[th] edition of the TNM is based in purely topographic criteria: N2 is applied for ipsilateral MLNI and N3 for contralateral and unresectable MLNI. Different authors have proposed many other prognostic classifications of MLNI

Methods
We retrospectively reviewed 45 resected LA-NSCLC previously treated with inductive chemo or chemoradiation in our center between October-2004 to June-2012. Three patients died due to early surgical complication were not analyzed. Systematic mediastinal dissection was performed in all of them. The prognostic impact of pN2 was statistically compared to number of resected nodes; number of affected areas; number of metastatic nodes; and ratio of metastatic/resected nodes

Results
Our series included 37 men and 5 women with a mean age of 64.64 years-old. There were diagnosed eight non-specified NSCLC, two adenocarcinomas, and 32 squamous carcinomas. Cis- and carboplatin-based chemotherapy was administered in 35 and seven patients, respectively. Concurrent radiation was administered in 21. pN2 MLNI clearly worsened both disease free (DSF) and overall survival (OS) of the patients (p= 0.006; and p= 0.018, respectively). Proportion of patients with pN2 MLNI was lower when radiation was applied (p= 0.033) and did not significantly vary according to chemotherapeutic regimen or histology. The number of resected nodes did not impact on survival (DFS p= 0.261, and OS p= 0.277), but OS was better in patients with a lower rate of pathologically detected nodes (p= 0.046). The number of metastatic nodes according to the Tokio score did not impact on survival. The Regina Elena and the Mount Sinai ratios of metastatic nodes according to the number of resected nodes neither impacted

Conclusion
The presence of pN2 MLNI according to the 7[th] edition of the TNM classification is the strongest mediastinal prognostic factor in our series of resected LA-NSCLC previously treated with inductive chemo or chemoradiation. No other tested classification of MLNI could predict significant differences in survival. The prognostic significance of the lower detection rate of resected nodes after inductive treatment should be explored. Figure 1. DFS and OS Kaplan-Meier curves according to MLNI. Figure 1

Background
Combined cytotoxic chemotherapy and radiation therapy is established as the standard treatment for patients with medically inoperable or technically unresectable stage III NSCLC. Multiple randomized studies and meta-analyses demonstrate that CChRT results in improved survival compared with sequential chemo-radiotherapy or radiotherapy alone. The aim of our study was to evaluate the effectiveness and toxicities of CChRT with bi-weekly Docetaxel (D) and Cisplatin (C) and thoracic radiotherapy, after one cycle D-C induction chemotherapy.

Methods
Between May 2009 and November 2012, 53 chemo-naive p with histologically confirmed inoperable locally advanced NSCLC, stage IIIAN2/IIIB (no pleural T4), PS 0-1 and adequate lung function (FEV1 > 1.1, V20 < 25%) were included: one cycle of D 75 mg/m2 on day 1 and C 40 mg/m2 days 1-2 followed at 21 days by CChRT with bi-weekly D 40 mg/m2 and C 40 mg/m2 for four courses, during conformal thoracic radiotherapy (66 Gys, 180 cGy/day). The primary objective was overall survival (OS); secondary objectives were progression free survival (PFS), response rate (RR) and toxicity. Median follow-up: 17,8 months.

Results
The p characteristics were: mean age 59,4 years (34-75); male/female 47/6; ECOG PS 0/1 in 17/36 p; squamous/adeno/large cell carcinoma: 53%/34%/13%; stage IIIAN2 15 p (28.3%) and stage IIIB 38 p (71.7%). All p were evaluable for response and toxicity. RR: 6 CR, 37 PR (RR 81.8%; 95% CI:71-92), 4 SD (7.6%) and 6 PD (11.3%). The median PFS was 14 months (95% CI:11-17) and median OS was 21 months (95% CI:9-32). The PFS at 1/2 years were 55%/32% and the OS at 1/3 years were 82%/50%. A total of 53 cycles of D-C induction chemotherapy were given; main toxicities (NCI-CTC 3.0) per p Grade (g) 1-2/3-4 (%) were as follows: neutropenia 1.8/15; anemia 11.3/0; nausea/vomiting 26.4/1.8; diarrhea 22.6/3.7; fatigue 35.8/0; there were three episodes of hospitalization: febrile neutropenia 2 p and g3 diarrhea 1p. Main toxicities per p in CChRT (D-C doses: 203, 3.8 per p; mean doses RT: 64,6 Gys) were g1-2/3 (%): neutropenia 28.3/5.6; anemia 62.2/0; esophagitis 50.9/3.7 and pneumonitis 32/0; nausea/vomiting 20.7/0; fatigue 37.7/3.7; there were four episodes of hospitalization: febrile neutropenia, 2 p and g3 esophagitis, 2 p.

Conclusion
CChRT with bi-weekly Docetaxel and Cisplatin and thoracic radiotherapy is a feasible treatment option for inoperable locally advanced stage III NSCLC, showing good clinical efficacy and tolerability with acceptable long-term survival.

Background
Inoperable stage III NSCLC is increasingly diagnosed in poor-risk patients for whom there is not yet a standard treatment. We conducted a randomized two-stage phase II study to assess whether sequential or concurrent chemoradiation was feasible, tolerable and showed efficacy.

Methods
Patients with inoperable stage IIIA and B NSCLC and at least one of the following conditions: age ≤ 75 years, ECOG PS=2, weight loss > 5%, creatinine clearance < 60 ml/min, a comorbid condition precluding the patient from being treated in a protocol for fit patients,were randomized to receive either carboplatin AUC 2.5 and vinorelbine 15 mg /m[2] both on days 1,8,22, and 29, and thoracic radiotherapy (TRT) total dose 60 Gy starting day 1 (CT arm) or, carboplatin AUC 5, days 1 and 22 and vinorelbine 25 mg/m[2] days 1, 8, 22, and 29, followed by TRT 60 Gy starting day 43 to 50 (ST arm). The primary end-point was response rate.

Results
From June 2001 to June 2006, 70 patients from 8 centers were included : 47 in CT arm and 23 in ST arm. Forty-eight of these patients were randomized during the first stage of the trial. By September 2004, due to a decrease in treatment compliance and an increase in early deaths in the ST arm, accrual was continued in the second stage of the trial only in the CT arm. Patient characteristics: median age 74 (49-84), Male 96%, Stage IIIB 65%; ECOG =2, 28%; Weight loss >5%, 29%; Creatinine clearence <60, 26%; Comorbidity, 70%. More than one poor risk inclusion criteria: 59 %. Fifty-eight patients completed treatment 93 % in CT arm, and 73% in ST arm. There were 2 CR and 25 PR (RR 60%) in CT arm, and 10 PR (RR 45.5%) in ST arm. Grade 3- 4 hematological toxicity was absent in CT arm and was 14% (neutropenia) in ST arm. Grade 3 and 4 non-hematological toxicities experienced by more than 5% of patients were asthenia (7%) and dyspnea (9%) in CT arm and anorexia (9%), asthenia(14%), and dyspnea (14%) in ST arm. Only one patient developed grade 3 esophagitis (CT arm) There were five deaths during treatment: two in CT arm and three in ST arm. Median PFS and overall survival rate were 6.7 (95% CI:4.9-8.5) and 16.8 months (95% CI 9.5-24), and 7.9 (95% CI:3.9-16.2) and 5.6 months (95% CI:2.7-8.9 ), for the CT arm and ST arm, respectively.

Conclusion
In poor-risk patients with inoperable stage III NSCLC, concurrent chemoradiotherapy outperformed sequential chemotherapy and radiotherapy, and was feasible, very well tolerated, and provided efficacy. The survival outcome with concurrent chemoradiotherapy was notably longer than anticipated.

Background
Concurrent chemoradiotherapy is the standard treatment for locally advanced non-small-cell lung cancer (LA-NSCLC). However, its cure rate remains unsatisfied, and further improvement in the treatment outcome is strongly warranted. S-1 (S), an oral fluoropyrimidine, is a new active agent possessing a radio-sensitizing effect. Additionally, combining S and cisplatin (P) offered an active and safe regimen for metastatic non-small-cell lung cancer. The objective of this study was to assess the efficacy and safety of S plus P with concurrent thoracic radiotherapy (TRT) for LA-NSCLC.

Methods
Patients with stage IIIA/IIIB, aged ≤75 years and PS 0-1, and without any prior chemotherapy were eligible for this study. Patients were treated with P (40 mg/m² on day 1, 8, 29 and 36) and S (40 mg/m²/dose b.i.d. on days 1-14 and 29-42) and TRT (60 Gy/30 fr over 6 weeks starting on day 1). Primary endpoint was respsonse rate, and required sample size was 48 patients.

Results
Between 2006 and 2009, 48 patients were enrolled (37 men; median age, 66 years; PS 0/1, 36/14; IIIA/IIIB, 23/25; sq/non-sq, 22/26). Partial response was observed in 37 patients (77%; 95% confidence interval: 63-88%). At a median follow-up of 54 months for the surviving patients, median progression-free survival and median survival time were 9.3 months and 31.3 months, respectively. No difference in efficacy (response and survivals) was observed stratified by histology (sq vs. non-sq). Toxicities were generally mild, including G3/4 neutropenia (44%), G3/4 thrombocytopenia (13%), G3 febrile neutropenia (8%) and G3 pneumonitis (4%). No one developed Gr3/4 esophagitis. No toxic deaths have occurred.

Conclusion
This chemoradiotherapy regimen yielded a favorable overall survival data. Also, it was well-tolerated in patients with LA-NSCLC as compared with concurrent docetaxel plus P with TRT therapy especially in term of TRT-related toxicities.

Background
Concurrent chemoradiotherapy is the standard treatment for unresectable stage III non-small cell lung cancer (NSCLC). S-1 has been shown to be significant efficacious for treating advanced NSCLC. Our previous phase II study reported short-term outcomes of cisplatin (CDDP)/S-1 chemoradiotherapy. Because CDDP/S-1 chemoradiotherapy is considered to have advantages over others in overall survival (OS) and toxicity, we analyzed its long-term outcomes by following up patients included in the phase II study.

Methods
Forty-eight patients (aged <75 years) with unresectable stage III NSCLC were evaluated. They were treated with CDDP (60 mg/m[2] on day 1) intravenously and oral S-1 (40 mg/m[2] twice daily on days 1–14); this regimen was repeated every 4 weeks for four cycles. A 60-Gy thoracic radiation dose was delivered in 30 fractions beginning on day 2.

Results
After a median follow-up of 6.3 years (range, 5.7–7.4 years), the median OS was 2.8 years [95% confidence interval (CI); 1.04–4.63 years], and the 3- and 5-year OS rates were 49.7% (95% CI: 35.6%–63.8%) and 33.0% (95% CI: 20.0%–46.6%), respectively. Out of the several variables evaluated as predictors of OS, including gender, age, stage, histology, and performance status (PS), only PS proved to be a statistically significant predictor in both univariate and multivariate analyses.

Conclusion
CDDP/S-1 concurrent thoracic radiotherapy is clinically feasible and highly efficacious. Despite our relatively small sample size, the benefits of this regimen revealed in this study warrant further research.

Background
Concurrent chemoradiotherapy (CCRT) has become the standard of care for patients with unresectable stage III non-small cell lung cancer (NSCLC). The comparative merits of two widely used regimens: carboplatin/paclitaxel (PC) and cisplatin/etoposide (PE), each given with concurrent radiotherapy, remain largely undefined.

Methods
Records for consecutive patients with stage III NSCLC treated with PC or PE and ≥60Gy chest radiotherapy between 2000-2011 were reviewed for outcomes and toxicity. Survival was estimated using the Kaplan-Meier method and Cox modeling with the Wald test. Comparison across groups was done using the student t and chi-squared tests.

Results
75 (PC: 44, PE: 31) patients were analyzed. PC patients were older (median 71 vs 63 years; p=0.0006). Other characteristics were comparable between groups. With PE, there was significantly increased grade ≥3 neutropenia (39% vs 14%, p=0.024) and thrombocytopenia (10% vs 0%, p=0.039). Radiation pneumonitis was more common with PC (66% vs 38%, p=0.033). Five treatment related deaths occurred (PC: 3 vs PE: 2, p=1.000). With a median follow up of 51.6 months, there were no significant differences in relapse free survival (median PC 12.0 vs PE 11.5 months, p=0.700) or overall survival (median PC 20.7 vs PE 13.7 months; p=0.989). In multivariate analyses, no factors predicted for improved survival for either regimen. Table 1: Non-hematological and hematological adverse events, by grade (CTCAE 4.0)

Adverse events	PC (n = 44)	PE (n = 31)
	n (%)	n	P~χ2~
Esophagitis 1 2 3 4	3 (7) 19 (43) 10 (23) 5 (11)	5 (16) 7 (23) 10 (32) 1 (3)	0.151
Pneumonitis 1 2 3 4 5	21 (48) 6 (14) 0 (0) 1 (2) 1 (2)	4 (13) 6 (19) 1 (3) 1 (3) 0 (0)	0.033
Neuropathy 1 2	1 (2) 1 (2)	0 (0) 0 (0)	0.485
Nephropathy 1 2 3	3 (7) 0 (0) 0 (0)	4 (13) 0 (0) 1 (3)	0.314
Nausea/vomiting 1 2 3	7 (16) 8 (18) 0 (0)	7 (23) 2 (6) 1 (3)	0.291
Chest infection 1 2 3 4 5	1 (2) 1 (2) 11 (25) 1 (2) 1 (2)	0 (0) 3 (10) 5 (16) 2 (6) 1 (3)	0.534
Neutropenia 1 2 3 4	4 (9) 5 (11) 6 (14) 0 (0)	2 (6) 0 (0) 8 (26) 4 (13)	0.024
Febrile neutropenia 3 4	5 (11) 0 (0)	5 (16) 1 (3)	0.394
Anemia 1 2 3 4	12 (27) 5 (11) 1 (2) 0 (0)	10 (32) 9 (29) 0 (0) 1 (3)	0.117
Thrombocytopenia 1 2 3	1 (2) 3 (7) 0 (0)	4(13) 1 (3) 3 (10)	0.039
Treatment-related deaths	3 (7)	2 (6)	1.000

Conclusion
PC was more likely to be used in elderly patients. Despite this, PC resulted in significantly less hematological toxicity but achieved similar survival outcomes as PE. PC is an acceptable CCRT regimen, especially in older patients with multiple comorbidities.

Background
cCTRT is the current standard treatment for good Performance Status (PS) unresectable locally advanced NSCLC. A phase III study demonstrated that Docetaxel consolidation does not improve overall survival (OS) after cCTRT (Hanna. JCO 2008). The role of consolidation chemotherapy after cCTRT is still investigational and our study was set up to evaluate the role of pemetrexed in this setting. A less toxic consolidation chemotherapy may enable a higher proportion of patients to comply to planned treatment which may improve outcome.

Methods
This was a single-institution prospective phase II study. Treatment comprised cisplatin (50 mg/m[2] days 1, 8, 29, 36), etoposide (50 mg/m[2] days 1-5 and 29-33) and concurrent thoracic radiotherapy starting on day 1 chemotherapy (66 Gy in 33 daily fractions; 3D conformal radiotherapy or IMRT) followed by consolidation pemetrexed (500 mg/m[2] on days 71, 92 and 133). The primary endpoint was 1 year OS. Secondary endpoints were progression-free survival (PFS), 2 yr OS, acute/late toxicity (CTCAE v3.0), compliance to treatment.

Results
35 patients were recruited between March 2008 and October 2010. Median age was 61 years (range 42-76). M:F ratio was 23(66%):12(34%). ECOG PS was 0:1 11(31%):24(69%). Histology: squamous 21(60%), adenocarcinoma 8(23%), undifferentiated 4(11%), other 2(6%). Stage: IIB 1(3%), IIIA 19(54%), IIIB 15(43%). All 35(100%) had PETCT staging. All 35 patients received concurrent chemotherapy (dose reduction in 3 patients) and 32 (91%) received the planned 66Gy (range 56-66 Gy). The number of patients who completed pemetrexed were: cycle 1=25 (71%), cycle 2=22 (63%), cycle 3=16 (46%). Radiation parameters: Gross Tumour Volume (GTV) was median 60.2 cm[3] (range 11.4-274.4 cm[3]), V~20Gy ~median 30.4% (range 10.5-35.3%), During the concurrent phase, grade 3/4 toxicity was noted for: neutropenia 17(49%) anaemia 1(3%), thrombocytopenia 1(3%), infection 8(23%), fatigue 6(17%), nausea±vomiting 4(11%), mucositis 3(9%), anorexia 3(9%). During the pemetrexed consolidation phase, the only grade 3/4 toxicities were: infection 5(20%), anaemia 3(12%), neutropenia 2(8%) and fatigue 2(8%). Acute radiotherapy toxicity (<3months): oesophagitis grade 3/4 10(29%) and late toxicity (>3months): pneumonitis grade 3/4 2(7%), oesophageal stricture 2 (7%), pulmonary fibrosis 1(3%). Median follow up was 25months. Median OS was 34months, with 1yr OS 77% (95% CI 60-88%), and 2yr OS 61% (95% CI 37-72%). Median PFS was 22months, with 1yr PFS 62% (95% CI 43-76%) and 2yr 49% (95% CI 31-65%). Of the 14 deaths, causes were, 1 suicide during radiotherapy, 2 treatment-related deaths (1 grade 5 pneumonitis and 1 grade 5 haemoptysis) and 13 due to lung cancer.

Conclusion
In an unselected locally advanced NSCLC population, staged with PETCT a median survival of 34 months can be achieved. The study reinforces the challenge of delivering consolidation chemotherapy and suggests that improved staging contributes to improved outcomes. Although there was failure to deliver all planned cycles of consolidation pemetrexed after cCTRT in 54% of patients, these are encouraging results that warrant further investigation.

Background
Purpose of this study was to evaluate the association between change in hemoglobin (Hb) levels during treatment and disease control and survival in locally advanced non-small cell lung cancer (LA-NSCLC) patients treated with concurrent chemoradiotherapy (CRT).

Methods
Medical records of 368 LA-NSCLC patients, who had been treated with definitive CRT at our department between dates January 2007 and December 2011, and whose pretreatment Hb levels were ≥12.0 g/L, and who had at least 3 Hb measurements during CRT, were retrospectively evaluated. All patients received 60-66 Gy thoracic 3-dimentional conformal radiotherapy concurrently with 1-3 cycle cisplatin/carboplatin-vinorelbine/taxane (q21) doublet chemotherapy. For all analyses, patients were divided into groups according to their nadir Hb level, percent Hb change, development of anemia (Hgb<12) and nadir Hb cut-off defined from ROC analysis. Primary end point was overall survival (OS), and secondary end points were progression-free survival (PFS) and locoreginal PFS (LRPFS).

Results
At a median follow-up of 24.7 months (range 19.7-27.5), Median OS, PFS and LRPFS for whole group were 23.8 (%95 CI=22.8-24.8), 11.7 (%95 CI=11.0-12.5) and 15.3 months (%95 CI=14.7-15.9), respectively. During CRT course, 138(37.5%) patients developed anemia, and median OS, PFS, and LRPFS were significantly worse in these patients [(27.3 vs. 18.3; p<0.001), (12.7 vs. 9.2; p<0.001), and (16.4 vs. 11.2 months; p<0.001)]. While there was no statistically significant survival difference according to median nadir Hb level and % Hb change, patients with Hb levels above the cut-off point defined from ROC curves (12.4 g/L) revealed better OS (27.7 vs. 20.4; p<0.001), PFS (12.8 vs. 10.2; p<0.001) and LRPFS (16.4 vs. 13.8 months; p<0.001). On multivariate analyses, only these two factors found to be independent prognostic factors (p<0.001 for Hb > vs. < 120 g/L, p<0.03 for Hb> vs.

Conclusion
Results of this study has shown that 37.5% NSCLC patients with normal pretreatment Hb levels had developed anemia during the course of CRT, which was found to be a worse prognostic factor independent from other co-existing factors in terms of locoregional tumor control and survival outcome.

Background
Cancer Care Ontario’s (CCO) Program in Evidence-based Care has been developing lung cancer practice guidelines since 1997. Based on randomized clinical trials and published meta-analyses, a modest but clinically significant benefit for the use of concurrent chemo-radiotherapy treatment (CCRT) in selected inoperable stage III NSCLC (good performance status, limited weight loss), was recommended in a guideline published in 2005 and revised in 2006.

Methods
In 2008, CCO began to measure concordance with guidelines and to publically report regional results through the Cancer System Quality Index (CSQI),a web-based public reporting tool released annually by the Cancer Quality Council of Ontario (CQCO), Guideline concordance is a measure within the Effective quality domain of CSQI and is used to track the consistency of cancer treatment services across Ontario. This measure links data within Cancer Care Ontario’s Activity Level Reporting Enterprise Data Warehouse and the Ontario Cancer Registry with information from the Canadian Institute for Health Information’s Discharge Abstract Data and National Ambulatory Care Reporting System.

Results
Of 1312 patients with unresected stage III disease diagnosed in 2010 and 1259 in 2011, only 30.3% and 31.8% respectively received CCRT defined as radiation and chemotherapy given within 180 days of diagnosis. An additional 33.9% received an alternative form of treatment in 2011: 83.6% of these patients were treated only with radiation, 66% of whom had palliative radiotherapy while 33.4% had radical (curative) radiotherapy. In 2010, a similar pattern of treatment was observed with 33.5% of cases receiving alternate treatment: 81.3% of whom received only radiation; 70% of these patients received palliative treatment while 27.5% received radical radiotherapy. In 2011, 34.2% received no treatment, a decrease of 2% from 2010. Variation in guideline concordant practice was evident between the 14 health service regions of the province (range from 23.3% to 44.5%) but only one was significantly greater than the Ontario rate (95% Confidence Interval (CI); effect size (d =0.56). Six of 14 regions had a decline in the concordance rate between 2010 and 2011. There was no difference in the rate of CCRT use by gender (28.4%) but there was a sharp decline in CCRT after age 65 (45% < 65 yr vs. 25% > 65 yr), (95% CI, 21.5-28.5; p=0001). Less CCRT was given to the lowest income quintile (Q)(22.9% vs. 29.9% for Q4, 95% CI 23.9-35.9; p=0.0001), to urban vs. rural populations (22.9% vs. 34.8%; 95% CI, 28.6-40.9; p=0.0001) and in those areas of the province with higher populations of immigrants (lowest tercile 28.4% vs. 18.6% for middle and 19.0% for the highest tercile, p=0.0001).

Conclusion
Concordance with the CCO guideline on CCRT in Stage III unresectable NSCLC is particularly low in older, lower income, urban and immigrant populations. The absence of weight loss and performance status data makes interpretation of this data difficult. Further study of the reasons for these variations in practice will be necessary to inform appropriate strategies to reduce these inequities.

Background
Trimodality therapy consisting of induction chemoradiotherapy (CRT) followed by surgery can be one of the treatment options for locally advanced non-small cell lung cancer (NSCLC). While recent randomized phase III trials failed to demonstrate a benefit from the addition of surgery in the entire population, the subset analysis of the intergroup trial 0139 indicates that trimodality therapy is beneficial for population who did not undergo pneumonectomy. This result strongly suggests that the status of disease may influence the prognosis even in same stage population. Thus, identifying prognostic factors and their inclusion in stratification are critical for the appropriate randomized study. In this study, we retrospectively examined the prognostic impact of tumor location in NSCLC patients with clinical (c-) N2 disease who underwent trimodality therapy in our institute.

Methods
Among patients who underwent induction CRT followed by surgery between 1999 and 2011 at our institution, a total of 76 NSCLC patients with c- N2/3 stage III were enrolled for this retrospective study. Induction CRT basically consisted of docetaxel and cisplatin with concurrent radiation at a dose of 40 - 60 Gray.

Results
A total of 76 patients consisted of 53 male and 23 female, 43 adenocarcinomas and 33 non-adenocarcinomas, and 44 c-Stage IIIA and 32 c-Stage IIIB. Primary tumors were located in right upper lobe for 33 patients, right middle lobe for 5, right lower lobe for 11, left upper lobe for 20, and left lower lobe for 7. For all population, lower lobe tumors showed significantly shorter overall survival (OS) and disease-free survival (DFS) times compared to non-lower lobe tumors (OS, p = 0.022; DFS, p = 0.0007). In a multivariate analysis, tumor location was independent prognostic factor for poor prognosis. Limited to pathologically proven N2/3 disease before induction CRT (n = 36), location of lower lobe tend to be poor prognosis compared to other location (OS, p = 0.068; DFS, p = 0.0075).

Conclusion
We showed that tumor arising from lower lobes is a poor prognostic factor in NSCLC patients with N2 disease treated with induction CRT. The status of tumor location should be considered in stratification in randomized trails that estimate the impact of the trimodality therapy.

Background
The potential prognostic value of survivin is variably reported in various stage of lung cancer. This study evaluated the correlation between tumor survivin expression before and after chemoradiation therapy (CRT) and prognosis in stage III non-small cell lung cancer (NSCLC) patients treated with platinum-based chemoradiation therapy followed by surgery

Methods
Medical records from 53 patients whose tissues were suitable for study were reviewed. Tissues were stained by immunohistochemistry and were estimated the degree of stain with scoring for survivin. Clinicopathologic parameters and survivin expression score were evaluated for a prognostic relationship with overall survival (OS) and time to recurrence (TTR).

Results
Pathologic complete response, residual nodal involvement and radiologic response after CRT were not related with OS or TTR. Tumor survivin expression on pre-treatment tissues was presented in 47 patients (88.7%). Pre-treatment survivin score was not related with TTR and OS (p = 0.249 and p = 0.601, respectively). There was no correlation between pre-treatment and post-treatment survivin score (p = 0.309). Downregulation of survivin and post-treatment survivin score (0 – 1) after chemoradiation showed significant improvement in OS (p = 0.04 and p = 0.033, respectively). Age, downregulation of survivin score, and post-treatment survivin score (0 – 1) were significant prognostic factors for survival by multivariate analysis.

Conclusion
Downregulation of survivin score and post-treatment low survivin score in stage III NSCLC treated with platinum-based chemoradiation therapy followed by surgery has a value of prognostic factor regardless pre-treatment survivin score. Further studies to evaluate the relation of survivin expression and platinum-based chemoradiation therapy are warranted in NSCLC.

Background
to analyse survival data, treatment results, toxicity and prognostic factors, the influence of PET-staging in the combination of daily low-dose cisplatin with high-dose radiotherapy in stage III NSCLC patients.

Methods
retrospective study between 2005 -2012 of PET staged patients with mainly stage III irresectable, locally advanced non small cell lung cancer. Concurent chemoradiation is given by using daily low-dose cisplatin (6mg/m2) combined with 24 fractions 2,75 Gy to a total dose of 66 Gy.

Results
mean follow-up 20 months, median follow-up 15 months (1, 7-88). Prognostic factors at multivariate analysis are for survival: PTV1 < 380 cc (p=0.018), ECOG scale (p=0.09). For local control: PTV1 < 380 cc (p=0.031). Our data show a linear relationship between the size of the PTV1 and survival. The median survival was 36 months and the 5-year survival rate is 33%. The incidence of distant free survival of 5 years is 27%. The local relapse free survival of 5 years is 51%.

Conclusion
results for survival and local control are good and show a 100% increase in 5 yr overall survival compared to our historical control series. In our data toxicity is low. In multivariate analysis a low PTV volume (<380 cc) is the only significantly favorable prognostic factor for local disease-free and overall survival. Possibilities to increase the biological radiation dose (EQD2= 74 Gy) and/or in combination with biological response modifiers should be studied in the near future. Figure 1. Overall survival for all patients Mean follow-up 20m, median follow-up 15 m (1,7-88) Figure 1 Figure 2. Overall survival stage III < 380cc and > 380cc Survival: 1=PTV1 < 380 cc, 2=PTV1>380 cc (p=0.018) Figure 2

Background
Concomitant chemoradiotherapy is the standard treatment of unresectable stage III non-small cell lung cancer (NSCLC). However, the optimal chemotherapy regimen is still controversial, particularly in lung squamous cell carcinoma. We have conducted a phase II clinical trial in a Chinese population to evaluate concomitant treatment using docetaxel/cisplatin chemotherapy and thoracic radiotherapy followed by docetaxel/cisplatin consolidation chemotherapy in unresectable stage III lung squamous cell carcinoma. The purpose of this study is to evaluate the feasibility and activity, and also assess its impact on progression-free survival (PFS).

Methods
A total of 32 patients were enrolled between January 2011 and January 2013. Patients received concomitant docetaxel 30mg/m[2] on day 1 and day 8, cisplatin 25mg/m[2] on day 1 to day 3 repeated every 3 weeks for 2 cycles and thoracic radiotherapy, followed by docetaxel/cisplatin for 2 cycles as consolidation therapy (docetaxel 60mg/m[2] on day 1, cisplatin 25mg/m[2] on day 1 to day 3 repeated every 3 weeks). Objective response rate according to the RECIST criteria was recorded and toxicity was evaluated using the NCI Common Toxicity Criteria. The Kaplan–Meier method was used to evaluate patient survival. Univariate analysis of patient characteristics and tumor responses was conducted using the Chi-square and Fisher’s exact test.

Results
Eight (25.0%) and 20 patients (62.5%) had a complete or partial response, respectively, while 3 patient’s disease remained stable and 1 patient had progression of the disease. The overall response rate (87.2%, 95% confidence interval (CI): 63–97%) exceeded the goal per study design. The median PFS was 11.0 months (95% CI: 5.6–16.4 months). This approach obtained likely better effects than history control group. Main toxicity (grade 3 or greater, %): neutropenia 10 (31.3%); thrombocytopenia 7 (21.9%); anaemia 8 (25.0%); nausea/vomiting 5 (15.6%); anorexia 7 (21.9%), dysphagia 4 (12.5%), radiation pneumonitis 3 (9.4%) and fatigue 4 (12.5%).

Conclusion
This data suggests that concomitant treatment with docetaxel/cisplatin and thoracic radiotherapy was well tolerated, with promising activity in a Chinese population with unresectable stage III lung squamous cell carcinoma. Although the data presented herewith appears promising, this study is relatively small, and more data from randomized trials are needed to further validate this regimen.

Background
To evaluate the role of postoperative radiotherapy (PORT) after curative resection and adjuvant chemotherapy for patients with pathological stage N2 non-small cell lung cancer (NSCLC).

Methods
We performed a retrospective review of 219 consecutive patients who underwent curative surgery followed by adjuvant chemotherapy between 2000 and 2011. Among 219 patients, 41 received PORT additionally. Propensity scores for PORT receipt were calculated for each patient and used for matching to patients without PORT. 118 patients in non-PORT group and 39 patients in PORT group were matched. Clinical and pathologic characteristics were well-balanced after matching. PORT was delivered using conventional technique (n=13) or three-dimensional conformal technique (n=26) with median dose of 54 Gy (range, 50-60). The median follow-up duration for matched patients was 47 months.

Results
During the follow-up, 58 patients (49.2%) experienced loco-regional failure in the non-PORT group and 12 patients (30.8%) in the PORT group. Distant metastasis occurred in 68 patients (57.6%) in non-PORT group and 22 patients (56.4%) in PORT group. PORT was associated with improved loco-regional control rate (LRC) (5yr LRC 67.0% vs. 48.4%, p = 0.047), but not disease-free survival (DFS) (5yr DFS 43.3% vs. 32.3%, p = 0.257). An exploratory subgroup analysis suggested a potential DFS benefit of PORT in patients with multiple stations of mediastinal lymph node metastasis (5yr DFS 42.8% vs. 16.6%, p = 0.023). Grade 3 radiation pneumonitis and esophagitis was seen in only one patient, respectively.

Conclusion
In pathological stage N2 NSCLC patients, more than half eventually developed distant metastasis despite adjuvant chemotherapy. PORT increased LRC in these propensity-matched patients, but did not DFS. However, patients with multiple stations of mediastinal lymph node metastasis appear to benefit from PORT.

Background
Bevacizumab and pemetrexed/cisplatin improves the response and survival in patients with advanced or metastatic non-small cell lung cancer (NSCLC); however, the role of these medications in the setting of induction therapy for NSCLC is not well defined. The purpose of this study was to evaluate the feasibility of induction combination therapy with cisplatin, pemetrexed and bevacizumab followed by surgery in patients with clinical stage II/IIIA nonsquamous NSCLC.

Methods
Patients with clinical stage II/IIIA nonsquamous NSCLC were enrolled. The induction chemotherapy consisted of three cycles of cisplatin (75 mg/m[2]), pemetrexed (500 mg/m[2]) and bevacizumab (15 mg/kg) on Day 1, administered every 21 days. At least six weeks after the last administration of bevacizumab, the patients underwent surgical resection. The primary endpoint was the complete resection rate after the completion of three cycles of induction chemotherapy. The sample size was set at 30. The feasibility of the treatment was considered to be confirmed if the complete resection rate was 80% (24/30) or more.

Results
A total of six institutions in Japan participated in this trial. The study was initiated in June 2010, and patient enrollment was completed in November 2012. Thirty-one patients were recruited, 30 of which were eligible. The median age was 64 years (range: 54-71), and the male/female ratio was 17/13. The PS0/PS1 ratio was 29/1, the adenocarcinoma/large cell carcinoma ratio was 30/0 and the clinical stage IIA/IIB/IIIA ratio was 5/3/22. Grade 3 toxicities included neutropenia (7%), nausea (7%), appetite loss (13%), hypertension (23%) and pulmonary embolism (3%). There were no grade 4 events, and 27 (90%) patients completed three cycles at the full dose of chemotherapy. All but one patient exhibited radiologic tumor reduction based on the RECIST criteria. The objective responses to chemotherapy was CR in 0% of the patients, PR in 37%, SD in 50% and PD in 10% (due to new lesions). The disease control rate (CR+PR+SD) was 87%. Five patients dropped out from the study before surgery due to the patient’s decision in one patient, adverse events in three and disease progression in one. The complete resection rate after the completion of three cycles of induction chemotherapy was 83% (25/80). Therefore, the results met our criterion for feasibility.

Conclusion
Induction chemotherapy using a combination of cisplatin, pemetrexed and bevacizumab in patients with resectable clinical stage II/IIIA nonsquamous NSCLC is therefore considered to be a feasible treatment modality. Figure 1

Background
To evaluate pemetrexed in combination with cisplatin in these patients, a randomized phase III study of concurrent cisplatin with pemtrexed or vinorelbine and late course accelerated hyperfractionated radiotherapy (LCAHRT) was performed.

Methods
Patients with unresectable stage III non–small cell lung cancer （NSCLC） were randomly assigned to two concurrent regimens. Arm1 included cisplatin at 25 mg/m2 on days 1-3, 22-24 and vinorelbine at 25 mg/m2 on days 1,8 and 22,29 with concurrent LCAHRT. Arm 2 used cisplatin at 25 mg/m2 on days 1-3, 22-24 and pemtrexed at 500 mg/m2 on days 1 and 22 with the same radiotherapy protocol. The primary endpoint was progression free survival (PFS), and secondary endpoints included overall survival (OS ) and toxicities. Kaplan–Meier analyses were used to assess survival, and toxic effects were examined using the Pearson Chi-Square test. All statistical tests were two-sided.

Results
A total of 104 patients were enrolled with 100 ones analyzed for 3 in arm 1 and 1 in arm 2 were not finished treatment according to the protocol. The characteristics of study population between the two arms were well balanced. The median PFS were 15 and 23 months , respectively （p=0.014）, while median OS were 25 and 25.5 months for arms 1–2, respectively （p= 0.270）. In arms 1 and 2, the median OS of squamous cell carcinoma was 22.5m and 23m, while non-squamous cell carcinoma 27m and not reached, respectively (p=0.216). The median PFS of squamous cell carcinoma was 17m and 20m, while non-squamous cell carcinoma 15m and not reached, respectively (p=0.020). Compared with lower radiation doses, patients treated with more than or equal to 70 Gy had a better OS and PFS (p=0.002) for both arms. Concerning toxicities, for arms 1 and 2, grade 3-4 depression of white blood cell was 20/48 and 11/52 (p=0.027), grade 2-3 hemotoblatin 8/48 and 5/52 (p=0.047), grade 1-3 radiation induced pneumonitis 40/48 and 31/52 (p=0.032). The differences of other adverse events including esophagitis, nausea, vomiting, platelet were not statistical difference.

Conclusion
Comparing with NP, concurrent pem/DDP at full systemic doses and LCAHRT was well tolerated, with promising activity and milder side effects in a Chinese population with inoperable stage III NSCLC in spite of histology types.

Background
EGFR tyrosine kinase inhibitors (TKIs) showed great survival benefit in the selected patients with stage IV non-small cell lung cancer (NSCLC) harboring TKI-sensitive EGFR mutations. Assuming that if the cases were properly selected, EGFR-TKIs would be integrated into treatment paradigms of stage III NSCLC as more effective systemic therapy, we designed this study to evaluate the efficacy and toxicity of erlotinib and chemotherapy in the combined-modality treatment of unresectable stage III NSCLC patients according to EGFR mutation status.

Methods
Patients over 18 years with unresectable stage IIIA (N2) or stage IIIB NSCLC, ECOG performance status 0–1, and adequate organ function are eligible. The mutational analysis of EGFR (exon 18–21) is performed using direct sequencing in tissue sample. EGFR mutation-positive patients initially receive 3 cycles of erlotinib and then are treated by concurrent chemoradiotherapy (CRT) with either erlotinib (Arm A) or irinotecan-cisplatin (IP) (Arm B). After CRT, patients in Arm A receive consolidation therapy with 6 cycles of erlotinib while those in Arm B are observed until progression. EGFR mutation-negative or unknown patients are treated either with 3 cycles of IP followed by CRT with IP (Arm C) or vice versa (Arm D). Erlotinib is given at 150mg daily with 3-week cycle. IP is given with cisplatin 30mg/m[2] (day 1 and 8) and irinotecan 60mg/m[2] (day 1 and 8) during radiotherapy (total 60 Gy/ 2.4 Gy/fr) or with cisplatin 30mg/m[2] (day 1 and 8) and irinotecan 65mg/m[2] (day 1 and 8) during induction or consolidation therapy with 3-week cycle. The primary endpoint is response rate (RR), toxicity, and survival estimation. Erlotinib and irinotecan were provided by Roche Korea and Pfizer, Inc., respectively

Results
From February 2008 to February 2013, 62 patients were screened and 50 patients were randomized into Arm A (n= 6), Arm B (n= 4), Arm C (n= 19), and Arm D (n= 21). The EGFR mutation status was positive in 10 (20.0%) patients, negative in 23 (46.0%), and unknown in 17 (34%). The median age was 65 years. The proportion of never smoker and adenocarcinoma histology was 80% (event/ total No. = 8/10) and 90% (9/10) in the EGFR mutation-positive group, 13% (3/23) and 48% (11/23) in the EGFR mutation-negative group, and 6% (1/17) and 18% (3/17) in the EGFR mutation-unknown group. The median follow-up time was 34.7 months (range, 3.8−53.8 months). The completion rate of planned treatment was 66.7% (4/6), 100% (4/4), 78.9% (15/19), and 76.2% (16/21) from Arm A to Arm D. For induction therapy, the RR to elrotinib of the EGFR mutation-positive group was 70.0% (7/10). Moreover, the subgroup harboring TKI-sensitive EGFR mutations (exon 19 deletion and exon 21 L858R mutations) showed the RR to erlotinib of 87.5% (7/8). The RR to IP induction therapy was 41.7% (5/12) in the EGFR mutation-negative group, 0% (0/4) in the EGFR mutation-unknown group, and thus 31.1% (5/16) in the EGFR mutation-negative or unknown patients.

Conclusion
The combined-modality treatment by molecular diagnostics was feasible in stage III NSCLC and accrual is ongoing.