Virtual Library

Background
Malignant pleural mesothelioma(MPM) is a locally aggressive disease characterized by a poor prognosis and increasing. MPM tumors are usually related to asbestos exposure, and the incidence is anticipated to peak between 2020 and 2030 because of the lag time between asbestos expousere and the development of the malignancy. MPM is difficult to detect at an early stage, and surgical and radiotherapeutic approaches are ineffective when used independently, because MPM spreads diffusely in the surrounding chest wall. No universally accepted treatment approach currently exists.

Methods
We examined whether combination treatment consisting of pemetrexed chemotherapy and photodynamic therapy (PDT) using the photosensitizer, NPe6, enhanced the antitumor effect in vitro and in vivo models. We also investigated preclinical treatment schedules. Four human malignant mesothelioma cell line, (MSTO-211H, H2052, H2452and H28) were assayed using the WST assy after treatment with pemetrexed and NPe6-PDT. The treatment schedule for the combination treatment was examined using nude mice.

Results
In nude mice injected with MSTO-211H cells and then treared using a combination of pemetrexed and NPe6-PDT (10 mg/kg NPe6, 10 J/cm2 laser irradiation), the tumor volume decreased by 50% but subsequently increased, reaching the pretreatment value after 14 days. Pemetrexed treatment followed by NPe6-PDT resulted in an 80% reduction in the tumor size and inhibited re-growth. NPe6-PDT followed by pemetrexed treatment resulted in a 60% reduction in tumor size but did not inhibit re-growth.

Conclusion
Pemetrexed reportedly inhibits multiple enzymes in the folate metabolic pathway, with TS being the main target. In non-small cell lung cancer cell line, high baseline TS expression levels confer resistance to pemetrexed, and the TS level is correlated with pemetrexed efficacy in a variety of solid tumors. These results suggest that the overexpression of TS protein induced by NPe6-PDT may be associated with the failure of pemetrexed to exert a tumoricidal action. Therefore, we concluded that NPe6-PDT followed by pemetrexed did not enhance tumor cell lethality in the in vivo model. Combination treatment, consisting of pemetrexed followed by NPe6-PDT, should be further investigated as a new treatment modality for malignant pleural mesothelioma. In the future, this combination treatment may contribute to a reduction in local recurrence and a prolonged survival period in patients with malignant pleural mesothelioma.

Background
Adenocarcinoma of the lung is a heterogenous group of disease. Even within the same patient, the tumor may show, various patterns of histopathology. In fact published data suggest that only 1/3rd of lung cancer are homogenous. Although known this factor is not taken into account in treatment planning and management. It is also known that treatment induces heterogenity and change in character of the tumor.

Methods
46 patients of Stage IV Adenocarcinoma of Lung were studied. All patients were confirmed as Adenocarcinoma by doing IHC TTF p63. All slides were viewed by two pathologists and all patients had EGFR mutation done.IHC marker for neuro-endocrine differentiation was done i.e. Chromogranin-A and Synaptophysin. Patients were treated with chemotherapy or EGFR-TKI. Patients were re-biopsied on progression. The same set of IHC studies was done.

Results

	IHC marker for adenocarcinoma	IHC marker for squamous cell	Neuro-endocrine marker
n=46
Pre-chemotherapy/TKI	41 (89%)	15 (32%)	17 (37%)
Post-chemotherapy/TKI	25 (54%)	25 (54%)	29 (63%)
n=18
Pre TKI	16 (89%)	5 (28%)	4 (22%)
Post TKI	8 (44%)	9 (50%)	14 (78%)
n=28
Pre-chemotherapy	25 (89%)	10 (36%)	13 (46%)
Post-chemotherapy	17 (61%)	16 (57%)	15 (54%)

Median duration of chemotherapy - 10 months Median duration of TKI - 16 months Change in character on IHC - seen in approximately 30% patients Infact change in post TKI occurring in upto 50% of patients with gain in Small Cell characters. Of the 50% patients who gained i.e. 10 patients - 7 of them also showed microscopic features of Small Cell Carcinoma of Lungs.

Conclusion
The study suggests that patients with advanced lung cancer have benefit from chemotherapy and TKI. On treatment and on progression there is a change in IHC and histopathology character in the patients. It occurs in close to 40% in such patients. These changes also call for changes in treatment. Hence re-biopy in such patients is essential.

Background
The CT-guided lung needle biopsy is a well-established and safety technique for diagnosis. A biopsy specimen often had loose connection and broke to tiny pieces before formalin fixation. Tumor invasion often involves the epithelial-mesenchymal transition (EMT) during which cells lose the lateral attachments to their neighbors and become more motile. The hypothesis is the fresh macroscopic appearance of specimens may relate pathological features and predict clinical features in patients with lung cancer.

Methods
The correlations between fresh macroscopic appearance of specimens and pathological findings or clinical outcomes were examined in patients who underwent CT-guided lung needle biopsy in our institution between May 2009 and May 2013. The intensity of fiber stained Azan staining (0, 1+, 2+, and 3+) and the percentage of positive cells (<1%, <25%, <50%, <75, and >75%) were assessed. The score of each case was multiplied to give a final score and the fibrosis was finally classified as low (<200) or high (>200). Comparisons of variables were performed by using Fisher exact tests.

Results
A total of 93 (86.1%) of 108 patients had adequate samples for diagnosis. The mean nodule diameter was 26 mm (range 4-75mm). CT findings revealed only three of 93 lesions showed ground-glass opacity, and all of them were in tight connection group. Macroscopically, 21.3% (n=23) specimens had loose connection, and 78.7% (n=85) specimens had tight connection. In loose connections, 73.9% (n=17) diagnosed as malignant and 26.1% (n=6) as benign, with sensitivity of 77.3%, specificity of 100%, and accuracy of 78.3%. In tight connections, 75.3% (n=64) as malignant and 24.7% (n=21) as benign, with sensitivity of 86.5%, specificity of 100%, and accuracy of 88.2%. There were 58 NSCLC samples, including 30 well or moderate (w/d or m/d), and 8 poorly differentiated (p/d) adenocarcinomas (Ad), 7 w/d or m/d, and 6 p/d squamous cell carcinomas (Sq), and 7 undifferentiated carcinoma. In 58 samples, 20.7% (n=12) specimens had loose connection and 79.3% (n=46) specimens had tight connection. In Azan staining, the tight connection group had 32 samples of high and 14 of low scores with the mean score of 213.6, and the loose connection group had 4 of high and 8 of low scores with the mean score of 118.6. The tight connection group had significantly higher scores than the loose connection group (p=0.042). The patients with loose connection had significantly higher rate of distant metastasis than those with tight connection (58.3% vs 21.7%, p=0.028). The median survival times are not reached in both groups.

Conclusion
Macroscopically loose connection specimens can afford to provide adequate amount of samples for diagnosis with sensitivity of 77.3%, and this appearance was negatively correlated with amount of fiber. Furthermore, the patients with loose connection tissue were associated with distal metastasis. The loose connection specimens may represent the status of EMT acquisition which is induced tumor initiation, growth, and metastasis. These findings suggest that the macroscopic appearance of tissue specimens obtained from CT guided biopsy can be an effective evaluation for prediction of metastasis in patients with NSCLC.

Background
Surgical adjuvant chemotherapy with platinum doublet is effective but not enough to suppress recurrence of NSCLC. We have applied combination chemoimmunotherapy with CBDCA+GEM and autologous tumor lysate transduced dendritic cell vaccination in patients with operable NSCLC in order to clarify that such a treatment can induce tumor specific immune response.

Methods
Eight patients with NSCLC whose pathological stages ranging from IB~IV were included in this study. The main purpose of the study is to obtain the principle of concept in which such vaccination can induce cellular immune response against autologous tumor cells detected by ELISPOT assay and establishment of CTL clones to recognize the tumor cell. Autologous tumor lysate was prepared from the surgical specimen aseptically by 5 times of freeze-thaw and sonication . DC cells were prepared from peripheral blood mononuclear cells obtained by apheresis, and in vitro culture with IL-4 and GM-CSF. Matured DC was induced by culture with TNF-α. Autologous tumor lysate-transduced DC was prepared by electroporation using MaxCyte cell loading system(MaxCyte[R]). More than 7x10[6] of the DC were injected intradermally at groin biweekly at 1, 3, 5, 7, 9 and 11week. Combination chemotherapy with CBDCA+GEM was administered at 0, 2, 4, 6, 8 and 10 week.

Results
Delayed type hypersensitivity skin reaction to the DC was observed in 5 of 8 patients after completion of the vaccination(6 times). ELISPOT assay revealed specific IFN-γ production in response to autologous tumor lysate-transduced DC following completion of the vaccination in 3 of the 5 patients. In one patient among them, CTL clones could be induced successfully in vitro, and identification of the Ag recognized by the CTL is now underway. No serious adverse effect was observed. Six patients have recurrent disease at 6, 6, 8, 10, 12 and 18months after operation. One patient(N0.6) died of the disease at 13 months after the operation due to systemic metastasis. However, the other patients are alive with(5 patients) or without(2 patients) evidence of the recurrent disease(ranging from 10~22 months as of February, 2013). Figure 1

Conclusion
The autologous tumor lysate-transduced DC vaccination is effective to induce tumor specific cellular immune response in some of such patients. Establishment of CTL clones recognizable autologous tumor cells can lead to identification of specific Ag coding genes which can be used as immunological targets.

Background
Malignant pleural mesothelioma (MPM) is an incurable and fatal malignancy of the pleural membranes. MPM has a poor prognosis and patients have a median survival of 9-13 months in clinical studies, with above 1,500 cases in the UK per year. No surgical approach has been shown to prolong survival and pemetrexed-cisplatin is now seen as the chemotherapy standard of care in the UK. However it is clear that new therapeutic strategies are urgently needed for MPM. Immunotherapy is potential new treatment approach to be considered in MPM, as the disease has been shown to respond to various immunotherapeutic strategies tested in animal models and early phase clinical trials. TroVax® consists of a highly attenuated vaccinia virus containing the human TAA 5T4 glycoprotein gene under regulatory control of a modified VV promoter, mH5. Velindre NHS Trust and the Wales Cancer Trials Unit (WCTU) have developed the SKOPOS trial to evaluate whether TroVax® is active in the treatment of MPM. SKOPOS is funded by the June Hancock Mesothelioma Research Fund, and the Velindre Cancer Centre Stepping Stones Appeal. Oxford BioMedica (UK) Ltd. is providing TroVax® vaccine free of charge, and also labeling and distribution costs. The trial is sponsored by Velindre NHS Trust, and coordinated by the WCTU.

Methods
A UK single centre, single arm, phase II trial. Eligible patients include locally advanced or metastatic, histologically or cytologically proven MPM, WHO performance status 0-1, life expectancy > 6months, at least four weeks since any previous therapy (surgery, radiotherapy). Enrolled patients will be treated with: TroVax® - intramuscular injection, dose 1 x 10[9] TCID ~50~/ml in 1ml, given on Day 1 of weeks 1, 3, 6, 9, 12, 15, 18, 21, 24. Folic Acid - 400μg oral daily from Day 2 of week 3 to Day 2 of week 16 B12 Vitamin – 1000μg intramuscular, Day 2 of weeks 3 and 12 Dexamethasone –4mg BD, Days 2-6 of weeks 4, 7, 10, 13 Pemetrexed - 500 mg/m[2] over 10 min, given on day 3 of weeks 4, 7, 10, 13 Cisplatin - 75mg/m[2] over 1h, given on day 3 of weeks 4, 7, 10, 13 The co-primary endpoints of the trial are i) cellular response to 5T4 and MVA antigens measured by intracellular cytokine staining (ICCS); and ii) antibody response to 5T4 and MVA antigens as measured by ELISA. Secondary outcome measures include safety, progression free survival, objective response rate and overall survival. The sample size was determined using Fleming’s single arm design. The outcome measure is the immune response to the 5T4 antigen. A success in this trial is defined as an increased response (at least a doubling in the 5T4 antibody or T-cell response) from that measured at baseline at any of the six follow-up time points. If an increased response is seen in at least 64% of patients then we will research the vaccine further in this group of patients. Using Fleming’s single-stage design, p1=0.40 and p2=0.64, setting alpha=0.05 (1-sided) and 80% power, 26 participants are required.

Results
Not Applicable

Conclusion
Not applicable

Background
The immune cells can prevent and inhibit tumour development but also may contribute to growth and progression of cancer. Myeloid-derived suppressor cells (MDSCs), a heterogeneous population of immature cells with immune suppressive properties, have been correlated with worse prognosis in several tumors. In addition, high levels of circulating CD4[+] T regulatory (Tregs) cells exhibit suppressive functional activity against anti-tumour T-cell responses and correlated with worse recurrence-free survival in NSCLC patients. In the present study, we investigated the expression and levels of MDSCs’ subpopulations and CD4[+]Tregs, their correlation to distinct immune cells, as well as to the clinical outcome of patients with advanced NSCLC.

Methods
Peripheral blood from 112 chemotherapy-naive patients with stage III/IV NSCLC and 27 healthy donors was analyzed with flow cytometry for the presence of monocytic and granulocytic subpopulations of MDSCs, CD4[+ ]Tregs, as well as the frequencies of distinct immune cells such as CD4[+ ]and CD8[+ ]cells, dendritic cells (DC) and B cells. The frequencies of these cells in the patients were compared to the healthy donors. The patients’ clinical outcome (PFS and OS) was compared according to the frequency of MDSCs and Tregs (high vs low expression as defined by their percentage above healthy donors).

Results
Two monocytic (CD14[+]CD15[-]CD11b[+]CD33[+]HLA-DR[-]Lin[-] and CD14[+]CD15[+]CD11b[+]CD33[+]HLA-DR[-]Lin[-]), and a granulocytic (CD14[-]CD15[+]CD11b[+]CD33[+]HLA-DR[-]Lin[-]) subpopulations as well as CD4[+] Tregs (CD3[+]CD4[+]CD25[+high]CD152[+]CD127[-]Foxp3[+]) were increased in patients compared to healthy donors (p<0.001 and p<0.007, respectively). Levels of MDSCs and CD4+ Tregs did not associated with tumor histology or stage of the disease. The levels of both subpopulations of monocytic but not of granulocytic MDSCs were reversely correlated with the levels of dendritic cells (DC) (r[2]=-0.3, p≤0.04) whereas the granulocytic subpopulation of MDSCs was reversely correlated with the levels of CD4[+] T cells (r[2]=-0.3, p=0.006). Increased baseline levels of both monocytic MDSCs’ subpopulations was associated with early relapse despite front-line platinum-based chemotherapy (p=0.05). The detection of baseline CD14[+]CD15[+]CD11b[+]CD33[+]HLA-DR[-]Lin[-] MDSCs within the normal levels was associated with longer OS compared to those with high levels (p=0.0035). Finally, patients with normal CD4[+ ]Tregs frequencies had a higher OS than those with high frequencies (p=0.05).

Conclusion
The data show that increased levels of monocytic MDSCs and CD4[+] Tregs in the peripheral blood of NSCLC patients are reversely correlated to the other normal immune cells. These cells could represent potential predictive/prognostic biomarkers since their increased levels were negatively correlated to the treatment outcome.

Background
Interleukin-15 (IL-15) is a potent pro-inflammatory cytokine, that plays a major role in stimulating immune effector cells following microbial and viral infection. IL-15 exhibits broad activity and induces the differentiation and proliferation of T, B and natural killer (NK) cells. IL-15 is primarily expressed by antigen presenting cells; however, in areas where there are high microbial loads, such as the lungs and colon, epithelial cells may also express IL-15 and its receptor (IL-15Rα). The expression of IL-15 and IL-15Rα by the lung epithelium facilitates immune responses by inducing local proliferation of NK and CD8+ T-cells in response to infection. Here we determine the status of IL-15 and IL-15Rα on lung cancers and assess their ability to stimulate the proliferation of these immune cells.

Methods
Western blotting, ELISA and qPCR were used to determine the expression of IL-15 and IL-15Rα in 6 human lung cancer cell lines and 1 normal human bronchial epithelial cell (HBEC) line. Further, mRNA from 146 primary lung cancers of all stages and tissues from 45 normal lungs were examined by multiplex qPCR for the expression of IL-15 and IL-15Rα. NK and T-cell proliferation assays were performed to determine the effects of IL-15 expression by the cell lines on these immune cells.

Results
IL-15 expression by lung cancer cell lines was significantly reduced compared to the HBEC cell line. Similar results were seen when we examined the expression of IL-15 in primary tumors, with lung cancer expressing less IL-15 than normal lung (P<0.001). When we compared IL-15 expression between stages, we found that increased stage correlated with a decrease in IL-15 expression with normal lung> stage I> stage II> stage III> stage IV. In contrast, IL-15Rα expression levels in the tumor samples were found to be largely unchanged across stage, P=0.417. Decreases in IL-15 with increasing stage may represent one mechanism in which lung cancers limit the immune response directed at the tumor and may aid in metastatic progression. In order to assess the immune effects of a reduction in IL-15 expression, we examined the ability of the lung cancer cell lines and HBEC to induce the proliferation of NK and T-cells following co-incubation. We found that the lung cancer cell lines significantly inhibited the proliferation of either NK or T-cells compared to HBEC.

Conclusion
Pro-inflammatory cytokines such as IL-15 are important for the induction of lung immunity. Decreases in IL-15 expression may reduce immune responses thereby aiding in the escape of lung cancer from immune detection and the dissemination of tumor. The differential expression of IL-15 across lung cancer stages and retention of IL-15Rα expression may make lung cancer a target for IL-15-based treatments and opens the potential for IL-15 to be used as a predictive biomarker in early stage patients.

Background
Tyrosine kinase inhibitors (TKIs) that target the epidermal growth factor receptor (EGFR) has become standard therapy in patients with EGFR activating mutations. Unfortunately, acquired resistance eventually limits the clinical effects and application of EGFR-TKIs. Although main mechanisms of acquired resistance have been detected with varying frequencies, including T790M mutation, amplicfication of other kinases and epithelial to mesenchymal transition(EMT), potential therapies for these tumors have not been modeled in vivo. Researches have shown that suppression of EMT and IL-6/STAT3 pathway may abrogate this acquired mechanism of drug resistance of TKI.

Methods
By using 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium (MTT), immunofluorescence, Western blot, cell tracking, invasion assay, ELISA assay and xenograft experiments, we analyzed the effect of metformin and TKIs on TKI resistance of tumor cells both in vitro and in vivo.

Results
Metformin, which is widely used as an antidiabetic agent, effectively increased sensitivy of TKI-resistant lung cancer cells to erlotinib or gefitinib. Metformin reversed EMT and decreased IL-6/STAT3 activation in TKI-resistant cells, while IL-6 addition in those cells bypassed the antitumor effect of metformin. Furthermore, overexpression or addition of IL-6 in TKI sensitive cells induced TKI resistance, which can be overcomed by metformin. Lastly, metformin-based combinatorial therapy effectively block tumor growth in xenografts involving TKI-resistant cancer cells, which is associated with EMT reversal and decreased IL-6/STAT3 activation.

Conclusion
Thus, the unexpected ability of metformin to reverse EMT and inactivate IL-6 axis further reveals that this biguanide, generally considered non-toxic and remarkably inexpensive, might be used in combination with TKIs in NSCLC patients harboring EGFR mutations to overcome TKI resistance. Our findings therefore offer preclinical proof of principle that combination of TKI and metformin may benefit patients with NSCLC.

Background
Lung cancer is the leading cause of all cancer deaths worldwide. Non-small cell lung cancer (NSCLC) is the most common type, accounting for 75–80% of all lung cancers. Bcl-2 is a central regulator of cell survival that is overexpressed in NSCLC and contributes to both malignant transformation and therapeutic resistance. We previously identified a small-molecule BH3 mimetic named S1 that exhibits nanomolar affinity towards Mcl-1, Bcl-2 and Bcl-XL. The purpose of this work was to study the key factors that determine the sensitivity of NSCLC cells to S1 and the mechanism underlying the resistance of this drug.

Methods
Acquired resistant cells were derived from initially sensitive NCI-H1975 cells. Western blot and co-immunoprecipitation were used to evaluate the contribution of Bcl-2 family members to the cellular response of several NSCLC cell lines to S1. Quantitative PCR and gene silencing were performed to investigate Bim down-regulation. PD98059 and MG-132 was used to inhibit the degradation of Bim.

Results
S1 can disrupt Bcl-2/Bim, Mcl-1/Bim and Bcl-XL/Bim complexes regardless of the levels of the anti-apoptotic proteins NSCLC cell lines. A progressive decrease in the relative levels of Bim characterized the increased de novo and acquired resistance of NSCLC cell lines. Furthermore, in resistant cells, acute treatment of S1 induced Bim phosphorylation on serine 69 and degradation via the proteasome pathway. ERK inhibitor PD98059 abrogated Bim phosphorylation and degradation and induced caspase activation and apoptosis. We showed that BH3 mimetics including S1 and ABT-737 induced ER stress and then activated MEK/ERK pathway in NSCLC cells. The function of MEK/ERK pathway in defining BH3 mimetics was illustrated: Bim was released from anti-apoptotic proteins by S1, ERK1/2 activation leaded to Bim sustained phosphorylation and then degraded by proteasome in naïve and acquired resistant NSCLC cells.Figure 1

Conclusion
We describe here a new mechanism for the regulation of Bim expression by phosphorylation protects NSCLC cells from BH3 mimetics induced apoptosis. These results provide significant novel insights into the molecular mechanisms for ERK1/2 mediated the crosstalk between the Bim regulation and ER stress to define BH3 mimetics in NSCLC cells.

Background
Lung cancer is the leading cause of cancer death worldwide. Non-Small Cell Lung Cancer (NSCLC) which is the predominant type is mostly advanced stage disease at diagnosis. Cytotoxic agents including chemotherapeutic drugs and radiation therapy, the mainstay of advanced NSCLC treatment are neither specific nor curative and are significant causes of morbidity in patients. This poor NSCLC outlook therefore requires a novel therapy as well as predictive markers of treatment response for an improvement. We propose Ataxia Telangiectasia Mutated (ATM), a critical player in the DNA double strand break repair pathway, as a potential predictor of treatment response in NSCLC. Radiation sensitivity is one of the hallmarks of ataxia telangiectasia (A-T), a condition due to ATM mutations. Our lab has previously demonstrated ATM deficiency in about 20% of resected NSCLC tumors and this is associated with poorer survival outcomes; however, no treatment guidelines currently take this into consideration.

Methods
We assessed in vitro the altered sensitivity of ATM deficient NSCLC cell lines to both targeted agents (e.g. Poly (ADP) ribose polymerase (PARP) inhibitors) and non-targeted agents (e.g. DNA damaging agents including ionizing radiation (IR) and the commonly used cytotoxic chemotherapies in NSCLC) using the clonogenic survival assay. A panel of NSCLC cell lines (NCI-H23, NCI-H226, NCI-H460, NC1-H522, NCI-H1793, NCI-H1395 and HCC 4006) were characterised for pre-existing ATM deficiency in terms of ATM protein expression levels and functionality using western blot. By examining the cells’ responses to IR, confirmation of ATM deficiency was achieved with clonogenic assay to assess cellular viability and western blot for the expression of IR inducible ATM-dependent phosphorylation sites on target proteins including phosphorylation of serine 1981 of ATM (p-S1981 ATM), serine 15 of p53 (p-S15 p53) and serine 824 of KAP1 (KRAB-associated protein 1).

Results
We identified NCI-H23 cells with pre-existing low ATM protein levels (11% relative to BT cells) and undetectable ATM protein in NCI-H1395 cells compare to our positive (BT cells) and negative (L3 cells) control lymphoblastoid cells derived from normal and A-T patients respectively. In addition, H23 and H1395 cells display altered ATM signaling as evidenced by no detectable level of p-S1981 ATM expression post-irradiation. There is increased sensitivity of H23 cells to DNA-damaging agents (such as IR, topotecan and cisplatin) and PARP inhibition compare to ATM proficient NSCLC cell lines.

Conclusion
Our results seem to delineate the therapeutic sensitivity of ATM deficient versus ATM proficient cell lines to both non-targeted agents (chemotherapy and radiation therapy) and to targeted agents (PARP inhibitors). ATM could serve as a potential marker in guiding the use of 1) targeted agents such as PARP inhibitors, and 2) conventional chemo-radio therapeutic agents in the treatment of NSCLC. We are in the process of establishing ATM knock-down cells from ATM proficient NSCLC cell lines. Results of the altered sensitivity of established ATM deficient NSCLC cells to our investigative agents will be discussed and presented.

Background
Background: Development of acquired resistance to cisplatin treatment is a major problem when treating patients with malignant pleural mesothelioma (MPM) and non-small cell lung cancer (NSCLC). Chemotherapy leads to tumor cell stress activation of glucosylceramide synthase (GCS) to eliminate ceramide by glycosylation and formation of glycosphingolipids (GSLs) such as globotriaosylceramide (Gb3), the functional receptor of verotoxin-1. Ceramide elimination leads to stimulated cell proliferation and blocked apoptosis, thus stimulating tumor progression. GSLs also transactivate multidrug resistance 1/P-Glycoprotein (MDR1) and possibly multidrug resistance protein 1 (MRP1) expression which confers tumour cell resistance by further preventing ceramide accumulation and stimulating drug efflux. We investigated if Gb3, MDR1 and MRP1 are co-expressed and co-localized in MPM and NSCLC cells with acquired cisplatin resistance and if GSC activity inhibitors DL-threo-1-phenyl-2-palmitoylamino-3-morpholino-1-propanol (PPMP) and cyclosporin A would reduce their expression and relieve cisplatin-resistance.

Methods
Methods: Cell surface as well as intracellular expression of Gb3, MDR1 and MRP1 was analysed by flow cytometry and confocal microscopy using specific protein antibodies on P31 MPM and H1299 NSCLC cell lines and corresponding sub-lines (P31res, H1299res) with acquired cisplatin resistance.

Results
Results: Gb3 and MDR1, but not MRP1 were co-expressed and partly co-localized on the cell surface, and Gb3 and MDR1 as well as MRP1 intracellular co-expressed but not co-localized in P31res and H1299res cells. P31 cells expressed minute cell surface Gb3 and the non-resistant cells had less cell surface and but similar intracellular expression of Gb3 and MDR1. Glycosphingolipid synthesis inhibitors PPMP and cyclosporin A radically decreased intracellular Gb3, MDR1 and MRP1-expression in all cell sub-lines whereas cell surface Gb and MDR1 expression was decreased only by PPMP but not by cyclosporin A.

Conclusion
conclusion: These results indicate that cell surface Gb3 that is co-expressed and co-localised with MDR1 is a likely tumour biomarker for acquired cisplatin resistance in MPM and NSCLC and that therapy with GCS activity inhibitors or Gb3 blockers affecting ceramide metabolism may overcome or substantially reduce acquired cisplatin drug resistance. Targeting the functional interplay between Gb3 and MDR1 might aid in the development of new drug therapies against acquired drug resistance in MPM and lung cancer.

Background
Background: Platinum-based drugs, such as cisplatin, are the standard treatment for aggressive malignant pleural mesothelioma (MPM) and non-small-cell lung cancer (NSCLC), but inherent as well as acquired resistance are major clinical problems leading to therapy failure and low median survival after diagnosis. Cisplatin exposure initiates the mitochondrial signaling pathway of apoptosis, by activation of BH3-only proteins i.e. pro-apoptotic members of the Bcl-2 family of proteins. Therapy failure may be the result of decreased apoptosis due to caspase-9-deactivation or over-expression of the anti-apoptotic proteins Bcl-X~L~ and Mcl-1. Affecting cisplatin resistance by targeting post- and off-target apoptosis signalling proteins with pro-apoptotic BH3-mimetics, would possible sensitize cancer cells to cisplatin treatment.

Methods
Methods: We investigated the expression of Bcl-2 family and other proteins involved in apoptosis signal transduction and the difference between the response to equiapoptotic cisplatin concentrations as well as the response to the pro-apoptotic BH3-mimetics ABT-737 and GX15-070, alone or in combination. To separate mitochondrial-dependent from –independent signalling we compared initiator-caspase-dependent parental P31 MPM cells with its acquired cisplatin-resistant (P31res) sub-line which has initiator-caspase-independent caspase-3-activated apoptosis,

Results
Results: On acquisition of cisplatin-resistance, the expression of the pro-apoptotic and anti-apoptotic Bcl-2-family proteins was either not changed or slightly decreased in P31res cell compared to parental P31 cells. A 6-h exposure to equiapoptotic concentrations of cisplatin, on the other hand, increased the expression of potent pro-apoptotic BH3-only proteins as well as the anti-apoptotic Bcl-x protein in P31 but not P31res cells whereas Bcl-x expression was almost annihilated in P31res cells. TUNEL results showed a synergic effect on apoptosis when cisplatin was combined with the BH3-mimetic GX15-070 in P31res, but only an additive effect in P31. The BH3-mimetic ABT-737 did not augment cytotoxicity or apoptosis either per se or when combined with cisplatin and/or GX15-070. GX15-070 efficiently inhibited anti-apoptotic Bcl-2-family-, inhibitors of apoptosis (IAP) - and heat shock protein (HSP) - family protein expression both with and without cisplatin in P31 cells, whereas preserved protein expression was noted in P31res cells after 6 h incubation with cisplatin. Sub-toxic concentrations of the IAP-inhibitor AT-406 and the HSP90-inhibitor 17-AAG with GX15-070 markedly potentiated cisplatin cytotoxicity in P31res cells.

Conclusion
Conclusion: P31 malignant mesothelioma cell acquisition of cisplatin-resistance led to deregulated Bcl-2 family protein expression and induced post-target apoptosis resistance to the BH3-mimetic GX15-070. GX15-070 had a synergistic effect on cisplatin-induced apoptosis in P31res cells. The synergy was due to efficient GX15-070 inhibition of expression of the anti-apoptotic Bcl-x protein despite apoptosis resistance by preserved IAP and HSP protein expression. Cisplatin therapy combined with GX15-070 in acquired cisplatin resistance was even more efficacious when combined also with inhibitors of IAP- and HSP- apoptosis signalling pathways.

Background
It has been found that HGF-dependent c-Met（HGFR） autocrine is activated in a wide variety of human primary and second malignancies. On the other hand, the metastatic growth potential of tumors can be activated through paracrine mechanism. c-Met dysregulation leads to lung cancer development through overexpression and mutation.

Methods
70 kinase enzymogram screening was proceeded by Z-lyte technique. MOA analysis was completed on the inhibited kinases. Cell vitality was determined at 24h, 48h, 72h after treatment through CCK8 method. Transwell system was used to observe the inhibition of cell migration. Difference of special gene expression was evaluated by Real-time PCR. Westernblot assay was used to compare the expression difference of c-MET and p-MET. HGF level in culture medium is determined by ELISA.

Results
SIM-89 can inhibit 3 kinases including c-Met（IC50=297nM）, AMPK, TRKA (IC50=150.2nM). SIM-89 has an ATP competive inhibition on c-Met. By Real-time PCR, SIM-89 has been found to inhibit STAT1, JAK1, c-Met gene expression in H460 cell. P-Met expression of A549, H441, H1299 and B16F10 cell can be inhibited by SIM-89. HGF level of supernatant in culture is significantly lower than control group. Vitality of NSCLC cell lines is inhibited dependent on time and concentration by SIM-89. Induced by HGF, migration of H460, H1299 cell is inhibited.

Conclusion
SIM-89 has significant inhibitive effect on c-Met, TRKA kinases. It also can inhibit proliferation, migration and HGF autocrine of NSCLC cell significantly. Further study in vivo should be carried to explore the pharmacokinetics of SIM-89.

Background
Lung cancer accounts for the greatest cancer related mortality world wide, with two thirds of all patients receiving radiation therapy as part of their cancer care. Despite advancements made, the overall 5-year survival in Canada remains less than 20%. Developments in nanotechnology have proven to possess exciting potential in the treatment of various malignancies. Nanoparticles with high atomic number, such as Gold (GNPs), facilitate surprising local enhancement secondary to gold’s strong photoelectric absorption coefficient. GNPs are capable of forming covalent bonds, enabling the creation of “targeted” radio-sensitizing agents. The goal of this study is to evaluate the in vitro and in vivo radiation potential and biodistribution profile of GNPs targeted against the epidermal growth factor receptor (EGFR) using the monoclonal antibody Cetuximab (GNP-cetuxumab) stabilized with thiolated polyethylene glycol (SH-PEG) for the treatment of non-small cell lung cancer (NSCLC). To date a comprehensive evaluation of such a platform has not been undertaken.

Methods
We examined the radiation enhancement of 50nm GNPs in vitro using SKMES-1 (High wild type EGFR expression) and h460 (low wild type EGFR expression) NSCLC cell lines, both possessing Kras mutations. MTS, clonogenic assays and flow cytometry were used to assess radiation effect using 4 groups (no GNPs, GNPs, GNPs bound to SH-PEG stabilizer, GNP-Cetuximab). In vivo biodistribution was conducted on balb-c nude mice bearing two flank subcutaneous SKMES-1 xenografts. One tumor received a single radiation exposure (200 kVp, 8 Gray) prior to tail vein injection of GNP-cetuximab or GNP-PEG in order to assess the effect radiation induced inflammation may have on GNP tumor uptake. Tumors and organs were harvested at various time points with GNP concentration determined using inductively coupled mass spectroscopy. In vivo radiation experiments were conducted on 3 flank tumor bearing groups (each group = 7): 1) radiation only, no nanoparticles 2) GNP-PEG, GNP stabilized by bound PEG and 3) GNP-Cetuximab. Four weekly radiation fractions (4Gy, 200 kVp) with weekly tail vein injection timing based on the biodistribution results were administered. Tumor growth kinetics was then evaluated.

Results
Significant in vitro radiation effect was observed in the GNP groups compared to the radiation only group. GNP-cetuximab and GNP-PEG demonstrated enhanced radiation effect as compared to unfunctionalized GNPs. In the biodistribution experiment, the peak intra-tumor concentration without pre-administered radiation in the GNP-PEG group was twice that of the GNP-Cetuximab group 5 days after tail vein injection. Tumor pre-irradiation resulted in a doubling of intra-tumor nanoparticle uptake in both groups. At the end of radiation therapy experiment, the GNP-PEG group demonstrated the greatest reduction in tumor growth as compared to the radiation alone group (52mm[3] vs 180mm[3 ]respectively, p<0.01).

Conclusion
Despite the superiority of GNPs bound to cetuximab in vitro as a radiation enhancer, the favorable tumor biodistribution of the GNP-PEG accounted for the most dramatically reduced tumor growth kinetics observed. The future utility of targeted nanoparticles requires further investigation in light of these findings. In this study we demonstrate the exciting in vitro and in vivo potential of GNPs in the radiation treatment of NSCLC.

Background
Constitutive activation of the phosphoinositide 3-kinase (PI3K)/Akt pathway has been found in non–small cell lung cancer (NSCLC) and promotes cancer progression. Myristoylated alanine-rich C kinase substrates (MARCKS) is a substrate of protein kinase C (PKC), and acts as a key regulatory protein controlling cell motility and signaling. We previously reported that elevated MARCKS phosphorylation (pSer159/163) potentiates lung cancer cell malignancy by upregulation of AKT/Slug axis.

Methods
Tissue array and immunohistochemistry were performed to analyze MARCKS phosphorylation in 110 pairs of NSCLC tumor cells and corresponding normal tissues. The enforced and silenced expressions of MARCKS were performed in lung cancer cells to verify the role of MARCKS expression and its phosphorylation. In vitro and in vivo anticancer activities of a MARCKS ED domain peptide (MPS) were confirmed by MTS, colony formation, flow cytometry, invasion, migration assays and in vivo subcutaneous and orthotopic implantation. The molecules regulated by MPS peptide were determined by Western blotting, PIP3 pool assay and co-immunoprecipitation.

Results
We demonstrated that elevated MARCKS phosphorylation is correlated with advanced-stage and lymph node metastasis of lung cancer. siRNA knockdown of MARCKS expression confirmed the importance of MARCKS and its phosphorylation in modulating PIP3 pool and cancer cell survival. Furthermore, we have identified that a small peptide, MPS, which mimics the basic effector domain (ED) of MARCKS is very effective in suppressing phospho-MARCKS and PIP3 levels in lung cancer cells. MPS peptide treatment is able to inhibit cancer cell viability, colony formation, migration and invasion in vitro, as well as tumorigenesis and metastasis in vivo. Interestingly, MPS peptide is very cytotoxic to cancer cells with highly activating PI3K/AKT signaling and malignant phenotypes, while MPS has no cytotoxic effect on normal human bronchial epithelial cells. In addition, a co-treatment of MPS peptide with epidermal growth factor receptor inhibitor erlotinib could reverse drug sensitivity of these tyrosine kinase inhibitor (TKI) resistant cells, H1650 and H1975, toward erlotinib.

Conclusion
These results suggest a therapeutic potential in lung cancer treatment by MPS peptide through the sequestration of PIP2 pool and the suppression of MARCKS phosphorylation and PI3K/AKT pathway.

Background
Malignant mesothelioma (MM) is an aggressive, fatal, tumour of the pleura or peritoneum and is strongly related to asbestos exposure. Clinically, there is no curative therapy for MM and the profound chemoresistance of MM is well documented, reportedly being due to the ability of MM cells to escape cell death. Unravelling the molecular mechanisms employed by MM cells to evade cell death will therefore provide new insights into key cell death pathways that may be targeted for successful therapy. Currently, the Bcl-2 repertoire is an undervalued and attractive pharmacological target for mesothelioma therapy. The aims of this study were to investigate the sensitivity of MM to the BH3-mimetic, ABT-737, and identify mechanisms of overcoming resistance to support personalized therapy.

Methods
Sensitivity to the highly selective BCL-2/BCL-XL antagonist, ABT-737, was investigated using FACs-based analysis and immunoblotting techniques in a panel of MM cell lines and tumour cells isolated from freshly resected MM tumours. In addition, patient-derived tumour explants were similarly analysed, providing a clinically relevant 3D mesothelioma model. Furthermore, to explore the possibility that tumour cell metabolism may modulate the sensitivity of MM cells to ABT-737, we investigated the effect of inhibiting glycolysis with 2-deoxyglucose (2-DG) on ABT-737 -induced apoptosis in relevant pre-clinical models of MM.

Results
The majority of MM cell lines and freshly-derived cultured tumour cells were resistant to ABT-737, suggesting a deregulation of cell death signalling at the level of mitochondria. Aerobic glycolysis (Warburg effect) is a process by which tumour cells gain not only growth advantage (providing much needed “building blocks”) but also an increased survival potential. Importantly we report that, in MM, glycolysis can be inhibited by 2-deoxyglucose(2-DG), a competitive inhibitor of hexokinase. Although exposure to 2-DG did not induce cell death, 2-DG induced a time- and concentration-dependent potentiation of ABT-737 -induced apoptosis in either established mesothelioma cell lines or newly-derived tumour cells from patients. BCL-2-family member profiling revealed that, while the predominant pro-survival members expressed in MM were MCL-1 and BCL-XL, 2-DG selectively decreased MCL-1 protein levels, a leading cause of resistance to ABT-737 in other tumour models.

Conclusion
High constitutive levels of MCL-1 most likely explain the inherent resistance of MM cells to the highly selective BCL-2/BCL-XL antagonist, ABT-737. Importantly, 2-DG-dependent down-regulation of MCL-1 provides a potential mechanism for overcoming resistance to ABT-737 in the clinical setting.

Background
Met receptor phosphorylation is associated with poor prognosis in human SCLC. Several Met inhibitors are being tested for the treatment of different neoplasms. Met activation has been shown to be an inductor of epithelial to mesenchymal transition (EMT) in a number of tumor models. The aim of our work was to investigate the effects of hepatocyte growth factor (HGF)/Met induced EMT in SCLC, to evaluate the role of Met inhibition in mesenchymal/chemorefractory SCLC models and to investigate the significance of EMT in human SCLC.

Methods
Biological features (growth, invasiveness, tumorogenesis) and chemosensitivity of SCLC models (H69) of HGF-induced EMT (H69M) were evaluated in vitro and in vivo (subcutaneous xenografts in BALB/c nude mice). Mice with mesenchymal chemoresistant SCLC xenografts were treated with etoposide, the Met inhibitor PF-2341066 (Crizotinib) and the combination. Human SCLC samples at diagnosis and relapse were evaluated by immunohistochemistry and immunofluorescence for EMT markers and Met status and correlated these with patient outcome. Association between clinical-pathological characteristics were tested with Chi-Square and Fisher tests. Differences in survival according to biomarker status were evaluated by log-rank and Cox regression models, assuming a 2-sided statistical significance p< 0.05.

Results
We identified that the activation of the Met receptor through HGF induced expression of mesenchymal markers (Snail1, SPARC, vimentin) and downregulation of E-cadherin. This derived in increased tumorogenesis, local invasion and chemoresistance in xenograft models. The combination of etoposide and PF-2341066, but not the Met inhibitor alone significantly decreased tumor growth in this chemoresistant/mesenchymal models. Moreover, Snail1, SPARC and vimentin expression in human SCLC specimens (N:87) was significantly associated with Met activation (co-localization by immunofluorescence). Expression of mesenchymal markers predicted worse survival (all p-values <0.05) in the multivariate analysis. In 5 paired biopsies, we observed upregulation of mesenchymal markers and p-Met in chemorefractory disease.

Conclusion
These results provide novel evidence on an important role of Met-dependent EMT in the adverse clinical behavior of SCLC and support clinical trials of Met inhibitors and chemotherapy in this fatal disease.

Background
Gefitinib or Erlotinib (EGFR-TKI) is the first choice of treatment for advanced NSCLC patients harbouring activating EGFR mutations. In addition to primary resistance, acquired resistance to EGFR-TKI eventually occurred in patients after an initial response. New agents have been developed to specifically inhibit the signaling pathways involved in mediating resistance. Because of heterogeneous resistant mechanisms, single agent usually had limited efficacy. Thus drug combination therapy may offer more benefits by synergistic interactions and avoidance of resistance emergence. We studied the combination of afatinib (an irreversible ErbB family inhibitor) and dasatinib (an inhibition of Src, BCR-ABL, PDGFR, Eph) in 8 different genetically characterized NSCLC cell lines to evaluate resistance mechanisms and molecular predicting biomarkers.

Methods
Growth inhibition was assessed by MTS assay. The interaction between two different drugs was evaluated by the method of Chou and Talalay. EGFR and k-Ras mutations were tested by direct DNA sequencing. EGFR, HER2, Src and downstream proteins FAK, Akt, MAPK42/44, Stat3, Stat5 expressions were measured by western blot.

Results
The efficacy of dasatinib against NSCLC cells in vitro is significantly more potent than gefitinib (p<0.001) and anti-EGFR monoclonal antibody cetuximab (p<0.05). Afatinib demonstrated increased activity against gefitinib and cetuximab resistance cell lines. Synergistic interaction (combination index, CI< 1) between dasatinib and afatinib was found in 7 NSCLC cell lines except A549. The efficacy of dasatinib in combination with afatinib is significantly stronger than that of cetuximab in combination with afatinib (p<0.001). In gefitinib resistant cell lines, growth inhibition by dasatinib was correlated with the ratio of p-Akt/t-Akt (p<0.05); total FAK expression is correlated to the growth inhibition by afatinib (p<0.05); CI results between dasatinib and afatinib were correlated with the active ratio of p-FAK925/t-FAK (p<0.05). In H1650 cell line, which was resistant to both dasatinib and afatinib, the combination of both drugs significantly inhibited the activity of p-EGFR845, p-FAK925, p-Src416, p-Akt473 and p-MAPK42/44 when comparing with that treated by afatinib or dasatinib alone (p<0.05). No significant inhibition was found on p-Stat3 and p-Stat5 by dasatinib. Afatinib was able to reduce the activity of Stat3 but not Stat5. No significant effect was shown on p-Stat3 and p-Stat5 by the combination of afatinib and dasatinib.

Conclusion
Our study showed synergistic combination of afatinib and dasatinib is more potent than cetuximab plus afatinib against gefitinib resistant NSCLC cells; it inhibited cell proliferation in H1650 cell line via affecting SFK/FAK, PI3K/PTEN/Akt, and Ras/Raf/MEK/ERK, but not JAK/Stat pathways. The level of p-FAK925/t-FAK may be a useful biomarker predicating synergism between afatinib and dasatinib for the treatment of gefitinib resistant NSCLC cells. P-Akt/t-Akt and total FAK expression may be related to sensitivity to dasatinib and afatinib respectively.

Background
Psammplin A (PsA), a novel DNA methyltransferase (DNMT) inhibitor and histone deacetylase (HDAC) inhibitor, was reported to induce radiosensitivity in lung cancer cell line. Concerning in vivo tissue distribution characteristics, PsA was found to be highly distributed to lung tissues. However, it appeared to be unstable in biological matrices. Here we report the DNMT inhibitory effect of PsA derivatives and their potential for radiation sensitization in lung cancer cell line.

Methods
A549 lung cancer cell line was used in verification of DNMT inhibitory effect with cultured cell DNA extraction kit. A549 cell line was exposed to radiation with or without a total of 9 PsA derivatives for 18 hours prior to radiation, after which cell survival was evaluated via clonogenic assays.

Results
These 9 PsA derivatives all showed DNMT inhibitory effect and inhibited cell proliferation at the ranges of IC50 30–120μM. Furthermore, radiation clonogenic assays revealed that these compound shows radiosensitizing properties in A549 lung cancer cell line.

Conclusion
This preliminary data support the further investigation of these derivatives for use as radiation sensitizing agents with potential for clinical application.

Background
Background: New therapeutic strategies are urgently needed for small cell lung cancer (SCLC) which accounts for approximately 29,000 cases annually in the U.S. SCLC tumors have rapid doubling times and a propensity for early development of widespread metastatic disease. There have been no therapeutic advances in recent decades. The microtubule-targeting agent eribulin is a mechanistically-unique inhibitor of microtubule dynamics. Eribulin binds with high affinity to a maximum of 15 distinct beta-tubulin binding sites inhibiting microtubule growth by depolymerization and sequestration of tubulin into non-productive aggregates. Eribulin is currently FDA approved for the treatment of metastatic breast cancer patients with a demonstrated significant increase in overall survival in patients that are refractory or resistant to multiple chemotherapy agents. We investigated the effects of eribulin in a panel of 15-human SCLC lines.

Methods
Methods: Growth inhibition by eribulin was assessed by tetrazolium based assays at 5-days post treatment with varying drug concentrations and the growth inhibitory (GI50) was calculated. Erbulin induced cell cycle arrest was monitored following propidium iodine staining and analysis by FACS. Apoptosis was determined by using the DNA binding dyes YOPRO and PI and analysis by FACS.

Results
Results: Eribulin inhibited the growth of 13 of the cell lines. Four SCLC lines had GI50s of < 1 nM and 9-lines had eribulin GI50 values of 1-6 nM. These GI50 are similar to those reported in breast cancer cell lines. The eribulin IC50 value in the remaining 2-lines was > 100 nM. We are currently exploring clinical and gene expression differences that may explain the high sensitivity and resistance of different cell lines. A two to three-fold increase in the % cells in the G2/M phase of the cell cycle were observed following an 18-hour exposure to eribulin at concentrations of ≤ 5 nM in all cell lines growth inhibited by eribulin. Apoptosis assays are ongoing and in vitro studies of eribulin in combination with radiation are planned.

Conclusion
Conclusions: Erbulin inhibited the growth of SCLC lines and induced a significant G2/M arrest. Confirmation of growth inhibition of SCLC cell lines in an in vivo nude mouse model would support human studies in SCLC patients.

Background
In vivo evaluation is essential for development of lung cancer treatment. However, the subcutaneous xenograft models are not closely reproducing microenvironment of lung cancer. Although orthotopic implantation SCID mouse model of lung cancer presents lymphatic metastasis to mediastinum or pleuritis carcinomatosa with progression of disease, it has been difficult to evaluate the efficacy of treatment without sacrifice of model mouse. Positron-emission tomography-computed tomography (PET-CT) with the glucose analogue [18F] fluorodeoxyglucose (FDG) has been recently applied for evaluating tumor response to anticancer therapy. We have evaluated the utility of FDG PET-CT in orthotopic implantation SCID mice model of lung cancer.

Methods
Animals: 6 weeks male SCID mice (n=12). Cell line: Ma44-3 cloned from Ma44 (human squamous cell lung cancer cell line). Under sufficient anesthesia, mice were placed in the left lateral decubitus position. A 1-cm transverse incision was made in the right lateral skin just below the inferior border of the scapula. After intercostal muscles were exposed, 2 x 10[6] tumor cells/ml with 400 μg/ml Matrigel® was injected into the right lung in a volume of 10 μl (2.0x10[4 ]cells) of medium. Four or 5 days after implantation (6 mice on day 4 and other 6 mice on day 5), the SCID mice were examined with FDG PET-CT and mice whose lung tumors were identified were randomized to treatment group and control group. Treatment group mice received intraperitoneal injection of cisplatin (7mg/kg) on day 6 after implantation. All mice were examined with FDG PET-CT on day 8 and 13 after implantation. Tumor volume and maximal standardized uptake value (SUV max) of the lung tumor were calculated for all mice. All SCID mice were sacrificed on day 13 after implantation for histopathologic analysis.

Results
Six mice whose lung tumors were identified at the first FDG PET-CT were randomized to treatment group (n=3) and control group (n=3). The average growth rates (day 13 versus day 5 or 6) of tumor volume and SUV max of the treatment group were 144% and 108%, respectively, whereas the average growth rates of tumor volume and SUV max of the control group were 1470% and 271%, respectively.

Conclusion
Tumor growth and inhibition were evaluated by FDG PET-CT in orthotopic implantation SCID mice model of lung cancer. This in vivo evaluation system is useful for development of lung cancer treatment.

Background
Targeted therapeutics to mutant-BRAF and MEK have demonstrated remarkable efficacy in BRAF[V600E] metastatic melanomas. Unexpectedly, similar responses to the BRAF inhibitor vemurafenib (PLX4032) were not observed in BRAF[V600E] colorectal cancers (CRCs). Approximately 3% of lung cancers carry BRAF mutations, with only half being BRAF[V600E]. Recent case studies have reported responses to vemurafenib in patients with BRAF[V600E] NSCLC, suggesting that inhibition of the MAPK pathway with targeted therapeutics may be a feasible treatment strategy for BRAF-mutant lung cancers.

Methods
A panel of NSCLC cell lines carrying BRAF[V600E], BRAF[G469A], BRAF[G466V] and BRAF[L597V] mutations were assessed for responses to the MEK inhibitors Selumetinib (AZD6244), Trametinib (GSK1120212) and PD-325901 using standard drug response assays. Cell lines were also analysed by Western Blot analysis at various time-points following treatment with MEK inhibitors to determine the effects on the signalling pathways within the cells.

Results
Although all the cell lines displayed some sensitivity to MEK inhibition, the level of cell death observed with MEK inhibitors as single agents was minimal. The responses to MEK inhibition varied in cell lines carrying the same BRAF mutation indicating that other genetic differences could influence responses to inhibition of the MAPK pathway. Inherent co-expression of specific Receptor Tyrosine Kinases such as EGFR was detected in some of the cell lines and an increase in P-EGFR was observed following MEK inhibition. Increase in P-EGFR has been reported as the mechanism of resistance to vemurafenib in BRAF-mutant CRCs. In cell lines carrying non-V600E BRAF mutations, MEK inhibitor treatment also resulted in an increase in P-MEK, revealing a release of the MAPK pathway negative feedback resulting in upstream activation of MEK. We are currently focusing on assessing various combination therapies with MEK inhibitors, such as RTK inhibitors or second generation RAF inhibitors to overcome the feedback effect and drug-induced enhanced RTK activity. We have observed synergy in some cases, such as with the BRAF inhibitor AZ628 and the MEK inhibitor PD-901 in the H1755 cell line carrying the BRAF[G469A] mutation. In other cases we have observed additive effects suggesting co-existing oncogenic signalling.

Conclusion
Our findings suggest that different combination therapies are effective in the various mutant-BRAF cell lines and efficacy may depend on co-existing oncogenic ‘drivers’ and compensatory signalling mechanisms. We predict that combination treatment strategies for effective responses in BRAF-mutant NSCLCs may need to be determined on a personalised basis following tumour characterisation for co-expression of additional activators of the MAPK signalling pathway.

Background
Inactivation of STK11/LKB1 is one of the most common genetic events in lung cancer, and understanding the cellular phenotypes and molecular pathways altered as a consequence will aid the development of therapeutic strategies targeting LKB1-deficient cancers.

Methods
We report the comprehensive analysis of gene and protein expression patterns associated with LKB1 loss in lung adenocarcinomas, through which we identify hallmarks of altered tumor metabolism and down-regulation of the PI3K/AKT pathway.

Results
Significant differences are observed between human tumors and those derived from a genetically engineered mouse model of LKB1 loss. A 16-gene signature is predictive of both mutational and non-mutational LKB1 loss in human tumors. Cell lines expressing this signature show increased sensitivity to MEK inhibition, independent of mutations in RAS and RAF family members. Restoration of LKB1 in lung cancer cell lines down-regulates the gene expression pattern, attenuates FOXO3, and induces resistance to MEK inhibition.

Conclusion
These findings identify characteristic phenotypic features of LKB1-deficient tumors and identify LKB1 loss as a novel determinant of MEK sensitivity.

Background
Respiratory gated radiotherapy (RGRT) is a modern radiotherapy technique, which is increasingly used technique to take account of respiratory motion. One potential risk of RGRT is the possibility of sublethal repair during treatment leading more tumor cell survival. This study assessd the influence of elongation of treatment time in respiratory gated radiotherapy by measuring cell survival in vitro.

Methods
Human lung cancer cell lines, H460 and H1299, were irradiated with 6MV photons using Varian 21EX linear accelerator in two different delivery models. Cells of conventional model irradiated continuously while the other cells irradiated with gated delivery. Doses of 2 Gy, 4 Gy and 8 Gy were studied. In conventional model, treatment times were 0.5 min for 2 Gy and 3min for 8 Gy including latency for multiple field. In gated delivery model, treatment times were 2.5 min for 2 Gy and 11 min for 8 Gy with 20% gating efficacy.

Results
H1299 cells were radioresistant than H460 cells (P<0.001). Elongation of treatment time caused significant increase in survival fraction in H1299 cell line (p=0.046). In H460 cell line survival fractions also increased but differences were not statistically significant (p=0.107). Relative differences between two delivery models in log scale survival fraction were not increased in higher dose.

Conclusion
The biologic effects of protracted delivery with gated technique were different between cell lines, suggesting influence of diversity of sublethal damage repair. More radioresistant cell lines were affected by elongation of treatment time. These results suggest that potential risk of sparing tumor in prolonged delivery time with gated radiotherapy should be considered in resistant tumor at the clinic.

Background
The pro-apoptosic Bcl-2 family member BIM is known to be a critical mediator of targeted therapy-induced apoptosis in lung cancer. Also, pretreatment level of BIM strongly predicted the capacity of EGFR inhibitors to induced apoptosis in EGFR-mutant cancers. But prevous studies were mostly done in two-dimensional (2D) cell cultures (monolayers).

Methods
So we used three-dimensional(3D) spheroid culture model that were proved to be a valuable platform to study acquired multicellular apoptotic resistance and effectively recapitulate some of the complexity of solid tumors such as mesothelioma. 3D spheroids with EGFR-mutant cell lines (H3255, HCC827, H1650, 11-18, H4006) were generated in polyHEMA-coated plates and TKI treatments were done both in 2D and 3D. The degree of apoptosis and expressions of BIM by immunoblotting before and after treatment were compared.

Results
Contrary to other solid tumors, spheroids showed more prominent apoptosis to TKI than monolayer. Spheroids expressed more BIM than did monolayers except H4006. BIM was not expressed in H4006 (2D, 3D), but BIM expression was found to be elevated after TKI treatment only in spheroids. Higher level of BIM in spheroids had the tendency of higher apoptotic response to TKI.

Conclusion
3D spheroids cultures are useful in studying apoptotic mechanism in EGFR-mutant lung cancer. BIM expression before and after treatment of TKI might be useful predictor in EGFR-mutant spheroids model. So therapies that upregulated BIM expression can improve the efficacy of TKI therapy.

Background
Current treatments for mesothelioma typically increase median survival by a matter of months. Progress in treatment has been hampered by lack of a suitable small animal model, which could guide clinical advances, given limited numbers of patients eligible for clinical trials. To this end we recently developed a transgenic mouse model of mesothelioma in which the viral oncogene, SV40TAg (TAg) is directed to mesothelial cells by use of the cell type specific mesothelin promoter. MexTAg mice develop mesothelioma rapidly and uniformly, but only following exposure to the natural carcinogen, asbestos. The model closely mimics the human disease and is thus ideal for both rapid analysis of novel therapeutic studies and for investigating factors that might act synergistically with asbestos to cause disease. Since all MexTAg mice develop mesothelioma following asbestos exposure the model is highly suitable for early intervention and cancer prevention studies. An effective cancer prevention strategy for the millions of people who have been exposed to asbestos could have enormous benefit worldwide. Epidemiological evidence indicates that supplementation with some dietary factors or use of common drugs such as statins and non-steroidal anti-inflammatory drugs is associated with a lower incidence of cancer.

Methods
not applicable

Results
We previously reported that dietary supplementation with a number of antioxidants did not alter the time to develop disease nor overall survival, despite the widely accepted hypothesis that asbestos catalyzed production of reactive oxygen and nitrogen species contribute to the development of this cancer. We have extended these studies to test whether vitamin D, non-steroidal anti-inflammatories, statins and some other candidate diets could alter the pattern of disease in the MexTAg model. Supplemented diets were provided at levels based on published data and began 2 weeks prior to asbestos exposure in order to maximize our chance of detecting a benefit. Preliminary data suggests a single agent or nutraceutical may not be enough to prevent the multiple pathways involved in tumorigenesis. This is somewhat supported by epidemiological data from the Wittenoom cohort of asbestos exposed workers and residents.

Conclusion
In conclusion, we think it is unlikely that antioxidants, anti-inflammatories or other nutrient-specific dietary supplements will moderate the rate of mesothelioma in asbestos exposed populations.

Background
Malignant pleural mesothelioma (MPM) is an aggressive cancer of the mesothelial cells and is becoming a worldwide health burden. Progress in the treatment of MPM remains difficult, underscored by the fact that no single treatment option has proven particularly effective and chemo-resistance is a common problem. However, epigenetic modifiers, such as histone deacetylase inhibitors (HDi) have emerged as a novel class of anti-cancer agent and are currently undergoing clinical trials in numerous cancers. There is also evidence to suggest that HDAC expression may play a role in the development of chemo-resistance. We investigated the levels of HDACs in MPM cell lines, patient samples and determined the effect of HDi on MPM cisplatin sensitive and resistant cell lines.

Methods
A panel of MPM cell lines (n=16) was screened for the expression of HDAC11, Class I (HDAC1/2/3/8) and Class II (HDAC4/5/6/7/9/10) HDACs at the mRNA level (RT-PCR) and at the protein level (Western Blot). The HDAC expression profile was determined in an isogenic cell line model consisting of a cisplatin sensitive (Parent) and resistant (CisR) MPM cell line, P31. In addition, HDAC expression was examined in panel of twenty patient samples (benign/biphasic/sarcomatoid/epithelial). Also, MPM TMAs were stained by imummo-histochemistry for HDAC5 expression. Furthermore proliferation assays (BrdU ELISA) were performed to determine the effect of two HDi on the p31 cell lines. The HDi used were Vorinostat (pan HDAC inhibitor) and 19i (selective HDAC5 inhibitor).

Results
HDAC11, Class I and II HDACs were detected to varying degrees at the mRNA and protein level within our cell line panel. In the P31 isogenic parent/cisplatin resistant, HDAC protein expression was decreased (HDAC2/3/4/5/7) in the CisR compared with the Parent. HDAC5 was significantly reduced at both the protein and the mRNA level (p<0.05). The expression of HDACs 1/2/3/5 were increased in the MPM patient samples (n=15) compared with benign (n=5) (HDAC2/3/5, p<0.05). HDAC5 staining in a cohort of MPM samples, demonstrated no significant association between survival and a low HDAC5 score. However females had a greater median survival than their male counterparts. Vorinostat and 19i significantly reduced the proliferative capacity of the p31 Parent and CisR. SAHA was more effective in the Parent cells compared with CisR. The effect of 19i was similar in both cell lines.

Conclusion
Altered HDAC expression was observed in an isogenic Parent and CisR MPM cell model. Work ongoing in non-small cell lung cancer (NSCLC) has also demonstrated significantly reduced HDAC5 levels in CisR compared with Parent. This may suggest that a reduction in HDAC expression is involved in cisplatin resistance. Reduced levels of HDAC5, in an MPM TMA, were not associated with survival. It should be noted, however that patient numbers were small and biphasic and sarcomatoid sub-types had the lowest expression levels of HDAC5. We are currently increasing our patient numbers for an additional IHC study. HDi significantly reduced the proliferative capacity of CisR and Parent MPM cells. HDAC levels may represent an important biomarker in stratifying patients for future epigenetic targeted therapies in MPM.

Background
Malignant pleural mesothelioma (MPM) is an aggressive inflammatory cancer associated with exposure to asbestos. Untreated, MPM has a median survival time of 6 months, and most patients die within 24 months of diagnosis. There is an urgent unmet need to identify new therapies for this disease. Receptor tyrosine kinases (RTK) are an area of active research in cancer therapy. The identification of “addicted” signalling networks could rapidly lead to candidate inhibitors (RTKi) that could potentially be uused to target MPM. We have previously identified RON/MST1R as a candidate therapeutic target in MPM. Additional RTKs identified by other studies include Axl and c-MET. The agent BMS-777607 is an orally bioavailable small molecule with the capacity to inhibit c-MET, RON/MST1R, Axl and Tyro3 (at nanomolar concentrations) and has just completed a Phase I/II trial (ClinicalTrials.gov Identifier: NCT00605618). This drug may therefore have applicability in MPM.

Methods
A panel of MPM cell lines inluding the normal pleural cells LP9 & Met5A were screened for expression of Tyro3, c-MET, RON/MST1R and Axl by RT-PCR. A cohort of snap-frozen patient samples isolated at surgery comprising benign, epithelial, biphasic, and sarcomatoid histologies were subsequently examined for the expression of these RTKs. The effects of BMS-777607 on MPM cellular proliferation and viability are being assessed.

Results
Expression of various RON/MST1R isoforms, c-MET, Tyro3 and Axl were observed in all cell lines. Significantly higher expression of all genes were found in the malignant tumour material versus benign pleura. The effects of BMS-777607 on cellular proliferation/viability are ongoing and the results will be presented at the meeting.

Conclusion
BMS-777607 represents a unique compound capable of targeting four RTKs whose expression are significantly elevated in maligant MPM. Given the many indications that these RTKs are associated with “addicted” RTK networks, this compound may therefore have potential clinical utility in the clinical management of MPM.

Background
Activation of the mTOR pathway is a common down-stream mechanism of resistance to Epidermal Growth Factor receptor (EGFR) tyrosine kinase inhibitors. CC-223 (Celgene) is an mTOR kinase inhibitor under clinical development. We evaluated CC-223 in combination with erlotinib to overcome resistance to EGFR tyrosine kinase inhibition in non-small cell lung cancer (NSCLC) cell lines and xenografts in nude mice.

Methods
A panel of 18 NSCLC cell lines was used to evaluate the effect of erlotinib and CC-223 treatment on cell growth using an MTT assay. Intermediately (IC50 >1 and <10 mM) and highly (IC50 >10 mM) resistant cell lines to erlotinib were used in analyses of additive/synergistic effects for the combination treatment with CC-223.

Results
CC-223 demonstrated anti-proliferative activity in NSCLC cell lines with various degrees of sensitivity as reflected in different IC50 values, ranging from 0.15 to 12 mM. In combination with erlotinib, CC-223 showed synergistic anti-proliferative effects in NSCLC cells resistant to erlotinib with combination indices as low as 0.1-0.2. In vivo studies in mice xenografts demonstrated a strong synergistic effect of the combination treatment of erlotinib and CC-223 with a 90% decrease of tumor volume compared to untreated and 88% compared to erlotinib treated for A549 cells. IHC analyses of apoptosis and proliferation in the tumors are ongoing; mature data will be presented at the meeting.

Conclusion
The mTOR kinase inhibitor CC-223 demonstrated anti-proliferative activity in a panel of NSCLC cell lines. In NSCLC cells resistant to the EGFR tyrosine kinase inhibitor erlotinib, combining erlotinib with CC-223 demonstrated a synergistic anti-proliferative effect. In vivo studies of tumors in xenografted mice also demonstrated synergistic anti-tumor effects with the combination treatment even in erlotinib resistant tumors. The present data suggest that mTOR inhibition using the mTOR kinase inhibitor CC-223 may be a therapeutic strategy to overcome resistance to EGFR tyrosine kinase inhibitors in NSCLC.

Background
Nesfatin-1 is an anorexigenic neuropeptide. We have previously reported that serum nesfatin-1 level is decreased in lung cancer patients. In this study, we aimed to investigate possible effects of serum nesfatin-1 level in survival and chemotherapy response of non-small cell lung cancer patients.

Methods
Twenty-eight non-small cell lung cancer patients were included in this study. Data were obtained from clinical records and our previously study about nesfatin-1. The patients were divided into two groups according to high nesfatin-1 level or low. Survival data were compared between these groups. Moreover, serum nesfatin-1 levels were investigated in chemotherapy responder and non-responder patients.

Results
Three of patients received supportive care alone. Twenty-five patients were treated with platinum-based doublet chemotherapy. In chemotherapy non-responder patients, serum nesfatin-1 level was more lower than in the patients achieved clinical benefit (0.38 ng/ml ±0.12 ng/ml vs 0.55 ng/ml ±0.21 ng/ml; p:0.044). Patients with low nesfatin-1 level had shorter overall survival (OS), but it is not statistically significant (median OS 7.3 months vs 13.8 months, respectively; p: 0.829). Serum nesfatin-1 levels were not also different in between patients survived more than 12 months and not (0.62 ng/ml ±0.22 ng/ml vs 0.46 ng/ml ±0.19 ng/ml; p:0.077).

Conclusion
This preliminary study suggests that serum nesfatin-1 level is not prognostic factor in non-small cell lung cancer, but may be a predictive marker for chemotherapy response.

Background
Despite adjuvant chemotherapy improving overall survival of resected Stage II and III non-small cell lung cancer (NSCLC), acute and late toxicities of chemotherapy have highlighted the need to better predict which patients will benefit from treatment. RLIP76 is a ubiquitously expressed multi-functional transporter that is associated with cisplatin and vinorelbine resistance. Our aim was to analyse the prognostic and predictive value of RLIP76 expression in NSCLC.

Methods
We identified 367 NSCLC patients resected between 1996 and 2009. A tissue microarray was created and immunohistochemistry (IHC) performed with an anti-human RLIP76 rabbit polycloncal antibody (Millipore, Temecula, CA). The intensity (0-3) and proportion of tumour cells (0-100) with staining was scored. The product of RLIP76 intensity and proportion of tumour cells staining was calculated (range 0-300) and divided into high (>100) and no/low expression (≤100). RLIP76 expression was correlated with clinical features and patient outcome.

Results
IHC was available for 285 patients, 173(60.7%) of which were male. Number of patients according to stage I, II, III and IV was 150(52.6%), 83(29%), 44(15.4%) and 8(3%), respectively. RLIP76 was overexpressed in 117(41%) specimens. There was no relationship between RLIP76 expression and stage, histology, sex or age. High RLIP76 expression was associated with an improved prognosis on univariate (HR 0.62, CI 9.44-0.90, p=0.012,Figure 1) and multivariate analysis (HR 0.57, CI 0.39-0.83, p=0.003). Adjuvant chemotherapy was also associated with an improved survival on multivariate analysis (HR 0.52, CI 0.33-0.82, p=0.005). When stratifying by RLIP76 expression, the benefit of chemotherapy was limited to patients with no/low expression (HR =0.44, CI 0.24-0.8, p=0.008)(Figure 2). No benefit of chemotherapy was seen in patients with high RLIP76 expression (HR=0.75, CI 0.34-1.63, p=0.5). Figure 1 Figure 2

Conclusion
High RLIP76 expression is associated with an improved prognosis in resected NSCLC.Interestingly no/low RLIP76 expression may predict for benefit of adjuvant chemotherapy. Further studies are needed to confirm these results.

Background
Cabozantinib (C) is a potent ATP-competitive inhibitor of MET and vascular endothelial growth factor receptor 2 (VEGFR2) along with KIT, RET, AXL, TIE2, and FLT3. Hepatocyte growth factor (HGF), the ligand of MET, and VEGF act synergistically to promote angiogenesis. There are currently no widely accepted prognostic or predictive biomarkers for anti-angiogenic agents.

Methods
This is a retrospective correlative biomarker study from the phase 1b/2 trial of C+/-E in stage IIIB-IV NSCLC. All pts had to fail prior therapy with E. Both drugs are oral and dosed daily. C dosing is in the free-base equivalent weight. In phase I, there was a 2 week lead in with E and the cohorts included: 1A (60 mg C+150 mg E), 2A (60 mg C+100 mg E), 3A (100 mg C+100 mg E), 4A (100 mg C+50 mg E), and 2B (40 mg C+150 mg E). In phase II, both drugs started simultaneously: Arm A (100 mg C) and Arm B (100 mg C+50 mg E). Pts were included in the study if a pre-E+C and post-C > day 29 plasma sample was available. The Milliplex 13-plex and Luminex 51-plex assays were used. For this preliminary analysis, select markers previously implicated in angiogenesis (bFGF, VEGF, sVEGFR1-3, IL-6, IL-8, IL-12, IL-17, PDGF-BB, ICAM-1, VCAM-1) and those of interest (ligands of KIT and MET- SCF and HGF, respectively) were analyzed. Log transformed mean fluorescence intensity (MFI) values and Wilcoxon Rank paired sum tests were used to detect changes from day 1-29. A change from baseline was noted to be significant if at least 15% (median) with α<0.05 (2-sided) and a trend if 10-15% (median) with α<0.08 (2-sided).

Results
73 pts included: 52 phase I and 21 phase II; median age 60 years; 23M/50F; 56.2% nonsmoker; 91.8% adenocarcinomas. The pts with samples from both time points were divided into two groups due to limited sample size and included: Group R (complete/partial response and stable disease> 6 months; n=22) and Group NR (stable disease< 6 months and progressive disease; n=51). The only marker that changed in a single direction in all subjects within a group was sVEGFR2 in group R. Overall, significant decreases were noted in sVEGFR1-3, IL-6, PDGF-BB, and trended in IL12p70 and IL-17. By subgroups: Group R had significant decreases in bFGF, VEGF, sVEGFR1-3, IL-6, IL-8, IL-12(p40+p70), IL-17, PDGF-BB, SCF, and trended in HGF (median 10.1% ↓, p=0.0275); and Group NR had significant decreases in sVEGFR2-3, IL-6, PDGF-BB, and trended in sVEGFR-1, IL-6, and IL-17.

Conclusion
Both groups R+NR had a decrease in sVEGFR-2, suggesting that this is a marker of treatment with C rather than a marker of response. However, overall group R had larger dynamic decreases of immune markers than group NR. HGF, which is targeted downstream by C and plays a role in angiogenesis and E resistance, had a trend to decrease in group R but not group NR. This study is retrospective with a small sample size, imbalanced numbers per response subgroup, and is exploratory in nature.

Background
The T790M mutation is considered to be the major mechanism of acquired resistance to epidermal growth factor receptor (EGFR) tyrosine kinase inhibitors (TKIs). However, its clinical implication in patients with non-small cell lung cancer (NSCLC) is yet determined.

Methods
NSCLC patients with acquired resistance to initial EGFR TKIs such as gefitinib or erlotinib were identified, and post-progression tumor specimens were prospectively collected for T790M mutation analysis. Clinical features including the pattern of disease progression (intrathoracic only versus extrathoracic), treatment outcomes for initial or subsequent TKIs, post-progression survival, and overall survival were compared between patients with and without T790M.

Results
Out of 70 cases, 36 (51%) were identified to have the T790M mutation in the rebiopsy specimen. There was no difference in the pattern of disease progression, progression-free survival for initial TKIs (12.8 and 11.3 months), post-progression survival (14.7 and 14.1 months), or overall survival (43.5 and 36.8 months) in patients with and without T790M. In total, 34 patients received afatinib after post-progression biopsy as a subsequent treatment. Among them, six (18%) achieved objective response. The median progression-free survival for afatinib was 3.7 months for the entire group, and 3.2 and 4.6 months for the groups with (n = 21) and without (n = 13) T790M, respectively (p = 0.33). Figure 1Figure 2

Conclusion
Although T790M had no prognostic or predictive role in the present study, further research is necessary to identify patients with T790M-mutant tumors who might benefit from new treatment strategies.

Background
Molecular testing in lung cancer has dramatically evolved over the past few years with a large number of targeted therapies and associated biomarkers. This has created a demand for tissue preservation. A careful consensus for prioritization of the different assays is needed to identify the right therapy for the patient. It is necessary to have protocols available that give access to results rapidly and accurately without depleting the sample throughout the process. We have designed a prospective study to demonstrate an efficient biomarker testing workflow (EGFR, ALK and KRAS) in a real clinical setting that preserves limited, valuable clinical samples

Methods
This prospective study was conducted at two independent pathology laboratories (Laboratorio de Dianas Terapeuticas of Madrid, Queen’s University of Belfast). Each site performed tissue sectioning for diagnosis and molecular testing according to laboratory protocols or to the assay specific package insert. Digital pathology was used to annotate specimens. NSCLC specimens were obtained by each lab (surgical resections and small biopsies of primary and metastatic lesions), excluding cytology specimens, with the exception of cells blocks. EGFR and KRAS mutation testing was performed using the cobas[®] EGFR Mutation Test and the cobas[®] KRAS Mutation Test (both CE-IVD), using a single 5µm section per test. FISH analysis was performed using the Vysis LSI ALK probe. Figure 1

Results
A total of 103 cases were included. 100% of the specimens were successfully tested for EGFR mutation status. Failure rates were 1.9% for ALK and KRAS analysis. Failed DNA extraction/PCR amplification was 3.9%/2.9% for EGFR and 5.8%/12.6% for KRAS. The prevalence of molecular alterations identified in EGFR (4% and 13.2%), ALK (4% and 7.8%) and KRAS (26% and 29.4%) was somewhat similar to that described in earlier studies. ALK rearrangements were mutually exclusive with EGFR and KRAS mutations. The median turnaround time was almost identical between sites: 6-7 days for all biomarkers (EGFR range 1-12 days, ALK range 2-14 days and KRAS range 1-17 days).

Conclusion
We have demonstrated that it is feasible to incorporate this protocol into the routine analysis of thoracic samples. The pattern of retesting to achieve full analysis in all samples suggests that, while EGFR testing is easily achievable with a one-section approach, the analysis of KRAS and ALK may require another round of testing in a small percentage of cases. Our results thus provide a pre-analytical and analytical rationale for comprehensive genotyping in patients with NSCLC.

Background
Major obstacles limiting the clinical success of EGFR TKI therapy in EGFR mutant non-small cell lung cancer (NSCLC) patients are heterogeneity and variability in the initial anti-tumor response to treatment. The underlying molecular basis for this heterogeneity has not been explored in patients immediately after initiation of therapy.

Methods
We conducted CT-guided core needle biopsies immediately prior to erlotinib treatment initiation and at 6 days and 60 days post erlotinib initiation in a patient with histologically confirmed NSCLC harboring an established activating mutation in EGFR. DNA and RNA from each of the frozen biopsies were analyzed by whole exome sequencing and whole transcriptome sequencing, respectively. High-resolution CT images were also obtained at the time of each biopsy to assess the degree of molecular and radiographic responses observed.

Results
Two established activating somatic mutations were identified in EGFR (p.G719A and p. R776H). Gene expression analysis revealed that several proapoptotic genes including BID, CASP3 and several growth inhibitory genes including GADD45B, GADD45G were upregulated at 6 days post erlotinib treatment, while at 60 days their expression levels decreased to below pretreatment levels. Other proapoptotic genes such as BAD and BAX and were noted to be upregulated most significantly 60 days, as was growth inhibitory gene CDKN1A. In contrast, the growth-promoting genes CCNB1 and CCND3 exhibited a progressive decrease in expression across time points. Whole exome sequencing demonstrated the progressive evolution of global copy number abnormalities. High-resolution CT scans revealed no interval radiographic change in tumor size after 6 days of erlotinib treatment, and a decrease in tumor size 60 days into therapy. No clinical complications were encountered.

Conclusion
This study is the first reported longitudinal integrated genomic analysis of EGFR-mutant NSCLC in a patient treated with an EGFR TKI. We documented the feasibility, safety and utility of this strategy to establish initial drug efficacy at the molecular level prior to any radiographic evidence of response (6 days), as well as evidence that acquired resistance can emerge early in the course of TKI therapy. Serial integrated genomic analysis is ongoing in other TKI treated patients to enhance the management of NSCLC patients on targeted therapy.

Background
Circulating tumor cells (CTCs) are cells that originate from a primary solid tumor and are found transiting the circulatory system. CTCs are expected to provide useful clinical information on biology of their primary, as ”liquid biopsy.” We have developed a novel cell-sorting system equipped with a disposable microfluidic chip (On-chip Sort, On-Chip Biotechnologies, Tokyo, JAPAN). At American Association for Cancer Reseach meeting 2013, we previously reported about its CTCs enumeration capability when performed in a clinical setting. Currently, On-chip Sort enables recovery of more CTCs than conventional cell sorting systems for further characterization.

Methods
In a preclinical study, PC-9 and H1975 human lung cancer cells harboring EGFR mutations (E746_A750del and L858R, T790M, respectively) were spiked into the blood from healthy donors. After samples were negatively enriched using anti-CD45-coated magnetic beads, the spiked cancer cells in the samples were captured by On-chip Sort. The captured tumor cells were subjected to mutation detection by ARMS/Scorpion PCR assay. A clinical feasibility study was then conducted in lung cancer patients harboring EGFR mutations.

Results
On-chip Sort performed recovery of the spiked PC-9 and H1975 cancer cells (5 cells in 4 ml of blood) in a preclinical experiment. Successively, we were able to detect the EGFR mutations from the captured cells. In a clinical feasibility study, 4 blood samples from lung cancer patients harboring EGFR mutation were collected and were evaluated. All the samples were successful in capturing CTCs by On-chip Sort. ARMS/Scorpion PCR assay detected the EGFR deletion mutation from two of the samples (50%).

Conclusion
The preclinical study and the results of the clinical feasibility study suggested the possibility of the On-chip Sort assay to detect EGFR mutations from peripheral blood of patients with lung cancer. Further investigation is going to be conducted to evaluate the correlation of EGFR mutations in captured CTCs and primary lesions.

Background
Clinical implications of KRAS mutations in advanced non-small cell lung cancer remain unclear.

Methods
Kras mutation status was identified by direct sequencing test in 844 specimens that were diagnosed of NSCLC at Samsung Medical Center from 2006 to 2011. This study included stage IV NSCLC patients who were treated with systemic chemotherapy. We retrospectively evaluated the prognostic and predictive value of KRAS mutations in patients with advanced NSCLC.

Results
Among 484 patients with available results for both KRAS and EGFR mutations, 39 (8%) had KRAS and 182 (38%) EGFR mutations, with two cases having both mutations. The median overall survivals for patients with KRAS mutations, EGFR mutations, or both wild types were 7.7, 38.0, and 15.0 months, respectively (P < 0.001). The KRAS mutation was an independent poor prognostic factor in the multivariate analysis (hazard ratio=2.6, 95% CI: 1.8–3.7). Response rates and progression-free survival (PFS) for the pemetrexed-based regimen in the KRAS mutation group were 14% and 2.1 months, inferior to those (28% and 3.9 months) in the KRAS wild type group.

Conclusion
KRAS mutation tended to be associated with inferior treatment outcomes after gemcitabine-based chemotherapy, while there was no difference regarding taxane-based regimen. Although the clinical outcomes to EGFR tyrosine kinase inhibitors (TKIs) seemed to be better in patients with KRAS wild type than those with KRAS mutations, there was no statistical difference in response rates and PFS according to KRAS mutation status when EGFR mutation status was considered. Two patients with both KRAS and EGFR mutations showed partial response to EGFR TKIs. Although G12D mutation appeared more frequently in never smokers, there was no difference in clinical outcomes according to KRAS genotypes. These results suggested KRAS mutations have an independent prognostic value but a limited predictive role for EGFR TKIs or cytotoxic chemotherapy in advanced NSCLC.

Background
Neoadjuvant chemotherapy is used to help downstage cancer, and is widely used in locally-advanced breast cancer. Studies of Ki67 proliferation index in breast cancer have been fairly extensively evaluated. Comparison of core and surgical specimens in breast cancers without exposure to chemo- or radiotherapy revealed technical variation of up to 20% in the Ki67 index score. In non-small cell lung cancer (NSCLC), however, little is known about the association of rate of change of Ki67 after neoadjuvant chemotherapy with radiographic response and clinical outcomes. We surveyed NSCLC from patients treated with neoadjuvant chemotherapy.

Methods
NSCLC patients treated with neoadjuvant chemotherapy were identified from 3 Hungarian hospitals and 1 community hospital in the United States. Matched pre-chemotherapy and post-surgical resection formalin-fixed, paraffin-embedded (FFPE) tumor specimens were collected and the Ki67 index was scored under an IRB exemption. We set an absolute difference of 20% between pre-chemotherapy and post-resection Ki67 scores as “meaningful” to avoid possible technical variation reported in the literature. Radiographic response to neoadjuvant chemotherapy by RECIST 1.0 criteria was also measured. Fisher’s exact test was used to measure the relationship between gender, histology, and type of chemo with “meaningful” Ki67 index. Logistic regression was used to test the relationship between Ki67 index decline rate and outcome subgroup (response/no response). Decline rate was defined as a ratio of decrease of Ki67 to its level at baseline.

Results
63 matched cases were identified. 46 cases were analyzable for pre-chemotherapy and post-operative Ki67, and chemotherapy regimen; 40 cases also had response criteria by RECIST. Of the 46 cases, the median patient age was 59 years (range 40-77), 23 were men, and 30 of 34 had a smoking history. There were 24 adenocarcinomas and 22 squamous cell carcinomas. Stages I, II, III, and IV were 2, 9, 31, and 4; respectively. All but two patients received a platinum-doublet, with 24 containing gemcitabine. 5 patients also received neoadjuvant radiotherapy. The mean Ki67 index scores were 35% (range 1-100%) pre-chemotherapy and 32% (0-100%) post-resection. 9 patients (19.6%) had a paradoxical “meaningful” increase in Ki67 index after neoadjuvant chemotherapy. Of the 40 patients with RECIST response data, there was 1 complete response, 34 partial responses, 4 stable diseases, and 1 disease progression. There were no statistically significant differences between gender, histology subtype, or type of platinum doublet administered associated with this paradoxical increase. There was no statistically significant difference in Ki67 decline rate between responders and nonresponders.

Conclusion
In this cohort of patients, there was a paradoxical “meaningful” increase in Ki67 index after neoadjuvant chemotherapy in ~20% of patients, without a clear association between histology or platinum doublet administered. For patients receiving neoadjuvant chemotherapy, the Ki67 index decline rate was not associated with radiographic response. Approximately 1/5 of NSCLC may have selection of tumor cells for a higher proliferative index when undergoing neoadjuvant platinum-based chemotherapy. The effect of this on progression-free and overall survival is being evaluated.

Background
Pleomorphic carcinoma of the lung is a rare epithelial tumor, and its clinicopathological characteristics and prognostic factors are still controversial. Up-regulation of E-cadherin repressors is known to increase tumor cell invasion and motility in non-small cell lung cancer cells. The aim of this study was to clarify the clinicopathological characteristics and prognostic factors, such as E-cadherin repressors (e.g., Zinc finger E-box-binding homeobox [ZEB], Snail, and Twist), in pleomorphic carcinoma.

Methods
We reviewed 2,328 cases of resected lung cancers that occurred between 1988 and 2011 and identified 62 patients with pulmonary pleomorphic carcinoma. Immunohistochemical staining for ZEB, Snail, and Twist was performed, and nuclear expression was estimated by the percentage of positive cells. The patients were divided into two groups; a diffuse expression group (≥75%) or a focal expression group (<75%). Clinicopathological variables and the expression of E-cadherin repressors were investigated retrospectively to analyze prognostic factors for the disease-specific survival rate of 42 patients after lung resection.

Results
The TNM pathological stages of pleomorphic carcinoma were classified as follows: 7 (11.2%) cases with stage IA, 15 (24.1%) with stage IB, 3 (4.8%) with stage IIA, 20 (32.2%) with stage IIB, 10 (16.1%) with stage IIIA, 4 (6.4%) with stage IIIB, 3 (4.8%) with stage IV. The 5-year disease-specific survival rate after pulmonary resection was 68.3%. Forty-seven (75.8%) tumors contained at least 10% spindle and/or giant cells, and the other fifteen (24.1%) consisted entirely of spindle cells and giant cells. An adenocarcinoma component was found in 34 (54.8%) cases, a squamous cell carcinoma component in 7 (11.2%) cases, an adenosquamous cell carcinoma component in 4 (6.4%) cases, and a large cell carcinoma component in 2 (3.2%) cases. The pleomorphic carcinoma patients were divided into five groups according to the predominant epithelial component, and there were no significant differences in the disease-specific survival rate between the groups. Those with a predominance of spindle-cell or giant-cell components also showed no difference in disease-specific survival. Nuclear ZEB-1 expression was positive only in the pleomorphic component. Diffuse expression of ZEB1 was found in seven patients. Snail and Twist were positive in epithelial and pleomorphic components, but at various levels; however, expression tended to be higher in the pleomorphic component. Thirteen patients had diffuse expression of Snail, and eight had diffuse expression of Twist. Using multivariate analysis, lymph node metastasis, pleural invasion, and diffuse expression of ZEB1 in the pleomorphic component all predicted poorer disease-specific survival (p=0.007, 0.022, and 0.016, respectively). Nuclear ZEB1 expression was higher in cases with lymphatic permeation compared to those without lymphatic permeation, but this association was not statistically significant (p=0.107).

Conclusion
Several unfavorable prognostic factors for pulmonary plemorphic carcinoma have been reported, such as massive necrosis, TNM stages, lymph nodes metastasis, pleural invasion, and complete resection. Our current study showed that lymph node metastasis, pleural invasion, and diffuse expression of ZEB1 in the pleomorphic component suggested a worse prognosis for pulmonary pleomorphic carcinoma. ZEB1 may promote the aggressiveness and invasiveness of this malignant tumor.

Background
Pemetrexed (PEM) inhibits multiple enzymes in the folate (F) pathway. Several studies show that genetic polymorphisms in these enzymes influence the efficacy and toxicity of PEM. We aimed to investigate the relationship between genetic polymorphisms associated with the F pathway and clinical outcomes of Japanese patients with advanced non-squamous non-small cell lung cancer (NSQ-NSCLC) treated with PEM plus cisplatin (CIS).

Methods
We analyzed 34 polymorphisms in 14 genes associated with the F pathway in NSQ-NSCLC patients treated with PEM plus CIS: ABCC11, ADA, ATIC, DHFR, ERCC1, FPGS, GGH, MTHFD1, MTHFR, MTR, MTRR, SHMT1, SLC19A1, and TYMS. These polymorphisms were compared with clinical outcomes such as response, toxicity, and progression-free survival (PFS) using Pearson’s χ[2] test and the log-rank test.

Results
All 56 patients were Japanese, with a median age of 62 years; 57.1% were male, 96.4% had an Eastern Cooperative Oncology Group Performance Status of 0–1, 96.4% had stage IV disease, and 94.6% had adenocarcinoma. The response rate, disease control rate, and median PFS were 32.2%, 78.6%, and 4.7 months, respectively. Of the 38 polymorphisms tested, none were associated with response or toxicity, but 2 single nucleotide polymorphisms (SNPs) (in the gamma-glutamyl hydrolase [GGH 452C>T] and methionine synthase [MTR 2756A>G] genes) were significantly associated with PFS. Patients harboring the GGH-C452C variant had significantly longer PFS (5.6 vs 2.8 months; p < 0.0001) than those with the C452T or T452T variants. Further, patients harboring the MTR-A2756A variant had significantly longer PFS (5.3 vs 3.7 months; p = 0.036) than those with the A2756G variant. In addition, among patients with the GGH-C452C variant, those harboring the MTR-A2756A variant had significantly longer PFS (5.9 vs 4.3 months; p = 0.044) than those with the A2756G variant.

Conclusion
SNPs in GGH and MTR seem to predict differences in PFS in NSQ-NSCLC patients treated with PEM plus CIS, and a combination of these 2 SNPs may predict differences in PFS more accurately. These results should be validated in larger, adequately designed prospective studies.

Background
The ICE COLD-PCR (Improved and Complete Enrichment COamplification at Lower Denaturation Temperature PCR) technology is capable of high sensitivity mutation detection. The ICE COLD-PCR (ICP) reaction contains a reference sequence oligonucleotide (RS-oligo) that hybridizes to both alleles but will dissociate from the mutant strand at the critical temperature for all mutation types: point mutations, insertions, deletions and indels. Following ICP, samples with mutations present at 0.1% in the original sample can be enriched and determined using standard Sanger sequencing. Next-Generation Sequencing (NGS) allows high-throughput analysis of somatic mutations in cancer using targeted resequencing panels of genes. Thus a broad mutation signature of tumors can be determined. However, the level of detection is ~4% for mutations unless a higher depth of coverage is used; this comes at the price of reducing the number of samples that can be analyzed on a chip. ICP enrichment of mutations prior to NGS mimics an increased "depth of coverage" without reducing the throughput per chip.

Methods
Proof of concept experiments were performed using low level mutations: KRAS G12R (0.5, 0.1, 0.05%) PIK3CA E542K and E545K (1.0, 0.5, 0.1%) and EGFR Exon 19 E746_A750del (0.5, 0.1, 0.05%). In addition, DNA isolated from FFPE tissues and matched plasma were amplified by standard PCR or enriched by ICP and subsequently analyzed for PIK3CA, KRAS, BRAF and EGFR mutations using the Ion AmpliSeq Cancer Hotspot Panel on the Ion Torrent Personal Genome Machine (PGM) with the Variant Caller Plugin.

Results

ICE COLD-PCR Amplification of FFPE DNA with Sequence Confirmation by Sanger Sequencing or Next Generation Sequencing using the Ion Torrent PGM
Gene	Mutation in Sample	% Starting Mutant	% Variant Frequency (Sanger)	% Variant Frequency (NGS)	P-value	Coverage	Ref Cov	Var Cov
EGFR Exon 19	#1 E746_A750_Del	0.50%	30	9.1*	1.00E-10	2,000	1,812	182
#2 E746_A750_Del	0.10%	10	3*	7.82E-10	2,000	1,940	60
#3 E746_A750_Del	0.05%	Background	No Variant
KRAS	#1 G12R	0.50%	70	69.18	1.00E-10	19,174	5,819	13,264
#2 G12R	0.10%	10	23.58	1.00E-10	10,004	7,513	2,359
#3 G12R	0.05%	Background	2.86	5.01E-04	11,027	10,603	315
PIK3CA	#1 E542K	1.0%	15	16.48	1.00E-10	28,899	24,113	4,763
#1 E545K	1.0%	25	24.23	1.00E-10	30,027	22,700	7,275
#2 E542K	0.5%	10	9.19	1.00E-10	15,680	14,225	1,441
#2 E545K	0.5%	10	11.78	1.00E-10	15,267	13,405	1,798
#3 E542K	0.1%	Background	2.02	1.00E-10	15,358	15,045	310
#3 E545K	0.1%	Background	3.17	1.00E-10	15,154	14,657	481
* difficult to detect deletions by NGS

Conclusion
ICP is a flexible technology that results in a PCR product enriched for any mutation present in the sample analyzed. This upstream enrichment of low level somatic mutations combined with high throughput analysis by NGS brings NGS one step closer for use in high throughput screening of circulating mutations for patient treatment selection and monitoring of disease.

Background
As non-small cell lung cancer (NSCLC) is a leading cause of cancer death, new prognostic and predictive markers are highly warranted. miR-210 is an essential regulator of the hypoxia pathway. The biological and prognostic importance of miR-210 has not been thoroughly studied in NSCLC. We aimed to explore miR-210 expression in both cancer and tumor stromal cells in a large representative NSCLC patient cohort to assess its impact on disease-specific survival (DSS).

Methods
Tissue micro arrays (TMAs) were constructed, representing both cancer cells and tumor stromal cells from 335 unselected patients diagnosed with stage I-IIIA NSCLC. In situ hybridization was used to evaluate the expression of miR-210.

Results
In univariate analyses, high cancer cell (p=0.039) and high stromal cell expression (p=0.008) of miR-210 were both significantly associated with an improved DSS. Co-expression of miR-210 in cancer and stromal cells combined was also a positive prognostic factor for DSS (p=0.010). In multivariate analysis, miR-210 in stromal cells (p=0.011), and miR-210 co-expressed in cancer and stromal cells (p=0.011) were independent prognosticators for DSS.

Conclusion
High expression of miR-210 in tumor stromal cells, and high miR-210 co-expressed in cancer cells and tumor stromal cells mediates an independent favorable prognostic impact in NSCLC. miR-210 may be a candidate marker for prognostic stratification and therapeutic interventions in NSCLC.

Background
Treatment decisions in stage I and II resectable lung adenocarcinoma (ADC) are heterogeneous due to low efficacy of treatment and a high frequency of co-morbidities in the patient population. Currently, pathological stage is the main determinant of adjuvant treatment recommendations. The cell cycle progression (CCP) score is a proliferation based expression profile that has been shown to add significant prognostic stratification within stage I and II patients. We have developed an integrated prognostic model of pathological stage and the CCP expression score in order to maximize the prognostic utility of both markers.

Methods
Cox proportional hazards models with pathological stage and the CCP expression score were created from two data sets: 256 patients (190 stage I, 66 stage II) from the Director’s Consortium microarray cohort and 381 adenocarcinomas (337 stage I, 44 stage II) from a clinical study set analyzed by qPCR. Expression microarray data were scaled to adjust for differences in experimental platforms. The primary outcome measure was cancer-specific death, defined as death from lung cancer or death with recurrence within five years of surgery. Coefficients for modeling were derived from the hazard ratio in the Cox PH model.

Results
Both pathological stage and CCP score were independent predictors of lung cancer death in both cohorts. The coefficients for pathological stage and CCP score were consistent across both data sets and did not differ significantly between the analysis of all patients and a subset of untreated patients. A combined score (CS) of stage and CCP score (0.33 * CCP score + 0.52 * stage) was created from the subset of untreated patients. When applied to untreated patients in the clinical ADC cohort, pathological stage alone provided estimates of five-year risk of cancer-specific death of 12.6% (stage IA), 22.6% (stage IB), 38.4% (stage IIA) and 60% (stage IIB). In the same cohort, the combined score could separate stage IA patients with five-year risk estimates ranging from 6% to 24%. Similarly increased discrimination of risk estimates were observed for stage IB (10% to 42%), stage IIA (21% to 63%) and stage IIB patients (32% to 75%).

Conclusion
A combination of pathological stage and the CCP expression score is a more effective predictor of post-surgical risk of cancer-specific death than either marker alone. A more precise risk assessment provides better guidance in balancing treatment related risks with disease-specific survival.

Background
Prognostic biomarkers for patients with non-small cell lung cancer (NSCLC) who have been treated with neoadjuvant therapy have not been fully assessed. Identifying biological prognostic markers may help to distinguish patients who are likely to benefit from additional postoperative chemotherapy. The purpose of this study was to analyze prognostic biomarkers in surgically resected NSCLC after treatment with neoadjuvant therapy, with special reference to the immunophenotypes of both the cancer cells and the stromal cells.

Methods
A series of 66 patients with NSCLC who were treated with neoadjuvant chemotherapy, chemoradiotherapy, or radiotherapy followed by complete resection at our hospital between April 1992 and December 2009 were reviewed. Among the 66 surgically resected specimens, case with viable tumor cells remained in the specimens were included in this study (n =52). We examined the expressions of geminin and cleaved caspase 3 (proliferation and apoptosis markers), E-cadherin and vimentin (epithelial mesenchymal transition related molecules), ALDH1 and CD44v6 (Stem cells related molecules) in the cancer cells. Furthermore in the stromal cells, the expressions of podoplanin and CD90 in cancer associated fibroblasts (CAFs) and CD204 in tumor associated macrophages (TAMs) were also examined.

Results
The 5-year disease-free survival rate of patients with high ALDH1 expression levels in their cancer cells was significantly lower than those with a low ALDH1 level (47.3% vs. 21.5%, respectively; P =0.023). The expression statuses of geminin, cleaved caspase 3, E-cadherin, vimentin, and CD44v6 in the cancer cells had no prognostic impact. The 5-year disease-free survival rate of patients with low or high podoplanin and CD90 levels in CAFs and CD204 levels in TAMs expression levels were not any prognostic impact in the stromal cells (37.9% vs. 29.1%, 33.8% vs. 37.5%, 38.4% vs. 29.0%, P =0.90, P = 0.75, P =0.98). In NSCLC without neoadjuvant therapy matching for clinical stage and histopathology (case control, n =104), the 5-year DFS rate of the patients with a high ALDH1 expression level was 48.3%, while that of the cases with a low ALDH1 expression level was 59.8%. The expression of ALDH1 in cancer cells was not correlated with the prognosis in NSCLC without neoadjuvant therapy (P =0.507). A multivariate analysis identified ALDH1 expression in cancer cells as significantly independent prognostic factors for disease-free survival in patients who received neoadjuvant therapy (P =0.045).

Conclusion
The presence of ALDH1-positive cancer cells was an independent recurrence predictor in patients who received neoadjuvant therapy, while CAFs and TAMs did not provide any predictors. Although prospective studies with a larger number of patients are required to confirm the prognostic significance of ALDH1 expression in cancer cells in validation populations with neoadjuvant therapy, our results suggest that the immunophenotypes of ALDH1 expression can serve as a guide to additional treatment after surgical resection in patients who received neoadjuvant therapy.

Background
The insulin-like growth factor (IGF) pathway is involved in the development and progression of many tumours and there is growing preclinical evidence that blockade of this pathway has anti-tumour effects in NSCLC. IGF receptors (IGFR) are another potential target for targeted treatment of NSCLC and a number of agents are already undergoing clinical trial. Biomarkers are needed to select patients most likely to derive clinical benefit from these agents. The downstream pathway components of IGF1R and MET activation include PI3K and AKT, which are other potential biomarkers currently being investigated in this patient cohort. IGF1R has also been implicated in acquired resistance to EGFR-TKI treatments. Only a few small retrospective studies have investigated the prognostic role of IGF1R in NSCLC and the relationship with EGFR mutations is not known.

Methods
IGF1R status was evaluated by chromogenic silver in situ hybridization (ISH) and immunohistochemistry (IHC) in tissue microarray sections from a retrospective cohort of 264 surgically resected NSCLCs and results were compared to clinicopathological features and patient survival. Patients were classified as IGF1R gene amplification (either presence of tight gene clusters, IGF1R to CEN15 ratio ≥ 2, or ≥ 15 copies of IGF1R per cell in ≥ 10% of analysed cells); high polysomy (≥ 4 copies of IGF1R in ≥ 40% of tumour cells); low copy number (< 4 copies of IGF1R in < 40% of cells). Patients were also grouped as IGF1R-positve (amplification or high polysomy) or IGF1R-negative (low copy number).

Results
High IGF1R gene copy number was identified in 77 cases (29.2%) in which there were 32 amplified IGF1R cases (12.1%) and 45 high-polysomy IGF1R cases (17%). Increased copy number of IGF1R was more common in squamous cell carcinomas (SCC) compared to large cell carcinomas (LCC) or adenocarcinomas (ADC) (p<0.05). There was no correlation between IGF1R gene copy number status and other clinicopathological features including patient age, gender, smoking status, tumour size, vessel, perineural or lymphatic invasion, grade or stage. IGF1R copy number alteration in primary tumours was highly correlated with IGF1R copy number status in metastatic tumours (p<0.01). High IGF1R protein expression was observed in 61/259 (23.6%) primary tumours and 14/215 (6.5%) normal adjacent bronchial mucosae. High expression of IGF1R protein was significantly associated with SCC in comparison with non-SCC primary tumours, as well as with lymphatic and vessel invasion. There was a moderate correlation between IGF1R copy number status (positive versus negative) and IGF1R protein expression (high versus low) (Cramer’s V=0.3, p-value <0.001). Both IGF1R copy number status and protein expression were not associated with patient overall survival in univariate analyses (p>0.05).

Conclusion
High IGF1R gene copy number and its protein expression are frequent in NSCLC, particularly in SCC. However, alterations of IGF1R are not associated with patient prognosis. IGF1R gene copy number can readily be assessed in formalin fixed paraffin embedded tissue and warrants further investigation as a potential biomarker of targeted therapy in NSCLC.

Background
Background: In order to evaluate gene abnormality in lung adenocarcinoma at an early stage, genomic DNAs of six in situ adenocarcinomas (Ad) and nine small but invasive adenocarcinomas (INVAd) were examined by array-comparative genomic hybridization. Finally, 3q26 was detected as a significantly amplified region in INVAd relative to in situ Ad. Among the genes located at 3q26, we selected the epithelial cell transforming sequence 2 oncogene (ECT2) and examined it for abnormality in lung Ad, since ECT2 is related to Rho-specific exchange factors and cell cycle regulators, and is also thought to have an important role in the regulation of cytokinesis.

Methods
Methods: The clinical implications of ECT2 abnormality were examined by quantitative real-time genomic PCR (qPCR) and immunohistochemistry (IHC) using 158 cases of varies types of LAd, resected at Tsukuba University Hospital. The results were verified by cDNA microarray and 10k SNP array using another set of early-stage LAds, resected at the National Cancer Center Hospital.

Results
Results: qPCR and IHC analyses revealed overexpression of ECT2 in INVAd, and this was correlated with the MIB-1 index. ECT2 overexpression was significantly associated with lymph node metastasis, pathological stage, vascular invasion and the histological subtype of Ad. Patients with LAd overexpressing ECT2 showed an unfavorable outcome in terms of both disease-free and overall survival (Figure). These results were verified using another set of 144 early-stage Ads resected at the National Cancer Center Hospital. Figure 1

Conclusion
Conclusion: ECT2 is a new marker that can be used for prognostication of patients with early-stage LAd.

Background
The MDM2 protein plays an important role in regulating cell proliferation and apoptosis by interaction with multiple proteins including p53 and Rb. c.309 (rs2279744) polymorphism (T>G) in the MDM2 promoter has been shown to result in higher levels of MDM2 RNA and protein. In order to evaluate the association of the MDM2 309T>G polymorphism and lung cancer risk, we carried out a case-control study in a Japanese population.

Methods
The MDM2 genotypes were determined in 469 lung cancer patients and in 682 healthy control subjects using Smart Amplification Process (SmartAmp). Statistical adjustment was made for sex and age.

Results
The distribution of the MDM2 309T>G genotypes was not significantly different between control and overall lung cancer cases. Subgroup analysis of KRAS G to T transversion adenocarcinoma indicated that G/G genotype of SNP309 may be associated with lung cancer carcinogenesis (adjusted OR = 2.42 95% C.I. = 1.01-5.82 p = 0.05) compared to T/T + T/G genotypes. G/G genotype has lower-level exposure to cigarette smoke than T/T + T/G genotypes (p=0.03) among squamous cell lung cancer patients. There was however no effect of either polymorphism on age at diagnosis of lung cancer or on overall survival.

Conclusion
Our results indicate that the MDM2 309T>G polymorphism is not significantly associated with lung carcinogenesis but may be associated with smoking related cancer of the lung.

Background
Multimodal treatment provides the best outcome for some but not all malignant pleural mesothelioma (MPM) patients. Therefore the identification of prognostic markers helping to select patients remains a subject of key importance. Frequent loss of NF2 tumor suppressor gene, a regulator of Hippo pathway and mammalian target of rapamycin (mTOR) has been well documented in MPM. We recently observed loss of expression of phosphatase and tensin homologue (PTEN), a tumor suppressor gene of the phosphatidylinositol 3-kinase (PI3K)/mTOR pathway, in 62% of MPM patients. In this regard, increased activity of these pathways stemming from loss of abovementioned tumor suppressor genes may serve as potential prognosticator as well as therapeutic target. Here we aim to assess prognostic significance of PI3K/mTOR and Hippo pathways for MPM patients treated with a multimodal approach.

Methods
Large cohort of MPM patients uniformly treated with induction chemotherapy followed by extrapleural pneumonectomy was employed in this study. Tissue micro arrays were constructed composing of paired samples from patients (n = 153) collected pre- and post-induction chemotherapy. All tissues were evaluated for apoptotic index (cleaved caspase-3) and proliferation index (Ki-67). Expression levels of biomarkers of PI3K/mTOR (PTEN, p(phosphorylated)-mTOR and p-S6) and Hippo signaling pathway (nuclear-YAP and nuclear-Survivin) were evaluated by immunohistochemistry and correlated with overall survival (OAS) and progression free survival (PFS).

Results
Survival analysis showed that high p-S6 expression and high Ki-67 index in samples of treatment naïve patients was associated with shorter PFS (p = 0.02 and p = 0.04, respectively). High Ki-67 index after chemotherapy remained associated with shorter PFS (p = 0.03) and OAS (p = 0.02). Paired comparison of marker expression in samples prior and post induction chemotherapy revealed that decreased cytoplasmic PTEN as well as increased p-mTOR expression was associated with a worse OAS (p = 0.04 and 0.03, respectively). No correlation was observed between expression level of nuclear-YAP with PFS or OAS.

Conclusion
Ki-67 proliferation index has prognostic value for MPM patients treated with multimodality approach in when analyzed both pre- and post- induction chemotherapy specimens. Our results further reveal the prognostic significance of expression changes of PI3K/mTOR pathway components during induction chemotherapy. If confirmed, these data suggest PI3K/mTOR pathway to be a target for selected MPM patients.

Background
Antibodies targeting epidermal growth factor receptor (EGFR), such as cetuximab, may potentially improve outcome in non-small cell lung cancer (NSCLC) patients with high EGFR expression. The EGFR expression may be heterogeneously distributed within tumors and small biopsies may thus not accurately reveal the EGFR expression and EGFR expression may also change during chemotherapy.

Methods
EGFR expression in diagnostic biopsies and resection specimen was compared in 53 NSCLC patients stage T1-4N0-1M0 treated with surgery without preceding chemotherapy (OP-group) and from 65 NSCLC patients stage T1-3N0-2M0 (NAC-group) treated with preoperative carboplatin and paclitaxel were analyzed regarding EGFR expression to evaluate the concordance of EGFR expression between samples.

Results
No significant change in EGFR expression H-score was observed when comparing serial samples from either the OP-group (p=0.934) treated with surgery or the NAC-group (p=0.122) treated with preoperative chemotherapy. Discordance between tumors dichotomized according to EGFR expression (high: H-score≥200; low: H-score<200) in diagnostic biopsies and immediate resection specimens was 25% in the OP-group and 33% in the NAC-group (p=0.628).

Conclusion
EGFR expression in 25% of diagnostic biopsies may potentially not be accurate compared to the prevailing pattern in the whole tumor based on the larger resection specimens. This may potentially be an obstacle for proper use of antibodies targeting the EGFR in NSCLC. EGFR expression does however not change significantly during paclitaxel and carboplatin and rebiopsies in order to decide on anti-EGFR antibody therapy following chemotherapy does not seem warranted.

Background
Activating mutations in the EGFR gene identifies NSCLC patients sensitive to treatment with EGFR tyrosine kinase inhibitors. EGFR mutations are more prevalent in Asians compared to Caucasian. However, estimates of mutation frequencies are only available from selected patient cohorts and the incidence in an unselected Caucasian population is unknown. This study assess the prevalence of EGFR mutations in an unselected population based cohort, and the correlation between mutation and gender, age, ethnicity, smoking habits, as well pathological data.

Methods
Patients diagnosed with NSCLC June 1, 2010 - September 30, 2011 in a well-defined population of 1.7 million in the Capital Region, Denmark, were included in this single center study. The type and location of the diagnostic material was registered. All specimens were pretreatment, no residual tumors were included. Smoking data were also available. The mutation analyses were investigated by therascreen EGFR RGQ PCR Kit (detection of 29 somatic mutations in the EGFR oncogene, Qiagen) (n =1121) or direct sequencing (n=62). Twenty-two patients were analyzed with both methods.

Results
658 men and 598 women were included. All but 4 were Caucasians. 6.2% were never smokers, 38.9% were ex- smokers and 54.9% were current smokers. 1161 (92.4%) patients had material sufficient for mutation analysis. More than 50% malignant tumor cells were available for analyses in 68%, 76% and 56% of the histological, cytological clot, and cytological smear material respectively. Manual macro dissection to ensure this high concentration of tumor cells was done on 32%, 4%, and 10% of the histological, cytological clots, and cytological smears material, respectively. Cytological material was used for 38 % of the mutation analyses. 5.4 % of tested patients harbored mutation in the EGFR gene (4.3% men/ 6.7 % women). The majority of mutations (55 cases= 87%) were activating mutations (29 exon 19 deletions, 24 L858R mutations and 2 G719X mutations), 3 were rare mutations (two L861Q and one S768I) with unclear clinical response to TKI, five were insertions in exon 20 (resistance mutation). Three of the mutations (all exon 19 deletions) were detected in the three included Asian women. 8.0 % of adenocarcinomas, 1.9% of squamous cell carcinomas and one single case of adenosquamous cell carcinoma were mutated. No other tumor types were mutated. DNA yield was similar whether extracted from cytological or histological samples and these material types yielded similar mutation percentages (6.55% in cytology and 4.66% in histology of material analyzed). 29.4%, 4.4% and 2.9 % of never, ex- and current smokers, respectively, were mutated (p<0.001). No significantly difference between ex- and current-smokers was observed. Mutation status was not related to age.

Conclusion
EGFR mutations analysis on cytological as well as on histological material was possible in 92% of patients with NSCLC. 5.4% of the patients harbored EGFR mutation. Adenocarcinomas were mutated more often (8.0%) than squamous cell carcinomas (1.9%). Mutations were found in never smokers as well as in former and current smokers. No difference in gender and age regarding mutation status was observed.

Background
Pleural fluid samples are useful when tumor tissues are not available for mutation analysis. Moreover, the advantages of using pleural fluid are that it is easily accessible, can be sampled by relatively non-invasive methods, and can be repeatedly sampled. Approaches for examining molecular biomarkers in body fluids such as effusion may be clinically helpful in predicting the response to EGFR-TKI treatment.

Methods
Fifty-seven patients with malignant pleural effusion were enrolled in the study. EGFR mutations were assessed by direct sequencing and PNA clamping using tumor tissues, cell blocks, pleural effusions, and sera. The diagnostic performance of pleural effusion was investigated.

Results
Among the 57 patients with malignant effusion, 37 patients were NSCLC, 11 were SCLC and 9 were extrapulmonary cancer. Twenty patients including 11 SCLC and 9 extrapulmonary cancer showed negative for EGFR mutational status. Among the 37 NSCLC patients, the diagnostic performance of pleural effusion compared with the combination of tumor tissue and cell blocks showed 89% sensitivity, 100% specificity, positive predictive value of 100%, and negative predictive value of 95% by PNA clamping, and 67% sensitivity, 90% specificity, positive predictive value of 75%, and negative predictive value of 86% by directing sequencing. A patient in whom an EGFR mutation was identified in pleural effusion only by PNA clamping showed a significant response to EGFR tyrosine kinase inhibitor (EGFR-TKI) treatment.

Conclusion
Pleural effusion had a diagnostic performance for the detection of EGFR mutations in NSCLC that was comparable to that of tumor tissues and cell blocks. The diagnostic performance of PNA clamping was good compared with that of direct sequencing. A more sensitive and accurate detection of EGFR mutations would benefit patients by allowing a better prediction of the response to EGFR-TKI treatment.

Background
Epidermal Growth Factor Receptor mutated (EGFR+) NSCLC patients have a median progression free survival (PFS) of approximately 12 months when treated with EGFR-tyrosine kinase inhibitors (TKI). One of the resistance mechanisms described is the T790M mutation. This mutation is reported in 49-65% of patients who are rebiopsied at disease progression. Here, we report on the incidence of T790M mutation in a cohort of patients who were sequentially rebiopsied.

Methods
EGFR+ patients or with TKI-response>24weeks and progressive disease on TKI’s were retrospectively analysed. Patients should have had at least 2 separate biopsies. All biopsies and treatments were collected from the medical record and pathological reports. Survival was calculated according to Kaplan-Meier. Overall survival (OS) was calculated from date of 1[st] diagnosis until death or June 2013, which ever was first

Results
68 patients with 2 biopsies or more were available for analysis. In the first biopsy at TKI-resistance; T790M mutation was detected in 34 patients (50,0%). 26/68 patients had later biopsies available; showing gain and loss of T790M in later biopsies (figure 1). Overall development of T790M was 57.4% (39/68). 7 patients had >3 biopsies available (figure 2). Patients developing T790M had numerically longer median OS of 3.8 years (range 2.8 – 4.9) as compared to median OS in T790M-patients (2.5 years, range 1.0 – 3.9) (P = 0.204).Figure 1Figure 2

Conclusion
57.4% of patients developed T790M. OS in patients developing T790M was longer than in patients not developing T790M, however the difference was not significant. Sequential data show that some patients ‘loose’ the T790M later on in the course of the disease. Data from this cohort suggests that T790M development is a dynamic process.

Background
The strategy for EGFR-TKI acquired resistance including continuous EGFR-TKI with chemotherapy or local therapy or chemotherapy alone. The aim of this study was to investigate the association of T790M mutation status with the efficacy of different modalities after acquired resistance to EGFR-TKI.

Methods
From Mar 2011 to Mar 2013, the advanced NSCLC patients who performed a rebiopsy after acquired resistance to EGFR-TKI in Shanghai Pulmonary Hospital were included into this study, Scorpion ARMS was used to detected the EGFR mutation status. SPSS 13.O was used to perform the statistical analysis.

Results
53 patients were enrolled in the study with a median age of 51.2 years old. 45.3% (25/53) were females and 54.7% (29/53) of patients had T790M mutation. Among them, 16 people with local disease progression received local therapy combined with TKI therapy, while 21 with a slow progress received chemotherapy plus TKI therapy. The median progression-free survival time (PFS1) of all patients according to RECIST criteria was 11.8 months. The median progression-free survival time (PFS2) of patients who received continuous EGFR-TKI treatment was 3.5 months (95% CI 2.689-4.311). Patients with T790M mutation had significantly longer PFS2 than those without T790M mutation (6.3 vs 3.0 months, p = 0.001), while there were no significant differences found in PFS1 between the two groups (13.0 vs 8.5 months, p = 0.999).

Conclusion
NSCLC patients who had T790M mutation after acquired resistance to EGFR-TKI benefited more from the continuous EGFR-TKI treatment and should be recommend to adopt this modality.

Background
In NSCLC, higher Thymidylate Synthase (TS) levels have been reported in both squamous and large cell carcinomas compared to adenocarcinoma. In clinical practice, Pemetrexed, a potent antifolate inhibitor of TS, showed a selective benefit in patients with "non-squamous" NSCLC. Two retrospective studies have shown that low TS protein levels are associated with better clinical outcome in NSCLC patients treated with pemetrexed. Aim of this study was to explore, in a series of advanced stage IV patients receiving pemetrexed-based regimens in first line of treatment, the association between TS mRNA and protein expression with overall survival (OS) and therapeutic response.

Methods
Two series of histologically confirmed non squamous-NSCLC, assessed in formalin-fixed and paraffin embedded specimens from patients treated with pemetrexed-based regimens were collected: the first series at San Luigi Hospital (n=64), the second series at Regina Elena National Cancer Institute (n=8). Due to the limited amounts of tissue specimens available, total RNA extraction was possible in 52 out of 72 cases. TS protein expression was performed using immunohistochemistry (mouse monoclonal TS106 antibody) and scored through H-SCORE method, considering both staining intensity (0 no staining; +1 weak; +2 moderate; or +3 strong) and percentage of tumor cells stained, resulting in semiquantitative H-scores ranging from 0 to 300. TS nuclear and cytoplasmic staining, respectively, were separately scored. Statistical analyses were performed using the STATISTICA10 software.

Results
The differential H-SCORE assessment showed a strong importance of TS localisation for clinical outcome prediction: in Cox regression analysis, a statistically significant association was observed between nuclear TS expression and OS (p < 0.009) indicating that lower nuclear TS expression levels were associated with longer OS. In addition, lower nuclear TS levels were significantly associated with a better response to therapy (p<0.001). On the contrary, TS cytoplasmic staining did not affect patients’ survival or clinical response (p>0.05). Four subgroups of patients, based on the dichotomized low/high TS expression in both nucleus and cytoplasm, were obtained: both high, both low, nucleus high/cytoplasm low and nucleus low/cytoplasm high. Significant differences in overall survival among these four groups were detected (p=0.017), confirming the strong and selective influence of nuclear TS, as compared to cytoplasmic TS, expression in clinical outcome. Moreover, Chi[2] test revealed a significant association between low nuclear TS and partial response to pemetrexed treatment, independently of cytoplasmic TS expression (p<0.001). No correlation between TS protein expression data and clinico-pathological data (age, gender) were identified. TS gene expression analyses are ongoing.

Conclusion
This retrospective study suggests that TS protein expression, selectively assessed at nuclear level, has a potential predictive role in advanced stage IV patients, receiving pemetrexed in first line of treatment. Patients with low nuclear TS expression showed prolonged overall survival and better response to therapy. Such preliminary results define TS assessment as a potential tool which may select the most appropriate group of patients to be treated with pemetrexed.

Background
Personalised treatments for lung cancer are being increasingly employed. Around 4% of pulmonary adenocarcinomas have rearrangements of the ALK gene resulting in oncogenic fusion proteins. Such cases are sensitive to the small molecule tyrosine kinase inhibitor Crizotinib, and detection of the ALK rearrangement is crucial to the treatment of these patients. Currently the only approved test for the detection of ALK-positive NSCLC is FISH. Various immunohistochemical assays have been proposed as alternatives or screening tools, which are cheaper, quicker and easier to perform and interpret than FISH. Validation of these immunohistochemical tests is therefore important.

Methods
15 FISH positive cases were obtained from local archives at the Royal Marsden and Royal Brompton Hospitals, with accompanying data on crizotinib therapy and clinical response. A further 14 FISH negative and FISH uncertain cases were also retrieved. 3 antibodies were optimised for immunohistochemical detection of ALK: D5F3, marketed by Ventana and using their proprietary automated system, ALK1 (Dako), and 5A4 (Abcam). All three antibodies were applied to sections from the archival cases. Antibodies were semiquantitatively scored on their intensity (0-3) and proportion of malignant cells stained (0-100%). Cutoffs for positivity/negativity were set by ROC analysis to optimise correct classification.

Results
Positivity according to the Vysis FISH assay (i.e. appropriately abnormal FISH signal in at least 15% of cells) was taken as the gold standard. The Ventana assay (D5F3) was markedly more intense but showed higher background than the other two antibodies. ROC analysis of staining intensity showed that the Ventana and Dako antibodies should be considered positive when staining is of 'moderate' intensity or above. The Abcam assay was discriminatory when any positivity was seen. Scoring of proportion did not improve specificity or sensitivity with any of the antibody tests. Subsequent analyses were of intensity scoring. Taking all cases together, all 3 antibodies were highly specific (100%) and sensitive (Ventana 86%, Abcam 79% and Dako 71%). In excision specimens, all 3 assays showed sensitivity of 83%. In cytology and small biopsy specimens, Ventana performed the best (specificity of 88% compared to 75% for Dako and 63% for Abcam) although the numbers are small. No ‘false positive’ (-ve FISH, +ve IHC) cases were seen; the only disparity between FISH and IHC were 'false negative' cases. Two cases were FISH-positive but universally IHC-negative (with all 3 assays). Of these, one was treated with crizotinib, and failed to respond.

Conclusion
IHC is a highly specific (100%) and sensitive (71%-86%) test for ALK rearrangement in archival FFPE tumour tissue. In this study of 15 ALK-rearranged tumours, the Ventana D5F3 assay performed the best, despite having relatively high background staining. Occasional cases are FISH-positive but IHC-negative. One such case was not responsive to Crizotinib, further supporting IHC as a test predictive of drug response. IHC has the potential to replace FISH for the detection of candidates for Crizotinib therapy. However, further work is required to direct the treatment of tumours with discordant FISH and immunohistochemical tests.

Background
Currently, biomarker testing for lung cancer is not uniformly integrated into the Canadian public healthcare system, despite its clear importance for patient outcomes. In order to better understand the current practice pattern for lung cancer biomarker testing, we assessed physician perspectives by specialty and region.

Methods
A national survey of Canadian specialists was conducted to understand their perspective on biomarker testing in lung cancer. An 11-item survey was developed to assess current practice and challenges related to biomarker testing. The survey was sent to 375 specialists (150 medical oncologists, 75 pathologists, 150 respirologists/thoracic surgeons) with a focus or interest in the management of lung cancer.

Results
The overall response rate for the survey was 36%, (38% of medical oncologists, 24% of pathologists, and 40% respirologists). It is understood that knowing tumour genotyping results at the time of the initial medical oncology consultation impacts patient outcome and influences the treatment decision (98%). To date, medical oncologists most commonly initiate molecular testing (67%), however, most respondents suggested a shared model for initiating the testing involving medical oncologists and pathologists is needed. The current perception is that less than two-thirds of patients have testing results available at the time of initial medical oncology consultation. Barriers identified to routine testing for all advanced lung cancer patients include cost, lack of funding for molecular testing, tissue availability and the quality of the tumour sample.

Conclusion
There was clear agreement that biomarker testing is important in determining the most appropriate initial treatment for patients with advanced NSCLC. There is a need for national consensus on who should initiate molecular testing. Moving forward, clear clinical guidance for pathologists needs to be established for molecular testing as part of the lung cancer diagnostic process. This includes defining the population to be tested, timing, and the tests to be performed. This may be facilitated by including more information on diagnostic sample requisitions, such as clinical suspicion of lung cancer primary (versus metastasis from another site), other cancer history, and other samples collected (and tested) previously or planned (e.g. planned resection). Pathologists need to incorporate routine EGFR, ALK testing into diagnostic lung cancer algorithm. This and greater clinical information on sample requisitions will minimize unnecessary tissue sections and allow the most efficient use of available tumour tissue for molecular testing. Incremental laboratory funding is required throughout the Canadian public healthcare system in order to provide the current standard of molecular testing required for NSCLC. Turnaround times need to be clearly established and monitored. Implementation of the College of American Pathologists (CAP) guidelines for transport from diagnostic lab to molecular testing lab (3 days), and turnaround for test results (10 days) will also improve the proportion of patients with test results available at initial consultation. Finally, feedback to clinicians, including sample quality, volume, whether or not testing was successful, and molecular results is necessary in a timely manner.

Background
Erlotinib is an epidermal growth factor receptor (EGFR) tyrosine-kinase inhibitor. Clinical trials proved its efficacy in patients with advanced-stage NSCLC. Although, activating EGFR gene mutations are currently the strongest predictor of erlotinib efficacy, the majority of NSCLC patiens harbor wild-type EGFR gene and moreover there is still a large proportion of patients in whom it is not feasible to acquire an adequate tissue for EGFR mutation analysis. We conducted this retrospective study aiming to prove the predictive role of tumor markers in patients with advanced-stage NSCLC treated with erlotinib.

Methods
144 patients with advanced-stage (IIIB, IV) NSCLC were treated with erlotinib (150 mg daily). Pretreatment serum levels of CEA, CYFRA 21-1, NSE and TK were assessed. Comparison of patients’survival (PFS and OS) according to the level of assessed tumor markers was performed using the log-rank test.

Results
Median PFS and OS for patients with normal CEA levels was 2.9 and 16.1 vs. 1.9 and 8.6 months in patients with high CEA levels (p = 0.046 and p = 0.116). Median PFS and OS for patients with normal CYFRA 21-1 levels was 3.4 and 23.8 vs. 1.9 and 6.1 months in patients with high CYFRA 21-1 levels (p < 0.01 and p < 0.01). Median PFS and OS for patients with normal NSE levels was 2.1 and 9.8 vs. 1.1 and 3.8 months in patients with high NSE levels (p = 0.051 and p = 0.022). Median PFS and OS for patients with normal TK levels was 2.1 and 23.7 vs. 2.0 and 8.5 months in patients with high TK levels (p = 0.110 and p = 0.037).

Conclusion
The study proved that high pretreatment levels of CEA, CYFRA 21-1 and NSE predict low efficacy of erlotinib treatment in patients with advanced-stage NSCLC. Assessment of these tumor markers is a good predictive tool for erlotinib treatment efficacy especially for those patients with unknown EGFR mutation status.

Background
Circulating tumor cells (CTCs) are cells that have spread from the primary tumor into the bloodstream, and they play a crucial role in the development of distant metastases. CTCs have been detected in several cancers and associated with aggressive disease. The aim of this study was to evaluate the correlation between the numeric variation of CTCs in the blood of patients (pts) with advanced adenocarcinoma (ADK) of the lung during chemotherapy (CHT) and the radiological response to explore their potential role as an early predictive indicator of treatment response.

Methods
Peripheral blood samples and computed tomography (CT) scans were obtained before any treatment (baseline) from 25 pts with advanced ADK who were candidates for first-line CHT (platinum-based combination with pemetrexed). Blood samples were collected every two CHT cycles, and radiological responses were concomitantly assessed by CT according to the RECIST v. 1.1 criteria. CTCs were isolated from blood and diluted in a buffer containing formaldehyde to lyse red blood cells and fix CTCs using a filtration-based device (ScreenCell[®], France) to sort CTCs by size. CTCs were isolated by size using a microporous membrane filter composed of polycarbonate material containing circular pores that are calibrated (7.5±0.36 μm) and randomly distributed throughout the filter (1×10[5 ]pores/cm[2]). At the end of filtration, hematoxylin and eosin (HE) staining and immunofluorescence (IF) using CK7 were performed to enumerate and characterize the CTCs. Moreover, variations in the CTC count were compared with tumor size variations observed in CT scans.

Results
Baseline CTC counts and CT scans were obtained from 25 pts, including 18 males and 7 females, with a median age of 68 (range: 45-81) years. H&E staining revealed that the CTCs were morphologically compatible with tumor cells and were present in all 25 pts at baseline (range: 2-25 CTCs/ml). Furthermore, the epithelial origin of the CTCs was confirmed by CK7 positivity (demonstrated by IF). CTCs and CT images were assessed in 19 pts after at least two CHT cycles; the best response was a partial response (PR) in 2 pts, progressive disease (PD) in 3 pts and stable disease (SD) in 13 pts. The CTC number varied with the tumor size in 77.8% (14/18) of pts; decreased CTC counts were observed in 87.5% (7/8) of pts with reduced tumor size, whereas increased CTC counts were found in 70% (7/10) of pts with increased tumor size. Notably, variations in CTC count and tumor size were concordant in 100% (5/5) of pts achieving a PR or with PD as the best response.

Conclusion
This study demonstrates the feasibility of isolating CTCs in all advanced lung ADK pts using a low-cost, size-based technique. Interestingly, the concordance between the CTC analysis and CT suggests that CTCs may represent an early predictive marker of disease outcome. To our knowledge, this is the first study suggesting a relationship between CTC variation and treatment response in lung cancer.

Background
Molecular biomarkers in tissues and body fluids represent a promising source to improve cancer diagnostics. Bronchoscopic Microsampling (BMS) of endobronchial epithelial lining fluid (ELF) is a potent method to investigate potential biomarkers in lung diseases. Recent studies report that mRNA derived from ELF can be used to improve the differentiation of malignant and benign pulmonary nodules. Tenascin C (TNC) is an important component of the extracellular matrix involved in tissue remodeling and cell signaling. It has been reported that TNC is differentially expressed in malignant pulmonary nodules. We investigated whether specific splice variations of TNC are differentially expressed in ELF collected by the BMS method in proximity of pulmonary nodules.

Methods
ELF was collected by the BMS method from subsegmental bronchi close to the indeterminate pulmonary nodule and from the contralateral lung. Diagnosis was confirmed by transbronchial biopsy or surgery. In this study 176 ELF samples were included from a total of 88 patients (65 NSCLC and 23 benign cases) with indeterminate pulmonary nodules detected by computed tomography. Quantitative real-time PCR (RT-PCR) assays were used to detect different TNC variant patterns.

Results
All patients underwent BMS without complications. Quantitative RT-PCR was reliably applied to all ELF samples. A significant increase of TNC transcript variants close to malignant nodules was observed for most of the used qPCR assays. A better accuracy was observed for two protein-coding variants and an untranslated transcript indicating a distinct tumor-associated TNC variant pattern and a lower intraindividual variance.

Conclusion
Our results indicate that the analysis of individual TNC variants might improve the accuracy in detecting malignant nodules compared to the detection of whole TNC transcripts.

Background
Accurate assessment of ALK gene rearrangement in NSCLCs is critical to identify patients likely to respond to crizotinib. Currently, the gold standard for identifying ALK gene rearrangements is FISH and the Abbott Molecular ALK break apart probe is commonly used. We evaluated a new ALK/EML4 TriCheck FISH Probe for the detection of ALK rearrangements and confirmation of EML4 as the inversion partner. In addition, we evaluated its use as an ancillary FISH probe for use in cases with subtle, equivocal or atypical ALK FISH patterns.

Methods
ALK FISH was prospectively performed on 29 routine diagnostic cases using the ALK/EML4 TriCheck Probe (ZytoVision) and the Vysis ALK Break Apart FISH Probe (Abbott Molecular). ALK immunohistochemistry (IHC) was performed using the 5A4 clone antibody (Novocastra) and D5F3 clone antibody (Cell Signaling Technology).

Results
Both probes were concordant in all cases, except for one case which showed an atypical signal pattern using the Abbott Molecular ALK probe. This case was technically negative using standard scoring criteria for the Abbott probe, despite positive ALK IHC, but was confirmed as positive using the ZytoVision TriCheck probe. Two additional cases which were equivocal (10-16% ALK rearrangement), were confirmed to be positive for ALK rearrangement using the ALK/EML4 TriCheck probe. Of the 10 ALK rearranged cases, 4 showed evidence of EML4 translocation.

Conclusion
The ALK/EML4 TriCheck FISH Probe is useful for the detection of ALK gene rearrangements, including those involving EML4 as the translocation partner, especially for borderline cases or cases displaying atypical signal patterns, where an additional unique ALK FISH probe can provide further confirmation of rearrangement.

Background
Dysfunction of Nrf2-Keap1 interaction by Keap1 or Nrf2 mutations may promote tumor growth, drug resistance, and poor outcomes in non-small cell lung cancer (NSCLC). However, both gene mutations are uncommon in Asian patients, suggesting that a different mechanism might be involved in lung tumor progression via altering Nrf2-Keap1 pathway.

Methods
Two strategies—treatment of NO scavenger and a nuclear location sequence (NLS)-mutated Nrf2 expression vector transfected into Nrf2-knockdown stable clones—were used to explore whether IKKβ could be elevated by cytoplasmic Nrf2 and promotes cell invasion. The prognostic values of cytoplasmic Nrf2 and IKKβ mRNA levels in tumors from NSCLC patients were estimated by Kaplan Meier and Cox regression model.

Results
We showed that mutated p53 upregulated Nrf2 transcription by de-repression of Sp1 binding to the Nrf2 promoter. Interestingly, cytoplasmic Nrf2 expression was markedly increased in p53-mutated cells compared with p53 wildtype cells due to a decrease in iNOS-mediated NO production. The stability of IKKβ protein was also increased by the interaction of cytoplasmic Nrf2 with Keap1, which blocked IKKβ degradation by the Keap1-mediated proteasomal pathway. An increase in IKKβ expression due to cytoplasmic Nrf2 was responsible for soft agar growth and invasion capability. Positive cytoplasmic Nrf2 immunostaining or high IKKβ mRNA expression in tumors from NSCLC patients may predict poorer overall survival and relapse free survival. Moreover, patients with cytoplasmic Nrf2 positive tumors had unfavorable responses to cisplatin-based chemotherapy.

Conclusion
Cytoplasmic Nrf2 may promote tumor aggressiveness, poor outcome, and unfavorable response to chemotherapy in NSCLC.

Background
More than 70% of patients with lung cancer are diagnosed with advanced-stage, with short survival times. The discovery of epidermal growth factor receptor (EGFR) mutations is a major scientific breakthrough; tyrosine kinase inhibitor (TKI) treatment for patients with EGFR mutations reported a median overall survival of more than 2 years. Nevertheless, chemotherapy has steadfastly remained the most validated therapies in non-small-cell lung cancer (NSCLC) not only for vast majority of patients without EGFR mutations, but for those with EGFR-TKI resistance. There are many influence factors in platinum resistance, among them DNA repair genes have been extensively explored. Excision repair cross complementing 1 (ERCC1) and/or breast cancer susceptibility gene 1 (BRCA1) is generally associated with sensitivity to platinum, but the results were inconclusive and controversial. Apurinic/apyrimidinic endonuclease 1 (APE1), a bifunctional AP endonuclease/redox factor, is important in DNA repair and redox signaling, may be associated with chemoresistance.

Methods
In total, 172 patients with advanced NSCLC who received platinum-based doublet chemotherapy, ages 34 to 84 years at diagnosis from 2007 to 2012 were enrolled into this study. We evaluated the impact of APE1, BRCA1, ERCC1 expression levels on response to platinum-based chemotherapy and median progression-free survival (mPFS) and/or median overall survival (mOS) outcomes, and protein expression levels were determined by immunohistochemistry (IHC).

Results
The mPFS was 8.8 months, mOS 12.8 months. A partial treatment response was found in 55 patients (31.98%). No significant association was found between BRCA1 protein expression and response rate, mPFS or mOS. Interestingly, patients whose tumors did not express APE1 had 50.00% response rate compared to 25.00% whose tumors expressed the protein (OR 0.34; 95% CI 0.17-0.69; P=0.002). Patients negative for APE1 had a significantly longer mPFS (10.3 vs. 8.0 months, P=0.016), but not mOS (17.1 vs. 10.9 months, P=0.263), than those positive for APE1. Patients negative for ERCC1 expression experienced better response to chemotherapy (OR 0.36; 95% CI 0.18-0.72; P=0.003), longer mPFS (9.8 vs. 6.8 months, P=0.001), and longer mOS (14.4 vs. 10.3 months, P=0.011). In particular, the expression of ERCC1 had a positive trend of correlation with APE1 expression (r[2]=0.23, P<0.01), and patients negative for both APE1 and ERCC1 had significantly higher response rate (OR 0.18; 95% CI 0.07-0.45; P<0.001), longer mPFS (12.0 vs. 6.8 months, P<0.001) and mOS (18.8 vs. 10.1 months, P=0.010), than those positive for both APE1 and ERCC1. Multivariate analysis adjusting for possible confounding factors showed that negative APE1 and/or ERCC1 expression was a significantly favorable factors for mPFS (HR 1.86, P=0.002 and HR 1.83, P=0.001; respectively) and mOS (HR 2.02, P=0.002 and HR 1.78, P=0.007; respectively).

Conclusion
The results suggest that dual-function protein APE1 may be a novel prognostic and predictive factor, and the combined detection of APE1 and ERCC1 expression might better individualize the efficacy of chemotherapy and improve survival in advanced NSCLC.

Background
It has been demonstrated that the addition of bevacizumab to paclitaxel plus carboplatin (CPB) in the treatment of advanced NSCLC improves survival. Even though, there is a high variability in drug efficacy between patients, leading to different response rates. ANGIOMET is an exploratory study promoted by the SLCG in advanced NSCLC, non-squamous histologies (NS-NSCLC) treated in first line with a combination scheme based in CPB, designed to investigate the relationship between angiogenic mediators and the outcome and response to treatment. The primary end-point was progression-free survival (PFS), and the secondary end-points are the follows: OS, response-rates and toxicity profiles.

Methods
In this multicentric study, patients with stage IIIB/IV NS-NSCLC (ECOG status 0–2) were included and treated in first line with CPB. Peripheral blood samples were collected before treatment administration and DNA was purified from the leukocyte fraction. Ten SNPs of VEGF-pathway genes were genotyped in 186 samples by RT-PCR in duplicate. SNPs were related to PFS and OS (Kaplan-Meir method, log-rank test) and to response rate.

Results
10 SNPs were determined in 186 DNA samples. In this preliminary analysis there were data from 108 patients valid for PFS and OS analysis. Baseline characteristics of the patients were: median age, 63 years [37-80]; 74.5% male; 94.1% ECOG PS 0-1; 14% never-smokers, 100% caucasian; 89.7% adenocarcinomas, 2.8% large cell carcinomas; median number of CPB cycles was 4. There was no response assessment in 27 patients (25%), 30.6% PR, 31.5% SD and 13.0% PD. The SNP rs833061 (CC) in VEGFA correlated with lower response rates to CPB than the other genotypes (p=0.07). SNPs in KRAS and VEGFR2 were associated with PFS and/or OS in our cohort. The KRAS SNP rs10842513 (TT+CT) was associated with shorter PFS compared with the CC genotype (median: 5.39 vs 6.81 months; p=0.04, respectively). The VEGFR2 SNP rs2071559 (AA) was significantly associated with longer PFS and OS (Table 1). No significant differences in PFS or OS were observed according to other SNPs analyzed. Table 1: PFS and OS for VEGFR2 SNPrs2071559.

		PFS
	%	Median (months)	95%CI	p
VEGFR2 (rs2071559)
AA	25.6	9,408	5,084 - 13,732	0.01
GG+AG	74.0	5,724	4,902 - 6,546
		OS
	%	Median (months)	95%CI	p
VEGFR2 (rs2071559)
AA	25.6	NR	----	0.001
GG+AG	74.0	12,270	8,760 – 15.780

Conclusion
These preliminary data indicate that genetic variation in VEGFR2, SNP rs2071559 variant AA, is associated with prognosis in advanced NS-NSCLC patients treated with CPB and may have predictive implications as biomarkers in patients treated with chemotherapy with bevacizumab. On behalf of the Spanish Lung Cancer Group (SLCG)

Background
VeriStrat (VS) is a multivariate protein serum test that classifies Non Small Cell Lung Cancer (NSCLC) patients in 2 categories VS Good or VS Poor according to the overall survival (OS) of patients treated with EGFR-TKIs. Recently, the prospective Phase III PROSE study showed that the VS algorithm is predictive of differential OS benefit for erlotinib (E) vs second line standard chemotherapy (CT): VS Poor classified patients had worse OS on E compared to CT, while there was no significant difference between treatments outcome in the VSG group. Aim of the current study was to evaluate if baseline Standardized Uptake Value (SUV) of FDG-PET may help to optimize treatment choice between E or CT IN VS Good classified pts.

Methods
Plasma samples were collected before the beginning of E from metastatic NSCLC patients. Acquired spectra were classified according to the VS algorithm. The FDG-PET was performed tha day before the beginning of E. Survival curves were estimated using the Kaplan-Meier method.

Results
Thirty eight NSCLC patients on E therapy with the following characteristics were analyzed: median age 62 years old, 63% were males, 53% had adenocarcinoma histology, response rate was 26%, median OS 10 mos and TTP 3.4 mos. Twenty-six (68%) were classified as VS Good, 12 (32%) as Poor. TTP and OS for VS Good and Poor were 4.1 vs 2.1 mos (HR 0.86, log-rank p=0.6) and 11.1 vs 4.1 mos (HR 0.45,log-rank p=0.02), respectively. All VS Poor classified patients had SUV ≥ 7 and had the worst TTP and OS; VS Good classified patients with baseline SUV level ≥ 7 had worse OS (10 mos) and worse TTP (2.1 mos) compared to those who were VS Good and had SUV<7 (OS 16 mos) (TTP 13.8 mos).

Conclusion
The study confirmed that VS Poor classified patients had significantly shorter OS than those classified as VS Good. Patients with VS Good profile and with baseline FDG-PET SUVs levels <7 may benefit more from EGFR-TKIs than VS Good patients with higher FDG-PET SUV, suggesting that FDG-PET analysis may be a clinically useful tool for EGFR-TKIs therapy selection.

Background
Epithelial–mesenchymal transition (EMT) and expression of stemness features are key issues for tissue invasion and metastasis during cancer development. OCT4 and NANOG are homebox transcription factors essential to the self-renewal of stem cells and are expressed in several cancers. The role of OCT4/NANOG signaling in tumorigenesis and as biomarkers in non-small cell lung cancer (NSCLC) is still elusive.

Methods
mRNA was isolated from 177 frozen samples, corresponding to tumoral and normal parenchyma of NSCLC patients in resectable-stage. OCT4 and NANOG relative expression was determined by RTqPCR using hydrolysis probes. Expression levels were normalized using GUSB as endogenous housekeeping gene. Statistical significance was considered for p<0.05.

Results
Patient’s median age was 64 years [26-87], 87.6 % were males, 75.2 % presented performance status (PS=0) and 47.3% had squamous histology. We found a significant positive correlation between OCT4 and NANOG expression (r[2 ]=0.61 p<0.0001, Pearson test). Higher levels of expression of NANOG were related to poor differentiation grade (p= 0.04). Survival analysis revealed that there is a trend to a poorer progression free survival in the subgroup of patients with higher levels (> median) of expression of OCT4 levels

Conclusion
The transcription factor OCT4 may have a role as prognostic biomarker in resectable NSCLC. (Supported in part by ISCIII (PI12/02838), RTICC (RD12/0036/0025), Ministry of Economy and Competitiveness and SEOM Grants 2012)

Background
Neoplastic pericarditis is found in 60-80% of patients (pts) with large pericardial effusion (pe) (>2 cm on echocardiography). The most frequent causes of neoplastic pe are metastatic lung or breast cancer. Malignant pe is combined with high relapse rate after pericardiocentesis and poor prognosis. Early recognition of malignant pe and intensive local (+/-) systemic treatment is able to improve life expectancy. Nevertheless in 30-40% of pts the malignant cause of pe is not recognized during the first episode of disease. The aim of the study was to introduce the new scoring system assessing the probability of malignant pe in the pts requiring pericardial fluid drainage due to large pe/tamponade.

Methods
146 pts, median age 57 years (21-88), 80 with benign cause of pe, 66 with neoplastic cause of pe ( positive pe cytology or neoplastic infiltration in pericardial biopsy specimen) treated in the National Institute of Tuberculosis and Lung Diseases in 1982 - 2008 entered the study. Metastatic lung cancer was diagnosed in 67% of neoplastic pe.

Results
Based on previous results, the most important features distinguishing between neoplastic and benign pe were: cardiac tamponade on echocardiographic examination, HR>90 beats/min on ECG, enlarged mediastinal lymph nodes (>1cm) on chest CT scan, CEA> 5ng/ml in pe, Cyfra 21-1>95 ng/ml in pe, bloody pe. The sensitivity, specificity, PPV and NPV are listed in table 1.

Parameter Diag. value	HR>90/min	Cardiac tamponade	CEA>5 ng/ml (pe)	Cyfra 21-1>95 ng/ml (pe)	Bloody pe	Lymph nodes>1 cm (CT)
sensitivity	0.82	0.67	0.63	0.64	0.86	0.93
specificity	0.55	0.65	0.94	0.95	0.57	0.70
PPV	0.60	0.61	0.91	0.93	0.66	0.66
NPV	0.79	0.70	0.71	0.71	0.80	0.94

The original scoring system (-3 up to + 3 points) was developed based on PPV and NPV of the above mentioned parameters. The diagnostic value of the proposed scoring system was high, ( ROC: AUC 0.926 95%CI 0.85-0.96) and exceeding the value of single parameters (at a cut off of 0 points – sensitivity was 0.84 and specificity - 0.91).

Conclusion
Assessment of the probability of malignant pe according to the proposed original scoring system was able to improve the diagnostic algorithm based on single parameters, thus indicating pts in whom more invasive diagnostic methods should be applied to recognize malignant pe.

Background
Mucin 1 (MUC1), a glycoprotein highly expressed in many malignancies, is being explored as an antigen for immunotherapy. How best to measure MUC1 expression in non-small cell lung cancer (NSCLC) and its prognostic value in NSCLC are under discussion.

Methods
Tissue microarrays (TMAs) were constructed using triplicate 1mm cores of formalin-fixed paraffin-embedded tumour and stained with 214D4 (recognizes protein core) and MA695 (recognizes carbohydrate epitope) anti-MUC1 antibodies. TMAs were assessed for polarization, cytoplasmic and membranous staining intensity (scored 0–3) and proportion cells positive (0–100%; scored 0–5), averaged for multiple cores. A composite score (intensity x cells positive) was derived (range 0–15).

Results
TMAs from 518 patients were analyzed: 69% male, 95% Caucasian, 7% never-smoking; 49% adenocarcinoma, 35% squamous cell, 7% large cell; 43% stage I NSCLC, 33% stage II, 23% stage III. Immunohistochemistry staining intensity, proportion positive cells and depolarization were very similar between antibodies, with high concordance in composite score (R2=0.71, p<0.0001). Polarization was discordant in 8%. Similar scores were seen for N0, N1 and N2 when assessed by either antibody. For 77 cases with paired primary/N2 nodal tissue, mean 214D4 composite scores were 10.7 and 10.1 and MA695 scores 9.5 and 9.4, respectively. Discordant staining in primary but not node was seen in 5.2% and 10.4% with 214D4 and MA695, respectively. For 27 cases with neoadjuvant chemotherapy vs no chemotherapy, mean 214D4 scores were 10.2 vs 10.1 and MA695 9.3 vs 9.6, respectively. Higher scores with each antibody trended toward improved survival (non-significant). Polarization was associated with improved survival (whole cohort) with 214D4 (80.6 vs 47.8 months; HR 1.37 [95%CI 1.078–1.742], p=0.01 log rank test) and MA695 (95.8 vs 45.7 months; HR 1.48 [95%CI 1.159–1.878], p=0.002). Polarization with 214D4 was strongly associated with improved survival for adenocarcinoma (HR 1.92 [95%CI 1.385–2.668], p<0.0001) but not for non-adenocarcinoma. Similarly, polarization with MA695 conferred a survival advantage in adenocarcinoma (HR 1.68 [95%CI 1.225–2.311], p=0.001) but not non-adenocarcinoma cases. Data on MUC1 immunohistochemistry and circulating soluble MUC1 will be presented.

	214D4	MA695
No staining	3.5%	6.2%
Mean intensity
- Cytoplasmic	1.8	1.7
- Membranous	2.1	2.2
Mean cells positive	3.9	3.6
Mean composite score	10.1	9.6
- Adenocarcinoma	13.1	12.0
- Squamous cell	7.1	7.0
- Large cell	7.2	7.7
Depolarization	66.7%	62.4%

Conclusion
Over 93% were MUC1 immunohistochemistry positive, with higher scores in adenocarcinoma. Composite scores for the two antibodies were highly correlated and depolarization largely concordant. MUC1 expression was generally maintained in paired primary/nodal tumour and was similar across nodal stages and following neoadjuvant chemotherapy. Polarization was associated with improved survival in adenocarcinoma. Further investigation is needed to determine which antibodies best predict outcomes.

Background
The immune checkpoint receptor programmed death-1 (PD-1) negatively regulates T-cell activation. In a phase 1 study, nivolumab (PD-1 blocking antibody) delivered durable responses in patients with advanced NSCLC, melanoma, and renal cell carcinoma. Immunohistochemistry (IHC) analysis suggested association between pretreatment tumor PD-1 ligand (PD-L1) expression and response to nivolumab in patients with melanoma (Grosso ASCO 2013 abstract 3016; Topalian New Engl J Med. 2012;366:2443). There is little published information associating PD-L1 expression with gene profiles, mutational status, or patient characteristics in NSCLC. Such data may be relevant in understanding which patient subgroups may be more likely to benefit from nivolumab therapy.

Methods
60 NSCLC formalin-fixed paraffin-embedded tumor tissue samples (Asterand: 30 squamous; 30 non-squamous) with matching RNA, frozen tissue samples, and patient characteristics/outcomes were utilized. Tumor cell membrane PD-L1 expression was evaluated by IHC using an automated assay based on a sensitive and specific PD-L1 monoclonal antibody (28-8). Tumors were defined as PD-L1 positive (PD-L1+) when ≥5% of the tumor cells had membrane staining at any intensity. PTEN/EGFR expression was analyzed by IHC. Mutations in isolated DNA were analyzed on the AmpliSeq[TM] cancer panel using the Ion Torrent platform. Gene expression was conducted on the Affymetrix platform and association with PD-L1 status analyzed using ANOVA.

Results
Of 59 tumor samples with available data assessed by IHC, 42% (25/59) were PD-L1+ and 58% (34/59) were PD-L1 negative (PD-L1-). There was no apparent association between PD-L1 protein expression and NSCLC histology: for squamous and non-squamous tumors, 38% (11/29) and 47% (14/30) were PD-L1+, respectively. No association was observed between PD-L1 status and PTEN or EGFR expression. PD-L1+ tumors, compared with PD-L1- tumors, showed higher expression of several immune-related genes, including interferon-gamma (IFNγ), IFNγ-induced cytokine, and other genes involved in immune-cell regulation. The PD-L1 gene was differentially expressed between IHC PD-L1+ and PD-L1- samples, with no continuous relationship noted. Genes associated with tumor progression and signaling pathways were over-expressed in PD-L1+ versus PD-L1- tumors, including proto-oncogene tyrosine kinase (MET), EGFR ligands, neuropilin-2, and alpha E-catenin. Analysis of key mutations from the Ampliseq panel indicated that rates of detectable mutations in our tumor panel differed from those reported in COSMIC (Catalog of Somatic Mutations in Cancer; http://cancer.sanger.ac.uk/cancergenome/projects/cosmic/), as p14/CDKN2A and CBL mutations were not observed. However, KRAS and TP53 mutation rates were consistent with COSMIC. IHC PD-L1 positivity was observed amongst KRAS mutation-positive (8/10) and KRAS wild-type tumors (15/43), and importantly amongst EGFR mutation-positive and EGFR wild-type tumors.

Conclusion
Current data suggest PD-L1 protein expression on NSCLC tumors may be associated with several factors, including expression of immune genes, expression of tumor progression markers, and driver mutations. Ongoing analyses within this tumor panel are exploring putative associations of PD-L1 expression with patient characteristics and outcomes. Findings could help define additional factors that may influence the likelihood of response to nivolumab therapy. Correlates to PD-L1 will be explored in greater detail as part of ongoing phase 3 trials of nivolumab in NSCLC.

Background
The expression profiles of serum micro RNAs (miRNAs) are known to predict overall survival (OS) of metastatic and operable non-small cell lung cancer (NSCLC). We hypothesized that circulating miRNAs is also correlated with survival in unresectable/inoperable NSCLC treated with radiation therapy (RT).

Methods
134 patients with inoperable/unresectable NSCLC treated with definitive RT (18-month minimum follow-up) were eligible. Serum samples were collected prospectively before RT commencement. 100 patients with enough serum and reliable miRNA profile quality were randomly divided into training and validation sets (50 patients each). MiRNA profiling was performed using real-time PCR-based array, containing a panel of 84 miRNAs detectable in human bodily fluids. Spiked-in cel-miR-39 was used for normalization. Stepwise regression Cox model building was used to build a miRNA signature on the training set, which was then assessed on the validation set both alone and with clinical factors.

Results
The median age was 67 years; 76% were stages III and 79% received chemoradiation; the median physical dose was 70 Gy. Stepwise regression modeling identified five miRNAs as jointly significant predictors. Using coefficients from Cox model fit, the miRNA signature was 0.53*log(hsa-miR-15b)+0.21*log(hsa-miR-34a)-0.27*log(hsa-miR-221)-0.27*log(hsa-miR-224) -0.07*log(hsa-miR-130b). This signature was a significant predictor of OS in the validation set (p=0.011). It retained statistical significance in a model also containing terms for GTV Volume and KPS, the only two significant clinical factors in univariate analysis in the validation set (p=0.012). Using computational methods (TargetScan6.2) for miRNA target prediction, the putative targets of these five miRNAs are known to modulate apoptosis, cell cycle control, DNA damage response and repair process (including nucleotide excision repair and DNA translesion synthesis), angiogenesis and epithelial-mesenchymal transition.

Conclusion
In this study, we developed a prognostic miRNA signature consisting of five miRNAs and validated in an independent dataset for unresectable/inoperable NSCLC treated with RT. This circulating miRNA signature may be used as a non-invasive biomarker, which may have prognostic or therapeutic implications for the future management of locally advance NSCLC patients. Larger sample size studies are needed to further validate our findings.

Background
Estimating survival is desirable to determine the best palliative approach for patients with malignant pleural effusion (MPE). Previous studies have shown that some pleural fluid biochemical parameters as glucose or pH are predictors of survival in such a population. Inflammatory infiltrate profile was already determined as a relevant predictor in solid tumors. However, inflammatory cells profile and lymphocyte activity was not systematically assessed as a survival predictor in MPE patients. Therefore, our objective was to evaluate whether cytology profile and Adenosine Deaminase (ADA) levels are relevant predictors of overall survival in patients with MPE who undergo pleurodesis.

Methods
Retrospective cohort study carried out in a tertiary university-based hospital. We included all patients who underwent pleurodesis for MPE treatment in our institution during the period: Jan/08 to Jun/11. We excluded patients that had no pleural fluid analysis registered in our database. The following data regarding pleural fluid analysis were retrieved from our database: glucose, LDH, total proteins and ADA, total cell count, leukocytes, macrophages, neutrophils, lymphocytes and oncotic cytology. All patients were followed up in our outpatient clinic until death, those who were lost to follow-up were contacted by telephone. Cox regression models were built to identify predictors of overall survival

Results
156 patients were included in this study (44 men, 112 women, mean age 58.9+-12 years). Primary neoplasms were: breast (83), lung (39), lymphoma (10), other (24). Median survival was 9 months. The final regression model was built using forward stepwise selection and the overall survival predictors identified that levels of ADA below 15 IU/L (HR:2.3, p=0.0008) or above > 40 IU/L. (HR 2.3, p=0.01) are associated with a shorter survival. An association with shorter survival was also found for patients with primary neoplasms different from breast, lung or lymphoma. (HR: 2,1, p= 0.007). No other predictor analyzed was associated with worse prognosis. Exploratory analysis showed that patients with ADA < 15 IU/L had low pleural fluid lymphocytes count and low protein concentration; paradoxically, those with ADA > 40 IU/L also had low lymphocytes count while other parameters were normal.

Conclusion
We concluded that ADA is a relevant predictor of survival in patients with MPE who undergo pleurodesis.

Background
Combined epidermal growth factor receptor (EGFR) tyrosine kinase inhibitor (TKI) therapy and cyclooxygenase-2 (COX-2) inhibition has been shown to potentiate responses in NSCLC, and may overcome resistance in wild type EGFR tumors. Several biomarkers have been used to evaluate effective targeting of the COX-2 pathway. COX-2 expression by immunohistochemistry (IHC) and urinary 11α-hydroxy-9,15-dioxo-2,3,4,5-tetranor-prostane-1,20-dioic acid (uPGEM) have identified patients who may derive benefit from COX-2 inhibition. We evaluated these markers in association with PFS and tumor response in our randomized, placebo controlled trial of erlotinib and high dose celecoxib in advanced NSCLC patients.

Methods
Pts with stage IIIB/IV NSCLC who progressed following at least one line of therapy or refused standard chemotherapy were randomized to erlotinib (150mg/day)/high dose celecoxib (600mg/2x day) v E/P. Urinary prostaglandin E2-metabolite (uPGEM) was assessed at baseline, week 4 and week 8 of therapy. Immunohistochemical evaluation of COX-2 was performed on archival tissue.

Results
87 pts had urine assessed and 38 pts had additional evaluable tissue following EGFR mutation analysis. PFS was significantly improved in patients with wild type EGFR tumors (3.2 v 1.8 months, HR = 0.54, p = 0.03). Previously established normal uPGEM levels for men and women were used to determine high and low baseline values. 59 pts (67%) had high and 29 pts (33%) had low uPGEM at baseline. Elevated baseline uPGEM was associated with improved PFS for patients who received high dose celecoxib (5.6 v 2.2 months, p = 0.09). Pts with low baseline uPGEM did not benefit from celecoxib. Further assessment of COX-2 and E-cadherin was performed by IHC and correlations will be presented.

Conclusion
UPGEM is an easily assessed biomarker that is associated with improved outcomes in patients treated with high dose celecoxib and erlotinib with advanced NSCLC. Tumor levels of COX-2 and E-cadherin may refine our ability to best define a population who will benefit from COX-2 and EGFR targeted therapy in future studies. Supported by NIH 1P50 CA90388, K12 CA01727, Phase One Foundation and medical research funds from the Dept of Veterans’ Affairs.

Background
KRAS mutations on codons 12, 13 and 61 result in the constitutive activation of protein, which may render tumor cells independent of Epidermal Growth Factor Receptor (EGFR) signalling and thereby resistant to tyrosine-kinase inhibitor (TKI) therapy in NSCLC patients. This study was aimed to evaluate the associations of KRAS and EGFR copy number alteration and mutations with response and time to progression (TTP) in EGFR TKI treated patients.

Methods
KRAS mutations on codons 12, 13 and 61 result in the constitutive activation of protein, which may render tumor cells independent of Epidermal Growth Factor Receptor (EGFR) signalling and thereby resistant to tyrosine-kinase inhibitor (TKI) therapy in NSCLC patients. This study was aimed to evaluate the associations of KRAS and EGFR copy number alteration and mutations with response and time to progression (TTP) in EGFR TKI treated patients.

Results
KRAS mutation was detected in 15 (17.8%) cases, EGFR mutation in 27 (32.1%) and EGFR amplification in 8 (9.5%). Significant differences were detected in response rates for wild-type compared with mutant KRAS ( 20 vs 0%, p=0.023), mutant compared with wild-type EGFR (59 vs 8%, p=0.007), and EGFR-amplified compared with non-amplified (71 vs 18%, p=0.005). Additionally, significant benefit from TKI therapy was observed for KRAS wild-type compared with KRAS mutated patients (median TTP 7 vs. 3 months, p=0.001), for EGFR-mutated compared with wild-type patients (14 vs. 4 months, p=0.004) and for EGFR-amplification in contrast to non-amplified cases (11 vs. 5 months, p=0.001). KRAS and EGFR mutations or EGFR amplification did not correlated with overall survival (18 vs. 19 months, p=0.406; 16 vs. 21 p=0.094; 25 vs. 17 months, p=0.103, respectively). Combined analysis of favourable status of three biomarkers strongly predicted benefit to TKI therapy (median TTP 15 vs. 3 months, p<0.001).

Conclusion
Combined analysis of KRAS mutation, EGFR mutation and EGFR amplification in EGFR TKI treated NSCLC might provide superior predictive information than single biomarker study in these patients

Background
Epithelial-mesenchymal transition (EMT) has been known to play a key role in stromal invasion of lung adenocarcinoma. Loss of E-cadherin and acquisition of vimentin are two critical steps in EMT, that are induced by Snail-1 and TWIST upregulation associated with overexpression of epidermal growth factor receptor (EGFR). However, roles of EMT-related proteins in EGFR mutants have not been fully elucidated. We investigated the inmunoexpression of EMT-related proteins in EGFR lung adenocarcinoma to demonstrate their key roles in tumor progression.

Methods
E-Cadherin and vimentin expression was assessed in 84 patients with EGFR+ LA to determine if these markers had the potential to predict clinical outcomes in patients treated with Erlotinib. The percentage of tumor cells with grades 0, 1, 2, or 3 membrane staining of E-Cadherin and cytoplasmic staining of vimentin was measured. We selected previously reported cut-off points shown to provide optimal stratification: ≥40% of tumor cells with staining of +2 and +3 for E.cadherin and ≥10% of tumors cell with any staining for vimentin. Overall response rates (ORR), clinical benefit (CB), time to progression (TTP), and overall survival (OS) were estimated, as well as variables that influenced OS.

Results
Mean age was 59.6 years (SD +/- 13.1) and 79.8% of patients were women. Mutations of EGFR, L858R and G719X in exon 19 were present in 61%, 31% and 6% respectively. Vimentin expression was strong in 9.5% (n=8) and E-cadherine expression was weak in 51.2%, moderate in 23.8% and strong in 23.8%. Highest positivity of E-Cadherin was related to exon 19 deletion (p=0-001) but not to L858R mutations (p=0.14). Strong vimentin reactivity was associated with history of smoking (p=0.03). OS was 12.3 [10-14], 27.0 [23-31],26.1 [20-32] and 33.5 [30-36] months when E-cadherine expression was negative, weak, moderate and strong (p=0.05). OS was 33 months [31-35] in vimentin-negative and 8.2 months [6-10] in vimentin-positive (p=0.001). Similar trends were observed for progression-free survival and response rate.

Conclusion
E-Cadherin and vimentin are valuable predictive biomarkers for EGFR+ patients. These results warrant further research on EMT in selected populations exposed to erlotinib.

Background
Malignant pleural mesothelioma (MPM) is a malignancy characterized by therapy resistance and poor outcome. The high mortality of MPM is largely due to its invasive growth, local recurrence and locoregional spread. The identification of molecular changes that lead to this phenotype is indispensable. ITGA7 is a laminin binding receptor and has been identified as a novel tumor suppressor based on its frequent mutation in other malignancies. However, the alterations of ITGA7 have not yet been studied in MPM.

Methods
ITGA7 mRNA levels of normal mesothelial and MPM cells were measured by q-RT-PCR. ITGA7 promoter silencing in mesothelioma cell cultures was quantified by methylation-specific PCR analysis and by sequencing. Migratory activity of MPM cells has been investigated by 2D videomicroscopy. In vitro cell proliferation and adhesion assays were performed on siRNA transfected cells to demonstrate the biological consequence of decreased ITGA7 expression. ITGA7 expression in MPM tissue specimens was analysed by immunohistochemistry (IHC) and correlated to clinical outcome data.

Results
ITGA7 is highly expressed by normal mesothelial cells while decreased in MPM cells. In most MPM cells the expression of ITGA7 was reduced through promoter hypermethylation. ITGA7 promoter is hypermethylated in a number of tested MPM cell cultures (n=13) and the ratio of promoter methylation inversely correlates with ITGA7 expression. MPM cells with high in vitro migratory activity demonstrated a significantly lower ITGA7 expression when compared to slow migrating MPM cells. Proliferation of normal mesothelial cells is inhibited by laminin and this inhibitory effect is abolished by downregulation of ITGA7 expression via siRNA transfection. Adhesion of normal mesothelial and MPM cells is enhanced by laminin, however, it is not decreased by downregulation of ITGA7. The clinical significance of ITGA7 protein expression was investigated by IHC in a cohort of 89 MPM surgical specimens. Importantly, patients with high ITGA7 expression had significantly longer median overall survival (448 days) than patients with low expression (247 days, log rank test: p=0,0281).

Conclusion
ITGA7 is a novel tumor suppressor in MPM and its expression is down-regulated in MPM cells by promoter hypermethylation. Moreover, low ITGA7 expression in tumor cells influences clinical outcome as a negative prognostic factor in human MPM.

Background
Malignant pleural mesothelioma is a lethal disease and hence the strong need for identifying new prognostic factors.

Methods
This is a retrospective study including all eligible patients with advanced malignant pleural mesothelioma (MPM) presenting to National Cancer Institute, Cairo University. Neutrophil lymphocyte (N/L) ratio was assessed before second line chemotherapy.2.5 was used as the cutoff point. Endpoints were assessment of correlation between N/L ratio and clinical response (CR), progression free survival (PFS) and overall survival (OS).

Results
52 patients (19 stage III and 33 stage IV) MPM were included and followed up during the period from July 2009 till November 2012 with a median follow up period of 2.6 months. 87.5% of patients with N/L ratio > 2.5 showed progressive disease versus 91.7% in patients with N/L ratio <2.5. (P-value=0.66).6 months PFS was 11% for patients with N/L ratio > 2.5 versus 14% for patients with N/L ratio <2.5. (P-value =0.001). 6 months OS was 72% for patients with N/L ratio > 2.5 versus 66% for patients with N/L ratio <2.5. (P-value =0.4).

Conclusion
N/L ratio is a potential prognostic marker for advanced MPM treated with second line chemotherapy.

Background
Several MEK inhibitors are in late-stage clinical trials in mutant BRAF melanoma, where novel patient selection biomarkers are not major impediments to clinical development. In non-mutant BRAF disease, there is a presumption that the optimum patient population may also be selected according to gain of function mutations in key genes, resulting in pathway activation. In the present work we have taken a different approach, based on two publications (Dry et al. Cancer Res 2010;70:2264–2273; Loboda et al. BMC Med Genomics 2010;3:26), attempting to define NSCLC patient populations for MEK inhibitors based on gene expression signature measurements of pathway output.

Methods
In silico analyses were performed to test the sensitivity of candidate transcriptome signatures for detecting KRAS mutations in NSCLC. NanoString assays were subsequently developed for the signatures and used in: i) cell line based cross-platform comparisons with Affymetrix technology ii) formalin-fixed paraffin-embedded (FFPE) NSCLC samples to determine measurable genes, variation in gene expression and the limit of quantification for the signatures iii) matched tumour samples from the same patients iv) a blinded cohort of 50 NSCLC samples with known KRAS mutation status.

Results
In silico data confirmed the published correlations of transcriptome signatures with KRAS status in NSCLC samples. NanoString data appear to be robust, demonstrating a strong correlation with the Affymetrix platform and reproducible signature scores across separate samples from the same tumour. Reproducibility is maintained across dilutions of the same isolated RNA sample and supports previous observations regarding the sensitivity of these signatures for detecting KRAS mutations in clinical samples. In addition, we showed that high expression of these signatures is not restricted to samples with KRAS mutation, confirming previous observations that RAS or MEK activation is not exclusively linked to KRAS mutation.

Conclusion
We have developed a clinically relevant, robust assay platform, determined biological variation within tumours and confirmed the link to KRAS mutation status in a cohort of blinded NSCLC samples. The NanoString assays provide a means to test the prognostic and predictive capabilities of the gene signatures in the samples routinely provided in clinical practice. We intend to test the concordance of the gene signature indices between primary and metastatic tumours from the same patients, the prognostic relevance of the signatures in first- and second-line NSCLC patients treated with standards of care and wherever possible in future clinical trials of MEK inhibitors.

Background
Epidermal growth factor receptor (EGFR) mutation testing is essential to determine treatment regimens for patients with advanced non-squamous non-small cell lung cancer (non-Sq NSCLC). However, it requires at least two weeks until the results of commercialized EGFR testing are available. Therefore, we developed a novel high-speed real-time PCR system to reduce the time required for EGFR mutation detection. The reaction time of this system is only 5 minutes, and EGFR mutation status is found out within a few hours after diagnostic bronchoscopy. We tried to detect exon 19 deletion and exon 21 point mutation(L858R) in bronchial lavage fluid (BLF) after transbronchial biopsy (TBB) using this system. The aim of this study is to assess the performance of this novel high-speed real-time PCR system.

Methods
Seventy five consecutive patients who underwent TBB in our hospital from November 2012 to May 2013 were enrolled. Samples were obtained from BLF after TBB. DNA was extracted using QIAamp Blood Mini Kit (QIAGEN Japan, Tokyo, Japan). EGFR mutations were detected with high-speed real-time PCR system (TRUST medical, Hyogo, Japan) (Method A). Once the patient was diagnosed NSCLC with histology of TBB samples, EGFR mutation status was validated with PCR-invader method (BML, Inc. Tokyo, Japan) (Method B). We evaluated the sensitivity and concordance between (A) and (B).

Results
Lung cancer was histologically diagnosed in 51 patients (adenocarcinoma/squamous cell carcinoma/others=42/6/3), while not in 24 patients. Detection rate of EGFR mutation in patients with lung cancer was 29.4%(15/51) with (A) and 31.4%(16/51) with (B), respectively. Concordance rate between two methods was 94%. Discordance was found in one (6%) sample, where minor mutation of Exon19 L747-P753 deletion and insertion S was found only with (B). When (B) was used as a standard, sensitivity and specificity of (A) were 94% and 100%, respectively. Time to detect EGFR mutation by (A) and (B) were 2 hours and 18 days (6-45 days), respectively.

Conclusion
A novel high-speed real-time PCR system enables us to apply rapid EGFR mutation detection for clinical use.

Background
Lung adenocarcinomas represent a morphologically heterogeneous tumor composed of an admixture of different histologic subtypes (lepidic, papillary, acinar, and solid subtype). The presence of a solid subtype component is reported to be associated with a poorer prognosis. The aim of this study was to evaluate the characteristic immunophenotype of the solid subtype component compared with the immunophenotypes of other components.

Methods
We analyzed the clinicopathological characteristics of stage I adenocarcinoma patients with predominant solid subtype disease. Furthermore, we immunostained adenocarcinomas with predominant lepidic, papillary, acinar, and solid subtype components (n = 23 each) for 10 molecular markers of tumor invasiveness and scored the results.

Results
Patients showing predominance of the solid subtype component (solid subtype adenocarcinoma) had a poorer prognosis than those showing predominance of the lepidic, papillary, or acinar component. Lymphovascular invasion was more often detected in solid subtype tumors than in others. The solid subtype component showed a significantly stronger staining intensity of laminin-5 expression than the lepidic, papillary, and acinar components (P\\0.001, P\\0.001, and P = 0.016, respectively). The fibronectin and vimentin expression levels were also significantly higher in the solid subtype component than in other components. This immunostaining character was validated by using mixed-subtype adenocarcinomas containing all four components in the same tumor.

Conclusion
This study concluded that the solid subtype component in lung adenocarcinomas exhibit the invasive immunophenotype, including increased laminin-5 expression, compared with the other components, which may be associated with a poorer prognosis.

Background
Recent data suggests a trend for prolonged progression-free survival (PFS) in patients with wild-type EGFR (wtEGFR) non-small cell lung cancer (NSCLC) treated with second-line docetaxel over erlotinib. In this same study, however, overall survival (OS) appeared unaffected. In the maintenance setting, erlotinib was also shown to improve both PFS and OS in wtEGFR advanced NSCLC patients. More recently, evaluating serum protein patterns by mass spectroscopy revealed inferior PFS with erlotinib compared to docetaxel in patients with a particular protein pattern. The individual proteins in the mass spectroscopy peaks were not identified. The objective of this study was to develop a multi-analyte serum panel of specific proteins with predictive value for erlotinib versus palliative chemotherapy in pretreated advanced NSCLC patients that were unselected for EGFR mutation status.

Methods
74 biomarkers were evaluated using Luminex immunobead assays against a total of 120 patients with stage IV NSCLC that were previously treated with chemotherapy and were unselected for EGFR mutation status. Patients were treated either with single agent erlotinib or platinum-based chemotherapy at the discretion of the treating physician. Patients were evaluated for disease progression using RECIST criteria and association of biomarker with survival outcomes was assessed using Cox proportional hazards regression model. The differential association of the biomarkers in the two treatment groups (erlotinib or chemotherapy) with survival outcomes was assessed using a proportional hazards (PH) interaction model.

Results
In univariate PH regression analysis, we identified 7 serum biomarkers (osteopontin, CA-125, sTNF-RII, PlGF, leptin, IGFBP5, and amphiregulin) which were strongly associated (p<0.01) and nine additional biomarkers (IGFBP4, sTNF-RI, CA 15-3, IGFBP1, sIL-2Rα, sFAS, VEGF-A, sIL-1RI, and sIL-4R) which had significant association (p<0.05) with PFS in the erlotinib subgroup (n=92). Similarly, 9 biomarkers (osteopontin, sTNF-RI, sTNF-RII, CA 15-3, sIL-2Rα, epiregulin, PILG, IL6 and CA 125) were found to be strongly associated with OS. In our assessment of differential association with PFS, we found eight serum biomarkers (sIL-2Rα, IL-8, sIL-1RI, Tensascin C, FGF2, HGF, Leptin, and TGF-α) with significant to strongly significant positive interaction terms, thus indicating differentially increased hazard of progression in the chemotherapy subgroup with high biomarker levels. Four markers (IL-8, TGF-α, HGF, VEGF-A) were found to have significantly positive interaction, indicating a decreased hazard of death in the erlotinib group with high biomarker levels; whereas two (sTNF-RII, PSA) had significant negative interaction with OS, demonstrating a increased hazard of death in the erlotinib group.

Conclusion
We identified a series of circulating surrogate biomarkers associated with epithelial-to-mesenchymal transition (EMT) that have promise for algorithm development to help physicians determine whether erlotinib as a single agent of chemotherapy is capable of improving outcomes in patients that progressed after first-line platinum-based chemotherapy. Current efforts focus on evaluating biomarker relationships with EGFR mutation status and algorithm validation in an effort to enhance survival in advanced stage NSCLC.

Background
Next generation sequencing is a powerful tool to investigate somatic changes in tumor DNA. However the challenge is to detect low-frequency variants (< 10% minor allele frequency, MAF) in DNA extracted from formalin-fixed-paraffin-embedded (FFPE) tissue. DNA extracted from FFPE is highly fragmented and chemically modified. To overcome these challenges, we have developed a novel technique that can distinguish true variants from fixation artifacts with high sensitivity and specificity by investigating each of the two DNA strands independently.

Methods
TruSeq Custom Amplicon technology was used to generate sequencing libraries and deep sequencing was carried out to an average depth of 20,000X with a minimum of 1000X. The targeted re-sequencing assay* investigates ~14 kb of exons in 26 genes commonly mutated in solid tumors.

Results
Testing of more than 200 samples with a MAF ≥5%threshold revealed the presence of a large number of potentially false positive calls when data from only one strand of DNA was analyzed, but this number was significantly reduced (e.g. >50% for G>A) when both strands were considered.

Conclusion
This technique can distinguish FFPE artifacts from true variants and therefore provides increased accuracy for the detection of low-frequency variants by NGS. *Research Use Only

Background
EGFR mutation status is widely accepted as an important biomarker in NSCLC. To assess the current status of EGFR mutation testing, including testing procedures, sample types, mutation prevalence across demographic/histological subgroups (adenocarcinoma and non-adenocarcinoma histologies), and impact of EGFR mutations on personalized therapy choice, a large, multinational, diagnostic, non-comparative, interventional study (NCT01788163; IGNITE) of EGFR mutation status in patients with locally advanced/metastatic NSCLC of adenocarcinoma and non-adenocarcinoma histology will be conducted across centers in the Asia Pacific region and Russia.

Methods
Approximately 3500 patients (age ≥18 years) with chemotherapy naïve, Stage IIIA/B/IV NSCLC (newly diagnosed or recurrent disease after surgical resection) that is not suitable for curative treatment will be recruited over ~18 months from Asia Pacific (including 25 centers in China) and Russia. To give similar precision for the estimation of EGFR mutation frequency in the key non-adenocarcinoma subgroup in Asia Pacific and Russia, 2500 patients from Asia Pacific (assumptions: 20% patients non-adenocarcinoma histology, with 10% EGFR mutation frequency [one-fifth of the 50% prevalence in adenocarcinoma]; precision ±3%) and 1000 patients from Russia (assumptions: 20% patients non-adenocarcinoma histology, with 4% EGFR mutation frequency [one-fifth of the 20% prevalence in adenocarcinoma]; precision ±4%) need to be screened. Provision of diagnostic tumor samples for testing will be mandatory for all patients; additionally, plasma samples, which contain circulating-free tumor derived DNA (cfDNA), will be collected for EGFR mutation testing of plasma from a subset of 2500 patients in Russia, China, Taiwan and Korea. EGFR testing will be performed according to local practices, with Exon 19 deletions and L858R point mutations assessed as a minimum. The first-line (all patients) and second-line (patients with mutation-positive NSCLC; estimated 50% will progress to second-line treatment by study cutoff) therapy choices will be recorded. The primary objective is to determine EGFR mutation (and subtype) frequency in patients with adenocarcinoma and non-adenocarcinoma NSCLC (overall and by country/region). Secondary objectives are: to describe first-line therapy following EGFR mutation testing and second-line therapy following discontinuation of first-line treatment in patients with EGFR mutation-positive NSCLC confirmed by tissue/cytology; to determine concordance between EGFR mutation status determined using tissue/cytology versus plasma; to summarize current EGFR testing practices (methods, sample types, mutation detection rate, turnaround time, and reasons for not performing testing); to determine correlations between EGFR mutation status (derived from tumor or plasma) and demographic data and disease status. The secondary analyses in this study will provide information on current testing and therapeutic practices in advanced NSCLC across the Asia Pacific region and Russia, as well as an assessment of the utility of cfDNA as a less invasive methodology for the assessment of EGFR mutation status in patients with NSCLC.

Results
Not applicable.

Conclusion
Not applicable.

Background
Non-invasive genotyping of cfDNA has been shown to be feasible using highly sensitive assays. However, for detection of uncommon genomic events, specificity must approach 100% or false positive results impair clinical utility. Digital droplet PCR (ddPCR) is a quantitative genotyping technology that emulsifies input DNA into ~20,000 droplets which are PCR amplified, fluorescently labeled, and read as mutant or wildtype in a droplet flow cytometer. Using this quantitative technology, we aimed to develop a clinical-grade assay for non-invasive plasma genotyping and serial disease monitoring.

Methods
Patients with advanced NSCLC known to harbor EGFR or KRAS mutations were studied in an IRB-approved fashion. Plasma was collected in 10cc EDTA-tubes. Extracted DNA was quantified with a PCR for LINE1 and genotyped using ddPCR. Specificity of EGFR genotyping was determined using patients with KRAS-mutant lung cancer as gold standard negative cases. Serial assessment was piloted on EGFR-mutant cases receiving first-line erlotinib.

Results
To minimize risk of false positive results, we identified the “normal range” for EGFR L858R and exon 19 deletions in specimens from KRAS-mutant lung cancers as 0-1 and 0-8 copies/mL of plasma, respectively. Using this threshold for positive, ddPCR for EGFR sensitizing mutations had 67% sensitivity and 100% positive predictive value (Figure 1). Sensitivity was 100% with LINE-1 levels between 60-60000 pg/mcL but was poor with higher or lower cfDNA concentrations. Serial assessment on erlotinib (Figure 2) demonstrated pretreatment detection of EGFR mutations with ddPCR, complete plasma response on erlotinib, and subsequent reemergence of plasma EGFR up to 16 weeks prior to objective progression. Figure 1 Figure 2

Conclusion
Plasma genotyping of cfDNA using ddPCR has 100% specificity when using a rigorously defined threshold for a positive result. Sensitivity is highest in specimens with optimal cfDNA concentration. Clinical development is underway to use this non-invasive assay to guide genotype-directed therapy.

Background
The oncologic efficacy of lobectomy for lung cancer by means of video-assisted thoracic surgery (VATS) compared with conventional thoracotomy has been reported, and VATS lobectomy is now considered to be one of the standard surgical procedures for lung cancer. In this study, we retrospectively evaluated the long-term prognosis and some problems after VATS lobectomy, comparing with conventional thoracotomy, for clinical stage IA non-small cell lung cancer (NSCLC) in our institution.

Methods
From July 2002 to June 2012, 160 patients were diagnosed as clinical stage IA NSCLC and they underwent lobectomy. Of those 160 patients, 114 patients underwent VATS lobectomy and 46 patients underwent lobectomy under conventional thoracotomy. Patients’ clinical characteristics, recurrent status and overall survival were recorded. Disease free survival (DFS) and overall survival (OS) were calculated by means of Kaplan-Meier analysis and statistical significance between the groups was analyzed by using log-rank tests. Cox proportional hazard regression was used to ascertain independent predictors of recurrence.

Results
Median follow-up time was 44.8 months. 5-year DFS was 88.0% in VATS group and 77.1% in thoracotomy group in clinical stage IA (p=0.1504), and 91.5% in VATS group and 93.8% in thoracotomy group in pathological stage IA (p=0.2662). 5-year OS was 94.1% in VATS group, whereas 81.8% in thoracotomy group in clinical stage IA (p=0.0268), and 94.8% in VATS group and 96.2% in thoracotomy in pathological stage IA (p=0.5545). Cox proportional hazard analysis demonstrated a lower risk of recurrent disease in patients without lymph nodes metastases in this series (p=0.0026). The accurate diagnostic rate of preoperative staging was 71.9% in VATS group and 56.5% in thoracotomy group (p=0.2611). Inconsistent factors between clinical and pathological stage were largely tumor size (12.3% and 17.4%), nodal statement (10.0% and 21.1%) and pleural involvement (15.0% and 15.8%) in VATS group and thoracotomy group, respectively. There were 27 recurrent lesions (22 cases) at the first time of recurrence after surgery in this study. Twelve lesions (11 cases) with distant metastases were detected in VATS group, whereas 8 lesions (5 cases) were occurred distant metastases in thoracotomy group. Interestingly, only one lesion with local recurrence was detected in VATS group, whereas 6 lesions (5 cases) in thoracotomy group were detected (p=0.0495).

Conclusion
There was no significant inferiority for DFS and OS in VATS group, and local control was also significantly better in VATS group, compared with thoracotomy group. On the other hand, the significant difference of OS between two groups in clinical stage IA and multivariate analysis for recurrence showed the insufficiency of accurate staging before surgery.

Background
Limited pulmonary resections including lung segmentectomy for peripheral small lung cancer have attracted attention in recent years. However, a surgical consensus has not been established. It has been pointed out that there are not only lymph flows to pulmonary hilum along pulmonary vessels or bronchi but also pleural lymph flows directory into the mediastinum or adjacent lung lobe. There are some lung cancer cases with pleural indentation less than twenty millimeters. In these cases, it is concerned that lymph flows carry metastases from the pulmonary segment directly into the mediastinal lymph nodes without passing through the hilar lymph nodes. In other words, skip metastases might be caused. However, there have been few reports investigating pleural lymph flows exceeding the lung segment. The present study was designed to evaluate whether pleural lymph flows exceeding the lung segment could be detected using indocyanine green (ICG) and a fluorescence imaging system intraoperatively.

Methods
Fourteen patients undergoing lung segmentectomy or lobectomy for a tumor were enrolled in this study. A jet ventilation is selectively applied under bronchofiberscopy to the burdened bronchus to develop an anatomic border between the inflated segment to be evaluated and the deflated area. A 1.0 ml solution containing the fluorescent dye ICG (2.5 mg/ml) was injected into three to five subpleural sites of the segment. Fluorescence imaging device (HyperEye Medical System, MIZUHO IKAKOGYO CO.,LTD. Tokyo, Japan) was used to monitor the ICG-containing lymph flows from the injection site for five minutes. We evaluated the presence of pleural lymph flows exceeding the lung segment.

Results
We observed pleural lymph flows in eight of fourteen cases (57.1%), and pleural lymph flows exceeding the lung segment in seven of fourteen cases (50.0%). There is no pleural lymph flow from superior segment of bilateral lower lobe exceeding the segment in studies of several segments. Figure 1

Conclusion
Pleural lymph flows exceeding the lung segment can be observed in vivo. Skip metastases may occur through subpleural lymph channels in subpleural lung cancer cases. We should pay attention to skip metastases when we perform limited pulmonary resections for such cases.

Background
Large chest wall resections with significant loss of the skeletal framework can result in flail chest, prolonged ventilator dependence, and major respiratory impairment. Limited case reports address the extreme situation of massive chest wall defects, defined as oncologic resection of 5 or more ribs. We review our institutional experience and compare patient demographics, surgical techniques, and clinical outcomes to evaluate which factors are predictive or protective of complications.

Methods
Patients information was prospectively entered into a departmental database and then retrospectively reviewed. All consecutive patients who underwent immediate reconstruction of massive thoracic neoplastic or oncologic-related defects (≥5 ribs resected) between 1994 – 2011 were included. Tumor defect and reconstructive factors were evaluated for possible relationships with complications. Logistic regression analysis evaluated predictive factors for surgical outcomes.

Results
A total of 59 patients (median age 53) were available for review. Rib resections ranged from 5 to 10 ribs (defect area 80-690cm[2]). Indications included lung malignancy (52.5%), sarcoma (33.9%), and squamous cell carcinoma (5.9%). Types of rigid and semi-rigid reconstruction included use of prosthetic implants (83%), methylmethacrylate (25.4%), bioprosthetic mesh (5.1%). Soft tissue reconstruction required free tissue transfer (6.8%) and local muscle flaps (45.7%). Diaphragm reconstruction was required in 18.6% patients. The overall complication rate was 62%; which was subdivided into pulmonary complications (48%), cardiac complications (12%), and wound complications (17%). On average, patients were ventilator dependent for 3.9 days, required ICU monitoring for 4.9 days, and were discharged after 15.6 days. Mean follow-up time was 36 months. The 90-day overall survival rate of patients after initial procedure was 89.4%; all deaths occurred within superior resections (p=.03). Average postoperative decreases in FEV1 and FVC were 6.8% and 5.3%, respectively. Patients with superior resections and those older than 60 years were more likely to have post-operative complications.

Conclusion
In patients with massive oncologic thoracic defects, complex reconstructions are associated with a high rate of complications. However, creation of a stable construct is possible to prevent debilitating respiratory impairment and minimize pleural complications. Frequently, massive defects may be reconstructed with local muscle flaps obviating the need for free flaps. Prospective multicenter trials are warranted to differentiate and establish superiority of specific techniques and implant devices within these rare but challenging cases.

Background
The soft-coagulation system VIO is a new device for tissue coagulation in which the output voltage is automatically regulated. This system controls the temperature below the boiling point, without generating sparks, thereby causing minimal damage to surrounding tissues. This study was designed to evaluate the usefulness of a soft-coagulation system for lung cancer surgery in the viewpoint of minimizing surgeons’ risk.

Methods
We used soft-coagulation system VIO (ERBE Elektromedizin, Tubingen, Germany) for major pulmonary resections in 223 consecutive patients with primary lung cancer from January 2009 through April 2013. Bipolar soft-coagulation mode was used for tissue coagulation around lobar vessels, fissure dissection, and lymph node dissection. Three general thoracic surgeons were enrolled in this study. The checkpoints included blood loss, incidence of intraoperative complications, and surgeons’ stress. The data among the 223 patients who underwent major pulmonary resections using the VIO system (VIO group) were compared with data on 122 patients with primary lung cancer who underwent major pulmonary resection in our institution between January 2006 and December 2008 using conventional electrocautery (CE group). Student’s t-test was used to examine intergroup difference in blood loss. The threshold of significance was set at P<0.05).

Results
The patients consisted of 116 men and 107 women with a median age of 70.3 years. The type of resection was lobectomy in 200 patients, bilobectomy in 5, pneumonectomy in 6, and segmentectomy in 12. All pulmonary resections were performed by thoracotomy. Thoracoscopy was used in all cases for assistance. The pathological stage was stage IA in 129, IB in 43, IIA in 8, IIB in 14, IIIA in 24, IIIB in none, and IV in 4. With the VIO system, coagulation of tissues around the lobar vessels was effective without injury to the pulmonary vessels or bronchus. The mean blood loss in the VIO group was 76.1 g (range 5-700 g), which was significantly lower than that in the CE group (mean 175.3 g, range 5-1580 g) (P<0.0001). There was no intraoperative complication. The interview has shown that all three surgeons experienced less stress in surgery with soft-coagulation system, compared to surgery with CE.

Conclusion
The results of this study showed that the VIO soft-coagulation system is safe and feasible for major pulmonary resections in patients with primary lung cancer. This device could contribute to improve safety during dissection of the lobar vessels and decrease the surgeons’ risk in lung cancer surgery.

Background
In patients with completely resected non-small cell lung cancer, recurrence-free and postrecurrence survival, and metachronous primary lung cancer, have not been well studied at the same time.

Methods
A total of 315 patients with non-small cell lung cancer who underwent complete resection between 2001 and 2005 were examined. Patients were routinely assessed with chest computed tomography scan and physical checkups every 4 months for the first 2 years, and every 6 months from the third to fifth year. After that, they were examined annually. Accordingly, surviving patients can be followed up for five years or more after surgery.

Results
Median recurrence-free survival was 15.7 months. Multivariate analysis showed that pathological stage and pleural invasion were associated with decreased recurrence-free survival (Fig1). Median postrecurrence survival was 18.7 months. Multivariate analysis indicated that male gender, pleural invasion, extrathoracic recurrence and supportive care for recurrence were associated with decreased postrecurrence survival (Fig 2). The cumulative rate of metachronous primary lung cancer at 5 years was 3.7 %, and it developed at even eight years after initial surgery. Figure 1

Conclusion
The long-term follow-up of patients with completely resected NSCLC revealed that recurrence-free survival was related to the pathological stage of the original lung cancer, but postrecurrence survival was not. Only pleural invasion of the original lung cancer was related to both recurrence-free and postrecurrence survival. Moreover, postrecurrence survival was related to both site and treatment of the initial recurrence. In brief, the more advanced stage a lung cancer is at, the earlier it recurs. However, after recurrence, postrecurrence survival is related to the recurrence site or type of treatment of recurrent disease, rather than original lung cancer stage. The incidence of metachronous primary lung cancer was stable over time after the initial surgery.

Background
Minimally invasive pulmonary resection has been accepted as a technique for performing lobectomy since the mid 1990s; however, despite a number of effectiveness studies, the percentage of VATS lobectomies remains low. By some estimates, VATS lobectomy accounts for only 8 to 20 % of all lobectomy cases. We evaluated the development of a robotic assisted lobectomy program and short term outcomes. We assessed the impact of this platform on the rate of minimally invasive resections.

Methods
All patients undergoing robotic lobectomy were consented and entered into an IRB approved outcomes registry at the time of initial surgical consent. Multiple variables including length of operation, hospital stay, duration of chest tubes, complications, and mortality were evaluated. Medical records of patients undergoing robotic-assisted lobectomy between September 2011 and May 2013 were reviewed. Case logs for the years 2010 through 20123 were tallied to provide aggregate data on pulmonary resection.

Results
Histology for these patients included adenocarcinoma (45), squamous cell (20), benign disease (13), carcinoid (7), metastases from other primary (7). Stages of NSCLC were: I (56%), II (16%), IIIa (14%) and IV (6%). Ten of 87 patients with NSCLC had induction therapy (9 chemo/rad, 1 chemo). Conversion to thoracotomy occurred in 10% overall, but was performed in 3 of 10 induction patients. LOS (6d v 5d) and operative time (275 min v 221min) were longer in the induction group. When analyzed by quartile, average duration of robotic assisted lobectomy decreased from 280 min to 170 min. LOS decreased by 2 days over the initial 100 cases. Mortality rate was 1%. The minimally invasive case percentage prior to robotic assisted resection was 24%. During the first year of robotic assisted surgery, this minimally invasive value increased to 78%. In the second year of robotic assisted lobectomy, 86% of resections were performed with minimaly invasive approach.

Conclusion
In 2011, we added a robotic assisted surgery program into a multi-surgeon comprehensive thoracic oncology service. As expected, procedure times and need to convert to thoracotomy decreased as experience was acquired despite the increasing complexity of procedures. Robotic assisted resection has applicability to all stages of NSCLC including patients receiving induction therapy. Most importantly, robotic assisted surgery provided a platform for significantly increasing the rate of minimally invasive lobectomy in our institution.

Background
Preoperative induction treatment combining chemotherapy or chemoradiotherapy with surgical treatment may improve the prognosis of locally advanced lung cancer patients. On the other hand, induction treatment is associated with a higher incidence of postoperative pulmonary complications. We investigated the patients who received preoperative induction treatment to assess the respiratory function before and after induction treatment, and association with postoperative respiratory complications.

Methods
Preoperative induction therapy followed by surgery was performed for cT3-4 or cN2-3 locally advanced lung cancer in 118 of the 1,820 patients undergoing lung cancer resection between January 1997 and December 2012. Sixty-nine patients with complete data on the respiratory function before and after preoperative induction therapy were analyzed. Pulmonary functions before and after induction therapy were analyzed. Predicted postoperative those pulmonary functions also analyzed in the patients with pulmonary complication (group PORC) and without pulmonary complications (group NPORC). Independent group t tests were performed and p value<0.05 was considered statistically significant.

Results
There were 58 males and 11 females and median age was 61 years old. There were 38 adenocarcinomas, 20 squamous cell carcinomas, and 11 other pathologies. There were 11 stage IIB patients, 41 stage IIIA, and 17 stage IIIB. All patients received multidisciplinary induction treatment. Forty-three patients received induction chemoradiotherapy and 26 patients received induction chemotherapy. There was no significant change in %VC, %FEV1 before and after induction therapy. %DLCO (p<0.05) and DLCO/VA (p<0.01) were significantly decreased after induction treatment. More decrease of %DLCO was observed after induction chemoradiotherapy than chemotherapy (p<0.03). After the induction treatment lobectomy was performed in 51 patients, bi-lobectomy in 7, pneumonectomy in 10, and segmentectomy in 1. Combined resection of chest wall was performed in 16 patients, vertebra in 5, left atrium in 5, superior vena cava in 2, and diaphragm in 1. Sleeve lobectomy performed in 6 patients, sleeve bi-lobectomy in 3 and sleeve pneumonectomy in 1. Complete excision rate was 91.3%. Pathological analysis revealed that the ratio of patients obtained Ef 2-3 response were 56% after chemo-radiotherapy and 27% after chemotherapy (p<0.01). Median survival rate was 44.7 months and 5-year survival rate was 36％ for all patients. Especially 5-year survival rate of patients who obtained Ef 2-3 response after chemo-radiotherapy was 67%. There was no operative death and morbidity rate was 35%. Respiratory complications occurred in 12 patients. There were 8 pneumonia patients and 4 persistent hypoxemia patients. Ppo%VC, ppo%FEV1, ppo%DLCO, and ppoDLCO/VA were significantly low in the PORC group.

Conclusion
Higher proportion of patients obtained Ef 2-3 response after induction chemoradiotherapy and these patients showed a more favorable prognosis. DLCO should be evaluated to select candidates for induction therapy. Predicted postoperative pulmonary function should be assessed before surgery to select patients and to avoid critical pulmonary complications.

Background
Non-small cell lung cancer (NSCLC) is a typical disease of the elderly, and is becoming increasingly. Surgical resection is standard treatment for early-stage NSCLC. Our objective was to evaluate on surgery of NSCLC in the elderly, and assess the problem after surgical resection for NSCLC. We analyzed type of operation, postoperative complication, overall survival and cause of death for older patients undergoing surgical resection for NSCLC. We compare the clinical features of surgery of NSCLC in 75-79 years old patients and 80 years or older patients.

Methods
Of a total 140 patients aged 75 years or older who underwent surgery for NSCLC at our hospital between 2000 and 2012, there were classified into 75-79 years old and 80 years or older.

Results
Surgery of NSCLC in 75-79 years old and 80 or greater was 83 cases and 57 cases. 56(67.5%) patients in 75-79 years old and 36(63.2%) patients in 80 or greater were men. Adenocarcinoma constituted the most common pathologic subtype in two groups, 56(67.5%) cases in 75-79 years old, 42(73.7%) cases in 80 or greater. 55(66.3%) patients in 75-79 years old and 47(82.5%) patients in 80 or greater had stage IA/IB disease. 56(67.5%) patients in 75-79 years old and 32(56.1%) patients in 80 or greater underwent lobectomy. 19(22.9%) patients in 75-79 years old and 14(24.6%) in 80 or greater underwent segmentectomy. 7(8.4%) patients in 75-79 years old and 11(19.3%) in 80 or greater underwent wedge resection. Only one patient in 75-79 years old underwent pneumonectomy. Morbidity rate in 75-79 years old was 26.5%, and in 80 or greater was 22.8%. Atrial fibrillation was the most common postoperative complications in two groups. Mortality rate in 75-79 years old was 3.6%, and in 80 or greater was 3.5%. 30 patients in 75-79 years old and 12 patients in 80 or greater died within postoperative five years, 11 of 30 patients died for lung cancer and 19 patients died for the other illness. 15 of 19 patients were men, 3 of 15 were another organ cancer and 4 of 15 were respiratory failure. 6 of 12 patients died for lung cancer and other patients died for the other illness. Overall survival at 5 years in 75-79 years old was 58.5%, and in 80 or greater was 66.8%. Overall survival at 5 years in 75-79 years old male was 54.1%, and in female was 67.1%. Overall survival at 5 years in 80 or greater male was 68.6%, and in female was 64.3%. Overall survival at 5 years of stage IA in 75-79 years old was 80.8% and in 80 or greater was 78.7%.

Conclusion
Surgical resection of NSCLC in 75 years or older is permitted, especially in particular good indication for stage IA. However, long-term survival rate in 75-79 years old male was the poorest. The reason is that the ratio to die of the other illness was the highest. We thought that the decrease of the death by the other illness in 75-79 years old male is the most important.

Background
To explore the new technique of combined use of virtual puncture by simulator with methylene blue stain localization of the pulmonary peripheral nodules. And, evaluate the efficacy on focus localization and the clinical useful value of this method. To explore the new technique of combined use of virtual puncture by simulator with methylene blue stain localization of the pulmonary peripheral nodules. And, evaluate the efficacy on focus localization and the clinical useful value of this method.

Methods
During July 2011 to February 2013, total 80 pulmonary peripheral nodules of 75 patients were virtually punctured before operation. The methylene blue injected after anesthesia according to the identified skin point, angle and depth of preoperative virtual puncture. Then, marked lung tissue wedge resected under VATS. The samples were examined by pathologist during operation. The proper surgical style was decided to the pathological results. All of the data, including focus diameter, distance of colored spot and the lesions, localization time, interval time from injection finished to stain spot observed , successful localization and complication rate, were recorded and analyzed.

Results
75 of 80 pulmonary peripheral noudles were successfully localized. The accuracy rate of localization is 92.65%. Forty two nodules were primary lung cancer, those patients were accepted the lobectomy and systematic lymph node dissection. Twenty five nodules were benign lesion, one was metastatic tumor. Focus diameter was 10.15±3.57mm, distance of stain spot and the lesions was 5.54±2.83mm, localization time was 22.44±5.19min, interval time of injection finished to stain spot observed was 16.93±2.17min. All cases had no complication.

Conclusion
The new technique, combined preoperative virtual puncture by simulator with methylene blue stain localization of the pulmonary peripheral nodules, has high accuracy rate and safety. It is useful to localize the pulmonary peripheral nodules, especially for the early stage lung cancer lesion which hard to be seen or palpated during VATS.

Background
While VATS lobectomy yields enhanced recovery and fewer complications than open approaches, outcomes for thoracoscopic pneumonectomy are understood less well.

Methods
107 consecutive pneumonectomy cases performed at a comprehensive cancer center from 1/2002 to 12/2012 were studied retrospectively. 40 cases were open, while 50 Successful and 17 VATS conversions were grouped together for an intent-to-treat analysis.

Results
Preoperative characteristics were similar except for greater age (64±10 vs. 60±10, p=0.07), female sex (57 vs. 30% p=0.007) and preoperative comorbidities in the VATS group (Table 1). Right side was similar (46% vs. 45% open, p=0.9) as was disease extent (Early Stage 1&2, 72 vs. 61% open, p = 0.24). Neoadjuvant chemotherapy use also was similar (34 vs. 40% open). All VATS pneumonectomy pulmonary arteries were controlled safely and there were no intraoperative deaths from bleeding or other technical mishaps. Pursuing a VATS approach yielded a similar number of complications (3.1±2.6 vs. 3.0±2.6, p=0.8). Completion pneumonectomy (13%VATS/8% open) patients stayed longer (median 7.5 vs. 5 days, p=0.05) but had better survival (median not reached vs 27 months, p=0.05) largely because of more favorable stage distribution. A learning curve was evident as the rate of successful VATS pneumonectomy rose from 26% to 63% by the second half of the series (p<0.001). VATS patients started adjuvant chemotherapy an average of 39 days earlier. Excluded from long-term analyses were 7 pneumonectomies (3% VATS/13% open) for emergent indications like hemoptysis that led to 3 deaths. Stage-matched pneumonectomy cases had similar survival curves between the two groups. Multivariate logistic regression analyses found only age and pathologic stage as independent predictors of overall and disease-free survival. While the subset of patients who required conversion from VATS stayed longer (7 vs. 6 days, p=0.07), their survival curves were superimposable on open operations for all stages. In fact, achieving a successful VATS pneumonectomy demonstrated a trend toward improved survival compared to open/converted cases for early stage patients (median survival 80 vs. 27 months, p=0.07).

Procedure	VATS n=67	Open n =40	p
Predicted Post-resection Diffusing Capacity	38±10%	36±12%	0.6
Comorbidities (number)	3.2±1.7	2.3±1.3	0.001
Nodes retrieved	25±14	24±11	0.87
OR time (min -median)	289	225	0.001
EBL (ml -median)	400	325	0.84
ICU (days -median)	3	2	0.24
Hospital Stay (days -median)	5	6	0.2
Non-Emergent Case Hospital Death	8%	6%	0.7
Stage 1&2 Survival (mo -median)	26	26	0.74

Conclusion
Attempting VATS pneumonectomy appears to be a safe strategy that does not compromise short-term or long-term oncologic goals.

Background
The role of surgical resection remains controversial for pathologic N2 (pN2) non-small cell lung cancer (NSCLC). The objective of our study was to determine the prognostic and predictive factors for recurrence-free survival (RFS) in patients with completely resected pN2 NSCLC.

Methods
Between 1990 and 2009, 2439 patients with NSCLC underwent curative surgical resection at the Cancer Institute Hospital. Of these patients, 311 (12.8%) were diagnosed as having pN2 NSCLC. Surgical resection was performed in 79, 71, 70, and 91 patients in 1990–1994, 1995–1999, 2000–2004, and 2005–2009, respectively. Overall survival (OS) and RFS were calculated using the Kaplan-Meier method, and survival outcomes were assessed using the log-rank test. Univariate and multivariate Cox proportional hazards models were used to identify factors influencing RFS.

Results
The patient population comprised 199 male and 112 female patients, with ages ranging from 16 to 84 years (median, 61.4). Lobectomies or bilobectomies were conducted in 263 cases, and pneumonectomies were performed in 48 cases. For the entire cohort, 5-year OS and 5-year RFS rates were 46% and 24%, respectively. The median follow-up time was 44 months. OS and RFS rates were 34% and 23% in 1990–1994, 42% and 34% in 1995–1999, 45% and 20% in 2000–2004, and 59% and 20% in 2005–2009, respectively. The recent improvement in OS was remarkable, whereas the RFS rate did not improve. Multivariate analysis identified 4 independent predictors for poor RFS: multiple-zone mediastinal lymph node metastasis (hazard ratio [HR], 1.551; P = .002), ipsilateral intrapulmonary metastasis (HR, 1.038; P = .002), tumor size > 30 mm (HR, 1.007; P = .046), and clinical N1 or N2 stage (HR, 1.039; P = .041). Patients were grouped according to the number of risk factors for poor RFS. Patients with none of the identified risk factors had an RFS rate of 32%, those with 1–2 factors had an RFS rate of 25%, and those with 3–4 factors had an RFS rate of 7% (P < .001).Figure 1

Conclusion
In patients with surgically resected pN2 NSCLC, OS tended to improve in recent years, whereas RFS was fairly constant over the study period. The overall prognosis of patients with surgically resected pN2 NSCLC remains poor. However, prognosis is considerably better in those patients with fewer risk factors. Accordingly, surgical resection could be beneficial in select patients.

Background
With the increase in life expectancy, surgical treatment of non-small cell lung cancer (NSCLC) in elderly patients became more frequent. The aim of this study is to evaluate the risk factors for short and long-term outcomes after pulmonary resection in the elderly patients with stage I NSCLC.

Methods
From October 1994 to December 2011, the patients who were surgically treated with curative intent and pathologically diagnosed as stage I NSCLC were included. The patients were divided into two groups; elderly group (≥70 years) and younger group (<70 years). Comorbidity and surgical factors were analyzed for thirty-day mortality, hospital stay and overall survival in both groups.

Results

The Risk factors for short and long term outcome after pulmonary resection in elderly patients with stage I NSCLC

Risk factor	30-day mortality (Pearson’s Chi-Square Test)		Hospital stay (Linear Regression Model)		5-year survival (Cox Hazard Model)
	HR (95% CI)	p-value	HR (95% CI)	p-value	HR (95% CI)	p-value
DLCO less than 70%	12.9 (1.8-93.6)	0.001	5.0 (2.7-7.3)	< 0.001	3.4 (1.5-8.0)	0.004
FEV~1~/FVC less than 70%		NS	2.5 (0.8-4.3)	0.005		NS
Open thoracotomy		NS	3.6 (2.3-4.8)	< 0.001		NS
Pulmonary tuberculosis		NS		NS	3.3 (1.5-7.5)	0.004
Interstitial pulmonary fibrosis		NS		NS	5.0 (1.4-18.0)	0.015
Creatinine higher than 1.5mg/dL		NS		NS	5.7 (1.3-25.3)	0.022
Extensive resection		NS		NS	4.4 (1.4-14.0)	0.012

Total 1,340 patients were enrolled and 285 patients (21.3%) were classified as the elderly group and 1,055 patients (78.7%) as the younger group. The thirty-day mortality was 8 of 1,340 patients (0.6%) and all of the patients were elderly. The only independent factor for thirty-day mortality in elderly group was diffusing capacity for carbon monoxide (DLCO) less than 70% of predicted (hazard ratio, 12.9; p = 0.001). The elderly group had significantly longer hospital stay (11.2 12.2 vs. 8.0 6.7 days, p < 0.001). Open thoracotomy (p < 0.001), DLCO less than 70% of predicted (p < 0.001) and percentage of one second forced expiratory volume over forced vital capacity (FEV~1~/FVC) less than 70% (p = 0.005) were significantly associated with longer hospital stay. In-hospital complication rate in elderly patients was also significantly higher (47.7 vs. 26.9%, p < 0.001). 5-year overall survival rates were 91.1% in the younger group and 66.2% in the elderly group. In the elderly group, previous history of tuberculosis (p = 0.004) and interstitial pulmonary fibrosis (IPF; p = 0.015), DLCO less than 70% of predicted (p = 0.004), preoperative creatinine higher than 1.5 mg/dL (p = 0.022), and more extensive pulmonary resection (p = 0.012) were the independent risk factors for overall survival. On the other hand, previous history of IPF (p < 0.001) and pathologic stage IB over IA (p < 0.001) were the independent risk factors in the younger group.

Conclusion
Pulmonary resection for the elderly patient requires caution, particularly in case of low diffusing capacity (DLCO < 70%) or airflow limitation (FEV1/FVC < 70%).

Background
Chylothorax complicating pulmonary resection (CCPR) is an infrequent but well-known complication in lung cancer surgery. Previous published studies on this topic were limited and standard thoracotomy for pulmonary resection (STPR) was the surgical approach. Video-assisted thoracoscopic pulmonary resection (VATPR) has become prevalent in the lung cancer surgery nowadays. The purpose of this study is to analyze the clinical data of CCPR after VATPR and evaluate their outcome after treatment.

Methods
Between January 2010 and May 2013, we retrospectively reviewed 728 primary non-small cell lung cancer patients who undergone VATPR and mediastinal lymph node dissection (MLND) in our institute. CCPR were noted in 18 patients (2.47%) who constitute the subjects of our study. We initially treated these patients conservatively with oral intake cessation and parenteral nutrition. If conservative treatment failed, reoperation with video-assisted thoracic surgery (VATS) for thoracic duct ligation would be performed. Daily pleural drainage amount, timing of surgical intervention, and treatment responses were recorded and investigated. The data collected were compared to other studies in which STPR was the main operative method.

Results
Among the 18 CCPR cases, all of them were adenocarcinoma on the right side of lung. Thirteen of patients received conservative treatment and 5 patients received reoperation for CCPR. All of them were successfully treated with cessation of CCPR without mortality. The average pleural drainage amounts per day in conservative treatment group and reoperation group were 206 ml and 433 ml. The presented study suggests that CCPR with pleural drainage amount less than 400 ml in the first postoperative day will subsided after conservative treatment. CCPR with pleural drainage amount more than 400 ml in the first or second postoperative day can also resolve if drainage amount below 400 ml was seen in the postoperative day 4 and thereafter. Reoperations would be undertaken for CCPR in cases with increasing amount of pleural drainage in the postoperative 4 after conservative treatment.

Conclusion
chylothorax, video-assisted thoracoscopic pulmonary resectionVATPR did not incur more CCPR than did STPR in NSCLC surgery. The average pleural drainage amount of CCPR in reoperation cases was less in our study than that in other studies. The timing of surgical intervention for CCPR following VATPR can be earlier if pleural drainage didn’t show trend of decrease after conservative treatment. Figure 1

Background
Although surgical resection is the standard treatment of choice for early stage non-mall-cell lung cancer (NSCLC), not a small percentage of patients recur even after complete resection, and post-recurrence survival (PRS) is dismal. Since epidermal growth factor receptor (EGFR) tyrosine kinase inhibitor (TKI) was approved for patients with advanced or recurrent NSCLC in 2002 in Japan, the number of long-term survivors after recurrence seems to be increasing. There have been few reports on post-recurrence survival in the era of EGFR-TKI. In the present study, we analyzed PRS in patients with completely resected NSCLC after EGFR-TKI was introduced in clinical practice.

Methods
From 1992 to 2010, 3058 NSCLC patients underwent complete resection in our hospital. Among them, 938 (31%) patients recurred. Of them, 503 patients, who recurred in 2002 or later and received anticancerous treatment, were the subject of this analysis. We retrospectively analyzed their clinicopathological characteristics, PRS, and identified favorable prognostic factors.

Results
The median age at recurrence was 68 years (range: 23-91), and 332 (58%) of them were men. There were 347 (69%) adenocarcinoma patients and 101 (20%) squamous cell carcinoma patients. The pathological stages of the primary cancers were I in 183 patients (36%), II in 120 (24%), and III in 200 (40%), respectively. Fifty-six patients (11%) underwent reoperation for recurrent lesions. Of the 216 patients whose EGFR mutation was examined, mutation was positive in 97 patients (45%). The median period of follow-up after recurrence was 49 months (range: 7-124). The overall 3- and 5-year PRS were 27% and 16%, respectively and their median PRS length was 19 months. The multivariate analysis revealed that adenocarcinoma, negative pleural invasion, locoregional recurrence, surgical resection of recurrent lesion, positive EGFR mutation were independent favorable prognostic factors. The 3-year PRS rate and median PRS length of the patients undergoing surgery and those with EGFR mutation were 60% (55 month) and 46% (35 month), respectively.

Conclusion
The PRS was still poor, but longer survival can be expected in selected patients, who have resectable recurrence or EGFR mutation.

Background
The value of robotics in surgical treatment of lung cancer is not well-defined. Our goal was to examine the surgical results of robotic-assisted pulmonary resections in a high risk profile veteran population. .

Methods
A retrospective analysis of a single VA facility’s robotic thoracic surgical experience from January 2011 to May 2013 was performed. A total of 70 consecutive patients had undergone robotic pulmonary resections, by a single surgeon, for treatment of non-small cell lung cancer (NSCLC). All preoperative, intra- and postoperative data including length of stay (LOS) and readmission rates were collected.

Results
60 lobar and 10 sublobar (wedge) pulmonary resections plus mediastinal/hilar lymph node staging had been performed. Mean number of lymph node stations sampled were 3.5 (range 2-7). Mean age was 68 (40-86). 33 (47%) patients were active smokers. 42 (60%) patients had hypertension, 34 (48%) had COPD, 15 (21%) had BMI >30, 14 (20%) had DM, 13 (19%) had documented coronary artery disease, 11 (16%) had history of alcohol abuse, 7 (11%) had renal insufficiency defined as creatinine > 1.3, and 3 (4%) had received induction therapy. Average preoperative FEV1 and DLCO were 76% and 68% of predicted, respectively. Stage distribution is shown in Table 1. Intra- and postoperative data are summarized in Table 2. Thirty day mortality was 1.4% (1). 20 patients sustained at least one complication (28.5% morbidity). Mean LOS for the entire cohort was 7 days; mean LOS for those 57 patients having undergone completely robotic resection was 6 days. Prolonged air leak was the most prevalent reason for an extended LOS. Table 1: Clinical and pathologic stage distribution.

Total N=70	Clinical Stage : N(%)	Pathologic Stage: N(%)
Ia	49 (70%)	34 (48.6%)
Ib	6 (8.6%)	16 (22.9%)
IIa	8 (11.4%)	6 (8.6%)
IIb	3 (4.3%)	4 (5.7%)
IIIa	3 (4.3%)	9 (12.8%)
IIIb	0	0
IVa	0	0
IVb	1 (1.4%)	1 (1.4%)

Table 2: Intraoperative and postoperative outcomes

	N	%
Intraoperative data
OR extubation	66	94
Conversion to open	13	18
Blood transfusion	1	1.4
Death	0	0
Average EBL	83
Postoperative data
Atrial fibrillation	10	14
Bronchoscopy	9	12.8
Prolonged air leak (>7days)	8	11.4
Blood transfusion	7	10
Pneumonia	5	7.1
Respiratory failure (reintubation)	2	2.8
Reoperation within 30 days	2	2.8
Readmission within 30 days	2	2.8
Pulmonary embolism	1	1.4
DVT	1	1.4
Average Chest Tube Days	4.1

EBL: estimated blood loss; DVT: deep vein thrombosis

Conclusion
This is the first report on feasibility and outcome of robotic thoracic surgery in a high risk veteran population. Our data suggest that robotic-assisted pulmonary resection for NSCLC can be performed with acceptable morbidity and mortality in this cohort.

Background
Postoperative cardiopulmonary complications represent a major source of morbidity and mortality in the acute phase after lung cancer surgery. However, clinical impacts of postoperative cardiopulmonary complications on long-term outcome were not well studied. The objective of this study was to investigate the effects of postoperative cardiopulmonary complications in the acute phase on cardiovascular and respiratory events in the chronic phase after lung cancer surgery.

Methods
From a prospective single-institution database of 496 consecutive patients who underwent a lung cancer surgery between August 2008 and December 2011, we retrospective analysed medical charts of all patients with curative surgery for non-small cell lung cancer. Patients with limited surgery (n=32) and postoperative mortality (n=4) were excluded from the analysis. The remaining 460 patients were analysed for the incidence of cardiovascular and respiratory events in the chronic phase after surgery. Results were compared between the patients with and without postoperative cardiopulmonary complications in the acute phase.

Results
Postoperative cardiopulmonary complications were identified in 90 (20%) patients; these patients included more advanced patients with pathological stage IB-III compared to those without cardiopulmonary complications (60% vs. 48%; p < 0.05). There were significantly higher cardiovascular and respiratory events in those with postoperative cardiopulmonary complications than those without (Table 1, 14% vs. 4%; p < 0.001). Also, there was significantly higher incidence of cancer recurrence in those with postoperative cardiopulmonary complications than those without (27% vs. 20%; p < 0.05).

Table 1. Cardiovascular and respiratory events in the chronic phase for the patients with and without postoperative cardiopulmonary complications in the acute phase after lung cancer surgery.

Variables	With cardiopulmonary complications (N=90)	Without cardiopulmonary complications (N=370)	P Value
All events	13 (14%)	14 (4%)	<0.001
Cardiovascular events	7 (8%)	7 (2%)
Acute heart failure	3	1
Arrhythmias	1	2
Coronary artery disease	0	3
Peripheral vascular disease	2	0
Cerebrovascular disease	1	1
Respiratory events	6 (7%)	7 (2%)
Pneumonia	5	6
Acute respiratory distress syndrome	0	1
Chronic respiratory failure	1	0

Conclusion
Postoperative cardiopulmonary complications in the acute phase were associated with the incidence of cardiovascular and respiratory events in the chronic phase after lung cancer surgery. It is important for those with postoperative cardiopulmonary complications to be careful about not only cancer recurrence but also cardiovascular and respiratory events in the long-term period.

Background
The small pulmonary lesions, the localization of which cannot be confirmed by the sight, are often removed surgically with help of markers. But it is the most certain to confirm local existence of tumors with help of palpation and remove them surgically. I examine the metastatic pulmonary tumor operation cases in our hospital and identify the usefulness of hand assisted thoracoscopic surgery (HATS).

Methods
Thirty nine patients underwent surgery from July 2007 to November 2012 (colorectal cancer 24 cases, soft tissue tumor 5 cases, gynecology territory malignancy 3 cases, others 7 cases). Tumors which existed in the hilum of lung, multiple tumors in the same lobe or segment of lung and tumors which were diagnosed as primary lung cancers before surgery were removed by lobectomy or segmentectomy. The localization of small tumors which cannot be seen by sight were tried to confirm by palpation in all cases. Ten lobectomies, 2 segmentectomies and 27 partial resections were performed. Standard thoracotomy was done in 5 cases, video-assisted thoracic surgeries were done in 30 cases and HATS was done in 4 cases. In HATS cases, a hand of surgeon was inserted into the thoracic cavities of patient through a subxiphoid skin incision.

Results
Localized confirmation was possible in all cases using an ocular inspection and palpation without markers. Fifty seven tumors were excised in 39 cases (an average of 1.46) and 9 other lesions (including intrapulmonary lymph nodes and granuloma) were excised simultaneously. The maximum tumor diameters are 3-75mm (an average of 15.7mm).

Conclusion
New lesions which cannot be discovered by preoperative CT were confirmed by HATS. Local existence confirmation by the palpation with a finger from a small open chest wound and palpation by HATS is very important to remove pulmonary lesions without fail.

Background
Previous papers have demonstrated that pulmonary lobectomy on octogenarians is safe and feasible. However, there is little data characterizing the survival or the severity of complications in these frail patients after lobectomy. Therefore we reviewed our experience with patients aged eighty and above undergoing lobectomy.

Methods
We performed a retrospective review of consecutive patients aged 80 or above that underwent lobectomy between 2004 and 2012. Chart reviews were performed evaluating comorbidities, clinical stage, perioperative and postoperative course, time to recurrence, and date and cause of death. All complications were graded per the Seely Thoracic Surgery morbidity and mortality classification schema.

Results
45 patients (mean 82.2 years) underwent lobectomy. PFTs averaged 86% predicted for FEV1. Pathologic stage IA comprised 26% (10 of 39) of our patients; IB 33%(13), IIA 8% (3), IIB 8% (3), IIIA 18% (7), IIIB 3% (1), and IV 5% (2). Of the 45 patients, 28 had complications (60%), but only 18% (8 of 45) were significantly morbid to the patient (grade IIIB or above). Perioperative mortality was 2% (1 of 45). The most common complication was arrhythmia. Median LOS was 6 days for thoracotomy patients, 5.5 days for VATS patients, and 4.5 days for robot. 78% were discharged home, and 16% were readmitted to hospital within thirty days. Six patients had recurrent disease that occurred at an average of 768 days. 50% of our patients are still alive. Only three of the seven known causes of death were from metastatic disease. Five year actuarial survival was 52.3%. Mean survival was 53 months, and median survival was 72 months.

Complications Separated by Grade

Grade	Definition	Example	Incidence
Grade I	Clinically Insignificant	Asymptomatic vocal cord paralysis, urinary retention	4% (2 of 45)
Grade II	Medical Therapy Only	AFib, esophagitis, new home O2	26.7% (12 of 45)
Grade IIIA	Interventions not requiring anesthesia	Percutaneous pleural catheters	13.3% (6 of 45)
Grade IIIB	Interventions requiring anesthesia	Return to OR	8.8% (4 of 45)
Grade IV	Critical illness, reintubation, organ failure	MI, PNA, chyle leak	6.7% (3 of 45)
Grade V	Death		2.2% (1 of 45)

Conclusion
Lobectomy on carefully selected octogenarians can be done safely regardless of approach with a low mortality. 60% experienced a complication but when graded in a validated system only 18% were considered significant.

Background
Stage for stage, women have superior survival compared to men with non-small cell lung cancer (NSCLC). This includes women undergoing resection, but little is reported regarding sex-based differences in operative morbidity, mortality, and length of hospital stay (LOS) following NSCLC resection.

Methods
We retrospectively evaluated all patients undergoing primary NSCLC resections at our institution from 01/07-12/08 to evaluate the prognostic significance of female sex on morbidity, mortality and LOS.

Results
A total 259 patients were included, 155 females and 104 males. The majority of resections were lobectomy (162/259) and by VATS (161/259). Women were significantly younger (70.6 vs. 74.0 years, p=0.006), reported fewer pack years smoked (30 vs. 44 years, p=0.001), had more adenocarcinoma histology (86% vs. 63%, p<0.001) and had higher percent predicated forced expiratory volume in 1 second (FEV1%)(84 vs. 77, p=0.014) than men. No other significant differences in pre-treatment demographics, tumor characteristics or surgical procedures were noted. A significantly reduced rate of post operative atrial fibrillation (5% vs. 13%, p=0.031), prolonged air leak (8% vs 16%, p=0.044), any complication (28% vs. 44%, p=0.007), and LOS (5.0 vs. 5.98 days, p=0.031) was noted for women. In multivariate analysis, which included age, pack years and FEV1%, male sex remained a significant predictor for the occurrence of a post-operative complication (p= 0.007). Figure 1

Conclusion
Sex-based differences in the morbidity associated with resection for NSCLC were noted, with a significant improvement in several important short –term outcomes for women compared to men. As the percent of female NSCLC patients increase, these data carry significant implications in procedural risk stratification and in comparison of modern surgical trials to historic controls.

Background
General anesthesia with single-lung ventilation is considered mandatory for thoracoscopic lobectomy for non-small cell lung cancer (NSCLC). Nonintubated thoracoscopic lobectomy has rarely been reported previously. The objective of this study was to evaluate the feasibility and safety of thoracoscopic lobectomy without endotracheal intubation.

Methods
From August 2009 through March 2013, 196 patients with clinical stage I or II NSCLC were treated by non-intubated thoracoscopic lobectomy using epidural anesthesia, intrathoracic vagal blockade, and sedation at National Taiwan University Hospital, Taiwan, and The First Affiliated Hospital of Guangzhou Medical College, China.

Results
The mean age of the patients were 59 years and 128 (65.3%) were female. Collapse of the operative lung and inhibition of coughing were satisfactory in the non-intubated patients, induced by spontaneous breathing and vagal blockade. Sixteen patients (8.2%) required conversion to intubated-single lung ventilation because of significant mediastinal movement, persistent hypoxemia, dense pleural adhesions, and bleeding. Two patients were converted to standard thoracotomy because of bleeding. The mean postoperative chest tube drainage and postoperative hospital stay were 3.6 days and 6.8 days, respectively. Postoperative complications were noted in 23 (11.7%) patients, including prolonged air leaks, arrhythmia, pneumonia, and bleeding. No operation mortalities were noted.

Conclusion
Non-intubated thoracoscopic lobectomy is technically feasible and safe. It can be a valid alternative of single-lung ventilated thoracoscopic lobectomy in managing selected patients with early-stage NSCLC.

Background
Background: Metastatic lesions are the most common malignancy of the lungs. In the past, pulmonary metastatectomy was reserved for cases of solitary or oligo-metastasis. Over the past decade, however, indications for surgical treatment of pulmonary metastasis have broadened for many cancers. Sarcomas have a predilection for spread to the lungs, often in the absence of metastasis to other organs. We examined our experience with an increasingly aggressive approach to pulmonary metastasis, to help define evolving parameters and expectations for clinical outcomes, exploring the perceived differences in the clinical patterns between sarcoma and other cancers.

Methods
Methods: We identified 262 patients who underwent a total of 361 R0 pulmonary metastatectomies, 336 of which were performed at UCSF Medical Center between 1996 and 2009. Sarcoma was the primary tumor in 118 patients undergoing 180 of these operations. Survival estimates were based on Kaplan-Meier analysis and compared using either a log-rank or Wilcoxon test. Predictors included surgical procedure; number/size of lesions; repeat resection; intervals to metastasis (DFI) and to recurrent metastasis; chemotherapy; cancer type; distribution of pulmonary and extra-pulmonary metastasis; patient age/sex. These predictors were compared using univariate and multivariate Cox proportional hazards modeling; multiple-predictor modeling started with a set of predictors based on historical/clinical significance, and stepwise forward selection determined which additional predictors were included until all p-values in the model were less than 0.1.

Results
Results: Despite an increasingly aggressive surgical approach, reflected in an increase in number of lesions, the percentage of patients with > 8 lesions, the number of patients with lesions < 1 cm, and in a decrease in DFI, the overall 5-year survival was 48% (median survival 4.7 years, 95%CI 3.5-5.5), and did not differ between the early and late periods of the study. Sarcoma patients, however, tended to be significantly younger (46 ± 16 yrs vs. 59 ± 14, P<0.001), and to have more lesions (4.0 ± 4.3 vs. 2.3 ± 2.3, P<0.001), a shorter DFI (2.5 ± 3.3 yrs vs. 3.6 ± 3.9, P = 0.004), more diffuse pulmonary involvement (43% bilateral disease vs. 29%, P = 0.02), and more frequent recurrence rate (80% vs. 51%, <0.001) than the non-sarcoma patients. Whereas lesion size (HR 1.2, P=0.004), age (HR 1.4, P<0.001), DFI (HR 2.1, P=0.008), and extra-pulmonary disease (HR 1.9, P=0.04) were all independent predictors of survival for non-sarcomas, only metastasis synchronous to the primary tumor (HR 2.7, P=0.007) and a need for anatomic resection (HR 2.5, P=0.006) independently predicted a higher mortality among sarcomas. Furthermore, a need for repeat resection did not impact the survival of sarcoma patients as long as complete resection remained feasible, whereas the 5-year survival of patients even with resectable recurrent non-sarcoma metastases dropped from 76% to 43% (P = 0.003).

Conclusion
Conclusion: Encouraging long-term survival can be achieved even with an increasingly aggressive surgical approach to pulmonary metastasis. Although sarcoma patients tend to present with rapidly progressive and extensive pulmonary disease, a tendency toward confinement of metastasis to the lungs may justify an even more aggressive surgical strategy for these patients.

Background
Recent advances in imaging modalities have enabled detection of small-sized lung tumors at earlier stages with resultant dramatic changes in therapeutic strategies. However, if preoperative pathological diagnosis is not possible, video-assisted thoracic surgery (VATS) for therapeutic resection as well as diagnostic excisional biopsies may be indicated. Small-sized lung tumors, such as bronchioloalveolar carcinoma, are difficult to localize during surgery by visualizing or palpating the lung surface because the lung is a soft and deformable organ and contains air. Accurate intraoperative localization of the tumor is critical to surgeons. We usually perform preoperative lipiodol marking for small-sized tumors, particularly those located deep in the lung, to create a “visible target” from an invisible and impalpable tumor. This visualization technique enables resection of marked lesions under X-ray fluoroscopy.

Methods
From May 2006 through March 2013, we performed preoperative lipiodol marking for 356 lesions in 215 cases in which unconfirmed lung tumors were less than 10 mm in diameter, located deep to the visceral pleura, or of ground-glass opacity. One to five markings with lipiodol were performed in each case. The mean diameter of the lesions was 7.7 ± 5.1 mm (2–33 mm), and they were located 10.1 ± 9.5 mm (0–54 mm) below the surface of the visceral pleura. CT-fluoroscopy guidance was used to inject 0.1–0.5 mL lipiodol in the vicinity of the tumor before surgery. During VATS, X-ray fluoroscopy was used to confirm lesion location and to guide resection of the lipiodol-marked lesion.

Results
The average duration of the marking procedure was 18.4 minutes per lesion. Regarding complications, pneumothorax occurred in 40 cases (18.6%), but there were no cases of air embolization and no histological modifications in or around lipiodol markings. Of the 356 lesions, 354 (99.4%) were detectable and safely resected. Pathological examinations revealed lung cancer in 54 lesions, atypical adenomatous hyperplasia in 8, metastatic lung tumor in 165, organized pneumonia in 115 and other benign lesions in 12.

Conclusion
Lipiodol marking with CT-fluoroscopy guidance before VATS is a useful technique for small and impalpable lung tumors.

Background
Knowledge about symptoms is an important outcome to evaluate following lung cancer surgery because patients want information about the usual course of recovery of their physical and mental health. Patients need information at hospital discharge about when they need to contact their clinician if symptoms persist. Because to our knowledge only three studies have evaluated the occurrence of symptoms in patients prior to and following lung cancer surgery, the purpose of this study was to evaluate for changes in symptom occurrence ratings and severity scores from the preoperative period to 1 month after surgery using a multidimensional symptom assessment scale (i.e., Memorial Symptom Assessment Scale (MSAS)).

Methods
Patients were recruited from three university hospitals in Norway. They completed a number of self-report questionnaires prior to and again at 1 month following surgery. The questionnaires provided information on demographic and clinical characteristics and symptoms. Patients’ medical records were reviewed for disease and treatment information. Descriptive statistics were used to present demographic and clinical characteristics. McNemar's test was used to evaluate for changes over time in symptom occurrence rates. Paired t-tests were done to evaluate for changes in severity scores for patients who reported severity scores either preoperatively, postoperatively, or on both occasions. Because the severity ratings were skewed, 1000 bootstrapped samples were taken for each analysis to provide unbiased estimates.

Results
The sample consisted of 129 (57%) men and 99 (43%) women who had a mean age of 65.8 years (SD=8.5, range 30 to 87). The total number of symptoms increased significantly from the preoperative ( =9.4, SD=7.2) to the postoperative ( =13.1, SD=6.8, p˂.001) assessment. Of the 11 symptoms that occurred in more than 50% of the patients at 1 month, 8 of them increased significantly from the preoperative to the postoperative period. Eight of the symptoms increased in both their occurrence rates and severity scores (i.e., lack of energy, pain, feeling bloated, lack of appetite, shortness of breath, feeling drowsy, dry mouth, sweats). Only one symptom had a significant reduction in its severity score, namely cough. Four symptoms were experienced by more than 80% of the patients at the 1 month assessment, namely: shortness of breath (85.5 %), lack of energy (83.8%), pain (83.8%), and feeling drowsy (82.5%).

Conclusion
Findings from this study suggest that patients experience a high number of symptoms after surgery. These findings can be used to educate patients about the normal course of postoperative recovery. In addition, clinicians need to assess for these symptoms and develop effective interventions to improve symptom management for this vulnerable group of patients.

Background
Small pure ground glass nodules (GGNs) have sometimes been detected by high resolution computed tomography (CT). Pathologically, most of these pure GGNs are bronchioloalveolar carcinoma (BAC) or atypical adenomatous hyperplasia (AAH). In published papers, the thoracoscopic resection of the pure GGNs was reported to be difficult, because they cannot usually be palpated. Therefore, various marking techniques, such as those using a hook wire, colored collagen, barium, lipiodol and so on, have been described for the localization of pure GGNs. However, these techniques are associated with the development of several complications, such as pneumothorax, hemorrhage, and serious air embolism. Moreover, they requires a lot of time. In this study, we evaluated the localization of small pure GGNs in thoracoscopic surgery using the endo-finger technique.

Methods
Patients with peripheral pure GGNs that were 20 mm and less in diameter who planned to undergo resection in our institute were candidates for this study. Preoperatively, no marking technique was performed. Thoracoscopy was performed in the lateral position under single lung anesthesia. Thoracoports were placed near the GGNs based on the CT findings. One finger was inserted through the port into the pleural cavity to palpate the lung to localize the GGNs (the endo-finger technique). After the GGNs were detected, they were resected by endostaplers with adequate safety margin.

Results
Since January 2005, twenty patients with thirty-four GGNs were enrolled in this study. The size of the GGNs was 5 mm or less in eight lesions, 6 – 10 mm in 14 lesions, 11 – 15 mm in nine lesions and 16 – 20 mm in three lesions. The depth of the lesions from the visceral pleura ranged from 0 – 14 mm. The main reasons for the resection were a need for another ipsilateral simultaneous operation in six cases, the patients’ requests in five cases and enlargement of the GGNs during the follow-up period in three cases. All but one GGN could be detected using the endo-finger technique with two or three thoracoports, and were resected. Some of the GGNs adjacent to the visceral pleura were visualized by thoracoscopy as color changes in the visceral pleura. There was one conversion to thoracotomy in one patient who had a severe pleural adhesion due to a previous ipsilateral lobectomy of the lung. No complications occurred in association with this procedure. The pathological diagnoses of the GGNs were BAC in 23 nodules, AAH in nine, hyperplasia of the alveolar epithelium in one and an inflammatory lesion in one. The surgical margins of all of the resected specimens were pathologically negative.

Conclusion
The endo-finger technique is safe and useful for the localization and resection of peripheral GGNs during thoracoscopic surgery. We suggest that the preoperative marking to detect GGNs can be replaced by the endo-finger technique in some cases.

Background
The American College of Chest Physicians(ACCP) has published the revised guidelines on preoperative physiologic assessment of lung cancer surgery. However, the relevance of the guidelines has not yet evaluated enough, partially because cardiopulmonary exercise testing recommended for achieving to the patients with poor predicted postoperative lung function is available in only limited institutions.

Methods
Initial analysis was conducted to examine the relationship between the maximum oxygen consumption (VdotO2 max) measured from cycling ergometer and the minimum saturation level (SpO2min) or maximum desaturation level (ΔSpO2) during 6 minute walking test (6MWT). In the next analysis, we modified the risk assessment flow chart in ACCP guidelines using SpO2min and ΔSpO2 instead of VdotO2 max. The patients with lung cancer who underwent lobectomy or more resection were retrospectively assessed using the modified flow chart.

Results
Using the data obtained from 39 patients with cardiac diseases and five healthy volunteers, we analyzed the correlation between VdotO2 max and variables of 6MWT. VdotO2max was significantly correlated to SpO2min and ΔSpO2 (r=0.469,p=0.001 and r=-0.458, p=0.002,respectively). Receiver operating characteristic (ROC) analysis revealed that both SpO2min and ΔSpO2 was available for the detection of the patients with VdotO2max of <15 mL/kg/min which is the borderline value between the average and increased risk group in ACCP guidelines (AUC 0.802, p=0.001 and AUC 0.802, p=0.001, respectively). When the cut-off value was set as SpO2min <91%, sensitivity and specificity was 85.2％ and 70.6％,respectively. When the cut-off value was set as ΔSpO2>4%, sensitivity and specificity was 74.1％and 76.5％,respectively. We introduced a new decision criteria of SpO2min <91% or ΔSpO2>4% instead of VdotO2max of <15 mL/kg/min in the final step of the flow chart in ACCP guidelines. A total of 965 patients were evaluated according to the modified flow chart and 883 (91.5%) patients were classified to the average risk group and 31 (3.2%) patients were classified to the increased risk group. Fifty-one (5.3%) patients were excluded because of the lack of 6MWT data. In regard to surgical outcome, there was a significant difference between the average risk group and the increased risk group in introduction rate of home oxygen therapy (0.7% vs. 25.8%,p<0.001) and cardiopulmonary oriented 90-day mortality rate (0.8% vs. 9.7%,p<0.001).

Conclusion
Figure 1In the clinical practice, the modified ACCP risk flow chart may be easier to apply and useful for the perioperative risk assessment of the patients with lung cancer being considered for standard resection.

Background
ICAN study (NCT01106781) investigated EGFR mutation status, clinical outcomes and recurrent risk factors in Chinese lung adenocarcinoma (ADC) patients after complete resection. Here we report analysis results for the profile of surgery, adjuvant therapy and 2-year disease free survival (DFS) rate in the real-world practice.

Methods
Patients were aged ≥18 years, with histological diagnosed lung ADC, and received surgical complete resection. Tumor sample EGFR mutation status was determined according to clinical routine practice. All eligible patients would be followed up to collect the clinical information and the outcomes till 3 years after operation.

Results
Of 571 patients from 26 sites, 315 (55.2%) patients were EGFR mutation positive. The most common mutations were exon19 deletion and L858R mutation found in 140 (24.52%) and 132 (22.59%) patients respectively. There were 50.79% of patients who received adjuvant therapy, in which 45.37% received chemotherapy, 4.55% received radiotherapy and 1.93% received TKI therapy, respectively. The 2-year DFS rate was 68.83%. There was statistically significant difference of 2-year DFS among the patients with different postoperative pathologic stage (P <0.0001). There was no statistically significant difference of 2-year DFS among the patients with age, gender, smoking history and EGFR mutation status. Operation method and adjuvant therapy correlated significantly with 2-year DFS, but was not significant when adjusted for postoperative pathologic stage.

Conclusion
The overall EGFR mutation positive rate in operable Chinese ADC was 55.2%. 2-year DFS rate was 68.83%. Postoperative pathologic stage had a statistically significant association with 2-year DFS, while age, gender, smoking history and EGFR mutation status didn’t show statistically significant association.

Background
The purpose of the study was to evaluate complications, risk factors and outcome of pneumonectomy for the surgical treatment of lung cancer patients.

Methods
We reviewed retrospectively the hospital records of 64 consecutive patients who underwent a pneumonectomy between January 2007 and December 2012. The majority were male (87, 5%), with a mean age of 63, 7 years (r: 38-80). Comorbidities and complications was assessed. Mean follow up was 22, 8 months.

Results
The number of left pneumonectomies was slightly higher (62.5%) than the right ones. Most of the patients were diagnosed of squamous cell carcinoma (57, 1%). 24 patients (37.5%) had N2 disease. Twenty-eight patients (44.4%) were treated with neoadjuvant chemotherapy and 6 (9.4%) with concomitant radiotherapy. Mean VEF1 was 74% (r: 38-125). Diabetes, hypertension and smoking were not significant risk factors. Four patients (6.3%) died within 30 days of surgery. Atrial fibrillation appeared in 9 patients (14%), major pulmonary complications in 4 (6.3%). Late bronchopleural fistula appeared in 4 patients (6, 3%) and the presence of brochopleural fistula was not related to an increase in mortality (p=0. 78). Mean overall survival was 38 months (1-year 72%, 3-years 50% and 5-years 24.7%). Mean survival was higher for a right pneumonectomy (31 vs 38 months). Survival according to size was T1: 33.3 months, T2: 46.3 months, T3: 28.1 months and T4 15.7 months. There was no difference between TNM stage, histological type, neoadjuvant or adjuvant treatment in survival. Survival was lower in patients who underwent chest wall resection (37.4 vs 9.9 months, p=0.035). Body mass index, diabetes, hypertension and arrhythmia didn´t show differences in overall survival.

Conclusion
The side of the pneumonectomy was not related to mortality. Bronchopleural fistula, hypertension and diabetes and arrhythmia were not related to an increase in mortality.

Background
Whether robot-assisted esophagectomy is a technically feasible and oncologically reliable operation for esophageal cancer has not been proven. This study aimed to evaluate short-term and long-term outcomes of robot-assisted esophagectomy.

Methods
Robot-assisted esophagectomy was performed in prone position for cervical anastomosis or lateral position for intrathoracic anastomosis. Thoracic procedures were performed by totally robotic technique and abdominal procedures were performed by robot or laparotomy. Two field lymph node dissection was performed in all patients and dissection along both recurrent laryngeal nerves was performed in the patients with T1b or more stages. Retrospective review on short-term and long-term outcomes for robot-assisted esophagectomy was performed.

Results
Robot-assisted esophagectomy was performed in 46 patients between 2008 and 2013, which was 16% of total esophagectomy cases during the same period. There were 43 men and 3 women and mean age was 63.9 ± 8.2 years. Preoperative clinical stages were IA in 19 patients (41%), IB in 8 (17%), IIA in 6 (13%), IIB in 9 (20%), and IIIA in 4 (9%). Neoadjuvant chemoradiation was performed in 5 patients (11%). Abdominal procedures were performed by robot in 29 patients (63%) and by laparotomy in 16 (35%). R0 resection was accomplished in 45 patients (98%) and mean operation time including robot docking time were 512 ± 104 minutes. Total 2-field lymph node dissection along bilateral recurrent laryngeal nerve was performed in 32 patients (70%) and mean number of dissected lymph nodes were 29.1 ± 14.1. Cell types of esophageal cancer were squamous cell carcinoma in 45 patients (98%) and melanoma in 1 patient (2%). Pathologic stages were IA in 9 patients (20%), IB in 19 (41%), IIA in 2 (4%), IIB in 11 (24%), IIIA in 4 (9%), and IIIB in 1 (2%). There were one 30-day mortality (2%) and postoperative complication occurred in 15 patients (33%); respiratory complication in 5 patients (11%), anastomosis site leakage in 5 (11%), and vocal cord palsy requiring treatment in 3 (7%). Overall 5-year survival was 88% and 5-year freedom from recurrence was 73%. Locations of recurrence were regional in 4 patients (9%), distant in 4 (9%), and there was no local recurrence.

Conclusion
Robot-assisted esophagectomy was technically feasible and oncologically reliable surgery in this study. Further studies based on large series of data are necessary to prove advantages of robot-assisted esophagectomy.

Background
Background. Subcentimeter lung cancers are still rare and their pathobiological behavior and management have not yet been fully clarified. In this retrospective study, we investigated the clinicopathological characteristics of patients with subcentimeter lung cancers.

Methods
Methods. From among 7,463 patients with primary lung cancers that were surgically resected at the National Cancer Center Hospital, Tokyo, from 1993 through 2011, 291 (4%) patients with peripheral lung cancers of 1.0 cm or less in diameter were studied retrospectively with regard to their clinicopathological characteristics including prognosis. Of these 291 patients, 141 (48%) were male and 150 (52%) were female, and they had a mean age of 62.0 years. According to the proportion of consolidation component within the tumor in preoperative imaging on high-resolution computed tomography (HRCT), the tumors were classified into 4 types; Type 1 (n = 50): non-solid ground-glass opacity (GGO) lesion, Type 2 (n = 89): part-solid GGO lesion including 50% or more GGO within the lesion, Type 3 (n = 62): part-solid GGO lesion including less than 50% GGO within the lesion, and Type 4 (n = 90): solid lesion with no GGO component.

Results
Results. Patients with Type 4 included significantly greater percentages of males and smokers than those with the other types. Pleural invasion and vascular/lymphatic permeation were significantly more frequent in Type 4 than in the other types. While none of the patients with Type 1 to 3 had lymph node metastases, these were found in 10% of the patients with Type 4. Overall, recurrence was observed in 13 patients (4.5%). Almost all of these patients with recurrence had Type 4 tumors. The lone exception was a Type 3 patient in whom local recurrence developed adjacent to a surgical staple line. The 5-year overall survival rates were 100% in Type 1 and Type 2, 98% in Type 3, and 88% in Type 4. Patients with Type 4 had a significantly worse prognosis than those with other types.

Conclusion
Conclusions. Subcentimeter lung cancers with a GGO component on preoperative HRCT (Type 1 to 3) can be considered “early” lung cancers. Thus, in these cases, limited resection may be warranted to achieve a cure because patients with Type 1 to 3 did not have lymph node metastasis. On the other hand, lobectomy should still be considered the standard operation of choice for Type 4 tumors.

Background
Bronchopulmonary carcinoid (BPC) comprise about 2-5% of lung tumors. BPC are malignant neoplasms with indolent course. Surgical treatment is the gold standard therapy.

Methods
We present our experience of 228 patients treated surgically for typical (TC) and atypical carcinoid (AC) from1998 to 2011 in National Tuberculosis and Lung Diseases Research Institute. We tried to determine the variables influencing the long-term survival of patients with BPC. Among 4467 patients, treated surgically for non-small cell carcinoma, BPC encompassed 5,1%. All cases were reviewed and classified according to the latest WHO classification (2004). The clinical course and survival analysis were performed.

Results
The number of BPC during last 14 years gradually growth, from 13,4 cases per year between 1998-2004 to 19,1 cases between 2005-2011. 102 cases (44,7%) were classified as TC and 126 (55,3%) as AC. There were 158 females (69%) and 70 men (31%) and the mean age was 52 ys. Men were significantly younger then females (49 vs 53). Symptoms were present in 143 patients, the most commonly were cough (31%), respiratory tract infection (31%), then haemoptysis (18%), dyspnoe (12%) and atelectasis (3,5%). No patients showed a carcinoid syndrome. There were no correlation between smoking status and BPC. Most of the tumors were central (73,7%), the remaining were peripheral. The mean diameter of BPC was 2,25cm (range 0,6-7,0cm). AC were significantly larger (2,54 vs 1,9) and centrally located were also larger the peripheral. Surgical treatment consisted of: 146 standard lobectomies (64%), 19 pneumonectomies (8%), 23 bilobectomies (10%), 23 sleeve lobectomies (10%), 2 sleeve-pneumonectomies (1%), 7 anatomic segmentectomies (3%), 2 wedge resections (1%), 6 – bronchoplastic procedures without lung resection (2,6%). Radical mediastinal lymphadenectomy was added in all cases exept 1. Involvement of lymph nodes was present in 35 cases (15,4%), N1 -22 (9,7%) and N2 – 13(5,7%). Infiltration of bronchial or vessel margin (R1) was revealed in 8 cases. The postoperative TNM stage contained IA (50%) disease, IB (31%), IIA (10%), IIB (2,6%), IIIA (6,1%), IV (0,4%). The stages IB, IIIA and IV more often related to AC. Four patients died in short time after operation (0,2 -2 months). Three of them were after pneumonectomies, one – sleeve-lobiectomie. In most cases tumor was localized centrally. The mean follow-up time for all patients was 69,3 months (range 0,2-172), with 204 still alive. Four patient died of tumor progression (3 from A, 1 from TC), remaining 8 patients from other causes, from 12 – the cause of death was unknown. Lymph nodes metastasis were seen in 5 cases (20,8%). Overall, 5, 10 and 14-years survival for TC was respectively 95%, 92,1% and 90,2%, for AC – 93%, 90,6% and 88,9%.

Conclusion
BPC demonstrate gradual growth during last years. Lung carcinoids are tumors with an excellent prognosis in most cases, even in the presence of metastases in lymph nodes and positive surgical margin (bronchial or vessel). Histologic subtypes did not infleunce on survival. Surgery currently represents the best treatment with good results and long survival but long-time observation is necessary.

Background
Combined pulmonary fibrosis and emphysema (CPFE) has been recently reported as a prognostic factor for patient with respiratory disorders. It might increase the risk of lung cancer. However controversisies remain as to surgical outcome in this population.

Methods
Retrospective study was performed on 981 patients who underwent surgical resection of lung cancer at our institute between 2008 and 2012. Findings on thin-section computed tomography which was available for all patients were reviewed. Based on the findings, patients complicated with CPFE were selected and clinicopathological features were investigated. Surgical outcome and prognosis following lung resection were also examined.

Results
CPFE was observed in 97 (9.1%) patients with resected lung cancer.　Patients with pulmonary fibrosis alone were 43 patients (43.8%), patients with emphysema alone were 148 patients (15.1%) and patients without abnormal shadow were 649 patients. Lung function test were as follows (CPFE/ Fibrosis/ Emphysema): vital capacity (VC); 3.3L/ 2.9L/ 3.5L, forced expiratory volume in one second (FEV1); 71.5%/ 77.4%/ 67.9%, diffuse capacity (DLco); 44.6%/ 58.1%/ 58.6%. Ninety day-mortality(CPFE/ Fibrosis/ Emphysema) was 10.2%/ 2.3%/ 1.1%. Risk factor of ninety day mortality in patients with CPFE was operative blood loss. CPFE patients also have higher risk of major complication after surgery (CPFE 44.3%, non-CPFE 8.1%). The statistically significant difference in survival was found with the Kaplan-Meier method (p<0.001). Survival at 2 years(CPFE/ Fibrosis/ Emphysema/ Normal) was 74.6%/ 88.2%/ 91.4%/ 93.6% and survival at 5 years (CPFE/ Fibrosis/ Emphysema/ Normal) was 58.6% /61.8% /72.9%/ 81.5%. Within CPFE patients, multivariable analysis of hazard ratio for prognosis showed following significant factors; pO2<70 (HR 13.52, p=0.001), lymph node metastasis (HR 10.89, p=0.002).

Conclusion
Surgery for patients with CPFE is feasible. Postoperative complications were frequently and prognosis is poor compared with either emphysema or fibrosis. Not only the status of lung cancer but also respiratory status is a risk factor of prognosis after surgery for primary lung cancer with CPFE.

Background
Metastasis to mediastinal lymph nodes is a significant negative prognostic factor for NSCLC in spite of a localized presentation. According to the results of Hata’s study on the vital lymphdrainage from each segmental bronchus, the predictable N3 stations of lymphatic spread of cancer cells are the contra-lateral mediastinal nodes. Therefore, we have considered that it would be valuable to perform Systemic extended bilateral mediastinal dissection and lung resection through a median sternotomy (ND3 operation, Hata’s method). And we had devised ND3 operation for patient with NSCLC of the left lung. So far, we reported effectiveness of N3α operation (Hata’s method) in patients with lymph node metastasis to the mediastinum in NSCLC of the left lung. From January 1990 ,we have also applied N3α operation (Hata’s method) to patients with NSCLC of the right upper lobe.

Methods
We retrospectively analyzed the clinical records of patients who underwent ND3 operation (R0 resection) for NSCLC of the right upper lobe. Patients who had incomplete resection (R1,R2), stageIV disease were excluded from this analysis. The patients with NSCLC who are estimated to be able to conventional radical operation and aged 75 year-old or less becomes the adaptation of ND3 operation. From January 1990 to December 2012, 195 patients of the right upper lobe primary underwent ND3 operation completely (R0 resection) and we reviewed these cases. Survival estimates were calucurated by The Kaplan-Meier method and compared using the log-rank test.

Results
Overall 5-year survival rate in 195 patients, including operation-related deaths and deaths due to unrelated diseases, was 55.4%. In these 195 patients, the five-year survival rate according to pathological stages were 90% in 35 pts of stageIA, 70.6% in 48 pts of p-stageIB , 57.5% in 19 pts of stageIIA, 47.9% in 18 pts of p-stageIIB, 42.1% in 45 pts of p-stageIIIA, 18.1% in 30 pts of p-stageIIIB. And the 5-year survival rate according to nodal involvement, 70.9% in 109 pts of N0, 46.2% in 22 pts of N1, 41.5% in 41 pts of N2, and 19% in 14 pts of N3γ. Operative mortality in 195 patients of right upper lobe primary NSCLC was 5 patients (2.6%).

Conclusion
In this nonrandomized comparison, the post-operative survival of patients with p-N2 NSCLC of the right upper lobe would be improved by our Systemic Bilateral Mediastinal Dissection. To improve survival rate, it is important to perform curative operation with complete　dissection of all station of mediastinal lymph nodes.

Background
The number of patients with non-small cell lung cancer (NSCLC) has been increasing for some decades. The patients are predominantly male and over 50 years old in age, whereas female patients under 50 years old are relatively small cohort in NSCLC. Therefore, the characteristics of these female patients remain to be unclear.

Methods
In this study, we examined the clinico-pathological characteristics of female patients under 50 years old who had received surgical treatments for NSCLC. Female patients who had been under 50 years old and received curative surgical resection for NSCLC in our hospital from January 2000 to December 2010 were involved in this study. The clinico-pathological characteristics of them were examined retrospectively and compared with those of relative male patients in the similar criteria. Both overall survival (OS) and disease-free survival (DFS) times after surgery were obtained by Kaplan-Meier analysis, and differences between two cohorts were analyzed by log-lank test.

Results
In total, 13 female patients with median age of 43 years old (range: 40-49 years old) and 12 male patients with median age of 43years old (range: 31-49 years old) were received curative surgical resection for NSCLC in this period. All of the female patients were never-smokers, whereas 11 male patients (91.7%) were current smokers. Twelve female patients were free from symptoms at the diagnosis of NSCLC, however, 5 male patients had some symptoms related to NSCLC (a rate of symptom-free patients; female vs male = 92.3% vs 58.3%). Pathological stages of female NSCLC were as follows; 1A: 9, 1B: 1, 2A: 1, 3A: 1, 3B: 1. All cases of female NSCLC (100.0%) were diagnosed with adenocarcinomas histologically; mixed subtype: 6, papillary: 2, bronchiolo-alveolar cell carcinomas (BAC): 5 cases. In males, 8 patients (66.7%) were diagnosed with adenocarcinomas including 1 BAC case. The rates of OS and DFS at 5 years after surgery of female patients were 100.0% and 74.6%, respectively, with the median observation period was 69.0 months (range: 17-148 months). On the other hand, the rates of OS and DFS at 5 years after surgery of male patients were 74.6% and 58.3%, respectively, demonstrating that OS in female patients seemed to be long as compared with that in male patients (P=0.302).

Conclusion
More patients who were free from symptoms, never-smokers and with BAC subtype were included in female than male patients in this study. These characteristics of female patients would have contributed to the better prognosis of them after surgery. In the NSCLC patients under 50 years old, the prognosis of female patients who received curative surgery is likely to be better than that of male patients.

Background
Bronchopleural fistula (BPF) after lung resection is a life-threatening complication. Thoracic surgeons should always consider the risk of postoperative BPF and the management to avoid worst scenario, but it is still controversial of which is the best way to manage BPF. We herein describe the results of BPF and explore optimal treatment.

Methods
Data on 2270 patients with lung resection for NSCLC over a period from 2000 to 2012 were retrospectively reviewed. Details regarding surgery and subsequent treatment were carefully reviewed. Followed information was recorded; age, sex, clinical diagnosis, associated condition, TNM stage, time from primary operation to rethoracotomy, and postoperative outcome.

Results
The overall BPF incidence was 1.1% (24/2270). There were 20(83.3%) male and 4(16.7%) female, mean age was 67.1 years. BPFs occurred after pneumonectomy in 2(8.3%), lobectomy in 20 (83.3%) and sleeve resections in 2 (8.3%). In side right was in 20 (83.3%) and left was in 4 (16.7%). The histological types were 9 adenocarcinomas, 9 squamous cell carcinomas, 6 others. The pathological stage were stage IA in 6 cases, IB in 4, IIA in 3, IIB in 4, IIIA in 6 cases. Mean postoperative day was 19.8. In initial treatment, fenestration was 12 cases, primary closure using various techniques was 8 cases, and completion pneumonectomy was a case. Massive hemoptysis causing death owing to bronchial pulmonary artery fistula (BPAF) was in 3 cases. Primary closure using various techniques succeeded in 3 cases, while the repair failed in the other 5 case and 2 cases subsequently developed further fatal complications. The mortality rate of primary closure was 25% (2/8) and success rate was 38% (3/8). The mortality rate of fenestration was 8.3% (1/12), and success rate was 91.6% (11/12). The overall mortality rate for postoperative BPF was 25% (6/24); 3 cases were BPAF, 3 cases were aspiration pneumonia.

Conclusion
BPF remains a major complication in the surgery of lung cancer because of its high mortality and morbidity rate. Especially, the mortality rate of primary closure using various techniques was high, and aspiration pneumonia with consequent ARDS is fatal complication. To avoid death related to BPF, the surgeons should consider the fenestration as soon as the BPF was diagnosed.

Background
Preoperative localization is necessary to perform thoracoscopic resection of small or deeply located solitary pulmonary nodules (SPNs). We recently developed a new localization technique using a self-made, fragmented platinum microcoil, and retrospectively compared the effectiveness of our technique with that of lipiodol.

Methods
Fifty two patients underwent thoracoscopic pulmonary wedge resections for 57 SPNs between January 2006 and June 2013. Self-made, fragmented platinum microcoils (Easimarker) were targeted to localize 30 SPNs [17 solid nodules, and 13 ground glass opacities (GGOs)] in 28 patients (Group A), and lipiodol was injected in 27 SPNs (17 solid nodules, and 10 GGOs) of 22 patients (Group B). Preoperative localization using both targeting materials was performed into, or just around the pulmonary lesions on the day of thoracoscopic surgery in the room of chest CT scanner. Localized SPNs were then, wedgely resected using fluoroscopy-aided thoracoscopic surgery (FATS). The intraoperative fluoroscopic exposure (radiation) time, diagnostically detecting rate of pathologic lesions, and other clinical data were collected.

Results
Mean size and depth of SPNs in group A and B were 10.6 ± 4.7 mm (range: 0.9 to 23) versus 7.9 ± 4.9 mm (1 to 21), and 10.9 ± 7.9 mm (1 to 30) versus 9.7 ± 8.4mm (1 to 28.2), respectively. CT-guided localizations were successfully performed in both groups. No mortality and major morbidity were observed in both groups. All lesions in both groups were completely resected and diagnosed histopathologically. The intraoperative fluoroscopic exposure time of group A (55.0±40.8 seconds) was significantly shorter than that of group B (105.7±109.0 seconds).

Conclusion
Our preoperative localization procedure using fragmented platinum microcoils appears to be effective and feasible in that it has shorter intraoperative time, less radiation exposure, and better accuracy of detecting SPNs. Once inserted fragmented microcoil into the pulmonary lesions stays firmly and more visible radiologically, through lipiodol tends to diperse outside the targeting lesion.

Background
The aim of this study was to clarify the prognosis of pulmonary adenocarcinoma patients after postoperative recurrence according to EGFR and KRAS mutations and recurrence site.

Methods
Between July 2002 and December 2011, a total of 297 consecutive patients underwent surgical resection for primary pulmonary adenocarcinoma. Among all the patients, we retrospectively evaluated 58 recurrent adenocarcinoma patients. They were divided into the groups according to presence of EGFR mutation and KRAS mutation, and compared clinicopathological features, recurrence sites and postrecurrence survival.

Results
EGFR, KRAS mutations were detected in 26 patients (45%), 11 patients (19%), respectively. Of the cases with EGFR mutations, L858R point mutation in exon 21 was most frequently observed in 18 cases, secondly deletion in exon 19 was in 8 cases. Initial recurrence was detected in distant in 25 (43%), local in 17 (29%), and both in 16 (28%). In EGFR mutant (EGFR+) cases, bilateral/contralateral lung recurrences were significantly frequently occurred. EGFR+ cases had significantly better outcome than KRAS+ cases and EGFR-KRAS- (Wild) cases. 2-year postrecurrence survival rate were 81%, 18%, and 47% in EGFR+, KRAS+, and Wild cases, respectively. Patients with distant organ recurrence (D+) showed significantly worse survival than those without distant recurrences in only Wild cases, but not significant in EGFR+ cases and entire cohort. Multivariate analysis revealed that only EGFR mutation and number of recurrent lesions were statistically significant independent postrecurrence prognostic factors. Figure 1Figure 2

Conclusion
Our results indicate there were distinct survival differences in recurrent adenocarcinoma patients according to driver mutations. Patients with EGFR mutated tumors could be expected of long survive regardless of presence of distant site recurrences, and patients with KRAS mutated adenocarcinoma had poor outcome after postoperative recurrence. The examination of driver mutations is essential for prediction of postrecurrence survival after surgical resection.

Background
Pulmonary resection by robotic-video assisted thoracic surgery (RVATS) has been performed for selected patients in specialized centers over the past decade. Despite encouraging results from case-series reports, there remains a lack of robust clinical evidence for this relatively novel surgical technique. The present systematic review aimed to assess the short- and long-term safety and efficacy of RVATS.

Methods
Nine relevant and updated studies were identified from 12 institutions using five electronic databases. Endpoints included perioperative morbidity and mortality, conversion rate, operative time, length of hospitalization, intraoperative blood loss, duration of chest drainage, recurrence rate and long-term survival. In addition, cost analyses and quality of life assessments were also systematically evaluated. Comparative outcomes were meta-analyzed when data were available.

Results
All institutions used the same master-slave robotic system (da Vinci, Intuitive Surgical, Sunnyvale, California) and most patients underwent lobectomies for early-stage non-small cell lung cancers. Perioperative mortality rates for patients who underwent pulmonary resection by RVATS ranged from 0 – 3.8%, whilst overall morbidity rates ranged from 10 – 39%. Two propensity-score analyses compared patients with malignant disease who underwent pulmonary resection by RVATS or thoracotomy, and a meta-analysis was performed to identify a trend towards fewer complications after RVATS. In addition, one cost analysis and one quality of life study reported improved outcomes for RVATS when compared to open thoracotomy. Figure 1 Figure 2

Conclusion
Results of the present systematic review suggest that RVATS is feasible and can be performed safely for selected patients in specialized centers. Perioperative outcomes including postoperative complications were similar to historical accounts of conventional VATS. A steep learning curve for RVATS was identified in a number of institutional reports, which was most evident in the first 20 cases. Future studies should aim to present data with longer follow-up, clearly defined surgical outcomes, and through an intention-to-treat analysis.

Background
Non-small cell lung cancer (NSCLC) of the thoracic inlet accounts for less than 5% of all lung cancers. Due to the lack of efficient treatment and the complexity of the anatomical structures commonly invaded, these tumors were deemed historically unresectable and fatal. In this study, we reviewed our surgical experience and long-term outcome after resection of NSCLC invading the thoracic inlet

Methods
All consecutive patients from a single center who underwent resection of NSCLC invading the thoracic inlet were reviewed with data retrieved retrospectively from the charts.

Results
A total of 65 consecutive patients with a median age of 61 years (range 32 to 76) underwent resection of NSCLC invading the thoracic inlet from 1991 to 2011. Tumors were located in the previously described (Reference) five zones of the thoracic inlet as follows: zone 1 or anterolateral (n=5, 8%), zone 2 or anterocentral (n=7, 11%), zone 3 or posterosuperior (n=12, 18%), zone 4 or posteroinferior (n=22, 34%) and zone 5 or inferolateral (n=7, 11%). Fifty-two (80%) patients had induction therapy, mostly two cycles of cisplatin-etoposide and 45 Gy of concurrent radiation. All patients underwent en bloc resection of the lung and chest wall. Lobectomy was performed in 60 patients (92%). A median of three ribs were resected (range 1 to 5) and included the first rib in all patients. Twenty-four patients (37%) had an additional vertebral resection of up to five levels (median 3). Considering the most extended vertebral resection, total vertebrectomy with anterior-posterior spinal stabilization was required in 6 patients (25%), hemi-vertebrectomy with posterior spinal stabilization in 15 (62%), and partial vertebrectomy without stabilization in 3 (13%). Arterial resections were performed in seven patients (11%) and included subclavian artery (n=5), vertebral artery (n=1) and combination of sublclavian and carotid arteries (n=1).The median postoperative length of stay was 11 days (range 4 to 173). Postoperative morbidity and mortality were 46% and 6%, respectively. Pathologic response to induction treatment was complete (n=19) or near complete (n=12) in 31 patients (49%). Pathologic stages were 0 in 19 patients (29%), IB in 1 (2%), IIB in 28 (43%), IIIA in 15 (23%) and IIIB in 2 (3%) patients. After a median follow-up of 20 months (range 0 to 193), 34 patients were alive without recurrence. The overall 3- and 5-year survivals reached 58% and 52%, respectively. Results of the Cox regression and log-rank/Breslow tests identified the site of tumor (zone 1/3 vs 2/4/5, p=0.050) and the response rate to induction treatment (complete/near complete vs partial, p=0.004) as significant predictors of survivals. A trend toward shorter survival was found in case of arterial resection (p=0.063).

Conclusion
Survival after resection of NSCLC invading the thoracic inlet in highly selected patients reached 52% after five years. Tumor location within the thoracic inlet and pathologic response to induction therapy were significant predictors of survivals. Reference: de Perrot M, Rampersaud R. Surgical approaches to apical thoracic malignancies. J Thorac Cardiovasc Surg. 2012 Jul;144(1):72-80.

Background
Video-assisted mediastinoscopic lymphadenectomy (VAMLA) has been developed to reduce the false-negativity rate of mediastinoscopy. We aimed to analyze the impact of VAMLA on survival in patients with operable IA-IIIA disease and inoperable IIIB (N2) tumor.

Methods
Between May 2005 and June 2013, 421 N2 patients with non-small cell lung cancer who had undergone standard mediastinoscopy or VAMLA were evaluated. Of these, 105 patients (24.9%) underwent VAMLA, whereas 316 patients (75.1%) underwent standard mediastinoscopy. The median number of resected lymph node was 29.1 in VAMLA group , while the median was 10.3 in mediastinoscopy group.All patients with N2 disease were referred to medical oncology and/or radiation oncology departments. The survival rates were calculated using Kaplan-Meier test. Of those, 26 patients (6.2%) were referred to neoadjuvant treatment.

Results
The 5-year survival rate of N2 patients who had VAMLA was 34.0%, whereas it was 18.0% in patients who underwent mediastinoscopy (p=0.03)in all patients. Survival analyses revealed T factor (p=0.04), N factor (p=0.01), multiplicity of nodal status(p=0.04), and lymphatic invasion (p=0.03), type of mediastinoscopy (VAMLA vs standard mediastinoscopy) statistically significant in entire study population. Multivariable Cox analysis confirmed N stage(p=0.01, hazard ratio, 4.1, 95% confidence interval;1.125-8.661)T stage (p=0.045, hazard ratio, 1.4, 95% confidence interval: 1.050-4.112)and type of mediastinoscopy as independent prognostic factors (p=0.02, hazard ratio, 2.1; 95% confidence interval:1.11-11.03)

Conclusion
VAMLA seemed to provide longer survival in operable and inoperable (T3-4N2) non-small cell lung cancer patients. This effect could be due to complete resection of mediastinal lymph nodes. Further studies in order to clarify the possible survival impact are warranted.

Background
There has been little investigation of temporal trends in outcomes following resection of early non-small cell lung cancer. Analyses are easily confounded by changes in patient characteristics and variations in background mortality when assessing all-cause survival. This study aimed to evaluate changes in patient characteristics, tumour factors and survival over time.There has been little investigation of temporal trends in outcomes following resection of early non-small cell lung cancer. Analyses are easily confounded by changes in patient characteristics and variations in background mortality when assessing all-cause survival. This study aimed to evaluate changes in patient characteristics, tumour factors and survival over time.

Methods
A retrospective analysis of 2816 consecutive pathological stage 1A to 3A patients, treated by surgical resection between 1984 and 2007 was performed. Patients were divided into four 6-year eras by date of surgery. Relative survival probabilities were estimated by era and TNM stage. Expected survival was calculated from national age, sex and period specific mortality rates. Multivariable regression using a generalised linear model with Poisson error was used to estimate the excess hazard of death in each era, using the 1984-1989 cohort as the baseline, controlling for age, sex, extent of resection, margin status, tumour stage and cell type.

Results
In later eras, patients were older, had a greater proportion of adenocarcinomas and stage 1A tumours. Relative 5-year survival rates for 1984-1989, 1990-1995, 1996-2001 and 2002-2007 were 45.4, 49.6, 48.5 and 57.9% respectively. There was a significant improvement in 5-year relative survival in the 2002-2007 cohort (Excess hazard ratio 0.62, p<0.001). Age ≥75, increasing TNM stage, positive margins and mixed cell type were also significant prognostic factors. The increased survival demonstrated in the most recent era can be attributed primarily to survival gains in stage IIa/b and stage 3a (Figure). Figure 1

Conclusion
Temporal trends in patient characteristics in this series mirror recent epidemiological data for non-small cell lung cancer. After controlling for known confounders and background mortality variation, improved survival was demonstrated for more recent patients. Advances in clinical staging and adjuvant therapy may explain these findings.

Background
In recent years, the opportunity to encounter a ground glass opacity nodule (GGN) by high-resolution CT is increased, and simultaneous multiple GGNs also are not uncommon. A GGN has been usually classified as pure GGN and part-solid GGN, the former seems to correspond to atypical adenomatous hyperplasia (AAH) or adenocarcinoma in situ (AIS) and the latter seems minimally invasive adenocarcinoma (MIA) or invasive adenocarcinoma (IA), but CT image and pathological findings do not necessarily match. Some GGNs are diseases unrelated to primary lung adenocarcinoma. We have examined the resected cases for simultaneous multiple GGNs on both sides of the lung.

Methods
Adaptation of resection for GGNs on our hospital is as follows. 1)10-15mm or more size, 2) larger solid component, 3) just below the pleura, 4) increase over time in size or density, 5) the purpose of pathological diagnosis, etc. The prevention of lung function is noted in the resection on both sides of the lung. In this four years, we performed surgery on seven patients with bilateral multiple GGNs for diagnosis and treatment. We investigated the clinical features and histopathological findings of the resected lung.

Results
The seven patients consisted of 40 to 70 years old, five women and two men. We performed lobectomy and partial (wedge) resection of three patients. Four patients underwent several wedge resections for pathological diagnosis and treatment. Two women did the two-term surgery on both sides of the lungs. Pathological diagnosis was adenocarcinoma (AIS, MIA, IA) in five cases, AAHs in one, and lymphoproliferative disease in one. In one patient, all three lesions from four wedge resections had different mutated patterns of EGFR. There was no recurrence or death in 13 to 58 months of the observation period.

Conclusion
Simultaneous multiple GGNs was more frequent in women than men. Surgical biopsy (wedge resection) seems to be necessary for definitive diagnosis because a GGN may not be related to lung cancer. Even if multiple cancers in bilateral lung are supposed, prognosis may be able improved by surgical removal of more invasive (advanced) lesions in GGNs.

Background
It is said that atrial fibrillation(Af) after lobectomy is seen in about 20%, sometimes difficulty in postoperative management. Landiolol is very short acting selective β1-blocker. It is reported that landiolol have the usefullness of not only treatment for Af, but also the prevention of Af after lobectomy. We had randamized control study about preventation of Af after lobectomy or more.

Methods
We divided into control group and landiolol administration group for patients to perform lobectomy or more. The patients of administration group are subjected prophylactic landiolol to 24hr continuous infusion from the start of surgery at 5γ. We analysed the 93 cases with informed consent in this clinical trial from June 2010 to April 2013. Finally, 2cases dropped out because of changing operative procedure for dissemination of lung cancer.

Results
45 cases is in landiolol administration group and 46 cases is in control group of 91 cases. Postoperative Af was occurred 10 cases. But, there were no occurrence of Af during landiolol infusion. 9 cases of 10 cases had Af during the 3days after surgery. The data is following: (administration group, control group) Event of Af=(6,4), postoperative days of Af=(1.5, 3.), age(years-old)=(70±8.7, 66.3±9.4), gender(M:F)=(25:20, 29:17), pasthistory of Af=(0, 5), operative site(R:L)=(29:16, 31:15), bleeding=(286±463ml, 212±308ml), operation time=(290±86min,272±87min), in-out balance in operation=1816±827ml, 1414±732ml), the rate of concomittant use of epidural anethesia=(91%:95%), operatibe approach(open:VATS)=(6:39, 10:36), operative procedure(lobectomy:bilobectomy:pneumonectomy)=(45:0:0, 40:3:3), upper mediastinal LN disection rate=(67%, 63%), #7 LN disection rate=(73%, 70%). Adverse effect of landiolol is hypotension(BP<80) in 3 cases(7%) and bradycardia(HR<60) in 1 case.

Conclusion
Landiolol administration group had more occurrence of postoperative Af, compared to control group. Considering the incidence of Af is high up to 3days after surgery, only 24hr continuous administration can not suppress Af after stopping infusion. β- action, by exciting sympathetic nerve, is important for the occurrence of postoperative Af. β- action is inhibited by landiolol during administration, for positive feedback, β-receptor are upregulation. It is thought that Af occurrence after stopping landiolol is increasing in administration group than control group, because the β-agonist binds to the upregulated receptor. It is possible that turned out to be different result, if prophylactic landiolol administration would continue until 3days after surgery, for Af is high up to 3days after surgery.

Background
Salvage therapy could be indicated for residual tumor and local recurrence of treated lung cancer. However, there is no report of the meaning of making full use of bronchoplasty, pulmonary arterioplasty, and combined resections of superior vena cava (SVC) in salvage surgery for lung cancer. In this study, we investigated perioperative complications of the salvage surgery for lung cancer according to the mode of operations.

Methods
We retrospectively reviewed 1320 consecutive patients who underwent lung resection for lung cancer at our institution from January 2008 to May 2013 and surveyed 18 salvage surgery cases among them. The mode of operation, perioperative complication and long-term outcome were investigated in detail.

Results
Twelve salvage surgical therapies were indicated for residual tumor after 10 chemotherapy and two chemoradiotherapy cases, and another six salvage surgeries were indicated for local recurrence after chemoradiotherapy. Radiation dose was 45 – 66Gy in seven chemoradiotherapy cases and 140Gy of proton therapy in one case. The number of mode of operation was as follows; one pneumonectomy with carinal resection, three pneumonectomies, one lobectomy with bronchoplasty and pulmonary arterioplasty and combined resection of the SVC (triple plasty), one lobectomy with bronchoplasty and combined resection of the SVC (double plasty), one sleeve bilobectomy, two sleeve lobectomies, eight lobectomies and one wedge resection(Table 1). Median operation time was 178.5 minutes (range 80-395). Median intra-operative blood loss was 130ml (range 5-1720). Average duration of hospitalization days after salvage surgery was 10.5 days. Regarding to operation time, intra-operative blood loss, and hospitalization days after operation, there was no significant difference between salvage surgery and conventional lung resection at our institute. Post-operative complications were as follows; three empyemas, three pneumonias, two pleural fistulas, and one chylothorax. We had to make an open window for one empyema case, but another complications were recovered safely and there was no 30-day mortality. Median follow-up was 9.5 months. There was no local recurrence but there were three distant metastases cases after salvage surgery. The longest survivor without recurrence after salvage surgery survives for 31 months.Figure 1

Conclusion
There were no critical complications and mortality in salvage surgeries after chemotherapy and chemoradiotherapy for lung cancer. Pneumonectomy, bronchoplasty, pulmonary arterioplasty and combined resections of the SVC are feasible even in salvage surgery for treated lung cancer.

Background
Systematic nodal dissection has become a standard part of a curative resection for the non-small cell lung cancer. Its value in lung metastasectomy is unknown. The aim of our study was to assess the frequency of lymph node metastases in the patients undergoing lung metastasectomy, survival of the patients with and without lymph node involvement and to consider, if and when routine nodal dissection should be recommended. Study was supported by the grants of the Ministry of Health of the Czech Republic IGA MZ ČR NS10095-4, NT/12085-3.

Methods
All consecutive patients selected for lung metastasectomy in 3 surgery departments from 7/2008 to 12/2011, were operated by standard technique of the lung metastasectomy with systematic nodal dissection according to the ESTS guidelines for intraoperative lymph node staging in non-small cell lung cancer. If wedge resection was done, N1 nodes were removed only as a part of the local procedure. Patients with mediastinal lymph node involvement detected by the preoperative CT or PET scan and patients with metastatic lung cancer were excluded from the study.

Results
There were 101 lung metastasectomies, for metastatic carcinoma in 87 patients, for metastatic sarcoma in 14 cases. Surgical procedures were as follows: 71 wedge resections, 27 lobectomies, 8 segmentectomies. Bilateral metastases were present in 22 patients; solitary metastatic lesion was found in 57 cases. Average diameter of the metastasis was 25.3 mm. Average number of the lymph nodes yielded by lymphadenectomy was 16.4. Metastatic involvement of the mediastinal lymph nodes was found in 9 cases (8.9%), metastatic carcinoma (colorectal in 4 cases) in 7 cases and sarcoma in 2 cases. Average DFI was 37.5 months. 3-years survival according Kaplan-Meier was 76% (0.76±0.06), for metastatic carcinoma 81% (0.81±0.06), sarcoma 46% (0.46±0.15); for colorectal carcinoma 92% (0.92±0.06). 3-years survival for patients with negative lymph nodes was 78% (0.78±0.06), with metastatic involvement 53% (0.53±0.25). Statistical analysis with two-sided log-rank test at the 0.05 level of significance showed better survival for the patients with metastatic carcinoma in comparison with sarcoma (p=0.0007) and better survival for patients without metastatic lymph node involvement (p=0.044).

Conclusion
Even in carefully selected group of patients, incidence of the mediastinal lymph node metastases is high and systematic lymphadenectomy should be considered in all cases. Because of a lack of robust data, we recommend routine nodal dissection only as a part of a clinical study. Due to the small numbers available, lymphadenectomy remains questionable in bilateral cases, for sarcoma metastases and in patients in high operative risk.

Background
Pulmonary solid nodules on chest computed tomography (CT) included inflammatory nodule, benign tumor, carcinoid tumor, small cell lung cancer, large cell carcinoma, large cell neuroendocrine carcinoma, squamous cell carcinoma, and poorly-differentiated adenocarcinoma, and so on. If the solid nodule was highly suspicious of malignant tumor or diagnosed as being primary lung cancer, a surgical resection would be recommended as expeditiously as practicable because of its metastatic potential. On the other hand, most well-defined ground-glass nodules including part-solid nodules were atypical adenomatous hyperplasias or lung adenocarcinomas. The development of spatial resolution on CT had provided improvement in predicting malignancy of these nodule shadow suspected of lung cancer. The aim of this study was to elucidate the preoperative chest thin-section CT (TSCT) findings of lung nodules which were inherent with metastatic potential from the perspective of recurrence and long term results.

Methods
We reviewed 392 primary lung adenocarcinomas with clinical T1N0M0 who underwent surgery between 2003 and 2007. Independent recurrence predicting parameters were extracted from the following ten parameters by using logistic regression analysis; sex, age, smoking index, preoperative serum carcinoembryonic antigen level, tumor location, maximum tumor size, consolidation tumor ratio (C/T ratio), tumor disappearance rate (TDR), the maximum size of consolidation at lung window setting (lung consolid), and the maximum size of consolidation at mediastinal window setting (med consolid). We evaluated extracted parameters by using receiver operating characteristic area under the curve (ROC-AUC) for recurrence prediction and identified these optimal cut-off levels of these parameters for prediction of whether patients had a good chance of being cured by surgical resection from their recurrence rate and survival.

Results
The median follow-up period was seven years. The 75 of 392 patients recurred. C/T ratio, lung consolid, and TDR were extracted as an independent recurrence predictor. ROC-AUC of these parameters was 0.70, 0.71, and 0.64 for predicting recurrence, respectively. If C/T ratio was 0.5 or less, distant metastatic recurrence was observed in only one of 75 patients and if lung consolid was 10 mm or less, it was also observed in only one of 82 patients. There were significant differences in overall survival and recurrence free survival among two populations divided by these cut-off levels of C/T ratio and lung consolid.

Conclusion
If the C/T ratio was 0.5 or less and/or lung consolid size was10 mm or less in cT1N0M0 lung adenocarcinoma patients, the recurrence rate was extremely low if they underwent a standard surgical resection. But when TSCT parameters of lung nodules exceeded these cut-off values, the recurrence rate wound increase and the prognosis would become worse even if they underwent complete resection. In these cases a prompt surgical resection would be recommended.

Background
The cuffed tunneled catheter has revolutionized outpatient management of recurrent pleural effusions. Some patients, however, refuse these tubes on the basis of prohibition of swimming/tub bathing or cosmetic considerations. The objective of this study was to develop a technique to serve such patients.

Methods
Four patients with recurrent effusions refused tunneled catheter placement and were not candidates or refused pleuro-peritoneal shunts. All required multiple thoracenteses, one weekly for over a year. One patient had a chronic benign symptomatic transudative effusion after successful nonoperative treatment of NSCLC. The second patient had recurrent chylothoraces secondary to Yellow Nail Syndrome. Two patients had malignant pleural effusions. In all cases, after assuring their effusions could be drawn through a 19 gauge needle, they were taken to the operating room where a tunneled catheter was placed, leaving the fibrous polyester cuff in place for tissue ingrowth, but trimming the valve and connecting that end of the catheter to a standard implantable single lumen infusion port (Figure 1). Figure 1 Figure 1 A) The configuration of the cuffed catheter connected to a single lumen infusion port B) The port-accessed catheter being tested after implantation of the catheter and prior to implantation of the port C) Accessing the catheter as it is done at home

Results
All procedures were performed uneventfully and without complications. The catheters continue to work well and are accessed by the patients at home.

Conclusion
This novel technique offers a safe palliative option for any effusion that can be aspirated through a 19 gauge needle, benign or malignant. The patients have been able to maintain their normal life style without the risk and inconvenience of serial outpatient thoracenteses or the limitations imposed by the presence of an exterior tube. Specifically, the proven benefits are that it enhances quality of life for those patients who wish to maintain the activities of swimming or tub bathing or have compelling concerns about the appearance of an external tube. There is also the theoretical benefit that this port-accessed drainage system may be less prone to infectious complications.

Background
Objective: In the United States, thoracic surgery quality, as measured in mortality, morbidity or processes (eg lymphadenectomy after lung cancer resection), is very heterogeneous between institutions and surgeons. Despite barriers involving surgeon specialties, payors and administrative systems, health care quality is measurable & implementable. We describe the Thoracic Surgery Initiative (TSI), a grass roots quality improvement effort within Providence Health & Services (PHS), which consists of 34 facilities providing healthcare over a large Western region (Alaska, California, Montana, Oregon & Washington).

Methods
Methods: The TSI was conceived and driven by a thoracic surgeon. A core team (surgeons, administrators, data manager) was formed, thoracic surgery (TS) service line specifics defined & identified (of 34 facilities, 14 perform TS), stakeholders identified & surveyed regarding interest & resources. A series of meetings, agreements and collaborations were formed to define and implement quality care with the following goals: decrease mortality & morbidity, clinical standardization & cost savings.

Results
Results: 2011- Feb 2012: (1) intra-mural grant for TSI development obtained, (2) development activities described above, (3) organizing meeting of stakeholders (40 attendees-12 hospitals) accomplished: (a) persuasion of site-specific & system leaders/physicians/administrators, (b) executive committee (EC), (c) operational calendar (bimonthly phone conferences). Surgeon-leader spent 265 hrs on project Aug 2011-Mar 2012. 2012 EC accomplishments: (1) formulated charter & mission statement, (2) agreed upon & defined TS clinical data elements, (3) determined system/site-specific data system costs, (4) endorsed electronic health record (EHR) as platform for standardization, (5) endorsed development of TS best practice. Dec 2012 meeting (42 attendees; 12 hospitals): (1) established data system requirements, (2) began system wide TS practice standardization using EHR (TS consult, op note, daily rounding note, discharge summary, clinic note, multidisciplinary thoracic oncology conference note), (3) consensus regarding required components of lung cancer screening program. 2013: Executive committee established best practice component candidates (performance status, clinical staging, lymphadenectomy, etc). Surgeons surveyed for importance of possible components. 38 components chosen for incorporation into EHR templates. Data system and EHR templates developed. Quarterly newsletter informs all TSI stakeholders.

Conclusion
Conclusion: Health care quality can be defined & implemented across a broad geographic area but requires dedicated physician leadership & support. The TSI serves as a model for other regions and systems to define & implement high quality thoracic surgery. Clinical data is required to monitor success.

Background
Despite increasing evidence of safety and efficacy of VATS lobectomy for early stage NSCLC, there has been persistent reluctance for this technique among thoracic surgeons in central Europe. Concerns have existed with regard to oncologic safety and the feasibility and accuracy of hilar and mediastinal lymph node dissection. Only recently the VATS approach has gained some increasing acceptance but no data exist regarding application and indication of the VATS approach for NSCLC in Germany.

Methods
A survey (12 questions) was performed among the thoracic surgical units in Germany. Questions focused on volume, indication, technique and distribution of VATS lobectomy for NSCLC.

Results
The response rate was 65% (25 units). Most units started the VATS lobectomy program only within the last 5 years and sofar performed up to 100 procedures, only 4 centers performed more than 300 procedures. Most centers consider NSCLC stages IA-IIB for a VATS approach, 4 units (16%) approach tumors up to stage IIIA by VATS. All units operate via an anterior approach, most (80%) use 3 incisions. In addition to standard lobectomies, 15 (60%) units also perform anatomical segmentectomies, 2 (8%) centers bronchial sleeve resections and 1 (4%) center pneumonectomies by VATS. 24 (96%) units perform systematic lymph node dissection, 3 centers perform VAMLA (Video Assisted Mediastinal LympAdenectomy). In most units (14, 56%) 2-3 surgeons participate in the VATS lobectomy program.

Conclusion
In Germany, the VATS approach for anatomical resections for NSCLC has been increasingly adopted only within the last 5 years. However, even in 2013, only approximately 15% of all lobectomies are performed by VATS, which is a low rate when compared to the US, Great Britain and some Asian countries.

Background
Stereotactic body radiation therapy (SBRT) is now the standard treatment for elderly patients with inoperable Stage I non-small-cell lung cancer (NSCLC). However SBRT with concurrent chemotherapy may be feasible and effective to selected elderly patients with NSCLC. This retrospective study was aimed to evaluate the safety and tolerability of concurrent SBRT and chemotherapy in patients aged 75 years or older.

Methods
We reviewed the records of 11 NSCLC patients who were 75 years or older when treated with SBRT and concurrent chemotherapy with curative intent from 2009 to 2012. Five patients had T1 tumor, the others had T2 tumors. The median age was 81 years with a range of 76 to 88 years. Eight (72%) patients had chronic obstructive pulmonary disease. The median number of Eastern Cooperative Oncology Group (ECOG) performance status of the cases in the beginning of treatment was 1 with a range from 0 to 2. The median delivered radiation dose was 48 Gy in 4 fractions. Concurrent chemotherapy regimen was carboplatin plus paclitaxel, carboplatin plus docetaxel, paclitaxel alone, pemetrexed alone, and S1 alone.

Results
All patients received SBRT on schedule. Out of 11 patients, concurrent chemotherapy was successfully accomplished as originally planned in 8 (72%) patients. Initial effect of SBRT plus chemotherapy could be evaluated in all cases. Complete local remission was achieved in 10 patients. One patient alone had local recurrence. Distant metastases observed in 4 patients. In 2 patients, chemotherapy was intermitted due to grade 3 neutrophil count decreased and anemia (Common Terminology Criteria for Adverse Events version 4.0). In another, chemotherapy was broken off because of Grade 2 radiation pneumonitis. During the treatment, no other adverse event was shown. No treatment-related death was observed.

	Grade 2	Grade 3
Neutrophil count decreased	6 (55%)	1 (9%)
Anemia	1 (9%)	2 (10%)
Radiation pneumonitis	1 (9%)	0

Conclusion
SBRT plus concurrent chemotherapy might be feasible in selected patients aged 75 years or older with Stage I NSCLC.

Background
Evidence-based Radiotherapy utilisation benchmarks [Delaney G et al, Cancer 2003] have been used as the basis for planning radiotherapy services both nationally and internationally. These utilisation models have been further expanded to estimate the benefit for each radiotherapy indication in individual cancer sites using evidence-based approach [Shafiq J et al, Radiotherapy and Oncology 2007].

Methods
Our study aimed at estimating the benefit of definitive or adjuvant radiotherapy to overall survival and local control of lung cancer patients if the entire lung cancer population are treated according to evidence-based treatment guidelines. The optimal radiotherapy utilization model previously reported for small-cell lung cancer (SCLC) and non-small cell lung cancer (NSCLC) was extended to incorporate overall survival and local control benefit from radiotherapy (radiotherapy vs no radiotherapy, radiotherapy and concurrent chemotherapy vs radiotherapy alone) from published data (1990-2010). Palliative benefits were not considered.

Results
The overall gains in 5-year local control and survival were 11% (95% CI 9.5%-12.3%) and 5% (95% CI 3.6%-6.7%) respectively if optimal radiotherapy was applied for all lung cancer patients. The optimal gains for all lung cancer from concurrent chemotherapy and radiotherapy over radiotherapy alone were 5% (95% CI 1.7%-7.6%) for local control and 4% (95% CI 1.0%-6.0%) for survival. The overall local control benefit for radiotherapy including concurrent chemotherapy for SCLC and NSCLC were 10% and 17%; 5 year survival benefit proportions were: SCLC 2% and NSCLC 10%.

Conclusion
Our model provides a quantitative estimate of benefit of curative radiotherapy for lung cancer at the population level if evidence-based guidelines were applied. The model predicted that guidelines-recommended application of radiotherapy treatment could save 900 extra lives per 10 000 cases of lung cancer for up to 5 years and prevent 1600 local recurrences in that period.

Background
Background: To identify predictive FDG-PET imaging factors for outcomes following Stereotactic Ablative Radiotherapy in early stage Non Small Cell Lung Cancer

Methods
Patients with inoperable T1 and T2 NSCLC and a baseline FDG PET-CT at a single centre (St James Institute of Oncology) treated with SABR for a single tumour between 2009 and 2012 were included. Prospective data was collected on a range of FDG-PET parameters (Tumour SUVmean, SUVmax, Tumour:Liver SUVmax and SUVmean ratios, Tumour:Mean Blood Pool SUVmax and SUVmean ratios and TLG –Total lesion glycolysis). Patient characteristics and outcome variables including stage, histology, PTV volume, performance status, dose, time interval between PET and SABR, response to treatment and patterns of failure collected and analysed. The PET parameters were analysed initially as a continuous variable.

Results
125 patients (72 female, 53 male), median age 75, 94 were T1 and 31 T2. Median follow-up 1.2yrs (range 0.28-3.3). Histology confirmation was possible in 40 patients. Relapse free survival at 2 years was 55%. Local (LR), regional (RR) and distant relapse (DR)-free rates were 94%, 89% and 83% respectively. Overall (OS) and cause-specific survival at 2 years was 57% and 81% respectively. On multivariate analysis, pre-treatment SUVmax was the only PET parameter that consistently predicted for recurrence (p<0.01), DR (p=0.012) and OS (p=0.026). This was consistent amongst patients with and without diagnostic pathology and no statistical difference was found between the two groups. Patients with SUVmax > 17.9 had significantly worse DR and OS (HR 9.99 and 6.68 respectively). SUVmax however did not correlate with LR or RR rates. The only other PET parameter with any statistically significant correlation was the TLG20 which showed some association with RR (p=0.038). Once SUVmax was established as the strongest predictor for outcome this was assessed as a dichotomised variable to assess predictability for DR. Patients with an SUVmax above17.9 were found to have significantly higher rates of DR (p=0.03)

Conclusion
In our population SUVmax is the strongest FDG-PET parameter to correlate with outcome following SABR for NSCLC. Number of events of DR were small and a cut off point predicting for this is difficult to interpret. This is consistent with previously published data. With SABR an emerging treatment modality for early stage disease, this may have future implications on the use of adjuvant chemotherapy.

Background
Hypofractionated radiotherapy (HRT) for pulmonary malignancies has been now commonly used, as a tumorcidal dose can be accurately delivered to the target without any consequential dose to adjacent normal tissues. However, radiation pneumonitis (RP) is still a major problem after HRT. To determine the significant parameters in developing RP, we retrospectively investigated data from patients treated with HRT using helical tomotherapy for lung metastases.

Methods
A total of 45 patients were included in the study and the median age was 53 years old. The median prescriptive doses were 50 Gy to internal target volume and 40 Gy to planning target volume in 10 fractions over 2 weeks. RP was diagnosed by chest X-ray or computed tomography after HRT and its severity was determined by CTCAE version 4.0.

Results
The incidence of symptomatic RP was 26.6%. Univariate analysis showed that mean lung doses, V5, V10, V15, V20 and V25 were statistically significant in developing symptomatic RP (P<0.05). However, only V5 was statistically significant in developing symptomatic RP in multivariate analysis (p=0.019). In the ROC curve, V5 was the most powerful predictor of symptomatic RP, and its AUC was 0.780 (p=0.004). In addition, the threshold value of V5 to develop symptomatic RP was 65%. The large distribution of a low dose resulted in a higher risk of lung toxicity.

Conclusion
To prevent symptomatic RP, we should limit the V5 to less than 65%, in addition to considering conventional dosimetric factors. However, a further clinical study must be done in oreder to confirm this result.

Background
The IAEA/RCA Project “Strengthening the Application of Stereotactic Body Radiotherapy” aims to increase the capacity to deliver SABR in Asia. Lung cancer is one of the most common cancers in Asia. Most patients present with locally advanced or metastatic disease, but increased access to diagnostic scanning will result in earlier detection of lung cancers suitable for SABR. The shorter treatment time is beneficial in reducing the burden of treatment and economic costs of radiotherapy treatment. The clinical efficacy and socio-economic benefits of SABR have led to rapid implementation in the US[1], Europe[2] and parts of Asia[3]. SABR lung demands specialised expertise, physical infrastructure, and a long-term commitment to rigorous quality assurance. There is significant heterogeneity in the resources and expertise between the RCA Member States. This project aims to identify and ameliorate obstacles to the safe and effective implementation of SABR within Asia. Figure 1

Methods
Each country developed a work-plan specific to their capacities and needs. Key strategies for implementation of the project are: Development of treatment protocols Regional training courses Advocacy with Government bodies responsible for policy-making and funding, and education of the wider medical community and public about the benefits of SABR Expert missions with on-site training Publications including educational materials and the results of implementation( utilisation rates, local control and clinical outcomes) Formation of regional training hubs Specific goals identified for Australia were the need to standardise protocols, advocacy for access to funding through Medicare for advanced radiotherapy technologies, and the safe implementation of SABR in regional centres. The Tripartite Collaboration being developed at a national level between RANZCR, AIR and ACPSEM and the NSW SABR Collaboration will address these concerns.

Results
Australia has already contributed physics expertise to facilitate the success of the first regional training course, held in Singapore in December 2012. Sydney will host the final regional training course in 2015 on SABR for lung and spine. We continue to contribute our technical and clinical expertise in developing training materials and resources for the project.

Conclusion
Collaboration between well-resourced and developing countries in Asia is helping to sustainably develop resources and expertise to improve access to SABR for lung cancer patients. This networking provides future opportunities for large scale clinical trials and research in diseases with a high prevalence in the region. 1. Pan H et al. Cancer 2011 117(19):4566 2. Palma D et al. J Clin Oncol. 2010;28(35):5153 3. Teshima T al. J Radiati. Res 53.5 (2012): 710

Background
Stereotactic Ablative Radiotherapy (SABR) is an alternative to surgery in patients with early stage non-small cell lung cancer (NSCLC) inoperable for medical co-morbidities (mainly cardiovascular or respiratory diseases) or who refuse surgery. We performed a prospective evaluation of lung function parameters, treatment-related radiological and clinical toxicity in a cohort of patients treated with SABR.

Methods
We prospectively recruited 26 patients from July 2012 to May 2013. All patients had a histological or cytological diagnosis of NSCLC (n=20) or a lung lesion in dimensional growth with PET positivity (SUV>2.5) (n=6). All patients had stage IA– IB and were judged unfit for surgery or refused it. Each patient did a 4D-TC with slice of 2.5mm/2.5mm, treatment consisted of a single Volumetric Modulated Arc Therapy and the fractionation schedule was dependent on tumor location. Pulmonary toxicity was assessed through the execution of pulmonary function tests and on chest Computed Tomografy (CT). All tests were synchronously performed before treatment and at regular intervals after SABR (the first control at 45 days, then every 90 days until progression). Lung function parameters were obtained performing spirometry, body plethysmography, determination of the diffusion lung capacity of carbon monoxide (DLCO) and arterial blood gas analysis.

Results
Of 26 patients enrolled, 17 performed the first evaluation at 45 days, 5 at 135. At 45 days the total lung capacity (TLC) slightly decreased from 5.87±1.50 Liters (L) to 5.62±1.42 (t=1.87; NS), whereas VC, FEV~1~ and FEV~1~/VC ratio showed minimal changes. At 135 days TLC in the 5 patients who ultimate this step showed a slight recovery to 5.75±1.75 L. The pulmonary diffusion capacity for carbon monoxide (D~L~CO), corrected for hemoglobin (Hb) levels, significantly decreased from 14.4±4.9 to 12.9±5.2 (mL min[-1]·mmHg[-1]) at 45 days (p<0.019) with a slight recovery at 135 day to 13.9±2.7. When D~L~CO was corrected for the measured Alveolar Volume (D~L~CO/VA) the change was not significant. The difference between plethysmographic TLC and the dilution VA (TLC-VA) increased at 45 days from 1.24±0.7 to 1.49±0.8 L, suggesting an increase in ventilation inhomogeneity of the lung. Arterial oxygen pressure decreased from 75.8±7.2 to 71.6±10.4 mmHg (p=0.056 NS) and the variation correlated with TLC-VA (r=-0.72, p<0.001) and DLCO variations (r=-0.67, p<0.03). We observed a low toxicity profile during the first evaluation at 45 days, with only 1 RTOG grade 2 and 1 grade 3 post actinic pneumonia, both treated with systemic corticosteroids. Only three patients reported fatigue as the only adverse event. At the first radiological re-evaluation we didn’t observe any progression disease, with a 59% rate of partial response.

Conclusion
Preliminary findings suggest that no major changes in lung function can be detected at 45 days after SABR. A slight reduction in D~L~CO can be observed, and this could reflect a transitory increase in pulmonary ventilation inhomogeneity caused by RT rather than a direct membrane damage. Study prosecution will hopefully clear the physiopathological evolution at several months after SABR, and further analyses will be carried out investigating for a potential correlation with radiological toxicity and dosimetric profiles.

Background
Central lung tumours pose a challenge for stereotactic ablative radiotherapy (SABR) due to proximity to vital organs and risk of potentially fatal toxicity. RTOG 0813 is an attempt to determine a safe dose for these tumours in an era where many institutions have multiple technologies that can deliver lung SABR. The purpose of this study is to use a novel scoring system to compare two different SABR platforms, robotic radiosurgery (RRS) and linac-based volumetric modulated arc therapy (VMAT), in a cohort of patients actually treated on 0813. The comparison is limited to target coverage and organ-at-risk (OAR) sparing capability for this technically challenging group of patients.

Methods
All 5 patients from our institution accrued to RTOG 0813 were selected for this study. Eight planners (4 VMAT, 4 RRS) with combined experience of >500 lung SABR cases re-planned each case for 60 Gy in 5 fractions. Patient setup, contouring details, and planning constraints were as per 0813. Monte Carlo planning was performed on Monaco v3.20 (Elekta Inc., MI, USA) for VMAT and Multiplan v4.5.0 (Accuray Inc., Sunnyvale, USA) for RRS on CyberKnife. An objective scoring system was designed that included each dose-volume 0813 protocol criterion. For each target requirement or OAR constraint a “structure score” was assigned whereby [Actual Plan parameter /Expected 0813 parameter] X priority factor = structure score. Priority factors (high 0.9, intermediate 0.6, and low 0.3) were assigned by 3 experienced lung SABR radiation oncologists for each of the 5 patients given that different OARs were of greater concern depending on exact target location. A ‘final plan score’ was the sum of all structure scores, with a lower overall score indicating a plan that best achieved target coverage and OAR avoidance in keeping with radiation oncologist priority. To reduce inter-planner bias more than one plan was created for each of the 5 patients using both modalities and only the best plans were selected for comparison.

Results
A total of 15 VMAT and 10 RRS plans were submitted for analysis, each satisfying the minimum 0813 protocol requirements. Using the scoring system, a final plan score was obtained for all 25 plans with a median VMAT score of 8.02 (range 5.52 to 10.09) and RRS score of 7.1 (range 4.98 to 12.41). The lowest scoring VMAT plan was then compared with the lowest scoring RRS plan for each patient. Analysis of target coverage parameters showed that both modalities had similar scores, indicating an equivalent ability to conformally cover the target. RRS plans had lower OAR scores (mean reduction of 1.3) compared to VMAT plans. Overall the plan scores for each patient (RRS: VMAT) were: Patient 1 (6.74:9.2), Patient 2 (6.69:7.32), Patient 3 (4.98:5.94), Patient 4 (7.69:8.92), Patient 5 (5.78:7.36).

Conclusion
When using a scoring system based on RTOG 0813 planning criteria to compare patient plans from two different lung SABR delivery systems, 5 of 5 patients planned using a robotic radiosurgery system had more favourable overall scores compared to VMAT linac delivery for centrally located tumours.

Background
The aim of this study was to compare the GTV volumes drawn manually on CT scans with GTV delineation on FDG PET scans utilizing automatic threshold (SUV 3) and gradient-based (PET Edge) auto-segmentation methods in lung tumors and discuss implications in radiation planning.

Methods
Twenty two patients with lung carcinoma treated with radiation therapy having discrete lesions with no adjacent consolidation or atelectasis and having PET scan done within 30 days of simulation CT were enrolled in this study. FDG-PET/CT and planning CT were transferred to the MIM software and fused using the deformable registration algorithm. For each patient three GTV’s were defined. GTV CT manually contoured on CT scan using lung window in lesions adjacent to lung parenchyma and mediastinal window when adjacent to mediastinum or chest wall. For GTV SUV3, circle of interest was created with a margin around the lesion, excluding blood pool (heart) and auto segmented with SUV value of 3. GTV-PET Edge was auto segmented using PET Edge tool centered on the hyper metabolic area. Statistical Methods: Spearman correlation coefficients were constructed to view relationships between variables, and sign tests were used for inference.

Results
Among 22 patients, only one was small cell and 21 were with non-small cell carcinomas (8 squamous cell, 11 adenocarcinoma and 2 poorly differentiated). As per the AJCC 7[th] Ed, 16-stage I, 3 stage II and 3 belong to IIIA. Median CT volume for all lesions was 4.385, (range 0.68-173.74), PET Edge median 4.235 (range 0.474-113.00), SUVs 3 median 4.845 (0.659-109.1). Correlation between CT and SUV 3, SUV 3 and PET edge, CT and PET Edge were 0.8690, 0.9105 and 0.8585 respectively. No significant differences between CT and SUV 3 volumes (p=0.5235) as well as CT and PET Edge volumes (p=0.3833). But PET edge volumes were significantly less compared to SUV 3 volumes (p=0.0525).

Conclusion
Lobectomy is the treatment of choice for early stage non-small cell lung cancers. In medically inoperable patients stereotactic body radiation therapy has become the new standard of care with 3 year overall survival of 60%, similar to lobectomy. For conventional Radiation therapy GTV’s are expanded to generate CTV for presumed microscopic disease (CTV). In SBRT GTV and CTV are identical. In phantom studies auto segmentation using PET edge tool shown to be superior to other methods and better correlated with pathology. In our study the median volume by PET edge is smaller compared to other two modalities. The difference is statistically significant between PET edge and SUV 3 but not between CT and PET edge. Hence, adaptation of PET edge tool may decrease the GTV and PTV volumes that will enable to spare the normal structures better in SBRT. Surgical pathologic studies determined CTV margins of 6mm for squamous and 8mm for adeno, beyond gross pathological tumor (Giraud et al). CT overestimates GTV volume in lung tumors with no additional or negative margins required to create CTV (Chan et al). We allude that adaptation of SUV 3 may be enough to generate CTV volumes for conventional radiation therapy.

Background
Thoracic lymph node recurrence after complete resection is common in non-small cell lung cancer but it mostly occurs along with distant metastases. The recurrent disease might be localized and curative intent radiation therapy is therefore the treatment of choice if no evidence of hematogenous spreading of the disease is observed. However, the treatment effect and the long-term outcomes of radiotherapy for lymph node recurrences have not been well reported. We sought to describe the treatment effect and the long-term outcomes of radiotherapy for postoperative lymph node recurrences.

Methods
Fifty patients that had developed thoracic lymph node recurrence after complete resection received curative intent radiotherapy between 1997 and 2009. The nodal stage at recurrence was N1 in 10 patients, N2 in 28 patients, and N3 in 12 patients. The diagnosis of lymph node recurrence was based on chest CT, FDG-PET, physiological examination, the value of CEA, and/or bronchoscopic sampling for cytology. The cytological evidence was obtained in 10 (20%) patients. Thirteen patients had symptoms associated with recurrent disease. Patients were treated using 3D conformal techniques. Conventional fractionation was used (2-3 Gy/fraction), and the total prescribed dose ranged from 50 to 80 Gy. The clinical endpoints included the tumor response, overall survival, progression-free survival, locoregional recurrence within the irradiated field, and any other recurrence.

Results
The planned total radiotherapy was completed in 49 patients. One patient refused further therapy at 56Gy/60Gy due to radiation esophagitis. The median follow-up time after radiotherapy was 41 (19-98) months among the survivors. The response to treatment was complete response in 65%, partial response in 24%, and progressive disease in 10% of the evaluated patients. The 1-year, the 3-year, and the 5-year progression-free survival rates were 49.1%, 28.2%, and 22.2%, respectively. The median progression-free interval was 12.0 months after radiotherapy. In ten patients, no additional recurrence was detected for longer than 3 years after radiotherapy. The median overall survival after radiotherapy was 37.3 months. Fourteen patients survived more than three years after radiotherapy. The 5-year overall survival rate was 36.1% (Figure). A multivariate analysis revealed that the absence of symptoms and the involvement of a single lymph node station were significant factors aFigure 1ssociated with a better overall survival.

Conclusion
Radiation therapy for thoracic lymph node recurrence after complete resection is safe and provides acceptable disease control. Early detection of lymph node recurrence may improve better disease control and increase the chance of curing the disease.

Background
The use of MRI for lung cancer volume delineation for radiotherapy is rare. This has been due to poor image quality as a result of physical and physiological factors such as low proton density, susceptibility effects and respiratory and cardiac motion. However as MRI technology has improved, imaging of lung abnormalities has become more feasible. A prospective study was therefore conducted to evaluate image quality for lung tumour delineation on a 1.5T (Tesla) and 3T MRI scanner. The aim of the study was to identify potential scan sequences that could be used clinically for tumour delineation for radiation therapy treatment.

Methods
Ten patients with lung cancer underwent MRI, five on a 1.5T GE scanner using a body phased array coil and five on a 3T Phillips scanner. Scans on the 1.5T scanner were undertaken with breath hold and scans on the 3T scanner were performed with respiratory and peripheral pulse gating to give optimal image quality. The thorax was imaged with T2 and T1 weighted sequences on both field strengths. Cine mode imaging to compare tumour motion was also acquired. Scan sequence was matched for the 1.5T and 3T scanners. The quality of images for lung cancer delineation was assessed by an experienced thoracic radiologist and thoracic radiation oncologist using a four point scale. A consensus score ranging from 1(superior) to 4 (inferior) was given for each sequence based on four categories; tumour edge detection, image artefacts, noise affecting edge detection and overall image quality. A score of 2 or below was considered clinically acceptable

Results
Both magnet strengths provided reasonable image quality to define tumour volume on lung MRI. The average score for overall image quality between the two scanners was 1.8 for 1.5T and 1.3 for the 3T scanner. For the 1.5T scanner the sagittal and coronal T2 weighted scan scored the best for overall image quality for tumour delineation (1.53), due to limited respiratory motion distortion. However these image planes are not compatible with radiotherapy planning systems. For the 3T scanner the axial T2 images scored best for overall image quality (1.05). For tumour edge detection the sagittal and coronal and T1 weighted images scored best (1.75) for the 1.5T scanner. The axial T2 weighted image and the sagittal cine mode performed best for tumour edge detection (1). Overall sequences on the 3T scanner were rated higher than those on the 1.5T scanner

Conclusion
It is feasible to utilise commercially available MRI sequences to acquire images of acceptable quality for the purposes of lung cancer delineation in radiotherapy. Both magnet strengths gave acceptable image quality for clinical use in radiotherapy, with the 3T magnet displaying slightly better image quality. A future study will compare lung cancer delineation between the current standard practice of CT&PET with MRI.

Background
Lung parenchymal recurrent or multiple lobe cancer is typically considered to have metastatic disease and treated with palliative approach such as chemotherapy. However, some of these patients may have multiple primary lung cancer (MPLC) that could be potential curable. Surgical resection has been the standard treatment for early-stage multiple primary lung cancer (MPLC). However, a significant proportion of patients with MPLC cannot undergo surgery. We explored here the role of stereotactic ablative radiotherapy (SABR) for patients with MPLC.

Methods
We reviewed MPLC cases treated with SABR (50 Gy in 4 fractions) for the second tumor. Four-dimensional CT–based planning/volumetric image-guided treatment was used for all patients. Patients underwent chest CT scanning every 3 months for 2 years after the SABR and then every 6 months for another 3 years. PET scans were recommended at 3–12 months after SABR. Toxic effects were scored according to the National Cancer Institute Common Terminology Criteria for Adverse Effects version 4.

Results
For the 101 patients treated with SABR, at a median follow-up interval of 36 months and median overall survival of 46 months, 2-year and 4-year in-field local control rates were 97.4% and 95.7%. 2- and 4-year rates of overall survival (OS) were 73.2% and 47.5% and progression-free survival (PFS) were 67.0% and 58.0%. Patients with metachronous tumors had higher OS and PFS than did patients with synchronous tumors (2-year OS 80.6% metachronous vs. 61.5% synchronous; 4-year OS 52.7% vs. 39.7%; p=0.047; 2-year PFS 84.7% vs. 49.4%; 4-year PFS 75.6% vs. 30.4%; p=0.0001). For patients whose tumors were both of the same histology (meaning that the second lesion could have been a satellite, a metastasis, or a recurrent lesion), the 2-year and 4-year OS rates were 76.4% and 51.2%, which were no different from the OS rates for patients with tumors of different pathology (2-year OS: 66.7% and 4-year OS: 40.5%; p=0.406). The 2- and 4-year OS of patients in whom both tumors were classified as stage I were 76.1% and 55.2%, which was better than the OS rates for the patients whose index tumors were of higher stage (2-year OS 66.7%, 4-year OS 26.6%; p=0.049). For patients whose index tumor was treated with surgery or SABR, the incidence of grade ≥3 radiation pneumonitis was 3% (2/71), but this increased to 17% (5/30) for patients whose index tumor was treated with conventional radiotherapy. Other grade ≥3 toxicities included grade 3 chest wall pain (3/101, 3%) and grade 3 skin toxicity (1/101, 1%).

Conclusion
1. SABR achieves an excellent long-term tumor control and promising PFS and OS in early-stage MPLC. 2. Toxicity could happen but within the scope of SABR in stage I disease. 3. Caution should be taken integrating SABR with prior conventional radiotherapy for stage II/III disease. SABR could be an effective alternative to surgery for curative treatment of early-stage MPLC tumors.

Background
Approximately 35,000 people die from lung cancer each year in the UK, the majority from non-small cell lung cancer (NSCLC). Patients with locally advanced (LA) NSCLC are often not suitable for chemotherapy or combined chemo-radiotherapy treatment because of patient or tumour factors. In these cases radical radiotherapy alone is used. Increased radiation dose may improve both local tumour control and survival. The radiotherapy dose is limited by surrounding organs, which include the lungs, heart, spinal cord and oesophagus. The maximum radiotherapy dose that can safely be delivered to the oesophagus is not known. The I-START trial was therefore developed, on behalf of the UK National Cancer Research Institute Lung Clinical Studies Group, to establish oesophageal radiation dose limits and to investigate the feasibility and effectiveness of a novel approach to dose escalation in LA-NSCLC. The study is funded by Cancer Research UK (C25518/A11535), sponsored by Velindre NHS Trust and coordinated by the Wales Cancer Trials Unit.

Methods
Patients with histologically or cytologically confirmed stage II to IIIb NSCLC, suitable for radical radiotherapy, are eligible for the trial. Enrolled patients will receive 20 fractions of radiotherapy over 4 weeks. The trial is split into two phases: Phase I: To establish the maximum tolerated (MTD) dose of radiotherapy to the oesophagus in patients where the oesophagus overlaps with the planning target volume (PTV). Phase I patients will be split into 2 groups depending on the length of the oesophagus lying within the PTV (Group 1A is where the overlap ≤6.5cm and Group 1B is where the overlap >6.5cm). Cohorts of 6 or 12 patients are recruited to both groups at sequentially increasing dose levels (58, 61, 63, 65Gy). Progression to the next oesophageal dose level will depend on the number of patients in a cohort with grade 3 or 4 acute oesophagitis, or other grade 3 or 4 toxicity, occurring up to 2 months after radiotherapy. Once the MTD to the oesophagus is established for each group, all participants will follow the Phase II protocol with the determined oesophageal dose limit. Phase II: Patients will receive a maximum of 65Gy in 20 fractions and the dose prescribed will be the highest dose achievable without exceeding defined safe dose limits for organs at risk. Where the oesophagus does not overlap with the PTV, patients can immediately be treated in Phase II, whereas patients whose oesophagus overlaps with the PTV can only be entered into Phase II once Phase I is complete, i.e. the MTD to the oesophagus has been established. The primary outcome of Phase II is the toxicity rate (grade 3 or 4) at 3 months. The I-START trial will determine whether this novel method of increasing the radiotherapy dose in patients with NSCLC patients is tolerable, safe and effective. If the results are positive, then this new treatment may be compared against the best currently available standard treatment in a future larger randomised (Phase III) trial.

Results
Not applicable.

Conclusion
Not applicable.

Background
Post-operative radiation therapy (PORT) is frequently given to patients with NSCLC who have microscopically positive margins (R1 resection) or gross residual disease (R2 resection) based on oncologic first principles. However, the data to support this practice is scarce. Here we report our institutional experience comparing patients who received PORT in the setting of R0, R1 or R2 resections.

Methods
Between 1999 and 2012, 203 patients with NSCLC were treated with PORT in 25-39 fractions to 45-70 Gy. All surgery and PORT were performed at our institution. Twenty-one patients had a sublobar resection, 158 had a lobectomy, and 24 underwent a pneumonectomy. PORT was given to R0 patients with pathologic N2 disease, R1 or R2 resections. Patients with negative margins were compared to patients with residual disease (R1 and R2 resections). Patients with tumor recurrence within the PORT field were recorded as local failures. Local failure-free survival (LFFS), disease-free survival (DFS) and overall survival (OS) were calculated using the Kaplan-Meier method. Univariate analysis was performed using the log-rank test.

Results
Fifty-five of 203 patients had residual disease after resection; 45 had R1 and 10 had R2 resections. The predominant histology was adenocarcinoma (80%). Stage at diagnosis was stage I-II in 17 patients, stage III in 186 patients. One-hundred and twenty-six (62%) patients received neoadjuvant and 39 (19%) adjuvant chemotherapy. Median age was 60 years and median KPS before PORT was 80. Median interval from surgery or adjuvant chemotherapy to PORT initiation was 1.6 months (range of 0-3.7 months). With a median follow up of 21 months, local failure occurred in 33/148 (22%) without and 14/55 (25%) patients with residual disease. Two- and 5-year actuarial LFFS rate were 79%/66% for R0 and 75%/58% for R1/R2 resections. Two- and 5-year actuarial DFS rate were 44%/30% for R0 and 48/26% for R1/R2 resections. Two-year and 5-year overall survival (OS) for all patients were 62.4% and 33.1%. LFFS, DFS and OS were not significantly different when comparing R0 to R1/R2 resections. Radiation dose, KPS, T-stage, N-stage, lymphovascular invasion, histology and chemotherapy were all not found to be significantly associated with any endpoint. OS was significantly worse for patients with R2 resection compared to R0/R1 resection (2-year OS 20% vs 64%; p= 0.002) and patients with age >65 (p=0.03). Multivariate analysis will be presented.

Conclusion
PORT results in equivalent local control rates after R1 and R2 resections when compared to R0 resections. This suggests that PORT has a significant role in local control of residual disease. However, OS was significantly worse for patients with gross residual disease postoperatively.

Background
Locoregional failure rates are high in patients with locally advanced non-small cell lung cancer (NSCLC), even when using concurrent chemoradiation. Recurrences have been shown to be predominantly located in the primary tumor, more specifically in areas with a high FDG-uptake that can be identified on a pre-treatment FDG PET-CT scan. Improved tumor control could be accomplished by dose escalation. The PET-boost trial is an ongoing randomized phase II trial investigating radiation dose-escalation using an individualized, accelerated schedule either to the entire primary tumor or to the regions of high FGD-uptake (>50% SUVmax) within the primary tumor. We present a preliminary analysis of the acute toxicity of the first 45 patients.

Methods
Patients with NSCLC stage IB-III with a primary tumor diameter ≥4 cm are eligible. Patients are treated with concurrent or sequential chemoradiation or radiotherapy only. Permitted regimens are: daily dose cisplatin (only in concurrent chemoradiation) or cisplatin-etoposide in concurrent and sequential chemoradiation. Eligible patients receive a planning PET-CT scan on which an IMRT plan is constructed up to a dose of 66 Gy in 24 fractions of 2.75 Gy to the involved lymph nodes and the primary tumor. In patients where normal tissue constraints allow further dose escalation to the primary tumor up to a minimal dose of 72 Gy of ≥ 3 Gy-fractions, 2 plans (with equal mean lung dose) are constructed: either giving the integrated boost to the entire primary tumor (Arm A) or redistributing the boost to areas of high FGD-uptake (>50% SUVmax) in the tumor (Arm B), up to a maximal prescribed dose of 129.6 Gy in 24 fractions of 5.4 Gy. All pts are followed according to study protocol. Toxicity is scored according to the CTCv3.0 criteria. Primary endpoint of this study is local progression-free survival at 1 year. Secondary endpoints are acute and late toxicity, overall survival and quality of life.

Results
Between 2010 and 2013 71 patients were registered of which 45 were randomized: 22 pts to arm A and 23 to arm B. In both arms, median follow up was 13.3 months. Patient and tumor characteristics were equally distributed in both arms. Thirty-seven patients (82.2%) had stage III lung cancer. Concurrent chemoradiotherapy was given in 25 patients (55.6%). Mean GTV was 154.2 cm ³ (range 26-416 cm³). Mean fraction size in both arms was 3.46 Gy (range 3.0-5.4 Gy). Baseline toxicity grade 3 occurred in 4 patients (8.8%) consisting of dyspnea in 1 patient, cough in 2 patients and renal dysfunction in 1 patient. During treatment grade ≥3 hematologic toxicity was seen in 6 patients (13.3%), whereas 2 patients (4.4%) suffered from cardiac toxicity grade 4 (ischemia/infarction). Seven patients (15.6%) had grade ≥3 dysphagia. 82.2% of the patients finished treatment according to study protocol. Radiation treatment was completed in all patients. Seven patients have died of which 3 (6.6%) due to pulmonary hemorrhage.

Conclusion
This first toxicity analysis of the multicenter phase II randomized PET-boost trial at a median follow up of 13.3 months did not reveal any unexpected acute or late toxicity.

Background
Thoracic radiotherapy is associated with significant acute toxicities including oesophagitis, anorexia and fatigue which can impact on the ability to achieve adequate nutritional intake, subsequently leading to weight loss and malnutrition. Malnutrition during cancer treatment is associated with poorer patient and treatment outcomes. Understanding factors associated with weight loss assists with the early identification and intervention of patients at nutritional risk. This study aimed to identify radiotherapy dosimetric factors associated with clinically significant weight loss (greater than or equal to 5%) in patients receiving treatment for non-small cell lung cancer (NSCLC).

Methods
A retrospective analysis of an existing cohort of 54 NSCLC patients treated with concurrent chemoradiotherapy for whom oesophageal dose distributions had previously been calculated. Weight change was calculated at any time point from the start up to 90 days from radiotherapy commencement to determine those with clinically significant weight loss. Chi-squared tests, Pearson correlation, Mann-Whitney U-test and logistic regression were used to examine associations.

Results
Four patients for whom weight was not available at the start or end of treatment were excluded leaving 50 patients for analysis. The prevalence of clinically significant weight loss was 22% (median weight loss 9.1%, range 5.9 – 22.1). Dosimetric factors associated with clinically significant weight loss were maximum dose to the oesophagus (z= -1.99, p=.046), absolute oesophageal length receiving 40Gy (r=.32, p=.03), 50Gy (r=.36, p=.01) and 60Gy (r=.45, p=.001) to the partial circumference, relative oesophageal length receiving 50Gy (r=.32, p=.02) and 60Gy(r=.44, p=.001) to the partial circumference. The odds of a patient receiving 40Gy (median length 10.6cm), 50Gy (median length 10.2cm) or 60Gy (median length 7.2cm) to the partial oesophagus experiencing clinically significant weight loss were 1.18 (95%CI 1.01,1.37, p=.04), 1.20 (95%CI 1.03,1.41, p=.02) and 1.32 (95%CI 1.09,1.60, p=.005) greater, respectively, than those with less oesophagus in the treatment field. Nine (82%) of the eleven patients who had clinically significant weight loss received a dose of 60Gy to at least 5cm of the partial circumference of the oesophagus.

Conclusion
The strongest dosimetric association with clinically significant weight loss was absolute oesophageal length receiving 60Gy to the partial circumference. A previous study identified an association between concurrent chemotherapy and late stage disease (stage III or IV) and clinically significant weight loss. Findings from both studies have been used to develop a model, currently undergoing validation, to assist clinicians in predicting NSCLC radiotherapy patients at high nutritional risk.

Background
To evaluate the clinical outcome and toxicity after radiotherapy using [18F] FDG-PET/CT based target volume delineation instead of conventional CT for locally advanced non-small cell lung cancer (LA-NSCLC).

Methods
Between January 2007 and May 2011, 248 patients with stage IIIA or IIIB NSCLC were referred to our institution for radiotherapy. Among them, twelve patients were excluded, including four patients upstaged to stage IV after PET/CT and eight patients lost follow-up. The remaining 236 patients were investigated, including 67 patients received RT with PET/CT-based planning and 169 patients with CT-based planning. Propensity-score matching (PSM) method was utilized to obtain two matched groups (1:1) with similar baseline clinical characteristics. Kaplan-Meier method was conducted to calculate the overall survival (OS), progression-free survival (PFS), and locoregional control (LRC). Log-rank test was performed to compare the survival between two groups. Response to initial treatment was also compared, as well as acute radiation toxicity, especially the radiation-induced lung injury (RILI).

Results
Two matched groups with comparable baseline clinical characteristics were created using the PSM method, including 67 patients with PET/CT-based planning (PET/CT group) and 67 patients with CT-based planning (CT group). The covariates that we selected for the propensity score model were gender, age, Charlson comorbidity index, Karnofsky performance status, smoking history, FEV~1~ (%), VC (%), location of the primary tumor (central or peripheral), presence of atelectasis, histology type, T stage, N stage and TNM stage. Patients in the PET/CT group received a higher total radiation dose than those in the CT group (60.7 Gy vs. 59.0 Gy, P = 0.067). After initial treatment, 55 (82.1%) patients in the PET/CT group and 41 (62.1%) patients in the CT group achieved complete remission (CR) or partial remission (PR), respectively (P = 0.010). After a median follow-up of 24.7 months, a significant superior locoregional control was observed in the PET/CT group compared with the CT group (P = 0.036), with a 2-year LRC of 54.2% and 40.1%, respectively. Log-rank test showed no statistical difference in respect to OS (P =0.319) and PFS (P =0.582) between the two groups. However, patients in the PET/CT group tended to have a slightly longer median survival time (25.4 mo vs. 21.7 mo) and median progression-free survival time (12.7 mo vs. 10.7 mo). Patients in the PET/CT group were less likely to develop grade ≥ 2 RILI than those in the CT group (9.0% vs. 19.4%, P = 0.083). Other acute toxicities were all comparable between the two groups.

Conclusion
Using PET/CT instead of CT for target volume delineation in radiotherapy for LA-NSCLC was associated with a better LRC and lower incidence of grade≥2 RILI. A possible explanation was that the more accurate target delineation in PET/CT-based planning allowed for an escalated total dose delivered to a relatively small target volume.

Background
Although loco-regional recurrences and second primary lung tumors are not uncommon following prior high-dose thoracic radiotherapy, only a minority of patients undergo reirradiation. Reirradiation performed at short intervals, and to low total doses, is generally associated with median overall survival (OS) of only 5-7 months. Few studies report outcomes following high-dose reirradiation. We describe institutional experience after high-dose, conventionally fractionated reirradiation.

Methods
High-dose conventional reirradiation was defined as fraction sizes of 2-3Gy and minimum total dose of 39Gy. A retrospective chart review of patients treated between Feb 2004-Feb 2013 was performed. Where possible, overlap in planning target volumes (PTV) and radiation doses were determined. New primary tumors were defined as new histology or reirradiation interval ≥5 years.

Results
24 patients were identified, 13 (54%) had recurrent disease, and 46% a new primary. Most (63%) had stage III NSCLC at initial and second treatments; median reirradiation interval was 51 months, and median follow-up from reirradiation 19.1 months. Median overall survival (OS) after reirradiation was 13.5 months, with 1-year survival 51%, median local progression-free survival (LPFS) 14.1 months and median distant progression-free survival (DPFS) 18.5 months. One-year disease-free survival was 47%. Three patients died from bleeding (2/3 had high-dose overlap in the mediastinum, of whom one had prior hemoptysis and was anticoagulated, the 3rd patient had extensive endobronchial therapy prior to reirradiation). Other post-retreatment toxicity was uncommon. The size of the second PTV (median 250cc) was prognostic. OS was 17.4 versus 8.2 months for patients with a 2nd PTV <300cc and >300cc respectively (p=0.02). Differences in DPFS (p=0.007) and for DFS (p=0.03) were also significant. LPFS was shorter when reirradiation interval was <24 months (p=0.02), however it was not different when groups were defined by the median interval of 51 months. Magnitude of PTV and dose overlap between the two treatments did not influence survival. Figure 1 Figure 1: Example of reirradiation for a new primary lung cancer. Planning target volume (PTV) and dose-cloud shown from treatment in 2004 (A, 23 fractions of 2.6Gy) and 2010 (B, 33 fractions of 2Gy) and the overlap of both treatments (C).

Conclusion
High-dose, conventionally fractionated reirradiation for new primary or recurrent lung cancer can deliver meaningful survival, especially for patients with a smaller PTV at the time of reirradiation. A shorter reirradiation interval may be associated with less chance of loco-regional control. Prospective studies are needed to confirm these findings, and establish reliable normal tissue tolerances for reirradiation.

Background
With the increasing penetration of stereotactic body radiotherapy (SBRT) into cancer practice, there is growing interest in applying this technique to patients with limited metastatic disease. We reviewed our single institution experience using SBRT in the treatment of oligometastatic pulmonary disease.

Methods
A retrospective review was performed to identify patients treated at our institution with SBRT for pulmonary metastases between 3/2008 and 1/2013. Treatment decisions were multidisciplinary and a biopsy was performed when feasible. The gross tumor volume (GTV) was non-uniformly expanded to create an internal target volume (ITV) to encompass tumor motion based on 4-dimensional computed tomography (CT). A 5 mm uniform expansion of the ITV was then applied to create the planning target volume (PTV). The most common dose/fractionation schedule was 48-50 Gy in 4-5 fractions. Cone beam CT was used for daily image guidance. Overall survival (OS), time to distant failure (TTDF), and local control (LC) were estimated from the end of the first SBRT procedure using the Kaplan-Meier method. Toxicity was scored based on CTCAEv4. Median follow up was 15 months (range 3-60).

Results
64 patients underwent 66 SBRT procedures to treat 74 lesions. There were 36 males (56%) and 28 females (44%) with a median age of 71 years (range 42-90). The most common primary disease sites were lung (n=23; 36%), colorectal (n=10; 16%), melanoma (n=9; 14%), and head and neck (n=5; 8%). The target lesion represented the only site of metastatic disease in 32 patients (50%); 20 patients (31%) had additional pulmonary metastases but no extra-thoracic disease; 12 patients (19%) had both pulmonary and non-pulmonary sites of metastases. Median lesion size was 2.3 cm (range 0.6-6.8). Median, 1-year, and 2-year OS was 21 months, 73%, and 49%, respectively. Distant metastatic disease progression was observed in 37 patients (58%) at a median time interval of 12 months. Patients (n=52) with no extra-thoracic disease at the time of SBRT had a significantly longer TTDF compared to patients (n=12) with concurrent extra-thoracic disease (median 13 vs. 5 months; 1-year failure rate 47 vs. 73%; p = 0.02) without a difference in OS (median OS not reached vs. 20 months; 1-year OS 75 vs. 63%; p=0.1). Local failure was observed for 4 lesions resulting in an 18-month LC rate of 88%. There were 11 toxicities observed in 10 patients including fatigue (n =1, grade 1), dyspnea (n=1, grade 1), dermatitis (n=1, grade 2), rib fracture (n=1, grade 2), and pneumonitis (n= 5, grade 2; n=2, grade 3).

Conclusion
SBRT is a reasonable treatment for patients with pulmonary metastases. High rates of local control can be achieved with acceptable toxicity. Only 2 patients (3%) developed grade 3 pneumonitis (oxygen indicated). In this high risk population, new distant metastatic progression was common, especially among patients who present for SBRT with known extra-thoracic sites of disease at the time of SBRT. However, among patients with solitary lung lesions or oligometastatic disease limited to the chest, survival was encouraging. We will continue to utilize SBRT in this population of oligometastatic disease with careful patient selection.

Background
Cavitation in NSCLC has anecdotally been associated with worse prognosis and survival. It has been postulated that rapid tumour growth and disordered angiogenesis leads to necrosis, cavitation and central hypoxia, with associated reduction in tumour radiosensitivity. This study investigated central photopenia on FDG-PET imaging (indicating reduced metabolic activity in necrotic areas) as an independent prognostic factor in patients with NSCLC treated by definitive radiotherapy.

Methods
Pre-treatment PET images for 172 patients (single institution, enrolled in multicentre prospective observational study TROG 99.05) with pathologically proven stage I-III NSCLC planned for definitive radiotherapy (minimum 50 Gy in 20 fractions) or chemoradiotherapy (93.4%) were independently analysed by two investigators at two time points one month apart. The presence and percent of tumour demonstrating central photopenia were scored using visual assessment. Central photopenia was defined as a non-peripheral volume within the primary tumour with an SUV <40% of the SUVmax. Survival was adjusted for known prognostic factors including performance status, T and N stage.

Results
Inter-observer agreement for the presence (Kappa=0.822, p<0.001) and proportion (intraclass correlation coefficient (ICC)=0.913 with 95% CI, 0.862-0.946, p<0.001) of photopenia and were good. Intra-observer reassessment showed fair agreement (Kappa=0.600, p<0.001 and ICC=0.752 with 95% CI, 0.634-0.836, p<0.001). Photopenia, present in 43% of the tumours was associated with larger tumour volumes (mean volume 197.3cm[3 ]vs 56.2cm[3 ]in solid tumours, p<0.001) and weight loss (51% of patients vs 31%, p=0.029). Median survival was shorter in the presence of photopenia but there was no significant difference in overall survival (OS) on univariate (HR=1.25, 95% CI, 0.89-1.77, p=0.200) or multivariate analysis (HR=1.14, 95% CI, 0.80-1.61, p=0.465) after adjusting for stage and performance status. Correlations of OS and the percent of photopenia were not statistically significant. Figure 1

Conclusion
The presence and proportion of photopenia on FDG-PET imaging can be reproducibly measured using visual analysis without requiring specialist workstation software. Central photopenia was associated with larger tumours, weight loss, and shorter median survival, but there was no statistically significant effect on overall survival after adjusting for known prognostic factors.

Background
Aim/Objective: Radiotherapy for Stage IIIA and B non-small cell lung cancer at Nepean Cancer Care Centre (NCCC) consists of a 3D conformal radiotherapy (3D CRT) plan dosed to 60Gy in 30 fractions. However, due to the size and location of many stage IIIA and B NSCLC lesions radical radiotherapy is not always clinically deliverable, with the limiting factor often being healthy lung toxicity. In these cases the prescribed dose may be reduced to 50Gy, or the treatment intent altered to palliation. One solution to achieve treating these lesions with a radical intent at NCCC is a hybrid (Volumetric Modulated Arc Therapy) VMAT planning technique. We present a comparison planning study of the Hybrid Volumetric Modulated Arc Therapy (VMAT) vs. 3D Conformal Radiotherapy (3D CRT). Hypothesis: It is expected that the hybrid VMAT technique will achieve better coverage of the Planning Target Volume (PTV) by 95% (57Gy) of the prescribed dose (PD) in addition to reducing the combined lung V20 (volume of lung receiving 20 Gy) and V30 (volume of lung receiving 30 Gy) doses when compared to 3D CRT technique.

Methods
Method: This study was conducted retrospectively on 3 patients diagnosed with NSCLC lesions staged IIIA or B previously treated at NCCC. The original 3D CRT plans were used for comparison. Gross Tumour Volume/ Clinical Target Volume (GTV/CTV and PTV) contours have remained unchanged. Original 3D CRT utilised on average 4 static beams, 8mm field margins, and dosed 60Gy in 30 fractions. The hybrid VMAT technique consists of AP/PA static beams, 8mm field margin only dosed to a portion of the prescribed dose, the remaining dose delivered using a partial VMAT arc. The two techniques were compared using PTV coverage, V20 < 30% and V30 <20% of the combined lung volume and mean lung doses.

Results
Results: Average results from the first two planning comparisons show coverage of the PTV by D95% is achieved with both hybrid VMAT (96.6%), and 3D CRT (99.2%). However improvements in coverage by D100% are seen using the hybrid VMAT technique compared to 3D CRT (88.7% vs. 79.5%). Hybrid VMAT combined lung doses were reduced, average results V30 = 19.5% and V20 = 24.8% compared to 3DCRT 23.9% and 30% respectively. Mean lung doses using the hybrid technique were also lower in comparison to 3D CRT, 15.6Gy vs. 18.5Gy, reduced on average by 2.9Gy. Average contralateral mean lung dose also followed this trend with hybrid VMAT 7.3Gy vs. 3D CRT with 9.4Gy average reduction by 2.01Gy.

Conclusion
Conclusion: Hybrid VMAT is a viable treatment option to achieve radical treatment intent to 60Gy in 30 fractions, adequate PTV coverage by D95% whilst obtaining acceptable V20 and V30 lung doses for patients with stage IIIA and B NSCLC.

Background
Hypofractionated image guided radiotherapy of extracranial targets has become increasingly popular as a treatment modality for inoperable patients with one or more small lesions, often referred to as Stereotactic Ablative Body Radiotherapy (SABR). Our institution is using SABR for lung, liver, spine and kidney tumours and is the lead in a multicentre clinical trial of radical SABR for early stage lung cancer. Current and future trends in patient safety and quality assurance (QA) programs are towards ensuring patient safety using the most efficient methods. There is limited published work on patient specific QA for lung SABR treatments on which to base risk management QA programs. Thus, we have performed a review of the first two years of lung SABR patient specific QA process with the aims of highlighting specific areas of uncertainty in lung SABR delivery with the aims of improving efficiency and effectiveness of our QA program. This presentation will detail the results of the review and the evolution of the QA program to a risk-management based approach.

Methods
SABR involves one or few fractions of high radiation dose typically delivered in many small fields or arcs. Tight margins are often applied to mobile targets through heterogeneous tissue density with non-coplanar beams. We have conducted thorough QA for individual patients similar to the more common IMRT QA with particular reference to motion management. Individual patient QA was performed in a Perspex phantom (Modus Medical) using a point dose verification and radiochromic film for verification of the dose distribution. The results for the first 33 plans were analysed with the aim of revising QA procedures for future lung SABR plans. The results from these plans were then used to highlight particular areas of delivery uncertainty which require attention during patient specific QA.

Results
While individual beams could vary by up to 7%, the total dose in the target was found to be within ±2% of the prescribed dose for all 33 plans. The QA process verified all aspects of the plan delivery including non-coplanar geometry, isocentre accuracy under couch rotation and internal target volume construction. The QA process highlighted the importance of accounting for couch transmission and demonstrated the need for accurate motion management strategies. The review of the first 33 plans lead to the creation of a risk-management based approach to QA of subsequent treatment plans. Particular emphasis is now placed on verification of small field dosimetry and motion management strategies for lesions with large motion.

Conclusion
QA is essential for complex radiotherapy deliveries such as SABR. We found individual patient QA helpful in setting up the technique and understanding weak points in the process chain. Ongoing review of the patient specific QA results has lead to improvements in efficiency in the process, facilitating a risk-management based approach to patient specific QA for SABR.

Background
Hypo fractionated radical radiotherapy using 55 Gy in 20 fractions is one of the most commonly used radical radiotherapy regimens in the UK. The shorter overall treatment time has the radio-biological advantage of minimal re-population. Despite its popularity in UK, there is paucity of supporting data in the reported literature. We present the mature outcome data of this regimen based on the single centre experience in North Wales Cancer Centre,UK.

Methods
A retrospective case note analyses of Non-Small-Cell Lung Cancer patients who underwent radical radiotherapy using 55Gy in 20 fraction (fraction size that 2.75 Gy/fraction), between 2001 and 2011, was carried out. We included the patients who had the total dose reduced to 52.5 Gy in 20 fractions to allow for normal tissue constraints. Patients receiving concurrent chemoradiation and adjuvant radiotherapy after surgical resection were excluded. The records were updated using the national registry of death in Wales. Univariate analyses of the effects of different subgroups on overall survival were performed.

Results

Table 1. Stage and Histological distribution

		No of patients
stage	IA	27
	IB	35
	IIA	8
	IIB	30
	IIIA	52
	IIIB	66
	IV	1
	Unknown	3
Histology	NOS	37
	Non-Squamous	44
	Squamous	95
	Unknown	46

Total 222 patients (128 males, 94 females) underwent radical radiotherapy between 2001 and 2011. Median follow up period was 267 weeks (range 74-637). Mean age at diagnosis was 68 yr (41-91). The stage and histological distribution is shown in table 1. Of the stage III patients 81 had induction chemotherapy before the radiotherapy. The median survival of the whole group was 124 weeks (95% CI 105-141) and the survival by histology and stages is shown in table 2. The good 2 yr but poor 5 yr OS of the Stage I patients may reflect their poorer general health at presentation that rendered them medically inoperable. Table 2. Overall survival by Stage and Histological Subgroups

		Median OS in Weeks (95% C.I.)	2 yr OS	5 yr OS
stage	stage 1	135 (116-210)	68%	28%	p= 0.47
	stage 2	135 (80-264)	58%	42%
	stage 3	105 (85-141)	50%	27%
Histology	NOS	73 (60-102)	30%	10%	p= 0.0004
	Non-Squamous	131 (114-216)	67%	25%
	Squamous	130 (97-162)	57%	35%
	Unknown	191 (131-325)	67%	38%

Conclusion
Radical Radiotherapy using 55 Gy in 20 fractions is a highly effective form of treatment for early and locally advanced stages of Non-Small-Cell Lung Cancer with results comparable to isofractionated radiotherapy published in the literature.

Background
Radiotherapy (RT) for non-small cell lung cancer (NSCLC) is associated with important side effects with acute esophagitis (AE) as one of the main acute toxicities. RT dose and concomitant chemo-radiotherapy as well as volume of oesophagus treated are known risk factors for development of AE.

Methods
This is a multicentre national protocol on RT for locally advanced NSCLC. From 2009-2013 117 patients were randomized between 60 Gy/ 30 F (arm A) and 66 Gy/ 33 F (Arm B), 5 FW. Navelbine[®] 50 mg 3 days a week was given concomitant with RT. Before randomization patients were treated with 2 cycles of carboplatin and Navelbine[®] as induction chemotherapy. During RT, patients were registered for side effects once a week. After RT follow up schedule was every 3[th] month from RT start. Side effects were registered according to NCI CTCAE version 3.

Results
A total of 117 patients were randomized in this protocol. Baseline characteristics are summarized in table 1. Since last patient was randomized August 2013, all CRF are not available for the moment. There are complete esophagitis data on 103 patients. Grade 0-1 AE were seen in 66 patients (37 in arm A vs. 29 in Arm B), grade 2 in 27 patients (11 vs. 16), 10 patients experienced grade 3 (3 vs. 7), and none grade 4 AE. In a logistic regression analysis with N2/3, age ≥64 years, histology, gender and dose as covariates; treatment arm B was the only significant covariate (p=0.02) for developing grade 2 esophagitis. Dose volumetric data will be ready for WCLC.Figure 1

Table 1		Number	%
Performance status	0	59	50
	1	57	49
	missing	1	1
Gender	Male	68	58
	Female	49	42
Histology	Squamos cell carcinoma	41	35
	Adenocarcinoma	54	46

Conclusion
From this study we conclude, that the risk of developing grade 2 esophagitis is related to dose 66 Gy but the severity is manageable. Acknowledgements Supported by CIRRO- The Lundbeck Foundation Center for Interventional Research in Radiation Oncology.

Background
During conventional radiation therapy, treatment image guidance is largely indirect relying on slow acquisition 3D volumetric imaging or the use of bony surrogates. Fiducial marker placement within/adjacent to lung tumours facilitates image guided radiation therapy by …….. Marker placement has been attempted percutaneously but is associated with pneumothorax in up to 45%, with frequent use of chest drain tubes. Furthermore, in-treatment imaging protocols are not standardized, and the impact of marker characteristics on accuracy of in-treatment imaging has not previously been reported. We describe our preliminary experience in bronchoscopic implantation and in-treatment tracking/imaging of two different types of lung fiducial marker.

Methods
Study design: Prospective observational case series of NSCLC patients undergoing radical radiation treatment . Bronchoscopic implantation: performed under conscious sedation using radial probe endobronchial ultrasound and fluoroscopic guidance to achieve tumour localization and placement within/adjacent to peripheral tumours. Post-implantation/ in-treatment imaging: Time-resolved 4D CT (Philips Brilliance+bellows system) for treatment planning and after completion of treatment to investigate marker movement. Throughout treatment delivery MV electronic portal images (EPI) were acquired plus kV planar and Cone Beam CT (CBCT) (Varian Medical System) images.

Results
Four patients with T1N0 NSCLC underwent bronchoscopic implantation of fiducial markers (two using Visicoil[TM] linear fiducial 10x0.75mm, two using SuperDimension® superLock™ 2-band 13x0.9mm markers. Confirmation of tumour localization was achieved with EBUS in all four patients. Two markers were placed in adjacent airways in one patient, and the remainder had a single marker placed within/adjacent to their peripheral tumour. No complications related to bronchoscopy or marker implantation were observed. No marker migration was observed over the treatment time for both marker types. Visibility of the markers in EPI was only possibly in selected beam directions though they were easily discernible in kV planar images (Figure 1a). While diagnostic CT scanning was able to demonstrate the markers in great clarity (Figure 1b), they caused significant image artefacts in CBCT. Figure 1 Figure 1: Image-guided radiotherapy images demonstrating: a) 4DCT image showing visicoil fiducial on maximum intensity projection images, tumour+motion contoured in red, & b) kV orthogonal image showing superLock™ 2-band marker.

Conclusion
Our preliminary experience indicates bronchoscopic implantation of fiducial markers is safe, and is achievable with a high degree of accuracy on initial imaging, and stability on subsequent in-treatment imaging. There is a fine balance of marker size minimising CBCT artefacts while allowing visualisation in EPI imaging which would be an ideal tool to verify gated radiotherapy delivery.

Background
Consideration of respiration-induced motion based on 4DCT for lung cancer yields individualized margin. The purpose of our study was to quantify the gain from 4DCT based radiotherapy in lung cancer and to identify the tumor characteristics of better benefit.

Methods
51 patients with 52 lung tumors who underwent 4DCT based radiotherapy were included. GTVs on 10 respiratory phases were contoured by single radiation oncologist. Internal gross tumor volume (IGTV) was obtained by combining the GTVs at ten phases of the respiratory cycle. No separate margins were used to account for microscopic tumor extension. Additional isotropic setup margin of 3mm to derived the PTV-4D from the IGTV. PTV-3D was generated by adding a isotropic margin of 10mm to the GTV in a single phase (20%) for upper or middle lobe tumor, and a margin of 10,10,15mm in later, AP and SI direction for tumor in lower lobe. The target coverage on PTV-4D and PTV-3D were compared respectively. Using linear regression, clinical and anatomic factors for the reduction of PTV-4D vs PTV-3D were identified.

Results
PTV-4D was significantly smaller than PTV-3D, by 89cm[3] on average. Approximately 17% PTV-3D (in 9 of 52 tumors) were not fully covered PTV-4D with up to 2%-18% slices lost. For the cases of target missed in PTV-3D, tumors were either irregular or had a larger mobility of ≥ 1cm in SI direction. For the cases of target encompassed in PTV-3D (43 tumors), the volume reduction of PTV-4D vs PTV-3D was associated with the GTV size and the tumor movement in SI direction. Taken median GTV (48cm[3], the corresponding diameter of around 4.5cm) as a cutoff, big or small tumors had an average PTV reduction of 141 cm[3 ](77-304 cm[3]) or 42cm[3] (11.5-107 cm[3 ]), respectively.

Conclusion
4D plan has a more accurate and safe target coverage than that of empirical estimated margins in 3D plan. Patients with tumor character one of irregular shape, big size or high-mobility can better benefit from 4DCT based radiotherapy. It will be of clinical significance when administration a higher dose to a bulky tumor at equal normal tissue constraint, or high-precision radiation was delivered. A small benefit in a large PTV should be paid more attention especially when 4D based radiotherapy transforming a palliative intent to a curative one.

Background
Radical radiotherapy, with or without chemotherapy, is a treatment modality for early and late stage inoperable lung cancer. We investigated patients treated with curative intent over the course of one year.

Methods
Patients who received radical radiotherapy for lung cancer from Dec 2010 to Nov 2011 at the Beatson West of Scotland Cancer Centre were reviewed retrospectively. Information was gathered from the patient clinical records and the radiotherapy planning records (ARIA). All patients with lung cancer who received radical conventional radiotherapy (a total dose of 45- 66Gy) with or without sequential or concurrent chemotherapy were included. Patient demographics, disease characteristics, radiotherapy planning treatment parameters and overall survival were analysed using standard statistical methods.

Results
A total of 179 lung cancer patients received conventional radical radiotherapy. Median age 70 years, 55% were female, with ECOG 0 25%, ECOG 1 62%, ECOG 2 13%. Patients Deprivation Score as per Scottish Index of Multiple Deprivation score 2012: 44% are 1 (most deprived), 26% are 2, 13% are 3, 9 % are 4 and 8% are 5(least deprived). 22%(39/179) are stage I, 19%(34/179) stage II, 58%(104/179) stage III, and 0.6%(2/179) are stage IV. 51%(91/179) squamous cell carcinoma, 25%(44/179) adenocarcinoma, 7%(13/179) small cell carcinoma, 12%(22/179) had no histology and 5%(4/179) were neuroendocrine and others. 99% completed their prescribed radiotherapy. The mean whole lung V20 and whole lung V5 were 22.1% and 49.4% respectively. Mean PTV was 409.6 ml. 62/179(35%) died within the first year of receiving their radical radiotherapy. Out of those who died 33/62 PTV <500ml, 26/62 PTV 500-1000ml, 3/62 PTV>1000. 26/62 were treated with radiotherapy alone and 36/62 had received chemoradiotherapy. On univariate analysis using log-rank, age (p=0.663), sex (p=0.437), performance status (p=0.403), deprivation score (p=0.133), stage (p=0 .117), V20 (p=0.084) and V5 (p=0.064) were not significant factors affecting survival. PTV (p=0.020) and chemoradiotherapy (p=0.001) were statistically significant variable affecting survival. On multivariate analysis using Cox-regression PTV remained significant with respect to overall survival.

Conclusion
The outcome of this audit demonstrates that PTV is a significant variable affecting the survival of the patients in the first year after treatment. It was of interest to us that clinical stage, whole lung V20 and V5 were not significantly associated with survival. We also observed a significant one year survival difference in patients who were treated with radical radiotherapy alone compared with chemoradiotherapy. Overall survival was also not associated with deprivation score with survival outcomes similar across the whole population. The radiotherapy PTV may be an important variable when deciding treatment independent of other radiotherapy planning parameters.

Background
Radiation pneumonitis is a common cause of morbidity and is a radiation dose-limiting toxicity in patients with locally advanced NSCLC treated with definitive radiotherapy. NSCLC patients are increasingly receiving intensity modulated radiation therapy (IMRT), in part in an attempt to reduce irradiation doses to organs at risk, like the lungs. It is unclear whether the incidence of pneumonitis in IMRT patients differs from that in patients receiving the more commonly available 3D conformal radiation therapy (3DCRT). This retrospective study reports on outcomes at the University of Pennsylvania.

Methods
All consecutive patients with non-metastatic locally advanced NSCLC treated with curative intent at the University of Pennsylvania between January 2003 and October 2011 with 3DCRT (n=208) and IMRT (n=58) were graded for post-treatment radiation pneumonitis using Common Toxicity Criteria (CTC) and SWOG criteria in this IRB-approved study. Any short- or long-course initiation of prednisone for dyspnea 1-10 months following treatment was scored as grade 3 pneumonitis. Associations between type of treatment and clinical and demographic factors and outcomes were assessed using Χ[2] and non-parametric equality-of-medians tests. Logistic regression was used to determine predictors of pneumonitis.

Results
Patient characteristics, including age, gender, race, marital status, smoking history and pulmonary function were well balanced across treatment groups (p = 0.18 – 0.97). Patients also had similar tumor TNM-stage, histology, differentiation, and involved lobe (p = 0.26 – 0.62). No differences in the proportion receiving concurrent chemotherapy, radiation prescription, lung V5, lung V20 and mean lung doses, or planning target volume (PTV) were identified (p >0.05 for all). Pneumonitis rates did not differ between the IMRT and 3DCRT patients. In 3DCRT patients, 54% had grade 2+ and 27% had grade 3+ CTC pneumonitis compared to 59% and 34% in IMRT patients, respectively (p >0.05). Additionally, 45% and 17% of 3DCRT patients had grade 2+ and 3+ SWOG pneumonitis compared to 45% and 16% in IMRT patients, respectively (p> 0.05). Lower lung lobe, single marital status, pack-years smoked, low pre-treatment pulmonary function DLCO status, increased lung V5, increased lung V20 and mean lung dose all associated with pneumonitis on univariate logistic regression for both 3DCRT and IMRT patients (p <0.05 for all). A multivariate analysis was performed but yielded no significant results.

Conclusion
Our findings suggest that treatment with IMRT is associated with similar rates of pneumonitis as treatment with 3DCRT for locally advanced NSCLC patients treated with definitive radiation therapy. As opposed to treatment modality, several factors were identified that were associated with pneumonitis risk, included lung lobe, marital and smoking status, pre-treatment lung function, and irradiation dose received by the lung. Thus, care should be taken to limit the lung V5, lung V20, and mean lung doses when administering curative-intent radiotherapy, regardless of treatment with IMRT or 3DCRT.

Background
Acute radiation pneumonitis (SARP) is a limiting factor on treating lung cancer with radiotherapy. Radiation dose to the lung was used to predict the occurrence of SARP, but the data were from American or European people, and might not suitable for Chinese people. This study was to identify predictive valueof different dosimetric parameters for SARP based on Chinese people.

Methods
147 NSCLC patients treated with concurrent chemotherapy and 3D-CRT between 2006 and 2010 was collected. Radiation pneumonitis(RP) was diagnosed according to RTOG criteria. Grade 3 or even severe RP was defined as SARP. Logistic dose response model was established, and Lyman - Kutcher - Burman normal tissue complication probability(NTCP) model was fitted. The predictive value of model was explored.

Results
The incidence of SARP is 9.5% (14/147). MLD, V20, V30, V40, and V50 (P = 0.017, 0.025, 0.010, 0.009 and 0.027, respectively) are determining factors for SARP.Our datasets shows that for SARP < 5%, MLD, V20, V30 V40 and V50 should be ≤ 16.77 Gy, V20 ≤ 34.15%，V30 ≤ 23.62%，V40 ≤ 18.57%，V50 ≤ 13.02%. ROC analysis show that areas under MLD, V20, V30, V40 and V50 curves is corresponding to 0.678, 0.661, 0.667, 0.677, and 0.651, respectively. In addition, the sensitivity and specificity of each parameter at cutoff values are: 78.0% and 48.1% for MLD; 42.9% and 82.0% for V20; 78.6% and 52.9% for V30; 71.4% and 61.7% for V40, and 57.1% and 67.7% for V50.As predictive value of each parameter alone is relatively week, using two or more parameters to predict SARP is recommended. By logistic regression, tumor locations is a determined factor for SARP. (P = 0.020, B = 2.042 95%CI= 1.121- 3.374). The incidence of SARP is greater in patients with tumors in right lower lung than other locations (22.2% vs 6.7%, P = 0.023). The best fit parameter value for logistic dose response model is shown as follow: b0=-6.66, b1=0.2520, TD50=26.43Gy, γ50=1.67. The fit curve shows that when MLD<17Gy, it is similar to QUANTEC curve. When MLD amount to about 17-18Gy, the curve becomes sharper, which implies that probability for SARP increases. The best fit parameter value for LKB-NTCP model is volume factor: n = 0.87 ± 0.40, slope factor: m = 0.27 ± 0.10, and radiation dose cause more than 50% complication TD50 (1) = 29.5 ± 8.0 Gy. Logistic regression and ROC analysis show that NTCP value is a determined factor for SARP.（Logistic regression: P=0.013；ROC Area under curve: 0.707，P=0.019）.

Conclusion
MLD, V20, V30, V40 and V50 are determining factors for SARP. As predictive value of each parameter alone is relatively week, using two or more parameters, or using NTCP to predict SARP is recommended. Patient with tumor in right lower lung were at higher risk to develop SARP. The predictive values of dosmetric parameters are better in patients with tumor in upper lobes. The models show that when MLD amounts to about more than 17Gy, the curve becomes sharper and SARP odd increases. Limit MLD under 17Gy if possible is recommended for chinese patients in the clinic.

Background
Non small cell lung cancer with brain metastases are usually associated with poor outcomes and survival and about 40-50% of all patients with lung cancer develop brain metastases during the course. The poor outcomes and relapses following WBRT alone indicate a need for new therapeutic options. As some patients with NSCLC have mutations in the EGFR and treating with WBRT with the anti-EGFR agent erlotinib in patients of NSCLC with brain metastases may benefit the patients in terms of disease regression and possible improvement in quality of life. Temozolomide has been already used alone or in combination with radiotherapy in the treatment of primary brain tumors and there are few studies showing its benefits in metastatic brain with WBRT. In this study we test the feasibility and efficacy of both of the drugs with WBRT.

Methods
A total of 12 elderly patients of biopsy proven adenocarcinoma lung with brain metastases (on MRI) were analyzed from July 2010 to March 2012. All the patients were planned for palliative radiation of 30 Gy/10#/2 weeks to local disease with WBRT of 20 Gy/5#/1 week with concurrent Temozolomide@ 75mg/m[2] followed by assessment for further therapy after 3 weeks of radiation. All the patients were evaluated 3 weekly for assessment of symptom relief and improvement or progression. After 3 weeks all the patients were started on Erlotinib@ 150 mg, daily and Temozolamide@ 150-200 mg/m[2], D1-5, 4 weekly.

Results
The patient's age ranged from 58 to 75 years. All the patients completed the scheduled radiation with oral steroids. Five patients progressed and died between 3 and 10 months. One patient defaulted during the radiation therapy and another after completion of 6 cycles of oral chemotherapy. Only 5 patients could complete the 12 cycles of oral chemotherapy with Temozolamide and Erlotinib, 4 weekly. Only two patient needed dose reduction of Erlotinib to 100 mg due to grade III rashes in 3rd cycle. The patients who improved after local and brain RT have shown to tolerate the further oral chemotherapy. These patients are still alive with the metastatic disease.

Conclusion
The combination of Erlotinib with temozolomide appears to be a promising therapy for treating brain metastases in NSCLC in terms of tolerability and efficacy. Further studies need to be done in our set up of patients.

Background
Half of Non-small cell lung cancers are diagnosed in locally advanced stage (LA-NSCLC) and warrant multidisciplinary treatments including chemotherapy, radiation, and surgery (S) in a not well-defined combination and sequence. We compared tolerance and effectiveness of different combos of inductive chemo (iCT) or chemoradiotherapy (iCRT) followed by S or consolidative radiation (RT)

Methods
We retrospectively reviewed 108 consecutive LA-NSCLC diagnosed in our center between October-2004 and June-2012 and treated with: 1) iCRT+S (N= 24); 2) iCT+S (N= 31); 3) iCRT+RT (N= 36); 4) iCT+RT (N= 17). Their tolerance, response, and outcome were statistically compared. Survival of five patients that progressed during inductive therapy was not analyzed

Results
Mean age of the patients was 66.2 years-old, 92% were male, and 85.1% ECOG-0. Histology was squamous carcinoma in 71.3%, non-specified NSCLC in 15.7%, and adenocarcinoma in 12%. iCT included platin-doublets with taxanes, vinorelbine, and gemcitabine. CBDCA-combinations were commonly used in elderly patients (15.6% vs. 31.8%, p= 0.001). Grade 3-4 toxicity was observed in 14.8% of inductive therapies, without significant differences between iCT and iCRT arms (p= 0.976). No patient interrupted therapies due to toxicity. Progression rate was higher with iCT than iCRT (8.3% vs. 0; p= 0.023). S was performed in 51 patients (pneumonectomy 30%, bi/lobectomy 56%). Severe S complications appeared in 13.7% of cases. Three patients in the iCRT+S arm died due to early postoperative complications. Complete pathologic responses were higher with iCRT than iCT (25% vs. 11.5%, p= 0.049). Resected patients presented better disease free (DFS) and overall survivals (OS) than those definitively radiated (27.9 vs. 12 months, p= 0.000; and 37.8 vs. 25.9 months, p= 0.009). Higher DFS and OS was found among patients of the iCRT+S arm (p= 0.000 and p= 0.049, respectively)

Conclusion
Those LA-NSCLC that achieved S after inductive therapy presented a better outcome that those non-resected. iCRT+S was tolerable, feasible, and obtained the higher response and survival rate of our series, although these results are biased by the better prognosis of resectable patients. Anyway, prospective trials are warranted to confirm the benefits of triple multidisciplinary approach. Figure 1. DFS and OS Kaplan-Meier curves of patients according to the treatment arm. Figure 1

Background
Nimotuzumab, a humanized IgG~1~ monoclonal anti-EGFR antibody, is approved and widely used in patients (pts) with head and neck cancer or malignant glioma in combination with radiotherapy (RT) in several countries. In previous clinical studies, nimotuzumab has demonstrated a very mild and low incidence of skin toxicity compared to other anti-EGFR antibodies. On in-vitro and in-vivo experiments using NSCLC cell lines, nimotuzumab showed a radio-sensitizing effect.

Methods
This open-label, multicenter phase II study evaluated the tolerability and efficacy of nimotuzumab in combination with concurrent CRT in pts with unresectable locally advanced NSCLC. All eligible pts received concurrent thoracic RT (60 Gy, 2 Gy/day, 6 weeks from day 1) and 4 cycles of chemotherapy (cisplatin 80 mg/m[2] on day 1, vinorelbine 20 mg/m[2] on days 1 and 8) once every 4 weeks as scheduled. Nimotuzumab (200 mg) was administrated once a week from cycle 1 to 4. The primary endpoint was tolerability in combination with concurrent CRT, which was measured by the percentage of pts who completed 60 Gy of RT within 8 weeks, completed 2 cycles of chemotherapy and received more than 75% of nimotuzumab.

Results
Between June 2009 and May 2010, 40 pts were enrolled from 7 institutions in Japan, and 39 eligible pts received the study treatment. The pts characteristics (n = 39) were as follows: 62 years (median); male/female, 34/5; stage IIIA/B, 21/18; PS0/1, 25/14. Thirty-four pts (87%) met the criteria for treatment tolerability, and 38 pts (97%) completed 60 Gy of RT within 8 weeks. Infusion reaction, >grade 3 skin rash, >grade 3 radiation pneumonitis, or >grade 4 nonhematological toxicity were not observed. The 2-year overall survival rate for the 39 pts was 76% (95% CI; 59-87%). The median PFS was 16.7 months; and 30 pts were alive at the cutoff date (Nov 2011). The 1-year PFS rate for pts with squamous cell carcinoma (Sq; n = 16) was 75%, while that for pts with non-squamous cell carcinoma (non-Sq; n = 23) was 41%. In terms of the first relapse site, in-field relapse rates were low for both Sq (3/16; 19%) and non-Sq (3/23; 13%). However, the distant relapse rate was significantly higher for non-Sq (15/23; 65%) than that for Sq (2/16; 13%) (p<0.01, chi-square test with Yates correction).

Conclusion
Addition of nimotuzumab to the concurrent CRT in this setting was well tolerated with clinical benefit to the patients. The low in field relapse rates may be attributed to the radio-sensitizing effect of nimotuzumab. These findings warrant further clinical evaluation of nimotuzumab/cisplatin/vinorelbine/RT in a phase III trial.

Background
Series on aggressive local treatment in selected patients with oligometastatic non-small cell lung cancer (NSCLC) are mostly retrospective, and prospective data are scarce (De Ruysscher et al, JTO 7:1547-1555, 2012). Although a precise definition is lacking, ‘oligometastatic NSCLC’ is considered an intermediate biologic state of restricted metastatic capacity with a limited number of metastases. The turning point between oligometastatic and polymetastatic is merely based on personal opinion and situated somewhere between 1 and 5 distant metastases. In the absence of clear definitions or clinical practice recommendations, a treatment decision is mainly driven by the opinion of each local multidisciplinary tumor board (MDTB).

Methods
As the consideration of and the treatment modality for oligometastatic NSCLC is a controversial area in respiratory oncology, in preparation of a recent dedicated workshop, we simulated a MDTB with international experts in the field. Multiple disciplines from 7 different centers participated in the MDTB, including pathology (1), nuclear medicine physician (1), thoracic surgery (3), radiation oncology (3), and respiratory oncology (3). Participants were asked to assess an electronic file describing 10 clinical ‘oligometastatic NSCLC’ cases, with 2 simple questions per case: 1. Do you consider this case ‘oligometastatic’ (Yes/No) and 2. What is your preferred treatment proposal.

Results
A full response was returned by all 11 specialists taking part in the simulated MDTB. Only 1 case was considered ‘oligometastatic NSCLC’ by all MDTB members. The presented cases were considered by a median of 78% (range 36-100%) of responders as true oligometastatic disease. Despite the fact that each responder gave only one treatment proposal, a median of 4 different treatment proposals (range 2-6) was made per case. Except for brain metastases, most team members would treat the locoregional thoracic disease before the distant metastases. No preference towards neo-adjuvant or adjuvant chemotherapy could be found. The option for surgery or radiation therapy as part of a combined modality treatment was mainly driven by the physicians’ preference.

Conclusion
Our simulated MDTB shows that oligometastatic NSCLC is an entity with many unanswered questions, and thus a major challenge for clinicians. Patients with oligometastatic NSCLC are in the need of 1. discussion at an experienced multidisciplinary tumor board to select patients for a radical combined modality approach; 2. multidisciplinary prospective research protocols to set better definitions of oligometastic NSCLC, evaluate the validity of a radical approach, and to optimize therapeutic modalities.

Background
Neoadjuvant chemotherapy followed by surgical resection is uniquely permits assessment of the in vivo response to therapy in patients with IIIA non-small cell lung cancer. Studies of neoadjuvant chemotherapy often focus only on those who are ultimately resected. We describe an “intention-to-treat” analysis of sequential patients with stage IIIA adenocarcinomas receiving neoadjuvant chemotherapy with intent of surgical resection.

Methods
Using natural language processing software, we searched the electronic medical record at Memorial-Sloan Kettering Cancer Center for “neoadjuvant,” “preoperative,” or “induction” in physicians’ notes. Cases were limited to those with stage IIIA lung adenocarcinoma deemed resectable by a thoracic surgeon and treated with neoadjuvant chemotherapy (without radiation), including those enrolled in prospective clinical trials. Event-free survival (date of diagnosis to recurrence, relapse or death) and overall survival (date of diagnosis to death) were assessed using Kaplan-Meier methods.

Results
From 2007 until 08/2012, 129 patients were identified. Median follow up is 25 months (range 1-76). The patient details are described. 94/129 (73%) were treated with cisplatin-based therapy.Figure 1 The CONSORT diagram below describes the treatment patients ultimately received. Figure 2 The median EFS and OS were 16 (95% CI 13-22) and 44 (95% CI 36-NA) months. OS at 1, 2 and 3-years were 77%, 55%, and 32%. EFS plateaued at 23%, estimating the rate of cure. Overall survival strongly favored surgical resection over salvage radiation (HR=0.5, 95% CI 0.16-1.05).

Conclusion
These data provide unique perspective on the outcomes of patients with IIIA adenocarcinoma treated with neoadjuvant chemotherapy with intent of surgical resection. EFS and OS compare favorably to historical outcomes in this stage of disease, demonstrating the value of the neoadjuvant approach. The inferior survival in patients treated with radiation as a “salvage” approach emphasizes the recommendation for definitive concurrent chemoradiation in those unlikely to be resectable.