Virtual Library

Background:
Characteristics of CT image, including tumor size and component were correlated to survival. However, most studies discuss the survival impact of image characteristics in early clinical stage populations. No articles were discussed the issue from the view of pathology stage, ie. actual disease presentations of non small cell lung cancer patients who presented as resectable disease. The aim of study was to analyze radiologic and pathologic findings of non small lung cancer patients who received curative resection in order to clarify he clinical correlation between image characteristics and survival impaction.

Method:
From 2010 January to 2014 May, 440 patients who underwent curative resection, ie. anatomic resection and mediastinal lymph node resection, were included and medical record were reviewed retrospectively. Chest CT characteristics, including tumor size and consolidation-tumor ratio, were re-evaluated and re-measured by radiologists. The correlation between image and pathology characteristics and its survival impaction were analyzed.

Result:
We identified tumor size presented in CT and pathologic measurement were highly coincidence. (p-value<0.001) The presentation of pure GGO and GGO predominant lesion were adenocarcinoma ( p< 0.001) Majority of these lesions ( 61/76, 80.2 %) were well differentiated. ( p<0.0001). In addition, pure GGO lesion has less risk for visceral pleura (p < 0.0001), angiolymphatic invasion (p < 0.002), and lower risk with N1 lymph node metastasis (p = 0.004) In addition, we found excellent disease free and disease free survival for patients who was identified GGO predominant lesion with size ≤ 2cm. (Figure 3) The difference of disease free (p= 0.0032, log rank test; Figure 3A) and overall survival (p= 0.0066, log rank test; Figure 3B) of these four subgroups was statistically significance. Figure 1



Conclusion:
Non small cell lung cancer patients who presented with ground glass opacity were corrected with well differentiated adenocarcinoma. In addition, less visceral pleura invasion, less angiolymphatic invasion, and less intrapulmonary lymph node metastases risk were also identified. Our results showed lung cancer patients who presented as GGO predominant lesions and size less than 2 cm may had excellent disease free and overall survival.

Background:
Perfusion CT parameters have showed promising results for evaluation of tumor response and follow-up. However, clinical implications in patients with lung cancer have been relatively limited due to the fact that perfusion CT is performed while the patient is breathing and high radiation dose. Therefore, the purpose of this study was to evaluate the effects of respiratory motion correction on perfusion assessment of lung cancer in a reduced dose perfusion CT using a 16-cm coverage scanner.

Method:
Eighty four patients with primary lung cancer who underwent a reduced dose perfusion CT (80 kVp, 80 mA) were enrolled in this study. Each perfusion CT study included 25 repeated dynamic CT scans obtained using the volume perfusion software and was reconstructed with hybrid iterative reconstruction at a strength level of 50%. Two observers measured blood flow (BF), blood volume (BV), and permeability of the entire tumor with and without the use of non-rigid registration algorithm. Single-measure intraclass correlation coefficients (ICC) were calculated and interobserver reproducibility for parameters obtained with two different manners were assessed through Bland-Altman analyses.

Result:
CTDIvol obtained with each of the dynamic perfusion CT in this study was 36.16 mGy and the estimated mean effective dose ranged from 2.02 mSv to 4.0 mSv. Using non-rigid registration, all ICC values for BF, BV, and permeability were increased (0.982~0.994 to 0.988~0.997) compared with those obtained before application of non-rigid registration. All ICC values of lower lung tumor (0.991~0.998) or tumor ≤ 3 cm (0.989~0.998) were higher than those of upper lung tumor (0.925~0.984) or tumor > 3 cm (0.975~0.996). Using non-rigid registration, all 95% limits of interobserver reproducibility were narrowed compared with those obtained before application of non-rigid registration, regardless of tumor location or tumor size, except those of BV for tumors located in upper lung and tumors > 3 cm.

Conclusion:
Perfusion assessment of lung cancer using a reduced dose perfusion CT scan is clinically feasible and application of respiratory motion correction using non-rigid registration can reduce measurement errors.

Background:
Outcomes in Non-Small Cell Lung Cancer (NSCLC) are poor but heterogeneous, even within TNM stage groups. To improve prognostic precision we aimed to develop and validate a simple model for the prediction of overall survival (OS) using patient and disease variables.

Method:
The study population included 1458 patients from three independent cohorts. Associations between baseline variables and OS were estimated in a derivation cohort from a prospective single-centre study (n=695) using Cox proportional hazards regression. Points were allocated to variables based on the strength of association to create the Lung Cancer Prognostic Index (LCPI). Model performance was assessed (by a c-statistic for discrimination and Cox-Snell residuals for calibration) in two independent validation cohorts (n=479 and n=284).

Result:
Three disease-related and six patient-related variables were found to predict OS: stage, histology, mutation status, performance status, weight loss, smoking history, respiratory comorbidity, sex and age. Patients were classified according to predicted LCPI score. Two-year OS rates according to LCPI in the derivation and two validation cohorts respectively were 84%, 77% and 68% (LCPI 1: score≤9); 61%, 61% and 42% (LCPI 2: score 10-13); 33%, 32% and 14% (LCPI 3: score 14-16); 7%, 16% and 5% (LCPI 4: score ≥15). Predictive performance (Harrell’s c-statistics) were 0·74 for the derivation cohort, 0·72 and 0·71 for the two validation cohorts.

Conclusion:
The LCPI contributes additional prognostic information which, in conjunction with other validated tools and evidence based management guidelines, may be applied to counsel patients, guide clinical trial eligibility, or standardise mortality risk for epidemiological analyses.

Background:
Lung cancer screening with low-dose computed tomography (CT) is better than chest X-rays but is non-specific. Accuracy is improved with positron emission tomography (PET), at a cost of additional radiation. We had previously reported on simulated low-dose PET imaging and demonstrated that 10x10[6] net true counts is sufficient to generate images with acceptable diagnostic quality. We now hypothesize that we can maintain image quality with a 92% reduction of fluorodeoxyglucose (FDG) tracer activity from 6 mCi to 0.5 mCi.

Method:
Nine patients have been scanned with two sequential PET/CT scans on the same day. The patient is first scanned with 0.5 mCi FDG and a low-dose CT protocol, followed by a routine PET/CT with 6 mCi FDG. PET data from the standard-dose scan were manipulated to emulate various noise (dose) levels, corresponding to nine pre-defined true count levels. Data were matched to the level of the low-dose scan, to compare noise statistics to a ground truth and to directly validate our methods. The data were reconstructed, with many independent noise realizations, and the images were reviewed. Ten lesions, in seven patients, were identified as having the size and uptake consistent with those found in early disease. For a given count level, the corresponding images were determined to be acceptable if lesion detectability was comparable to that found in the full-statistic image set. Detection performance was determined automatically by machine learning, namely, convolution neural networks trained by 4 previous observer responses.

Result:
Lesion detection accuracy was evaluated in 4458 total image sub-volumes. Regions containing both target lesions (2627 samples) and healthy lung background (1831 samples) were used to assess sensitivity and specificity for the task at all noise levels. The table shows data across the 4 observer models.

Mean Sensitivity & Specificity by True Count Groups

True count/ millions	<0.5	0.5-1	1-2	2-5	5-10	10-20	>20
Mean sensitivity/%	0.35	18.85	62.35	85.40	95.73	96.23	96.42
Mean Specificity/%	6.64	6.93	20.43	66.54	93.55	96.87	98.25

Conclusion:
Low-dose PET can provide good performance for lesion detection within the true count range 5-10×10[6].

Background:
The incidence of primary lung cancer in developing countries has increased and it constitutes the most common cause of cancer related mortality. Some radiological features can be predictive for different histologies and have prognostic utility. One of the features of imaging in the Fleischner Society guideline is common occurrence in the upper lobe with predilection for the right lung.

Method:
A retrospective audit was done to compare the characteristics of primary lung cancer in patients in our environment. Images and patient reports were obtained from our local database archives or from the hospital records. The location of the primary was defined as peripheral if located in the outer third of the lung parenchyma on axial CT or PET/CT images, and defined as central if located anywhere else. Lobe location was also recorded for each primary.

Result:
We reviewed 42 patients with histologically confirmed lung cancer that were referred for PET/CT imaging. The age range was 37- 80 years. Sixty nine percent (n=29) were males 31% (n=14) were females. The most common histologic subtype was adenocarcinoma (48%), followed by squamous cell lung cancer (40%), and small cell lung cancer (12%). Most of the primary pathology were in the right lung (60%, n= 25) as compared to the left lung (40%, n=17). The upper lobes were more commonly affected (64%, n=27), followed by the lower lobes (33%, n=14), and right middle lobe (3%, n=1). Sixty percent (n=25) of tumors were centrally located, of which 64% were of the histologic subtype squamous cell ca. Our findings also showed that 26% (n=11) of patients had metastases to ipsilateral lobes, and 24% (n=10) had metastases to the opposite lung. There were 22% (n=22) of patients that had only nodal metastases and 12% (n=5) had both nodal and visceral metastases. Nineteen percent of patients (n=8) had no evidence of nodal or visceral metastases.

Conclusion:
Our findings conform to those found in literature that primary lung cancer involves the right lung more often and involves the upper lobes more commonly. It also showed that squamous cell lung cancers are mostly central and most adenocarcinomas are peripheral in location.

Background:
Aim of the study was to correlate epidermal growth factor (EGFR) mutation status with computed chest tomography (CT) image patterns at the diagnosis in patients with lung adenocarcinoma.

Method:
Patients diagnosed between January 1, 2015 and March 3, 2017 with available CT and EGFR mutation status were analyzed. The maximal diameter of the nodules, shape (mix, solid) , presence of an air bronchogram and calcification, specula, incision, round shape, central necrosis, pleural connection, UH pre e post were evaluated.

Result:
87 patients (52 males, median age 66 years, range 43-87) were enrolled. 9, 36 and 42 patients had stage II, III and IV, respectively. EGFR mutations were found in 22 patients (25.3%; 1 exon 18, , 15 exon 19 deletions and 6 exon 21 mutations) and were more frequently among females (45.2% versus 18.6% males, p=0.020) and never smokers (77.8% vs 11.8% smokers, p<0.001). The median maximal diameter of the nodules was smaller in mutated patients (30.5, range 15-74 mm) than in wild-type patients (45.0, range 13-110 mm; p=0.010). Tumors with exon 21 mutations were smaller than wild type tumors (p=0.014). Mutations were statistically more frequent in solid pattern (30.6% vs 0 mixt, p=0.010). A higher number of mutations, not statistically significant, were also found in presence of tumors with regular margins (31.9% vs 17.5% spiculati, p=0.14), no incisions (30% vs 5.9% with incision, p=0.059), no central necrosis or calcification (26.3% vs 14.3% with central necrosis or calcification, p=0.67), no calcification (26.3% vs 14.3% with calcification, p=0.67). Median tumor density was 35 UH in wild type vs 40.0 UH in mutated tumors (p=0.29) and 70 UH in wild type vs 80 UH in mutated tumors (p=0.17), before and after contrast, respectively.

Conclusion:
Adenocarcinomas with EGFR mutations were associated with smaller tumors and with solid pattern.

Background:
There is no consensus on the best program for postoperative follow-up and surveillance after curative resection for non-small cell lung cancer (NSCLC) patients at this time. We reported that the diagnostic capability of [18]F-fluorodeoxyglucose positron emission tomography/computed tomography (FDG-PET/CT) was high in postoperative NSCLC patients without clinical and radiological evidence of recurrence (ICVTS 2012). However, there were some problems, such as radiation exposure and its cost. The aim of this study was to propose a more feasible follow-up and surveillance program using FDG-PET/CT for asymptomatic postoperative NSCLC patients.

Method:
Between January 2005 and December 2013, a total of 172 NSCLC patients (98 males and 74 females; mean age: 67.8 years) underwent potentially curative operations, and follow-up FDG-PET/CT was performed in patients without clinical or radiological evidence of recurrence for at least once a year in principle. A total of 481 FDG-PET/CT studies were entered and retrospectively reviewed.

Result:
The histological type was adenocarcinoma in 129 patients, squamous cell carcinoma in 36 and others in 7. The pathological stage was Stage 0 in 2 patients, IA in 106, IB in 28, IIA in 15, IIB in 7, and IIIA in 14 (UICC 7[th]. edition). Surgical procedures were pneumonectomy in 5 patients, lobectomy in 139, segmentectomy in 14, and partial resection with a sufficient margin in 14. The mean time interval between the initial surgery and latest follow-up was 49.1 months. During the follow-up period, the mean number of times that FDG-PET/CT was performed per patient was 2.8. FDG-PET/CT correctly diagnosed recurrence in 31 of 32 (96.9%) patients and 41 of 42 (97.1%) recurrent sites. The sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV), and accuracy were 96.9, 97.1, 88.6, 99.3, and 97.1%, respectively. In 6 (3.5%) patients, other diseases were detected and treated appropriately. On the other hand, FDG-PET/CT performed within 3 years after the operation detected significantly more FDG-positive lesions compared to after 4 years (10.0 vs. 2.0%, respectively, p=0.007). Additionally, in 147 patients with recurrence after curative operations between 2005 and 2013 in our institute, recurrence occurred in 123 patients (83.7%) within 3 years.

Conclusion:
FDG-PET/CT has a high capability to detect recurrence in asymptomatic NSCLC patients after a potentially curative operation. However, it might be sufficient to perform FDG-PET/CT up until 3 years after the operation. A large scale multi-institutional randomized control trial is warranted to ascertain the benefit of surveillance with FDG-PET/CT.

Background:
“Lung cancer associated with cystic airspaces” is a rare radiological entity that is becoming more frequently encountered on imaging studies and has been gaining more attention since the widespread use of CT for lung cancer screening. The goal of this retrospective study is to investigate and correlate clinical, imaging, histopathological and molecular findings in patients presenting with this type of lung cancer.

Method:
Between January 2014 and April 2017, 13 patients presented at the Multidisciplinary Thoracic Oncology Tumour Board with this rare entity. Clinical, histopathological and molecular data were collected and imaging studies were reviewed for the presence of emphysema, size, morphologic classification and findings on [18]F-FDG-PET.

Result:
Median age at the time of diagnosis was 69 years (53-86 years) with a male/female ratio of 8:5. Ten out of 13 patients were smokers. Eleven patients (11/13) had no previous oncological history. Two patients with previous oncological history both had a history of head-and-neck and stage IA lung cancer. Imaging findings showed emphysema in 7 cases. Four patients had type I, 1 patient type II, 4 type III and 4 type IV morphology. Median diameter for the type I, II, III and IV lesions was 20 mm (17-43), 20 mm, 60 mm (25-67) and 46 mm (37-77) respectively. Lesions were more frequent in the right upper (4/13) and lower lobe (4/13). FDG-PET-scan was available in 11 patients and showed high uptake in all patients who presented with a solid aspect of the associated tumour. Four patients (4/13) presented with stage IV at diagnosis. Other stages varied: IA (4/10), IB (1/10), IIA (1/10), IIIA (2/10) and IIIB (1/10). Adenocarcinoma was found in 11 patients (11/13) and squamous cell carcinoma in 2 (2/13). Molecular genotyping for EGFR, ALK and ROS-1 was available in 10 patients (10/13). None of the patients showed positivity for ALK-immunohistochemistry. A single patient showed an exon-18 mutation in the EGFR gene. One patient showed a translocation at the 6q22 breakpoint of the ROS-1 gene. In one patient, a p.Gly469Ala B-RAF mutation was present. Two patients showed a mutation in the exon 2 of the KRAS gene (exon2 c/35G>C;p.Gly 12Ala and c/35G>T;p.Gly 12Val).

Conclusion:
In this series, 5 out of 10 patients with a “lung cancer associated with cystic airspaces” showed a molecular alteration. This suggests that targeted molecular profiling is mandatory in this subgroup. Larger series are needed to confirm these findings.

Background:
Visualizing the spread of cancer cells in lung cancer surgery is sometimes difficult. γ-Glutamyl-transpeptidase (GGT) is a cell surface-associated enzyme that is overexpressed in various type of human cancers. γ-Glutamyl hydroxymethyl rhodamine green (gGlu-HMRG), an activatable fluorescent probe, is non-fluorescent under a neutral pH and normal cellular environment. However, it becomes highly fluorescent upon reaction with GGT. We evaluated ex vivo fluorescent imaging of lung cancers using the GGT-activatable fluorescent probe.

Method:
Between April 2011 to November 2014, 116 resected cancer cells (91 primary lung cancers, 21 pulmonary metastases, and 4 pleural disseminations) were prospectively included in this study. Each tumor was analyzed by first taking a baseline image before gGlu-HMRG was sprayed onto the freshly resected specimen (termed N0; fluorescent intensity of normal lung, T0; that of lung cancer), and then by taking fluorescent images 30 min after spraying (N30 and T30) with the Maestro In-vivo imaging system (PerkinElmer Inc.). Positive fluorescent activity was defined as follows: in cases where fluorescence was observed only in tumor tissues, ΔN(=N30-N0) < 0 and ΔT(=T30-T0) < 0, in cases where fluorescence was observed in both normal and tumor tissues, ΔN > 0 and ΔT/ΔN > 1.

Result:
Figure 1In primary lung cancer, 61 of 91 (67%) cases rapidly developed fluorescent activity. In cases with pulmonary metastases, 15 of 21 (71.4%) cases showed positive fluorescent activity. Four disseminated pleural nodules all showed positive fluorescent activity (100%). Age, gender, tumor size, tumor marker, histology (adenocarcinoma (Ad) vs. non-Ad, squamous cell carcinoma (Sq) vs. non-Sq), pleural invasion, and angio-lymphatic invasion were not significant factors influencing fluorescent intensity.



Conclusion:
Fluorescence imaging with gGlu-HMRG may become one of the most powerful tools for accurate staging by rapidly detecting cancer cells and thus become highly useful for cancer resection.

Background:
We analysed the correlation between the new pathologic classification of lung adenocarcinoma and radiologic findings of early invasive pulmonary adenocarcinomas versus preinvasive lesions appearing as PGGN on HRCT, and evaluated the values in the diagnosis of pathologic classification of lung adenocarcinoma with PGGN on HRCT.

Method:
Retrospective analysis of 123 lesions (16AAH, 35AIS, 35MIA, 37IA) with PGGN on HRCT with T1N0M0 lung adenocarcinoma or AAH from January 2014 to June 2014 in shanghai chest hospital. There were 93 females and 30 males, with a median age of 58 years old. Statistical relationship between the 2015 World Health Organization Classification of the lung adenocarcinoma and radiologic findings of PGGN were analyzed, then sceened out the best predictors, created a modal and verified it.

Result:
The Pearson correlation coefficient( P<0.05) between pathological types and all CT scan morphologic features showed a significant correlation. The logarithm linear correlation cient showed the CT feathures(lobulation, spiculation, pleural indentation, aterial gathering, bubbles/air bronchogram, shape,margin,internal uniformity) had a positive correlation with pathological types excluding tumor-lung interface. These scale variables as maximum lesion area on CT scan, lesion size in cranial-caudal direction, average density of lesion and the corresponding lung’s average background density were significant correlation with pathological types. Multinomial logistic regression analysis showed that the best predictors were spiculation, internal uniformity, lesion size in cranial-caudal direction,average density of lesion, gender in turn. Then the multinomial logistic regression model was built, a likelihood ratio test showed that 70.7% of the cases were classified correctly overall, and the predicted value of AAH was up to 92.9%. Figure 1



Conclusion:
The HRCT characteristics of PGGN were significant correlated with the new pathologic classification of lung adenocarcinoma. The pathologic types of PGGN should be evaluated by HRCT, and the best predictors were speculation, internal uniformity, lesion size in cranial-caudal direction, average density of lesion and gender.

Background:
Indeterminate lung nodules detected during lung cancer screening with low dose computed tomography (CT) present a challenge in distinguishing between malignant and benign disease. Our aim for this study is to compare the sensitivity and specificity of flour-deoxy-glucose positron emission tomography (FDG-PET), volume doubling time (VDT) and a combination of both in the diagnostic workup of indeterminate lung nodules in lung cancer screening.

Method:
The Danish Lung Cancer Screening Trial (DLCST) is a randomized controlled trial with heavy smokers between 50-70 years of age. The screening group underwent 5 annual rounds with low dose CT scan of the thorax. When a lung nodule was detected, participants could either be referred to diagnostic workup, 3-month follow-up or continue with the screening program. We included participants who had a 3-month follow-up scan. Before the follow-up scan was conducted, the participants received a FDG-PET scan. Nodules that were resected or stable for at least 2 years were included. FDG-uptake was categorized from most likely benign (uptake less than background uptake in the mediastinum) to most likely malignant (uptake as mediastinum or higher). VDT was calculated from nodule volume measurements from two time points closest to the FDG-PET scan date. We used a commercially available and validated semi-automated nodule evaluation software. Based on VDT the nodules were divided into three groups. Regressing nodules (VDT<0), slow growing nodules (VDT > 365 days - less likely malignant) and fast-growing nodules (VDT < 365 days – most likely malignant). Finally, we divided combined outcome into three groups: 1. Both tests suggest benignancy, 2. One of the tests suggest malignancy and 3. Both tests suggest malignancy. We used receiver operating characteristic (ROC) curves to compare sensitivity and specificity for the ability to predict a malignant or benign nodule.

Result:
A total of 87 lung nodules in 76 individuals were included. 68/87 (78%) were solid nodules. 41.5% were malignant. Nodule size ranged between 5 mm – 20 mm in largest diameter. The sensitivity and specificity of VDT alone were 61% and 90% respectively. For FDG-PET the sensitivity was 62% and specificity was 90%. Combined use of both tests showed an improvement in test sensitivity to 82% and a specificity of 79%.

Conclusion:
Combined use of FDG-PET and VDT is recommended in the diagnostic workup of indeterminate lung nodules in lung cancer screening.

Background:
Primary lung cancers show more complicated outline on computed tomography (CT) images of the chest than metastatic lung tumors or benign inflammatory lung disease. This feature is proving to be useful for clarifying the diagnosis of pulmonary nodules. The degree of complexity of pulmonary nodules seen on chest CT images is usually described subjectively by such designations as spiculated, irregular, and so on. Up to now there have been no established methods for evaluating a tumor outline numerically. In this study, we applied fast Fourier transform (FFT) analysis to pulmonary nodules on CT images to evaluate variations in the outlines of primary versus metastatic lesions or benign inflammatory lesions.

Method:
Sequential cases of 72 histologically proven primary lung cancers (group PL), 54 metastatic lung tumors (group MT), and benign inflammatory nodule (group BN) were included in the study. The average ± SD diameters of tumors in groups PL, MT, and BN were 18.9±7.4 mm, 12.2±6.1 mm, and 18.0±5.3 mm, respectively. For the measurements, the outline of each tumor on chest CT images was described using polar coordinates. The data were converted to rectangular coordinates, yielding wave data of the tumor outline. The FFT was then used to analyze the wave data. The complexity index (Cxi) and high frequency percent (HF%) were determined for each tumor. The Cxi was defined as the sum of the amplitude of all harmonics over a fundamental frequency. The HF% was defined as the percent of the high frequency components against Cxi.

Result:
The Cxi was 10.3±6.7 mm, 3.2±2.4 mm, and 8.0±3.8 mm, in groups PL, MT, and BN, respectively. The Cxi was significantly smaller in group MT than group PL (p<0.0001) and group BN (p=0.0003). Cxi was significantly correlated with tumor diameter in each group. Discriminant analysis showed a significant difference (P<0.0001) in Cxi between groups PL and MT with regard to tumor diameter. “Cutoff=0.127*DT+2.23” provided the cutoff value that yielded the highest diagnostic accuracy for distinguishing primary lung cancers from metastatic lung tumors. Use of this cutoff line resulted in a sensitivity of 95.8%, specificity of 81.5%, and accuracy of 89.7%. The HF% was 1.82±2.23%, 3.78±3.93%, and 5.29±1.28, in groups PL, MT, and BN, respectively. The HF% was significantly (p<0.0001) smaller in group PL than group BN.

Conclusion:
Fast Fourier transform analysis of tumor outlines appears useful for distinguishing primary lung cancer from the other diseases.

Background:
Visceral pleural invasion (VPI) is an important prognostic factor for patients with peripheral lung cancer, but its accurate preoperative evaluation with CT is difficult. Pleural tags, which are defined as one or more linear strands that extend from the tumor surface to the pleura, are sometimes seen in peripheral lung cancer that does not abut the pleura. However, studies evaluating the correlation of pleural tags with VPI are limited. The aim of this study was to evaluate the association of pleural tags with VPI of peripheral lung cancer, especially focusing on the type of pleural tags.

Method:
A total of 183 patients were retrospectively analyzed. They underwent pulmonary resection in our hospital between January 2009 and December 2015 for peripheral lung cancer ≦2 cm in solid tumor diameter and dose not abut the pleura. Forms of pleural tag were classified as follows; linear tag, linear tag with soft tissue component at the pleural end, and soft tissue cord-like tag. The relationships between VPI and clinical factors including the forms of pleural tag were analyzed using chi-square for independence tests, and then multivariate logistic regression models were applied.

Result:
Histologically, 23 (12.6%) patients were proven to have VPI. Overlap and multiple forms of pleural tags were frequently seen. Multivariate analysis revealed that tumors with linear tag with soft tissue component at the pleural end had significant association with VPI (p = 0.008). Sensitivity and specificity were 47.8% and 83.8%, respectively, in the prediction of VPI with this type of pleural tag. Presence of multiple linear tags also was associated with VPI (p = 0.036).

Conclusion:
Specific forms of pleural tag can predict VPI of peripheral lung cancer ≦2 cm in solid tumor diameter and dose not abut the pleura.

Background:
Erlotinib treatment is offered to NSCLC patients in a palliative setting also in EGFR wildtype (EGFR-wt) patients because we know that some of these patients will benefit, However, in order to evaluate the effect of treatment, we wait for 8-12 weeks before performing a CT scan, since the effect is rather subtle especially in EGFR-wt patients. The purpose of this study was to evaluate which method for evaluating change in 18F-FDG uptake is the best predictor for survival.

Method:
18F-FDG-PET/CT scans from 56 NSCLC patients (48 EGFR-wt and 8 EGFR mutated) performed before and after 7-10 days of erlotinib treatment were analysed with four different methods for response to treatment including visual evaluation, and three semi quantitative methods measuring % change in SULpeak, SULmax and TLG, with a range of cut-off levels determining response, stable disease and progression. A direct comparison of the semi-quantitative parameters was performed using univariate cox regression, linear regression and ROC analysis for progression free survival (PFS) and overall survival (OS) < the median. Kaplan-Meier analysis was used to estimate PFS and OS for the response categories.

Result:
Both %SULpeak, % SULmax and %TLG were correlated to PFS and OS. The strongest linear correlation was found for %TLG (R=0.51, p< 0.001). The ROC analysis showed the highest AUC for predicting OS for %TLG (0.70 (0.56-0.85) with a sensitivity of 0.68 and a specificity of 079. All the semi-quantitative methods showed a statistical overall difference in PFS for the three response categories at some cut-off levels for %SULmax both at 15 and 25%, for %SULpeak at 20 and 25% and for %TLG at 45/75, 50, 30, 25 and 20% cut-off. Visual evaluation failed to differentiate between response categories. For OS %TLG at 4 different cut-off levels and SULpeak at the three lowest cut-off levels showed a statistical overall difference in OS, both visual and %SULmax did not.

Conclusion:
The choice of method for analysis is not clear-cut, but %TLG as suggested by the PERCIST 1.0 is not inferior to other methods, and visual evaluation seems to be the least sensitive at this very early time-point. A lower cut-off level for discriminating between response categories seems to be relevant, we find that 20-25% change for both response and progression is optimal.

Background:
The use of 18F-FDG-PET/CT is increasing for radiotherapy planning. We wanted to investigate whether 18F-FDG-PET/CT scans at this time-point can be a tool for selecting patients with locally advanced NSCLC who are likely not to benefit from radical chemo-radiotherapy in order to avoid the potentially harmful side effects in patients undergoing futile treatment, by predicting progression free survival (PFS) and overall survival.

Method:
18F-FDG-PET/CT scans before and after 2 cycles of induction chemotherapy (carboplatin/vinorelbine) from 91 patients enrolled in a Danish randomized controlled multicenter study (the NARLAL study) was retrospectively evaluated using visual evaluation. Following the induction chemotherapy patients received radical radiotherapy (66 or 66 Gy) with concomitant vinorelbine treatment as according to the NARLAL protocol. Estimates of median OS and PFS was calculated with Kaplan-Meier analysis for the response categories resulting from the visual evaluation, both considering all four response categories and separate analyses for the SMD and PMR groups. Statistical significance was tested using the log rank test, and a significance level of 0.05 was applied.

Result:
The median OS was 25.3 months (95% CI: 21.0-29.6 months), median PFS was 8.6 months (95% CI: 7.9-9.4 months). Overall survival did not differ for PFS or OS when considering all four response categories (p = 0.424 for PFS and p= 0.245 for OS). Considering only partial metabolic response (PMR) and stable metabolic disease (SMD) no difference for PFS, but we found a median OS of 22.6 (95% CI: 18.9-26.3) months for PMR and 32.4 (28.4-36.3) months for SMD (p=0.036). Stratified by histology, the effect was seen in the adenocarcinomas only.

Conclusion:
The median overall survival in patients with a partial response by 18F-FDG-PET/CT evaluation is shorter than for patients with stable disease, in adenocarcinomas. This is surprising and may suggest a differentiation of the treatment for patients with locally advanced NSCLC.

Background:
In patients with inoperable, locally advanced NSCLC, radical chemo-radiotherapy is the preferred treatment, and 18F-FDG-PET/CT is increasingly used for radiotherapy planning. Despite the introduction of the PERCIST criteria for evaluation of response with 18F-FDG-PET/CT, various methods are being used making comparisons of studies a difficult task. The aim of this study was to investigate which method for analyses of 18F-FDG-PET/CT provided the best prediction of survival.

Method:
18F-FDG-PET/CT scans before and after 2 cycles of chemotherapy from 91 patients enrolled in a Danish randomized controlled multicenter study was retrospectively evaluated using four different methods. The change in 18F-FDG-uptake was measured in a single lesion a) SULpeak (according to PERCIST) and b) SULmax (according to the EORTC criteria), c) total disease evaluation qualitatively (visual), and d) total lesion glycolysis measured in all involved measurable lesions with a 45% decrease as cut-off for response and 75% increase as cut-off for progression as according to PERCIST (%TLG). Univariate cox regression was used to evaluate a direct correlation between 18F-FDG uptake at baseline, follow-up and %change and survival (PFS and OS). Estimates of median OS and PFS was calculated with Kaplan-Meier analysis for the response categories from all four methods and, a log rank test performed for evaluation of the statistical significance. A Bonferroni correction for comparison of four methods was applied rendering a p < 0.013 the considered level for statistical significance.

Result:
There was no clear correlation between survival (neither PFS nor OS) and % change in FDG-uptake (independent of analysis method), but TLG at baseline showed the strongest correlation to both PFS (p= 0.02) and OS (p= 0.03). %TLG was the only method to show a statistically significant difference in OS, (p=0.004). Visual evaluation and change in SULpeak both succeeded in discriminating between the two largest response groups, PMR and SMD, with median OS for visual evaluation: PMR: 22.6 (95% CI: 18.9-26.3) months versus 32.4 (95% CI: 28.4-36.3) months for SMD; SULpeak: PMR: 20.6 (95% CI: 18.0-23.2) months versus 29.7 (95% CI: 24.2-35.1) months for SMD. Change in SUL max and TLG showed no difference.

Conclusion:
There was no clear correlation between PFS or OS and percentage change in FDG-uptake. The response categories for SULpeak and TLG (PERCIST criteria for cut-off), and visual evaluation predicts OS at various levels. SULmax categories as according to the EORTC criteria did not successfully predict OS. None of the methods predicts PFS.

Background:
The incidence of lung cancer is dramatically increased. To review the clinical and imaging pattern of lung cancer, diagnosed and treated with breast cancer within 15 years. To evaluate clinical/pathologic stages, doubling time, and lung RADs categories, related with hormone receptor of breast cancers.

Method:
Retrospectively reviewed medical record of breast cancer patients, since 2007. 17 female patients(19 breast cancers)were included. 19 lung lesions were detected. Age range was from 43 to 78 years old(mean 63.8). We reviewed TNM stage, lung RADS, tumor doubling time and the status of hormone receptor. One patient had bilateral breast cancers, mucinous carcinoma (left), and invasive ductal carcinoma(right). Two nodules were shown, invasive ductal carcinoma and one is DCIS.

Result:
The pathology of breast cancer was invasive ductal carcinoma(15), DCIS(2) mucinous ca(1), papillary ca(1). The T stage of breast cancers was T1(17), T2(2) . The subtype of lung cancers was adenocarcinoma, mostly. T stage of lung cancer is T1(16), one is AIS, T3(1), T4(2). The node metastases were not shown and 2/17 showed distant metastases. The primary tumor dimension was from 8mm to 29 mm. The interval between two tumors from 11 days to 14 years. Lung RADs categories were 4A/B(14), 3(2), 2(3) and final category were increased in 5 (from 2 to 4A or B(4 cases), 3 to 4A/B(2 cases)). In 7 interval growing, the tumor doubling time was from70 to 2016 days (mean 712 days).Figure 1



Conclusion:
The receptor (-) breast cancer were higher T stage of lung cancer and lymph nodes metastases were more common. The growth rate of lung nodules are slow as matched with lung RADS category. Volume doubling time of lung cancers ranging from 70 to 2016 days, and exceptionally short in one of them. RADS up scaling in 7 cases are all T1 stage lung cancers, favorable response for the treatment.

Background:
Widespread CT scan availability generates the challenge of managing small pulmonary lesions generally undetected on chest radiographs. Preoperative nodule localization is critical for adequate diagnosis and an accurate resection. Metallic hook wire localization under CT is a widely used method but can be associated with pneumothorax, hook related discomfort, and dislodgement. CT-guided percutaneous barium marking (CT-GPBM) can be simpler, low cost procedure, without risk of dislodgement, and radiopaque on fluoroscopy, increasing resection accuracy. Barium marking through bronchoscopy has been previously reported. This study describes our experience with CT-GPBM in the diagnosis and management of ground glass abnormalities and small pulmonary nodules.

Method:
Between January 2013 and May 2017, 36 patients underwent preoperative CT-GPBM and VATS resections at our institution, and were retrospectively reviewed. All cases were discussed in multidisciplinary rounds. After CT localization and local anesthesia, a 21-gauge Chiba needle was utilized, under CT guidance, to inject 0,2 mL of 140% barium sulfate suspension. A control CT confirms the marking and verifies possible complications. CT-GPBM and VATS resection are performed in the same day. VATS resection is performed under fluoroscopy guidance. Finally, the specimen is submitted to fluoroscopy to confirm a complete resection.

Result:
Thirty-six different cases were analyzed (41.7% male / 58.3% female). Median age was 65 (ranging from 32 to 91). 18/36 were never smokers. Most lesions were identified on routine follow-up CT (21/36 - 58.3%). Many had previous history of cancer (24/36 - 66.7%). 31/36 (86.1%) were single nodules - median size of 14mm (ranging from 5 to 47). Solid lesion was the most common radiologic finding (21/36 - 61,8%), while the others were semi-solid or ground glass abnormalities. Pathological findings confirmed primary lung cancer in 41,6% of cases (15/36). Adenocarcinoma was the most common histology. Nine patients had benign diseases. Metastatic lesions represented 33,3% (12/36), most derived from colorectal cancer. No patient had reported complications.

Conclusion:
This analysis shows that CT-GPBM is an effective, easy and safe pre-operative (VATS) localization procedure, allowing accurate diagnosis and resection of small or faint pulmonary nodules, avoiding thoracotomies in benign diseases and metastatic tumors. Data will be updated for presentation.

Background:
Percutaneous transthoracic needle biopsy (NB) has been widely used for the preoperative diagnosis of lung nodules. It has been proposed that the risk of pleural recurrence is high following lung resection in patients who underwent preoperative NB for sub-pleural nodules (Kashiwabara, et al. Cancer Invest 2016; Wang, et al. Sci Rep 2017). The aim of this study is to investigate the prognostic impact of preoperative NB for pleural recurrence in patients with early-stage lung adenocarcinoma (ADC).

Method:
Patients who underwent lung resection for pathologic stage I (≤3cm) lung ADC were included in the analysis (1995-2014, n=992; NB group 626 patients and no-NB group 366 patients). We compared the clinicopathologic characteristics and recurrence free probability (RFP, separately analyzed for any, locoregional, pleural, and distant recurrence) between NB and no-NB groups. The risk of pleural recurrence was evaluated in tumors both with and without visceral pleural invasion (VPI).

Result:
The NB cohort was associated with older age and larger tumor size compared to the no-NB cohort (p<0.05). There was no statistical difference in the incidence of VPI (VPI in NB, 12% vs. VPI in non-NB, 15%, p=0.2). In RFP analysis by Kaplan-Meier method with log-rank test, there was no statistical difference between NB and no-NB groups (NB vs. non-NB: 5-year RFP for any recurrence, 86% vs. 86%, p=0.8; locoregional recurrence, 93% vs. 94%, p=0.7; pleural recurrence, 98% vs. 96%, p=0.14; and distant recurrence 94% vs. 93%, p=1). In tumors both with and without VPI (n=128 and n=864, respectively), the risk of pleural recurrence was not higher after NB (Figure; 5-year RFP for pleural recurrence [NB vs. no-NB]: VPI positive, 93% vs. 83%, p=0.3; VPI negative, 98% vs. 97%, p=0.4). Figure 1



Conclusion:
Preoperative needle biopsy was not associated with an increased risk of pleural recurrence following lung resection.

Background:
The EarlyCDT-Lung test detects autoantibodies to abnormal cell surface proteins in lung cancer patients. It has been studied in UK[1] and USA[2] in diagnosing lung cancer patients and has a sensitivity of up to 47%, specificity up to 90% and positive predictive value of 16%. It is a potential diagnostic test and when combines with radiological (Computed Tomography) test, the positive predictive value can be enhanced up to 72%[3]. We want to study the performance of the test in a predominantly Asian ethnic city. 1. Caroline J. Chapman
 et al. EarlyCDT®-Lung test: improved clinical utility through additional autoantibody assays. Tumor Biol. 2012; 33:1319–1326 2. JR Jett et al. Audit of the autoantibody test, EarlyCDT®-Lung, in 1600 patients: An evaluation of its performance in routine clinical practice. Lung Cancer 2014; 83:51–55 . 3. P Massion et al. Autoantibody Signature Enhances the Positive Predictive Power of Computed Tomography and Nodule-Based Risk Models for Detection of Lung Cancer Journal of Thoracic Oncology 2017; 12:578-584

Method:
Patients who had been referred to our clinic for follow up for lung nodules from March 1st to May 30th, 2017 were included. EarlyCDT-Lung test was performed for those with size greater than 0.8cm or size less than 0.8cm with smoking pack-year more than 30 years. EarlyCDT-Lung test sensitivity and specificity were evaluated. Confirmed lung cancer patients were retrospectively compared with non-lung-cancer patients.

Result:
10 patients (Chinese: Caucasian=9:1) were included during the study period (Mean age=51.5; male gender=9 (90%); Smoker=2 (20%). 5 patients were diagnosed lung cancer (Mean tumor size=2.9cm; 1 with Stage IA; 1 with stage IB; 1 with stage IIIA, 2 with stage IV; with adenocarcinoma 5/5, EGFR+ve 2/5, ALK+ve 1/5. Mean age=54.8; male gender=4 (80%); Smoker=0 (0%). 5 patients were diagnosed non-lung-cancer (Mean tumor size=0.8cm. Mean age=48.2; male gender=5 (100%); Smoker=2 (40%). EarlyCDT-Lung test was positive in 1/5 lung cancer patients, while 0/5 in non-lung-cancer patients, giving rise sensitivity 20% (1/5), while specificity 100% (5/5).

Conclusion:
This pilot study gives the initial experience of EarlyCDT-Lung test for Asian with high specificity in limited number of subjects. A larger study is warrant for a more accurate result.

Background:
Non-small cell lung cancer (NSCLC) is the common type of histology among all cases of lung cancer. Studies have shown that PET/CT is more accurate than CT for the detection of nodal status due to the increased FDG uptake in very small nodes. In fact, tumor-node-metastasis staging of the mediastinum is currently one the most common indications of PET/CT in lung cancer. The aim of this pilot retrospective review was to assess the role PET/CT as a non-invasive procedure for the visualised lymph nodes in lung cancer.

Method:
Twenty-nine patients [NSCLC (86%), NET (10%), SCL (4%)] underwent PET/CT imaging and were followed for a period of at least 3 years. There were 15 females (51.72%) and 14 males (48.28%) aged between 38-78 years (mean +/- SD= 61.24+/- 9.86) and were referred for staging (86%), restaging (10%) and response to therapy (4%). All patients underwent 18F-FDG PET/CT with a mean time of 71.2 minutes after tracer injection. Tumor and lymph node uptake were evaluated with both visual and quantitative assessment.

Result:
Seven patients (24%) died at 12 months despite treatment induction and by the end of two years follow-up, just a little more than half (51.72%) of the patients died. The majority of those who died were males (r=-0.38; p=0.041). Active lymph nodes were seen in one nodal station in 19 patients. They were also seen in two nodal stations in 12 patients and in three nodal stations in 5 patients. The common stations were as follows: Hilar (33.3%), right lower Para tracheal (22.2%) and subcarinal (19.4%). While no significant association was seen between the primary location of tumor and the occurrence of lymph nodes, patients with positive nodes on PET/CT staging tend to have a reduced survival than those without visualized lymph nodes [odd ratio= 2.33, 95% CI 1.60-9.02; p=0.08]. However, when the analysis was only done for the NSCLC group, a significant reduction of survival was noted in patients who had positive lymph nodes (p<0.01).

Conclusion:
This small cohort confirmed the predominance of NSCLC as described in the literature. While this histological predominance could result from a referral bias, it clearly showed that the male gender was associated with a reduced survival. However, a larger cohort must be studied in a prospective analysis to confirm the significantly reduced survival among NSCLC patients with positive lymph nodes during staging, for PET/CT imaging to become a robust predictor of survival.

Background:
In this study, we built three 3-dimensional convolutional neural networks (CNN) for recognition of epidermal growth factor receptor (EGFR) mutation status in Chinese patients with lung adenocarcinomas based on non-enhanced computed tomography (CT) images.

Method:
From October 2008 to December 2015, 405 patients with lung adenocarcinomas were included in this retrospective study. Their pathological phenotypes and EGFR mutation status were gained from surgical resections. Their CT images used in this study were taken before any invasive operation. Tumors with a diameter smaller than 8 mm or have ground glass component were excluded. Region of interest that includes tumors were segmented manually by clinicians and preprocessed to have uniform size and grey-level range before applied to CNNs. The three CNNs have 4 convolutional and 1 full connection layers between input and output layers. The inputs size of three CNNs are 21×21×21, 31×31×31, and 41×41×41, respectively. The outputs of the CNN are the probabilities of mutant and wild status. The CNN classifier’s performance was then validated using an independent set and evaluated using area under curve (AUC) values of the receiver operating characteristic.

Result:
405 patients diagnosed with lung adenocarcinoma staging I to IV were included in this study (195 male, 210 female; 61 smokers, 344 non-smokers). The patients received surgery based treatment and their tumor stage was based on pathological reports. EGFR mutations (mainly 19del and 21L858R) were found in 198/320(61.9%) and 56/85(65.9%) patients in training and validation sets, respectively. The CNN showed an AUC of 0.767 (95% confidence interval: 0.668-0.866, p<0.001) in the validation set. The sensitivity and specificity are 62.5% and 89.7% at best diagnostic decision point. These results were highest among published results of only using images to recognize EGFR.

Conclusion:
The CNN showed potential ability to recognize EGFR mutation status in patients with lung adenocarcinomas and could be improved in the future works to help make clinical decisions.

Background:
There were many reports and evidences for part-solid nodules with ground-glass opacity (GGO) in small-sized lung adenocarcinoma, and the component of GGO has been known to be a factor of malignant potency of the tumor, although the relation between radiological appearance of >2cm or >3cm lung adenocarcinoma and clinicopathological features were less noted.

Method:
A total of 136 patients with >2cm lung adenocarcinoma with >0.75 of consolidation to tumor ratio (C/T ratio) who underwent lung resection at Shinshu University Hospital from February 2003 through December 2010 were assessed. Among these patients, 83 with pure solid appearance in preoperative thin section computed tomography (C/T ratio = 1.0) were placed into Solid group, and 53 with small GGO surrounding the solid type tumor (0.75 < C/T ratio < 1.0) were placed into Subsolid group in this study. We retrospectively analyzed the clinicopathological features and prognosis after surgery in each groups.

Result:
The maximum standardized uptake value (SUVmax) of the tumor in preoperative radiological assessment using [18F] fluoro-2-deoxyglucose positron emission tomography (FDG-PET) were significantly higher (p=0.0048) in Solid group (7.65±4.36) than Subsolid group (4.82±3.32). The presence rate of vascular invasion (Ly or V) was 54.2% in solid group and 39.6% in Subsolid group. The numbers of node positive patients were 19 with N1, 27 with N2 in solid group, and 8 with N1, 7 with N2 in Subsolid group respectively, and the rate was significantly higher in Solid group (p=0.0019). The prognostic analysis of pathological stage 1 patients (29 patients in Solid group and 37 in Subsolid group) revealed that the overall survival and recurrence-free survival was significantly poorer (p=0.044 and p=0.019 respectively) in Solid group than in Subsolid group.

Conclusion:
The small GGO surrounding the solid type adenocarcinoma as C/T ratio >0.75 indicated lower malignant potency and better prognosis than pure solid tumor without GGO component. Even small GGO component was a clinical factor of favorable oncologic outcomes, it may contribute to regarding the decision making for surgical strategy in preoperative assessment.

Background:
In this study we aim to investigate predictivity of alveolar spread in primary lung cancer by using preoperatively scanning methods

Method:
In order to re-evaluate alveolar spread, pathology preparations of 45 patients had operated for primary lung cancer diagnosis and scanned preoperatively with PET CT all in our hospital which interpreted by same nuclear medicine specialists implicated in this study. As using the patients pet CT findings, CTV (computerized tomography volume), MTV (metabolic tumor volume), TLG(total lesion glycolysis), SUDmax, SUDort values and their relation with alveolar spread analyzed

Result:
Preoperatively PET-CT scanned all in our hospital 45 patients has included and cause of couldn't reach their pathological preparations, 6 of them exluded from the study. 21 of 39 patients were men (53.8%) , 18 of them were women (46.2%) and mean age was 66.67/+-7.88 (42-80). We didn't detect any relation between CTV, MTV, TLG, SUDmax, SUDort values and alveolar spread (p>0.05). However when the CTV/ MTV ratio analyzed, alveolar spread was statistically more common in the group of patients had ratio lower than 1. (62.9% versus 0% , p:0.01)

Conclusion:
High local recurrence risk in sublober rejected patients with alveolar spread has indicated in various studies. Regarding this matter, we recommend re-evaluation of the patients for sublober rejections whose CTV/MTV ratio is lower than 1.

Background:
Spinal cord compression is one of the most disabling complications in lung cancer. It may be caused either by the tumor itself or by metastatic involvement of the spine and meninges. The goal of this work is to appreciate characteristic of spinal cord compression in lung cancer.

Method:
Retrospective review of the MRI data of patients with lung cancer presenting with neurological symptoms caused by medullary or cauda equine compression. Only patients with abnormal findings were retained. All exams were performed by a 1.5 T machine.

Result:
From January 2014 to December 2016, only 24 patients (23 males and 1 female) were referred to the imaging department of our institution for medullary compression exploration. Their average age was 57 years. Seventeen patients had adenocarcinoma. Three patients were classed stage IIIB and 18 were classed stage IV. Medullary compression was caused by direct compression (12 cases) of the tumor or by a spine metastases (12 cases). One spine region was involved in 18 patients with dorsal location in 13 patients. In three cases, the compression concerned the cauda equina. Myelopathy was noted in 15 of the 21 patients with spinal cord compression.

Conclusion:
Spinal compression is not frequent in lung cancer. It is mainly observed in advanced stages. It may occur by direct tumor invasion or by vertebral metastatic extension.

Background:
Lung cancer has a poor prognosis unless patients are operated. However, few patients may undergo surgery. Recurrences are not rare and they increase the mortality. The objective of this study is to appreciate the frequency of recurrences after surgery in non-small lung cancer (NSCLC) in our institution and to correlate them to TNM classification and histological subtypes.

Method:
Retrospective review of 127 patients (106 men and 21 women) aged between 19 and 80 years with NSCLC that were operated. All patients had chest, abdomen and brain computed tomography (CT) in their follow up. We precised for each patient the histological type of the lung cancer, the c and p TNM stages, the resection type and the other treatments prescribed, the recurrence site and the delay between the surgery and the recurrence.

Result:
Recurrences were observed in 35 patients (27.5%). They had either a lobectomy (n=30), a pneumonectomy (n=4) or a wedge (n=1). Recurrences were located to the chest in 36 cases (half of them occurred in the operated lung) and in extra-thoracic organs in 13 cases; the brain was the mostly involved. The delay between surgery and recurrence ranged between 1 month and 7 years. Fourteen patients had a recurrence within the first year after surgical removal of their tumor. Recurrences were mostly observed in patients with T4 stage (23%), N3 stage (66.6%), stage III cancer (49%), and with adenocarcinoma. An under-estimation of the cT and cN stages was observed in 5.6% and 27.5% of patients, respectively.

Conclusion:
Recurrences in patients with surgically removed lung cancer are not rare. They depend on its histological type and on the cT and cN stages.

Background:
PET scans are used during diagnosis and staging of lung cancer. The role of PET scan in guiding therapy for advanced stage non-small cell lung cancer (NSCLC) is not proven, but it continues to be used during the treatment course at many centers. We studied the Surveillance, Epidemiology, and End Results (SEER) Program database and Medicare claims data to evaluate the use of PET scan in advance stage NSCLC patients in the United States and the impact on patient outcome.

Method:
The SEER-Medicare database was queried to capture patients with stage IV non-small cell lung cancer diagnosed between the years 2000-2011. The cohort of patients that received PET scan after diagnosis were analyzed and compared with the cohort that did not receive PET. The univariate (UV) association between covariates and overall survival (OS) were compared by log-rank tests. Time dependent Cox Model was used in multivariable (MV) analysis, with time from diagnosis to first PET scan as time-dependent variable, while the other covariates as time-independent. All analyses were performed using SAS Version 9.4.

Result:
A total of 52,712 eligible patients with stage IV NSCLC were identified between 2000-2011, out of which 13,873 (26.3%) had received PET scan. Characteristics of PET cohort: median age 74 years, 53% male, 87% white and 82% from metro locations. 87% of the patients that received PET were diagnosed between 2006-2011. In the first year after diagnosis, 70% of the patients had 1 PET, 16% had 2 PETs and 14% had 3 or more PETs. About 64% of the patients had received their first PET scan within 2 months of diagnosis and 19% had it between 2 to 6 months. The average Medicare cost associated with patients that received PET was significantly higher than that of patients that did not receive PET scan ($60,417 vs. $34,287; p<0.001). Chemotherapy and radiation were given in a higher proportion of patients that received PET versus those that did not receive it (56% and 45% versus 26% and 36% respectively; p<0.001). Though univariate analysis revealed that a PET scan within a year of diagnosis was associated with better 1-year survival (HR 0.87, P<0.001), this did not translate into overall survival advantage on multivariable analysis (HR 0.99, P=0.56).

Conclusion:
The utilization of PET scan in stage IV NSCLC patients was associated with higher cost, but without a tangible improvement in survival compared to those that did not have a PET scan.

Background:
Questions have been raised about the appropriate treatment of lung cancers manifesting as subsolid nodules (nonsolid nodules (NSNs) and part-solid nodules (PSNs)), as these have very high reported survival rates and have been observed in up to 10% of screening participants. Our goal in this report is to summarize the publications on survival of patients with resected lung cancers manifesting as PSNs and to further the development of consensus definitions of the CT appearance and the workup of such nodules.

Method:
PubMed/MEDLINE and EMBASE databases were searched for all studies/ clinical trials on CT-detected lung cancer in English before Dec 21, 2015 to identify surgically-resected lung cancers manifesting as PSNs. Outcome measures were lung cancer-specific survival (LCS), overall survival (OS), or disease free survival (DFS). All PSNs were classified by the percentage of solid component to the entire nodule diameter into: Category PSNs < 80% or Category PSNs ≥ 80%.

Result:
Twenty studies reported on PSNs < 80%: 7 reported DFS and 2 OS of 100%, 6 DFS 96.3-98.7%, and 11 OS 94.7-98.9% (median DFS 100% and OS 97.5%). Twenty-seven studies reported on PSNs ≥ 80%: 1 DFS and 2 OS of 100%, 19 DFS 48.0%-98.0% (median 82.6%), and 16 reported OS 43.0%-98.0% (median DFS 82.6%, OS 85.5%). Both DFS and OS were always higher for PSNs<80%.

Conclusion:
A clear definition of the upper limit of solid component of a PSN is needed to avoid misclassification because cell-types and outcomes are different for PSN and solid nodules. The workup should be based on the size of the solid component.

Background:
Poor adherence to the recommended guidelines in diagnosing and staging patients with non-small cell lung cancer (NSCLC), with negative downstream effects has been previously shown. In addition, studies have demonstrated benefits of staging with PET, including a reduction in number of non-curative resections performed and a higher rate of identifying M1b disease. Staging with brain MRI has demonstrated a yield up to 10% for detecting metastasis in patients with negative clinical examinations. We aimed to assess the adherence to imaging guidelines for PET and brain MRI in the staging of NSCLC patients prior to treatment within our healthcare system.

Method:
We reviewed patients who underwent initial work-up for primary NSCLC during 6/2013–6/2015, in a hospital network of 7 institutions. Clinical stage II-IV patients were stratified by imaging performed prior to the initiation of treatment. Evidence-based clinical practice guidelines referenced include the American College of Chest Physicians (ACCP) 3[rd] edition and the National Comprehensive Cancer Network (NCCN) 7[th] version. Both ACCP and NCCN recommend a PET scan for suspected cIb-III; while ACCP recommends a brain MRI for suspected cIII-IV, and NCCN for suspected cIb-IV.

Result:
The figure demonstrates compliance rates for the 283 included patients. Of cII patients, 7% (2/30) did not receive a PET scan and 43% (13/30) did not receive a brain MRI; while, 11% (6/56) of cIII did not receive a PET scan and 20% (11/56) did not receive a brain MRI. Figure 1



Conclusion:
Variable compliance with imaging guidelines for the use of PET and brain MRI imaging for the staging of our NSCLC patients was seen. Lack of appropriate imaging for NSCLC staging may lead to inappropriate management decisions resulting from incomplete staging information. Quality initiatives are necessary to ensure guideline compliance.

Background:
Fluorodeoxyglucose Positron emission tomography (FDG PET) is a main tool in diagnosis and staging in patients with non small cell lung cancer (NSCLC). Metabolic parameters as standardized uptake value (SUV) and total lesion glycolysis (TLG) were reported to have independent prognostic and predictive value in different stages of disease and were shown to correlate with tumor activity and tumor burden. However, little is known regarding the correlation of these parameters with histologic subtypes of NSCLC. In this study we aimed to explore associations of FDG PET metabolic parameters in patients diagnosed with NSCLC and histologic subtypes of NSCLC.

Method:
We retrospectively evaluated 87 consecutive patients who underwent FDG PET scans in the workup of pulmonary nodules highly suspicious for malignancy. At total of 62 of them were found to have NSCLC; 44 with adenocarcinoma (ADC) and 18 with squamous cell carcinoma (SQCC). Measurements of hounsfield units (HU), region of interest (ROI), SUVmax, SUVmean, volume of interest (VOI) and TLG were obtained. Follow up for final staging at diagnosis and overall survival (OS) were obtained from electronic medical records. Statistical analysis was made with T-test and Mann-whitney u test.

Result:
In patients with SQCC, the mean ±standard deviation (SD) SUVmax and TLG (±SD) were found to be significantly higher than in patients with ADC (Table 1). Differences in OS showed a trend favoring SqCC but no significance was found (complete results from 16 more patients are pending). Figure 1



Conclusion:
FDG PET Metabolic characteristics such as SUVmax and TLG significantly differ between histological subtypes in NSCLC. This report joins others on the utility of functional imaging studies in NSCLC in assessing not just disease burden but also prognosis and function. Further analyses of the PET parameters prognostic impact according to histology subtype is ongoing and would be presented at the meeting.

Background:
In part-solid nodule of lung adenocarcinoma, the diameter of the solid part in computed tomography(CT) scan correlates with the diameter of the pathological invasive part. However, there are some cases revealing dissociation between them. We analyzed clinical factors predicting the dissociation of the diameter between radiographical solid part and pathological invasive part in adenocarcinoma less than 3 cm.

Method:
Among 291 cases with a lung adenocarcinoma smaller than 3 cm, we identified 91 cases whose solid part in preoperative thin-slice CT scan was less than 5 mm. Based on pathological diagnosis of invasive part, we divided these cases into Adenocarcinoma in situ/Minimally Invasive Adenocarcinoma(AIS/MIA) group (less than 5 mm) and Massive invasion group (5mm or larger), and retrospectively analyzed the clinicopathological factors. We also performed logistic regression analysis to detect the factors predicting the dissociation between radiographical and pathological findings.

Result:
Of 91 cases, 67 cases were in AIS/MIA group (AIS: 57, MIA: 10) and 24 cases were in Massive invasion group. In univariative analysis, cases of Massive invasion group were significantly higher in Brinkman index, CEA, age, and total tumor size than those of AIS/MIA group (p = 0.02, 0.01, 0.04, 0.03 respectively). With these detected four factors, we performed logistic regression analysis after determining threshold by ROC curve, which resulted in Brinkman index equal or larger than 400, and age equal or elder than 67 as significant predictive factors for Massive invasion group (p < 0.01, p = 0.05 respectively). Among 11 cases positive for these two factors, 7 cases (63.6 %) were in Massive invasion group.

Conclusion:
In the cases of radiographical AIS/MIA, the diameter of pathological invasive part tends to exceed 5 mm if Brinkman index equal or larger than 400, and age equal or elder than 67.

Background:
Pulmonary emphysema is known to be related to an increased incidence of lung cancer. The risk factors of emphysema and lung cancer are smoking history and envirolmental pollution. Low-dose chest CT is useful method for imaging both: emphysema and solitary pulmonary nodules (SPN).

Method:
We have enrolled a total of 602 healthy volunteers to the Pilot Silesian Study of Early Lung Cancer Detection with LDCT. Retrospectively we analyzed the baseline round of LDCT scans. One person was excluded due to presence of lung cancer symptoms. Gradual accrual period lasted for two years, from 2010 to 2012. The positive result was defined as a presence of solid or part-solid nodule equal to or larger than 5 mm or non-solid equal to or larger than 8 mm. Participants were stratified by exposure to tobacco smoke into five groups: A) current smokers with a history of at least 40 years of smoking - chronic smokers; B) current smokers with a history of smoking at least 40 cigarettes per day - intensive smokers; C) former smokers who quit smoking fewer than 5 years ago; D) former smokers who quit smoking at least 10 years ago; E) exposed to a second-hand tobacco smoke. With the statistical methods we analized corelations between this five groups depending on tobacco exposure and demographic features.

Result:
We identified 1016 lung nodules in 265 individuals. A total of 410 nodules coexisted with emphysema in 99 individuals. In group with emphysema we diagnosed 3 early lung cancers, unlike to the group without emphysema, in which we diagnosed 4 cancers (HR=0.7 95% CI 0.16-1.02). Subsolid morphology and diameter of a nodule larger than 5 mm were associated with higher risk of malignancy. Emphysema and lung nodules were more often found in the group of chronic smokers (>40 years) - 8.1% (A), in comparison to the group of intensive smokers – 0.8% (B). There were no statistically significant differences in both subcohorts of former smokers (for group C and D, 2.1% and 2.3%, respectively).

Conclusion:
In the subgroup of chronic smokers emphysema and lung nodules were more likely to occur. Additional risk factors for the presence of lung nodules and emphysema were: age (above 65 years) and occupational exposure (especially coal miners and welders).

Background:
Lung Cancer (LC) is the leading cause of cancer-related death worldwide. The Lung Cancer Screening Trial screened 26,723 patients via Low-Dose chest CT-scan. Screening results were positive in 24.2%, however 96.4% were false positive. Therefore, a biomarker helping to discriminate false positive from true positive screening results will limit unnecessary interventions. The aim of this study was to analyze the feasibility of the DNA methylation markers prostaglandin E receptor 4 gene (PTGER4) and short stature homeobox 2 gene (SHOX2) as potential biomarkers in liquid biopsies additional to CT-scan findings suspicious for LC.

Method:
A total of 73 patients with a chest CT-scan suspicious for stage I-IIIA LC were included in this study. LC was histologically confirmed in 50 patients. In the remaining 23 patients benign diagnoses were established. Circulating cell-free DNA was extracted, bisulfite converted and purified from 3.5 ml plasma, using a commercially available kit. Triplicates of bisulfited DNA were assayed with a 45-cycle rtPCR assay detecting methylated DNA of SHOX2, PTGER4 and ACTB as reference assay. Assay result was called positive for PTGER4 if at least two out of three PCR results had cycle threshold (Ct) values below 45; for SHOX2 if the Ct value was below 35.

Result:
PCR-results were valid in 18 (4 organising pneumonias, 5 pneumonias, 3 post-pneumonic residues, 3 hamartochondroma, 1 mycosis, 1 coal-workers pneumoconiosis, 1 aspergilloma) of 23 benign samples (control), and in 43 of 50 LC samples. PTGER4 results were negative in all controls and positive for 18 of 43 LC. These results translate into a specificity of 100 % and a sensitivity of 41.86% for LC detection via PTGER4. The SHOX2 assay was negative in all controls and positive in 9 LC patients, resulting in a specificity of 100 % and a sensitivity of 20.93% for SHOX2. Overall 21 of 43 LC samples were positive in either PTGER4 or SHOX2 and all controls were negative for both measurements. In combination both markers showed a specificity of 100% and a sensitivity of 48.84%. This specificity would imply a positive predictive value of 100%, further validation in a larger cohort is required.

Conclusion:
DNA methylation of PTGER4 and SHOX2 was highly specific for patients with chest CT-scan findings suspicious for LC. This could help to identify high-risk patients, who should be directly transferred to tissue-based diagnostics, surgery or stereotactic radiation. In the other cases the next diagnostic step should be decided clinically.

Background:
In conventional cytology discrimination of the reactive mesothelium cells and malignant cells is the most important diagnostic problem. Cell block (CB) technique is helpful in discriminating cytological abnormalities or well differentiated adenocarcinoma.

Method:
95 consecutive patients (52% male, mean age 67±12yrs) with malignant pleural effusion in lung cancer were enrolled. 44% of patients had Eastern Cooperative Oncology Group performance score 0-1 and 66% ≥2. In all patients pleural fluid smears were sent for conventional cytological and CB immunocytochemical examination.

Result:
In 69 (73%) cases malignant pleural effusion was the first sign of lung cancer and in 26 (27%) cases developed in disease progression. First pleural fluid immunocytochemistry examination confirmed malignancy in 80 (84%) cases: adenocarcinoma – 70 (88%), small cell carcinoma – 7 (9%) and non small cell carcinoma – 3 (3%). First cytological examination was positive in 81 (85%) smears. Adenocarcinoma cells were identified in 51 (63%) cases although of them CB technique confirmed adenocarcinoma in 49 (96%) smears and reactive mesothelium cells in 2 (4%) smears. Out of 30 (37%) specimens that showed malignant cells in cytological examination, CB method confirmed: adenocarcinoma 15 (50%), small cell carcinoma 6 (20%), non small cell carcinoma 3 (10%) and reactive mesothelium cells 6 (20%). Whether first cytological smear was negative, second smear was positive in 4 (27%) cases and identified 1 (25%) adenocarcinoma and 3 (75%) malignant cells in cytological examination, notably out of them CB examination showed reactive mesothelium in 2 (50%) samples, small cell carcinoma in 1 (25%) sample and adenocarcinoma in 1 (25%) sample.

Conclusion:
Our study aims that pleural fluid cell block immunocytochemistry is more accurate method identifying lung cancer and should be considered as the initial diagnostic approach for malignant pleural effusion in primary investigation in lung cancer, especially in patients with worse performance status.

Background:
There has been increasing recognition that lung nodule measurement on CT scans is imprecise and that an understanding of the extent of this imprecision is necessary when trying to determine whether actual change in volume has occurred. The various factors that influence this are numerous with two of the most prominent being the overall quality of the CT scan (including all of the adjustable parameters) and the size of the nodule.

Method:
We have developed an automated system whereby a calibration device is scanned on a given scanner with a given protocol and then the system can automatically predict the extent of measurement error for a given size solid nodule. We compared this approach to empirically derived results obtained from a database of 117 screen-detected stable nodule ranging in size from 2.2 to 18.7 mm that were scanned twice on the same CT scanner using the same protocol. Automated volumetric analysis was performed using commercial software. This allowed us to determine the relationship between standard deviation of the measurements versus nodule size. We then scanned our calibration device using the same scanning protocol as was used on those nodules to automatically calculate the size and standard deviation relationship.

Result:
Predicted solid nodule volume standard deviation compared with empirically derived values across a range of nodule sizes was within 20% (see figure)Figure 1



Conclusion:
Results from our automated approach were highly correlated with results obtained from scans obtained in actual clinical practice. The ability to predict extent of error specific to a given scanner and scanning protocol is an essential step in understanding whether change has occurred and has implications for both diagnosis and therapy assessment, including predicting when a follow up scan should be obtained. This type of information will ultimately become a necessary component of all quantitative imaging programs.

Background:
Lung cancer mimicking organizing pneumonia (LCOP) is a novel radiological entity of lung adenocarcinoma that could be misdiagnosed as inflammatory lesions.. However, the characteristic biological and genetic features of LCOP are not fully clarified.

Method:
We used thin-section CT images to select cases of (LCOP) among surgically resected lung adenocarcinoma patients. We compared the clinicopathological characteristics and the immunophenotypes of LCOP (n = 44) and other lepidic-predominant adenocarcinomas (non-LCOP, n = 56). We also analyzed the genomic mutation features of LCOP (n = 4) by whole-exome sequencing (WES).

Result:
All LCOP lesions were lepidic-predominant invasive adenocarcinoma. Patients with LCOP had significantly superior recurrence-free survival, compared to non-LCOP patients (95.5% and 74.4%; P = 0.006, respectively). Vascular invasion and lymph node metastasis were less frequent in LCOP than in non-LCOP patients (P = 0.001 and P = 0.03, respectively). The cancer cell expression levels of aggressiveness-related molecules, including ezrin, ALDH-1, laminin-5 were similar between LCOP and non-LCOP. On the contrary, the number of tumor promoting stromal cells, i.e., podoplanin-positive cancer-associated fibroblasts and CD204-positive tumor associated macrophages, was significantly lower in LOCP (P = 0.021 and P = 0.037, respectively). WES revealed that ABCB1, DNAH3, MSI2, and SLITRK2 were specifically mutated in LCOP.

Conclusion:
Our results indicate that LCOP is characterized by fewer tumor-promoting stromal cells, which may contribute to the better prognosis of LCOP patients. Moreover, recognition of specific somatic mutations of LCOP patients may provide information regarding the development and progression of this type of lung cancer.

Background:
Ever since its first success in large scale image recognition problem, deep convolutional neural networks (DCNNs) have shown their capabilities in solving many challenging visual perceptual tasks, such as image classification, segmentation, object detection. In some cases, DCNNs have already achieved near-human performance. In medical image analysis, DCNNs have also been successfully applied to lesion detection, segmentation, and diagnosis. The power of DCNN lies in its ability to learn a hierarchical representation of raw input data, without hand-crafted features.

Method:
The focus of this study was to improve the performance of DCNN in automatic detection of pulmonary nodule on CT scan. In particular, nodules whose size is less than 4mm were considered. A total of 171 scans were collected at Zhongshan Hospital Fudan University, which was first process by the proposed DCNN system (12Sigma). The detection results were then carefully reviewed by an experienced physician, all false positives and missed nodules were manually labeled. To refine the system, all false positives and true nodules of sizes 4mm and under were selected.

Result:
The data set was randomly split into training and test sets, where the training set consists of 90% (154 scans) and the test set consists of 10% of data (17 scans). Figure 1 shows the FROC curves on the test set before and after re-training. Overall, the detection sensitivity were improved at all false positive levels, but the improvement was most significant at low FP rate region. For example, when FP is 0.5 per scan, the detection sensitivity increased from 0.36 to 0.49. Figure 1



Conclusion:
This improvement suggests that, even with limited data, the deep neural network can learn from its mistakes and be easily tuned to be more sensitive to small nodules. We believe that when more data is colllected, more significant improve in high FP rate region will be observed.

Background:
Pathways to lung cancer diagnosis are complex. General Practitioners (GPs) play a vital role in ensuring that people diagnosed with lung cancer receive timely access to appropriate specialist care. Australian guidelines recommend that the first specialist appointment should take place within two weeks of the initial GP referral. However, local data indicate that up to 60% of patients are not referred within two weeks, while state-based data shows that 11% of patients in New South Wales (NSW), Australia, do not see a lung cancer specialist at all; these patients may be missing out on curative and palliative treatments. We designed a primary care intervention to address these gaps and conducted a feasibility study in three sites across NSW. The RoaDmaP pilot study aimed to develop, implement and evaluate the Referral Decision Prompt (RDP), an intervention that supports GPs to refer patients with a suspicious lung lesion on CT scan for specialist care. Specific objectives were to evaluate feasibility and acceptability with GPs, radiologists and radiology practice staff.

Method:
The RDP intervention was developed with key stakeholder input and was tailored for three participating radiology practices. Eligibility criteria included any patient aged over 18 attending for a chest CT scan. Recruitment and consent procedures were tested to enable collection of patient’s medical data from GPs and hospitals to confirm lung cancer diagnoses. The RDP template was integrated into CT scan reports for patients with a suspicious lung lesion. A process evaluation with GPs, radiologists and radiology staff was undertaken through brief surveys and qualitative focus groups.

Result:
Over seven months, 445 patients had a GP-referred chest CT scan; 400 were given study documents; 293 consented (73%). Twelve patients had a suspicious lesion, eight of whom received a confirmed lung cancer diagnosis. Sixty-six percent had a specialist appointment within two weeks. The intervention was rated as highly acceptable by radiologists, radiographers and practice staff; four GPs participated in the evaluation, three recalled the RDP and two used it to change their referral practices.

Conclusion:
The RoaDmaP study presents an innovative approach to identifying lung cancer patients prior to diagnosis and has the potential to improve timely and appropriate specialist referral. The process evaluation findings indicate that this simple, low-cost intervention is feasible and can be easily integrated into radiology reporting practices. It is highly acceptable to radiologists and practice staff, and may positively influence GP referral practices.

Background:
Lung cancer is the first cancer responsible for metastases. They can be either intra-or extra-thoracic. Their diagnosis when the cancer is locally non-advanced modifies completely the treatment. The aim of this work is to find out variables influencing metastases occurrence in T1-T2 classed lung cancer.

Method:
Retrospective review of data of 63 patients (60 males and 3 females) with confirmed lung cancer, which were classed T1 or T2 according to the TNM 7 in the work up of their disease and who were explored in the imaging department of our institution between January 2011 and December 2016. Patients were subdivided into two groups: with or without metastases. The variables tested to find out a relationship with metastases occurrence were age, T1 or T2 status, N status and histological type of the cancer. A significant statistical relationship was retained when p was ≤0.05.

Result:
The mean age of the included patients was 61 years (ranging between 43 and 73 years). Adenocarcinoma was the most common histological type (71%). Twenty seven patients (43%) had metastases; 21 of them (77%) had extra-thoracic ones. Comparing patients with and without metastases, no significant relationship was found between the T1 or T2 stage (p= 0.147), the age (p=0.745) and metastases occurrence. In the opposite, they were statistically related to N2-N3 lymph nodes stage (p=0.032) and to adenocarcinoma (p=0.006).

Conclusion:
Metastases are quite frequent in T1-T2 classed lung cancer. Their occurrence was related to the N stage and to the adenocarcinoma.

Background:
Brain metastases frequently occur in patients with small-cell lung cancer. Therefore, prophylactic cranial irradiation (PCI) is recommended for good responders of initial treatment. The purpose of this study was to investigate influence of PCI on pattern of brain metastases as a first recurrence site (BMFR) after radical treatment for limited disease small-cell lung cancer (LD-SCLC).

Method:
This retrospective study included LD-SCLC patients treated with thoracic radiotherapy and concurrent chemotherapy between January 2006 and December 2014. Induction chemotherapy was permitted. Thoracic radiotherapy was performed with accelerated hyper-fractionated radiotherapy (twice daily, 45 Gy in 30 fractions over 3 weeks) or conventional-fractionated radiotherapy (once daily, 50 Gy in 25 fractions over 5 weeks). Regimen of chemotherapy consisted of intravenous platinum-etoposide.

Result:
One hundred and sixty-two patients were included in this study. The median follow-up duration for surviving patients was 38 months (range, 6-105 months). Among 123 patients (76%) who died, 104 patients died due to disease progression, 11 died due to unknown cause and 8 died due to other cause. Ninety-three patients (57%) underwent PCI, and the 3-year disease specific survival (DSS) rates were 20%(12-35) in patients without PCI and 43%(33-55) in those with PCI (p<0.001). Concerning the recurrence pattern, the frequency of BMFR was significantly higher in patients who did not underwent PCI compared with those who did (49% vs 25%, p=0.008) although no significant difference in frequency of all site of recurrence was observed between patients without PCI and those with PCI (83% vs 73%, p=0.109). Regarding the impact of BMFR on prognosis, patients with BMFR exhibited significant shorter DSS than those whose first recurrence sites was other than brain metastases (3yr-DSS 6% vs 22%, p=0.007), and patients who developed BMFR without PCI exhibited significant lower DSS compared with those who developed BMFR with PCI (3yr-DSS 0% vs 17%, p=0.005). In addition, 68% of patients who did not underwent PCI exhibited multiple BMFR lesions with 5 or more while 12% of patients who did (p<0.001).

Conclusion:
The results of this study indicated that BMFR exhibited significant negative impact on prognosis after radical treatment for LD-SCLC and the difference in the number of metastatic lesions in BMFR between patients with PCI and those without PCI might affect clinical outcomes.

Background:
Tri-modality management using chemoradiotherapy followed by surgical resection is the current standard of care for patients with pancoast tumors. Surgical resection is considered the key element to improve patient outcomes, however it involves a lengthy operative procedure with surgical/anesthetics related side effects, long period of rehabilitation, and significant additional cost. In the era of image-guided radiotherapy, we are able to paint the dose around the tumor, targeting it with a high dose of radiotherapy, while avoiding sensitive organs such as spinal cord and brachial plexus. Here we present our institutional experience treating advanced pancoast tumors with chemoradiotherapy.

Method:
Patients are usually staged with chest-abdomen CT scan, upper chest/brachial plexus MRI (if needed), brain MRI and a PET scan. Patients are treated in a supine position, using head and neck mask as the immobilization device; and cone beam image guidance. The radiation dose is 45-70 Gy in 25-35 fractions concurrent with at least 2 cycles of chemotherapy. Patients with resectable tumors have surgery done within 4-6 weeks upon completion of chemoradiotherapy. Patients are followed every 3-6 months with CT scan and/or upper chest MRI for 3 years, then yearly after.

Result:
Of 205 consecutive patients with stage 3-4 NSCLC treated with radical dose radiotherapy at Southlake regional cancer center, 9 patients with pancoast tumor were detected. Two patients (2/9) died; one from a heart attack before completion of his treatment, and the other one from tumor progression (6 months after completion of tri-modality management). All the other 7 patients are still alive and free of disease (table 1).

Patients	Gender	Age (year)	Tumor size (cm)	Tumor pathology	Tumor invasion	Sugery	RT (Dose/fraction)	Date of completion RT (D/M/Y)	Date of last FU
1	F	61	6.2	Squamous carcinoma	T2, vessels, bone marrow	No	66 Gy/33 fr	06/02/2010	13/10/2016
2	M	62	11.3	Adenocarcinoma	C6-T3 ribs	No	70 Gy/35 fr	05/10/2010	25/01/2017
3	M	72	9	Adenocarcinoma	rib, vessels, neck	No	70 Gy/35 fr	25/01/2011	29/06/2016
4	M	66	9	Squamous carcinoma	C7-T5 vessels	No	66 Gy/33 fr	26/01/2015	11/01/2017
5	F	58	5.6	Adeno carcinoma	ribs	Yes	45 Gy/25 fr	21/03/2016	25/04/2017
6	F	55	6.6	Squamous carcinoma	T2-T4 rib chest wall soft tissue	No	66 Gy/33 fr	04/04/2016	08/06/2017
7	F	74	4.3	Adenocarcinoma	T3-T4 rib	Yes	66 Gy/33 fr	26/05/2016	20/06/2017

Conclusion:
Selected patients with pancoast tumors treated with high dose radiotherapy, using image guidance, concurrent with chemotherapy may have long term disease free survival. A multi-institutional study is warranted to conclude the management recommendation for these rare tumors.

Background:
Evidence regarding optimal follow-up (FU) strategies for patients after curative radiotherapy for NSCLC is limited, resulting in variable FU practice. The aim of this study was to describe the patterns of FU care for patients undergoing curative radiotherapy +/- chemotherapy.

Method:
A retrospective study was conducted of patients with Stage I-III NSCLC, undergoing a course of curative radiotherapy (a minimum dose of 50Gy), between 1/1/2007-31/12/2011 at three institutions. Data was collected from oncology records, including patient demographics, tumour characteristics, treatment and follow-up. At each FU, the reason for FU (routine or symptomatic), specialist seen and imaging performed were recorded, until an event (recurrence or new primary) was diagnosed. The censor date was 31/12/2016. Analysis of FU included univariate chi-square tests for categorical variables, t-tests for continuous variables, multivariate logistic regression analyses, and Kaplan Meier survival curves.

Result:
Two-hundred-and-eighty-three patients (183 males, 100 females) were identified with a median age of 72(36-91) years. Eighty-four (29.7%) were Stage I, 47 (16.6%) were Stage II, and 152 (53.7%) were Stage III. Pathology was large cell in 91 patients (32.2%), squamous cell in 100 (35.3%), adenocarcinoma in 68 (24.0%), and NSCLC NOS in 24 (8.5%). One-hundred-and-sixty-five (56.5%) patients received radiotherapy alone and the remaining 123(43.5%) received chemoradiotherapy. The average frequency of FU visits per year was 5.12, median number of FU visits to first event was 6, and median time to first event was 11 months. 73.7% of FU were routine, while only 16.2% were symptomatic. 1641 imaging tests were performed, equating to an average of 5.8 scans per patient, with only 98 resulting in a diagnosis of an event. Overall, recurrences were diagnosed in 175 patients of whom 85 were symptomatic and 90 diagnosed on routine imaging. New primaries were diagnosed in 23 patients, 15 with symptoms and 8 on routine imaging. Subsequent treatment was curative in 25 (14.3%) patients with recurrent disease and 18 (85.7%) with new primaries. Univariate and multivariate analysis determined that the method of diagnosis of an event (symptomatic vs routine) had no statistically significant impact on the intent of further treatment (curative treatment 13.8% v 7.7%, p=0.089) or on overall survival (2y OS 49.5% vs 51.6%, p=0.772).

Conclusion:
Following curative radiotherapy, patients undergo frequent FU with regular imaging. Despite this only a minority of patients who develop recurrence are suitable for curative treatment. The use of routine imaging did not impact on further curative treatment or improve overall survival.

Background:
The microscopic extension, known as clinical target volume (CTV), of the primary tumor in locally advanced non-small cell lung cancer for chest radiotherapy planning has been well studied. However, the CTV of the primary tumor in limited-stage small cell lung cancer (SCLC) for thoracic radiotherapy planning has not been reported. In this study, we tried to quantify CTV of primary tumor in limited-stage SCLC with pathological approach.

Method:
Patients with stage T1-2N0-1M0 small cell lung cancer, treated with two cycles of etoposide plus cisplatin neoadjuvant chemotherapy or without neoadjuvant therapy were eligible for this study. Routine radical lobectomy and mediastinal lymph node dissection were performed for these patients. After operation, the intact lung lobe specimens were perfused with 10% neutral formalin and inflated as the natural state in the body. Specimens should be fixed for at least 12 hours. The fixed specimens were sectioned at 3mm thickness along the cross-sectional position of the body, and for each slice, the gross tumor and its surrounding 2cm lung tissue was embedded with paraffin and sliced with 4μm thickness. HE routine staining and CD56 immunohistochemical staining were applied. The slides were scanned by KFBIO reading software, and the pathologists confirmed the microscopic extension of the surrounding tumor. The minimal distances between microscopic nidus and the edge of gross tumor were measured.

Result:
Eight patients were enrolled, of whom 4 received 2 cycles of neoadjuvant chemotherapy and 4 received surgery without any neoadjuvant therapy. In patients who received neoadjuvant chemotherapy, one was diagnosed as squamous cell carcinoma after surgery and was excluded for further evaluation of CTV, while 1 patient achieved pathologically complete remission of tumor.The median range of CTV in patients received or did not receive neoadjuvant chemotherapy were 0.4mm (0.05mm-2.76mm) and 1.7mm (0.08mm-12.7mm) respectively, a margin of 1.4mm and 10.2mm could cover 95% of microscopic nidus (P=0.00).

Conclusion:
The preliminary results of this study indicated that to take into account 95% of the microscopic nidus, a 1.4mm and a 10.2mm margin should be need for CT-simulation based thoracic radiotherapy planning for limited-stage SCLC patients who were treated with or without induction chemotherapy. However, the sample size of this study was small and more cases are needed to justify the results.

Background:
The management of NSCLC-patients has evolved towards a more personalized care approach. Currently, the outcomes of concurrent chemoradiation (CCRT) in NSCLC patients are evaluated after the end of treatment, while individual treatment response during treatment is not taken into account. To pursue adaptive radiotherapy, it is important to distinguish responses during treatment and correlate this with outcome. Therefore, the aim was to identify subgroups that show a distinct treatment response during CCRT and correlate this with treatment outcome.

Method:
NSCLC-patients treated with CCRT between 2007-2013 were included. Treatment consisted of 66Gy/24 fractions with concurrent daily Cisplatin. Deformable image registration of the planning-CT to all Conebeam-CTs acquired during treatment was performed, and the gross tumor volumes on the ConeBeam-CTs were measured. Latent Class Growth Modeling was used to identify subgroups showing a distinct treatment response of the primary tumor during CCRT. Cox survival analysis was performed to assess the association of subgroup ‘membership’ with overall survival (OS) and progression free survival (PFS).

Result:
402 patients were included. Median follow-up was 63 months and median OS was 23 months (95%CI 20-26 months) and median PFS was 18 months (95%CI 15–21 month). Six different patterns of treatment response were identified (Figure1). Group 1&2 showed a relatively stable pattern during treatment. Group 5 showed tumor progression in the first week followed by sharp decrease in tumor volume. All other groups showed a decrease in tumor volume from start of treatment. Remarkably, the groups 1&5, that didn’t show a decrease in tumor volume from the beginning of treatment, had a significantly improved OS and PFS. Figure 1



Conclusion:
Six groups showing a distinct treatment response during CCRT were identified. This is an important finding to be able to pursue adaptive radiotherapy. First analyses revealed an association between tumor response during treatment and OS and PFS. In-depth analyses are warranted.

Background:
In Bangladesh lung cancer stands in the top 3 in both male & female patient. Most of the patients are diagnosed as advanced stage. Radiotherapy plays an important role in the management of lung cancer. Motion management is a big challenge in GTV & ITV delineation, RT planning and minimizing the dose to OAR (Organ at risk). In Bangladesh there are 18 LINAC (12 in private & 6 in Govt. centers) is available. But, only one centre has the facility to do 4DCT simulation (United Hospital Limited). So it is time to consider that slow CT (computerized tomography) simulation can help to delineate GTV more precisely than axial CT simulation in Radiation Center of a developing country like Bangladesh.

Method:
: In this study we included ten case of carcinoma of primary lung who have been treated using Radiotherapy. Most of the patients are in stage II or stage IIIA. All patients were treated with Radiotherapy In 3DCRT technique concurrently weekly cisplatin 30mg/m2. All the patients were underwent went 4 modes of CT scan Axial, Helical, Slow & 4D-CT using GE discovery 16 Slice PET-CT scanner. And KV-CBCT for the treatment verification was taken at regular interval. For standardization all the patients underwent different mode of scan using 2.5mm slice thickness and. Slow CT were performed using axial mode scan by increasing the CT tube rotation time as per the breathing period. In 4D-CT scan Respiratory cycle was divided in to 10 phases and scans were performed throughout the entire respiratory cycle. Maxmium Intensity Pixel (MIP), Minimum Intensity Pixel (MinIP) & Average Intensity Pixel (AvIP) were derived from the 10 Phases. GTV or ITV volumes were delineated for all the patients in all the scan modes (ITVAX - Axial, ITVSP - Spiral, ITVSl– Slow, ITVMIP - 4DCT) in the treatment planning system All the GTV or ITV volume were measured, documented and compared with the different modes of the CT scan. We have also done 4 modes of CT scan in moving phantom too. Phantom volume was delineated in all modes of scan (MIP, Slow, Axial and Helical) in treatment planning system and the difference in volume was compared.

Result:
The mean ± sd (range) for MIP, slow, axial, & helical were 36.5 ± 40.5 (2.29-87.0), 35.38 ±39.52 (2.1 – 82), 31.95 ± 37.29 (1.32 – 66.9) & 28.98 ± 33.36 (1.01 – 65.9. Overall underestimate of helical scan and axial scan compared to MIP is 21% and 12.5%. CBCT and slow CT volume has a good correlation with the MIP volume. Ratio of 4DCT and Slow CT scan is 0.97. While we compare volumes (Ratio) of moving phantom of different modes of CT scans, the findings were MIP 50.29cc (1.0), Slow CT 48.83cc (0.97), Axial 35.6 cc (0.71) and Helical 31.68 cc (0.63). Volume comparison in moving phantom also showed good correlation and match in both 4DCT and Slow CT scan. From this study we can draw inference that Slow CT volume has a very good correlation with the MIP volume, the ratio of the Slow CT vs MIP is 0.97 ± 0.12, where as The ratios of GTV(MIP) to GTV (Axial) and GTV(helical) were 0.87 ± 0.14 and 0.79 ± 0.12, respectively, which showed a marked difference in GTV volumes.

Conclusion:
Though, it was limited number of patients in this observational study. But, it showed a very crucial finding to improve the GTV or ITV delineation and to limit the dose to OAR. As slow CT Scan can be done in any CT simulator machine, where there is no facility to do 4DCT scan. Although, 4DCT scan is a gold standard for GTV delineation for the motion management of lung cancer, but, in the absence of 4DCT scan, slow CT can be used. We should need multiple centers, large number of patients’ data to validate this finding

Background:
The literature on dose-volume parameters and pneumonitis is extensive, whose results are inconsistent, both for the best predictive metrics and significant comorbid factors. To develop an improved functional equivalent uniform dose (fEUD) with perfusion factors of single photon emission computed tomography (SPECT) images as predictors of radiation peneumonitis (RP) in patients undergoing curative radiotherapy (RT).

Method:
Figure 1Functional lung imaging was performed using single photon emission computed tomography (SPECT) for perfusion imaging. Perfusion factors were defined as the mean percentile perfusion levels of the four areas of the top to 75%, 75% to 50%, 50% to 25%, 25% to 0%, respectively. fEUD were calculated from perfusion factors in computed tomography (CT) and SPECT fused images and standard dose-volume parameters were extracted from radiotherapy treatment planning. Total lung (TL) volumes minus GTV, V5 and V20 (the percentage of normal lung tissue receiving more than 5Gy and 20Gy), fEUD (equivalent to 2Gy/fx, α/β=3.3, a=1), and mean dose of TL were analyzed to evaluate correlations between RP, which was evaluated using Common Terminology Criteria for Adverse Events (CTCAE) version 4.03. Statistical significance was defined as a P-values of <.05.



Result:
A total of 28 patients treated with intensity modulated RT or 3D conformable RT were analyzed, grades≥3 RP were observed in 6 patients. There was only a significant difference in fEUD between patients with RP and free (p=0.007). Insignificant tendency of difference was showed in the mean dose of TL (p = 0.052). V5 and V20 of whole lung were almost identical between the two groups (p=0.161 and 0.197), as the most values are below the recommended thresholds from published papers.

Conclusion:
SPECT-based equivalent uniform dose presented more significant to predict RP compared to standard dose–volume parameters and mean dose.

Background:
Early Stage lung cancer can be treated with surgical resection and SABR. In clinical practice, it may be important to find a characteristic matrix for precision radiotherapy. The characteristic matrix for a lung tumor should include: 1) a time factor that incorporates the motion pattern of the tumor, 2) spatial factor which is related to volumetric information including location and surrounding environment, and 3) intrinsic characteristics which are related to the biologic information and could be represented by imaging characteristics. In this investigation, the typical heterogeneous characteristics matrix of lung tumors treated with SABR were studied.

Method:
A total of 38 patients with 41 lung tumors were retrospectively reviewed. All targets were treated with SABR. 4-Dimensional Computed Tomography (4DCT) was used for simulation. The time factor parameter was defined by the breathing cycle signal recorded by an infrared reflector. The internal target volumes (ITV) were generated by the cine image set through phase binning method. The intrinsic characteristics of tumors were described by the image information inside the tumor, which was represented by the ITV. This target image information was described by the minimum Hounsfield Unit (HU), maximum HU, mean HU, and standard deviation.

Result:
For spatial location, there were 23 tumors in the right lung and 18 tumors in the left lung. For phase surrogate signal, the mean amplitude was 0.7 cm with standard deviation (SD) = 0.3 cm. Mean breathing cycle was 3.5 seconds (SD = 1.1). Mean prescription isodose volume (PIV) was 38.3 cc (SD = 37.7) and the mean corresponding diameter was 3.8cm (SD = 1.1cm). Mean ITV volume was 13.4 cc (SD = 20.4) and the corresponding diameter was 2.4 cm (SD = 1.2). With regard to imaging information, ITV minimum HU was -883.1 (SD = 120.3); ITV maximum HU was 222.3 (SD = 487.9); ITV mean HU was -224.9 (SD = 188.2). PTV minimum HU was -937.1 (SD = 62.2); PTV maximum HU was 437.1 (SD = 506.1); PTV mean HU was -482.3 (SD = 192.6).

Conclusion:
A characteristics matrix was computed for SABR treated lung tumors based on time, imaging, and spatial information. The matrix could be further expanded and assessed for intra-fractional and post treatment analysis. Other imaging modalities such as MRI and incorporation of immune system stimulation mechanisms can be combined with big data analysis methodologies for patient treatment options in precision medicine through similarity matching.

Background:
Despite wide adoption of intensity-modulated radiation therapy (IMRT), thoracic re-irradiation of lung cancer has been challenged by the tolerance doses of normal tissues such as lung and esophagus. We retrospectively analyzed the local control rate and the toxicity of thoracic re-irradiation using highly conformal radiotherapy.

Method:
Between September 2008 and November 2015, 50 patients received thoracic re-irradiation with 3D conformal therapy (n=6, 12%), linac-based IMRT (n=11, 22%), helical tomotherapy (n=31, 62%), and stereotactic body radiotherapy (n=2, 4%). Total lung mean dose (~Total~D~mean~) was calculated by the sum of lung D mean of initial radiotherapy (RT) and that of re-RT.

Result:
Twenty-two patients had distant metastasis at the time of re-RT. Median follow-up time was 14.0 months (range; 1.4 - 65 months) from re-RT. At the time of analysis, local progression after re-RT was observed in 40% (n=20) of patients. The 1-year and 2-year local progression-free survival (LPFS) were 60.3% and 47.1%, respectively. Median time to local progression was 8.5 months (range; 1.4 - 65 months) and median survival from re-RT was 13.9 months (range; 1.4 - 65 months). Median total dose was 70.3Gy (range; 33 - 131 Gy) with median fraction size of 2Gy (range; 1 - 10.9 Gy). Local control rate of 3D-conformal RT and that of IMRT was 50% and 61.4%, respectively. When we compared the LPFS of patients who received re-RT using IMRT, smaller PTV volume was associated with greater LPFS with marginal significance: the 1-year LRFR of PTV<57.5 cm[3 ]and PTV≥57.5cm[3]were 74.5% and 51.7%, respectively (p=0.054). Acute grade 1 and 2 radiation pneumonitis were observed in 16% (n=8) and 10% (n=5), respectively. No acute grade 3 or higher radiation pneumonitis occurred. ~Total~D~mean~of patients with and without acute radiation pneumonitis showed significant difference: 14.8Gy with acute radiation pneumonitis and 12.0Gy without acute radiation pneumonitis (p=0.048). Acute grade 2 esophagitis was observed in 6 patients (12%), and no grade 3 or higher esophagitis occurred. Chronic grade 2 radiation pneumonitis occured in 5 patients, and no chronic grade 3 or higher complications was observed

Conclusion:
The stratification of patients receiving thoracic re-RT is crucial in extending the indications and improving the efficacy of IMRT. Based on appropriate stratification, thoracic re-RT of lung cancer can be safely delivered using highly conformal radiotherapy with excellent local control and acceptable toxicity.

Background:
A potential benefit of integrating MR imaging in the radiotherapy planning and treatment process is the improved soft tissue contrast. However, this benefit may be obscured if there are substantial variations in the way that both target and organs at risk (OARs) are contoured, either at the time of initial simulation or at each fraction. The aim of this work is to develop standardised MR imaging sequences for target and OAR contouring in lung cancer patients for integration into the planning and treatment process for MR image-guided radiotherapy.

Method:
11 lung cancer patients were recruited across 3 European centres using a standardised MR protocol with various soft tissue contrasts as well as numerous respiratory management techniques. All MR images were acquired at 1.5 T with 3.5 mm slice thickness. 7 radiation oncologists and 3 radiologists reviewed the MR images alongside CT and FDG-PET-CT, rigidly registered to tumour area, to determine the most adequate sequences for OAR and tumour visualisation/delineation.

Result:
The most adequate MR sequences were found to be: Radial 3D T1, Turbo Spin Echo (TSE) fat-suppressed (fat-sat), DIXON TSE and T2 TSE (table 1). The sequences aided in visualisation of tumour, nodes and OARs including oesophagus, proximal bronchial tree, trachea, heart, lungs, spinal cord/canal and brachial plexus. The TSE (fat sat) and DIXON TSE gave strong signal from tumour, nodes and brachial plexus, aiding in their visualisation. The motion averaging characteristics ofthe Radial 3D T1 improve image consistency over slices, whereas the TSE sequences can include motion induced distortions.

Table 1. The most adequate sequences identified for tumour, node and OAR visualisation. These sequences have been acquired with varying contrast and different respiratory management techniques.

Sequence	Respiratory Management	Purpose	Coverage
Radial 3D T1 (fat sat)	Radial k-space sampling	Tumour, nodes, OARs, spatial reference	Ful thorax
TSE (fat sat)	Respiratory triggered	Tumour, nodes, OARs	Tumour (if tumour below aortic arch)
DIXON TSE	Multiple signal averages	Brachial plexus, tumour, nodes	Tumour and Brachial Plexus (if tumour above aortic arch)
T2 TSE	Respiratory triggered	Tumour, nodes, OARs	Full thorax

Conclusion:
A set of MR sequences suitable for lung cancer radiotherapy planning have been identified which provide adequate visualisation of tumour, nodes and OARs. The next step is to acquire these sequences in a set of patients and assess intra-/inter-clinician variability in target and OAR delineation, and develop an MR lung contouring atlas.

Background:
Early mortality after (chemo)radiotherapy can be caused by treatment-related toxicities and thus by delivered doses to normal lung and heart tissues. However, prediction models for mortality incorporating dosimetry are lacking. This study explores the prognostic value of common dosimetric features.

Method:
Two prospective cohorts containing 218 and 181 curatively treated stage I-III lung cancer patients from 2003-2007 and 2013-2016 periods, respectively, were studied. Prescribed dose was 66Gy/2Gy (concurrent chemotherapy), 66Gy/2.75Gy (sequential or no chemotherapy) or a similar schedule. Clinical (WHO performance status, age, T stage, N stage and primary gross tumor volume (GTV)) and dosimetric (mean lung dose (MLD) and mean heart dose (MHD)) covariates were analysed. Cox regression models of survival and a logistic regression model for the 12 month mortality endpoint were optimized using forward stepwise selection (p<0.05).

Result:
Median follow-up time was 80.2 and 20.2 months in dataset 1 and 2, respectively. MHD (HR=1.023, p=0.001) and WHO performance status (HR=1.25, p=0.03) were selected in the Cox model for dataset 1. Tumor volume (HR=1.0015, p=0.001), WHO performance status (HR=1.023, p=0.02) and MHD (HR=1.0030, p=0.03) were selected in dataset 2. Adding time-dependent covariates revealed a decreasing GTV HR over time in dataset 1 (p=0.02), while MHD risk did not significantly change with time. Worse survival observed in a high MHD subgroup indeed only starts after 8 months (Figure 1). 12 month survival modeling included the covariates MHD (optimal cut-off 22Gy) and GTV (AUC=0.71). In dataset 2, these covariates and cut-off resulted in a model with AUC=0.63. Figure 1



Conclusion:
Mean heart dose is an independent risk factor for early mortality in two cohorts with different treatment periods and techniques. The best classifier for 12 month mortality risk was obtained with the MHD<22Gy constraint, which could be used in model-based implementation of proton therapy.

Background:
A large proportion of patients with locally advanced/metastatic Non-Small Cell Lung Cancer (NSCLC) present disease progression despite aggressive treatments and will further receive immunotherapy. Pneumonitis is an uncommon but potentially fatal toxicity related to immunotherapy treatment. The association with the history of radiotherapy and the risk of developing pneumonitis have not been well described. We associated the history of radiotherapy with the development of pneumonitis in patients receiving immunotherapy.

Method:
Clinical information was retrospectively obtained from histologically confirmed advanced NSCLC patients treated from February 2013 to February 2017. Clinical and radiologic features of pneumonitis were collected from patients receiving immunotherapy. We sought associations between pneumonitis incidence and clinical characteristics.

Result:
Of 59 patients who received immunotherapy 61.7% were treated with nivolumab, 29.9% with pembrolizumab and 6.7% with the combination durvalumab plus tremelimumab. Immunotherapy treatment was administered in first line in 26.6% of patients, 28.3% received in second line and 36.7% in third or more line of treatment. Twenty-five of the 59 patients (41.7%) received previous radiotherapy, 16 of them (26.7%) to the lung and 9 (15%) to the thoracic spine. Fifteen (15/59) patients (25%) developed pneumonitis; this occurred irrespective of line of therapy in which immunotherapy was received (first line: 38.5%; second line: 33.5%; third line or more: 26.7%). No association was found between line of treatment and pneumonitis development. Median time from therapy initiation to pneumonitis was 4.5 months (range 18 days - 13.1 months). For any grade of pneumonitis, the percentage was of 25% (15 patients) of which 48% (12/25) had received radiotherapy, Grade >2 pneumonitis was seen in 10 patients (17%) and 32% (8/25) had history of radiation therapy. All Grade 3 pneumonitis events (n=4) presented in patients with previous lung radiotherapy. The incidence and severity of pneumonitis was higher in patients who had received radiotherapy (OR, 95% CI: 6.8 (1.6 – 28.5); p=0.009).

Conclusion:
The incidence of pneumonitis related to immunotherapy treatment could be underestimated. We observed an increase in the risk and severity of pneumonitis in patients with previous radiation therapy and subsequent treatment with immunotherapy, regardless of used drug or line of therapy. In these patients, we recommend close clinical and radiologic follow-up.

Background:
At our institution, we commonly treat brain metastases (BM) adjacent to critical structures with a smaller dose prescription (DP) to reduce the likelihood of toxicity. We sought to evaluate the impact of DP on LF and RN for small- to medium-sized BM (≤ 2 cm) from lung cancer.

Method:
A prospective registry of BM patients treated with gamma knife SRS between 2008 and 2016 was interrogated to determine per lesion rates of LF and RN. Each lesion was followed until LF or RN or at last MRI follow-up. Defined criteria were used to differentiate LF from RN. Whole brain irradiation (WBI) was a censoring event.

Result:
From 1,465 potential subjects, 345 small- to medium-sized BM from 151 lung cancer patients were evaluated. Median radiographic follow-up was 10.2 months. Median lesion volume and diameter were 0.17 cm[3], and 0.81 cm, respectively. The DP for 71 lesions (21%) was 15 Gy, and ≥ 20 Gy (median 21 Gy; 20-24Gy) for 274(79%). Most lesions were ≤ 1 cm (65%). Median number of SRS was 2 (1-4) and 36 patients received salvage WBI. Sixteen lesions (4%) developed LF and 12 (3%) developed RN. Freedom from local failure at 1 year (FFLF) for 15 Gy, and ≥ 20 Gy, was 80%, and 95%, respectively (p=0.02). FFLF for lesions ≤1cm, and >1 cm, was 95%, and 78%, respectively (p<0.01). Freedom from RN at 1-year (FFRN) for DP 15 Gy, and ≥ 20 Gy, was 98%, and 96%, respectively (p=0.3). FFRN for lesions ≤ 1cm, and > 1 cm, was 98%, and 93%, respectively (p=0.01). FFLF and FFRN for lesions ≤1 cm and >1 cm, according to DP, are shown in Table 1.

	Lesion size
	≤1 cm		P value	>1 cm		P value
	DP			DP
	15 Gy	≥20 Gy		15 Gy	≥20 Gy
FFLF	88.8%	96.4%	0.42	53.4%	88.2%	0.08
FFRN	100%	98%	--	98%	92%	0.43

Conclusion:
Our results suggest that, particularly for lesions >1 cm, DP ≥ 20 Gy correlates with improved FFLF, and similar FFRN rates, compared to DP 15 Gy.

Background:
Stereotactic body radiotherapy (SBRT) is the standard of care in patients with medically inoperable early stage NSCLC and an effective method of treatment of lung metastases (LM) in oligometastatic patient.We evaluated local control (LC), overall survival (OS), cause-specific survival (CSS), and related toxicity in the both group of patients treated in our center.

Method:
Between April 2012 and September 2016, 107 lung lesions were treated with SBRT; 62 early NSCLC(58%) and 45 LM(42%). Three risk-adapted fractionation schemes that ensure DBE>100Gy were considered: 3x18Gy, 5x11Gy and 8x7.5Gy. Kaplan-Meier(KM) curves were used to evaluate LC,OS and CSS. We analyzed toxicity and predictive factors.

Result:
The median follow-up was 16 months(range 6-44). For primary NSCLC: LC,CSS, and OS at 2 years were 100%,91% and 64%, respectively.19 patients died, 5 because of a lung cancer and 14 due to concurrent disease. Regional relapse was observed in five patients(7.6%).Six patients(9.2%) developed distant metastases. There was no statistically significant difference in OS and CSS when comparing peripheral-central tumors or T1-T2 tumors. However, T2 and central tumors showed lower survival rates. For lung metastatic lesions: LC,CSS, and OS at 2 years were 69%, 66% and 66%, respectively.The primary tumors were:colorectal 17(38%), lung 15(33%) and others 13(29%).There was no statistically signficant difference in survival among primary tumor or 1-3 metastatic lesions, but colorectal primary tumors and more than one lung metastases had lower survival.We compared LC and CSS in both group of patients. There was a statistically significant difference in local control(p=0.002) and CSS(p=0.027) among the primary tumors or lung metastases(Figure 1).According to RTOG,≤G2 lung and skin acute toxicity was 5% and 3% respectively.Lung late toxicity ≤G2 was 22.4%. No patients developed >G2 toxicities. Figure 1



Conclusion:
With a 2-year LC rate >95% with limited toxicity, SBRT confirms as state-of-the-art treatment for medically inoperable early stage NSCLC and effective options for oligometastatic patients.

Background:
The aim of this study was to perform a survival comparison between stereotactic body radiotherapy (SBRT) and sublobar resection (SLR) in patients with stage I non-small cell lung cancer (NSCLC) at high risk for lobectomy.

Method:
All patients who underwent SBRT or SLR because of medical comorbidities for clinical stage I NSCLC from January 2008 to December 2015 were reviewed retrospectively. Propensity score matching was performed to reduce selection bias between SBRT and SLR patients based on age, gender, performance status, tumour characteristics, pulmonary function. Overall survival (OS) and recurrence-free survival (RFS) were estimated with Kaplan–Meier method.

Result:
Forty-nine patients were matched into each group of SBRT and SLR (include 27 underwent segmentectomy, 22 underwent wedge resection). There were 32 and 30 men with median age of 67 and 69 years, respectively. Median follow-up was 25 months. In terms of treatment-related mortality, the 30- day mortality rates for the SBRT and SLR groups were 0 and 2.0 %, respectively. Patients treated by SBRT had a tendency to increase 3-year OS compared with SLR (94.0% versus 78.1%; P= 0.064).There was no difference between two groups in 3-year RFS (61.8% versus 65.7%; P=0.864). In a subanalysis, 3-year OS after SBRT was greatly better than wedge resection subgroup (94.0% versus 67.4%; P= 0.026), but there was no significant difference between SBRT and segmentectomy in 3-year OS (94.0% versus 88.0%; P= 0.212).

Conclusion:
SBRT had a tendency to increase OS compared with sublobar resection in patients who are not medically fit for lobectomy with clinical stage I NSCLC. However, OS after SBRT was better than wedge resection subgroup. SBRT can be an alternative treatment option to segmentectomy for patients who cannot tolerate lobectomy because of medical comorbidities. A randomized prospective study is necessary to determine survival in compromised SBRT and sublobar resection.

Background:
Radiotherapy is carried out five days in a week. Endostar is a target drug, its adminstration is d1-14 in three weeks. We changed endostar adminstration mode as the same of radiotherpy, five days in a week. Now we evaluate the efficacy and safety of different administration dose of endostar combined with concurrent chemoradiotherapy in ad vanced non-small cell lung cancer(NSCLC).

Method:
From September 2010 to December 2015，40 patients with local advanced NSCLC were recruited. All the patients received radiotherapy with 60Gy and two circles chemotherapy with docetaxel 75mg/m[2] d1 and cisplatin 75mg/m[2] d1～3，combined with two different administration dose of endostar(75mg/m[2] d1-14 in three weeks for two cycles and 75mg/m[2] d1-5 in one week for six cycles concurrent with radiotherapy. Tumor response were evaluated with RECIST1.1 criteria. Acute toxicities were evaluated with NCICTC3.0 and RTOG criteria.

Result:
In group one，2 patients achieved complete response，10 partial response，3 stable disease，5 progressive disease. Overall response rate was 60%. The 1-year progressive-free-survive rate was 40%. The 1-，2-year survival rates were 70% and 25%. The median overall survive is 19.9 month. In group two，overall response rate was 75%. The 1-year progressive-free-survive rate was 50%. The 1-，2-year survival rates were 80% and 45%. The median overall survive is 22.7 month. In group 2 there had a good survival and patient compliance, but there were no significant difference between two groups (P>0.05).

Conclusion:
Endostar combined with concurrent chemoradiotherapy is useful and well-tolerance. Administration mode of endostar combining with chemoradiotherapy, which is five day in a week for six weeks, had a better result and patient compliance.

Background:
To analyse the relation between clinical outcomes and dosimetric indices for SBRT lung treatments.

Method:
96 lung lesions were treated with SBRT(6MV photons 3DCRT). Planning CT included whole lung. 4DCT of tumor area was used to obtain a MIP-based ITV, with three risk-adapted fractionation schemes [3x18Gy, 5x11Gy, 8x7.5Gy(BED>100Gy)]. In treatment delivery the tumor was centered online using CBCT and its movement validated by fluoroscopy adjusting the gating limits to the breathing amplitude. Toxicity and dosimetric indices for PTV and OAR were evaluated and correlated with the clinical outcomes 6 months after radiotherapy

Result:
Table1 shows the dose constraints as well as the results of the dosimetric indices. It was found that 95%/75% of the patients developed G=0 acute/late toxicity, 3%/0% G=2 acute lung/skin toxicity, 3% G=2 late lung toxicity and no patients developed G>2 toxicities. Figure 1 displays the correlation between V~100~, V~90~, CI and the clinical outcomes after 6 months of radiotherapy. Only the CI was statistically significant(t-test p=0.017) as an indicator of the ratio between complete/partial responses(mean CI=1.1/1.05) Figure 1 Figure 2





Conclusion:
The lung SBRT technique is safe(no G> 2 toxicity has been reported) even for cases with OAR compromise. The CI has been statistically significant as a predictor of complete tumor remission at 6 months.

Background:
Continuation of Epidermal Growth Factor Receptor Tyrosine Kinase Inhibitors (EGFR TKI) in stage IV Non-small cell Lung cancer (NSCLC) patients harboring sensitive mutation (Exon 19 deletion /Exon 21 L858R mutation) upon first progression according to RECIST criteria were shown to be effective in ASPIRATION study and can prolong the use of EGFR TKI for 3 months till frank disease progression. Local ablative radiotherapy (LAR) on oligo-progression is increasingly advocated but the actual effect is to be determined.

Method:
Medical and radiotherapy records of patients given LAR from 2012-2017 were screened at a single centre. Patients with stage IV NSCLC harboring sensitive epidermal growth factor receptor (EGFR) activating mutations having extra-cranial oligo-progression and given LAR were included in the study. Patients’ demographics, site of oligo-progression, radiotherapy sites and dose/fractionation schedules were captured. Durations from starting of first line EGFR TKI to LAR was calculated (PFS1). Local progression free survival (L-PFS), overall progression free survival from LAR to further progression that led to stop of EGFR TKI (O-PFS) and overall survival (OS) were analyzed with Kaplan-Meier method.

Result:
There were 15 eligible patients with total 17 sites of oligo-progressive sites treated. There were 6 male and 9 female patients. The mean age was 59.6 years (36.5-82 years). All were treated with first-generation EGFR TKIs. The median duration PFS1 was 13.0 months (6.0-36.1 months). Treatment sites included 13 lung lesions and 4 bone lesions. The mean equivalent dose (2Gy) was 105Gy (64.5-122Gy). The median follow up time was 13.3 months. Ten out of 15 patients had CEA drop after treatment, with the median duration from treatment to first drop of CEA being 1.7 months. The median L-PFS and OS were not reached. The median PFS2 was 9.7 months (2.2-15.1 months). Eight out of 10 patients had second line/ third line treatment with either afatinib/ osimertinib chemotherapy or immunotherapy. Toxicities of radiotherapy were minimal and only grade 1 pneumonitis or pain flare documented. Duration of PFS1 was not found to affect duration of O-PFS.

Conclusion:
LAR appears to be a reasonable treatment approach in the event of oligo-progression in patients with advanced NSCLC harboring activating EGFR mutations. Longer follow-up and a larger cohort are underway to assess its impact on survival.

Background:
Small-cell lung cancer (SCLC) accounts for approximately 15% of all lung cancers, and is characterized as being extremely aggressive, often displaying rapid tumor growth and multiple organ metastases. In addition, the clinical outcome of SCLC patients is poor due to early relapse and acquired resistance to standard chemotherapy treatments. Several molecular targeted therapies were evaluated in SCLC; however, they failed to improve the clinical outcome. The receptor tyrosine kinase MET is a receptor for hepatocyte growth factor (HGF). Although aberrant activation of HGF/MET signaling is known as one of the crucial mechanisms enabling cancer progression and invasion, the mechanisms in SCLC have not been elucidated clearly. The aim of the present study was to evaluate the effect of inhibiting the HGF/MET pathway on tumor progression in SCLC with multi-organ metastasis.

Method:
We used eight human SCLC cell lines to elucidate the effect of MET inhibition on tumor progression. MET inhibitors, crizotinib and golvatinib, were used in this study in vitro and in vivo.

Result:
We found that the HGF/MET signaling was aberrantly activated in chemo-resistant or chemo-relapsed SCLC cell lines (SBC-5, DMS273, and DMS273-G3H) by the secretion of HGF and/or MET copy number gain. HGF/MET inhibition, induced either by MET inhibitors (crizotinib and golvatinib), or by siRNA-mediated knockdown of HGF or MET, constrained growth of these SCLC cells via the inhibition of ERK and AKT signals. We also revealed that treatment with either crizotinib or golvatinib in vivo suppressed the systemic metastasis of SBC-5 cell tumors in NK cell-depleted SCID mice, predominantly via cell cycle arrest.

Conclusion:
These findings reveal that the therapeutic potential of targeting the HGF/MET pathway for inhibition, to constrain tumor progression of SCLC cells exhibiting aberrant activation of HGF/MET signaling. We suggest that it would be clinically valuable to further investigate HGF/MET-mediated signaling in SCLC cells.

Background:
Small Cell lung cancer (SCLC), which belongs to neuroendocrine tumor (NT), is known for its highly aggressive clinical features and poor prognosis. The molecular underpinnings that may prognosticate survival and could increase our understanding of tumor development and progression are still poor understood in its highly aggressive malignancies. We aimed to describe the proteomic characteristics and biomarker profiling of high-grade NT, SCLC versus low-grade NT, carcinoid tumor (CT) by the state-of-the-art proteomics techniques utilizing clinical specimens.

Method:
In order to obtain the molecular-based differences between SCLC and CT, we performed proteomic analyses for these histological types in lung NTs. Based on definite histological diagnosis, cancerous cells were collected by laser microdissection from FFPE tissues of 6 patients each with SCLC and CT. To elucidate their protein expression profiles, a mass spectrometry-based quantitative proteomic analysis was conducted. After the different profiles were identified using a hierarchical clustering, we applied protein-protein interaction (PPI) network and gene-set enrichment analyses to assess how significantly proteins are expressed and which molecular pathways are activated in SCLC and CT.

Result:
A total of 1,991 proteins were identified from cancerous cell in FFPE tissues throughout these NT subtypes. By hierarchical clustering of the protein expression profiles, 12 cases were clearly classified into two groups of 6 SCLC and 6 CT. We finally identified 11 for SCLC and 44 for CT as tumor-specific proteins, respectively. In brief, our curation-based analysis of PPI networks built from proteins expressed significantly in SCLC cells has revealed that the activations of both mismatch repair system, and its resulted cell cycle pathway but suggested a fatal failure in disruption of mismatched DNA sequences, which suggested an occurrence of carcinogenesis progression simultaneously with an activity trying to return to normal, and which might suggest a heterogeneity of SCLC cells. Proteins expressed significantly in CT cells were the molecules known representatively as the biomarkers of NTs such as neurosecretory protein VGF, chromogranin-A, secretogranin-1 and -2. Besides those, we have found that the certain key network modules controlling cellular proliferation and apoptosis, affecting a series of pathways, as well as regulating genomic instability and DNA damage and repair, which also promotes cell transformation and tumorigenesis.

Conclusion:
Our clinical proteomic profiling of both SCLC and CT revealed their corresponding nature of aggressiveness and extent of prognosis, and detailed curation of PPI networks of interestingly suggested candidates of their therapeutic targets, which might reflect their disease mechanisms.

Background:
SCLC has a high response rate to first line chemotherapy but this is often short lived. Patients who relapse six months after first line chemotherapy can be retreated with platinum +/- etoposide and patients that relapse early are treated with CAV (cyclophosphamide, doxorubicin and vincristine), Topotecan or best supportive care (BSC). There have been no clinical trials comparing oral Topotecan with iv CAV.

Method:
A retrospective case note review was undertaken of 296 SCLC patients who received second line chemotherapy for recurrent SCLC in the West of Scotland. Results were analysed using STATA version 14.

Result:
SCLC patients that relapsed after first line platinum-based chemotherapy received CAV (n=161), platinum +/-etoposide (n=61), oral topotecan (n=67) and 7 patients received other cytotoxic agents. Median survival for patients who received CAV was 5.1 months (CI 4.1 – 5.7), 6.3 months (CI 5.2-8.6) for platinum/etoposide and 3.0 months (CI 1.9-3.6) for Topotecan. As expected platinum +/- etoposide was superior to CAV (Log rank p = 0.016) and Topotecan (p <0.0001), as these patients had demonstrated platinum sensitivity. CAV appeared superior to Topotecan (p = 0.001) and was better tolerated. 79% of Topotecan patients required a dose reduction versus 29% of CAV patients. 16% of Topotecan, 8% of CAV and no platinum +/- etoposide patients died within 30 days of receiving chemotherapy. 39% of Topotecan and 19% of CAV patients experienced either neutropenic sepsis or myelosuppression requiring a dose reduction or dose delay. There was no difference in time to progression after first line chemotherapy in the CAV and Topotecan populations. Figure 1



Conclusion:
In this West of Scotland retrospective data collection, patients treated with oral Topotecan appear to have a lower median overall survival and experienced more toxicity than those receiving iv CAV.

Background:
Small-cell lung cancer (SCLC) is a devastating illness with a median five-year overall survival of 5%. Focal Adhesion Kinase (FAK) is a non-receptor tyrosine kinase which regulates integrin and growth factor signaling pathways involved in cell proliferation, survival, migration, and invasion. FAK is overexpressed/activated in many cancers, including SCLC. We hypothesized that FAK overexpression/activation in SCLC contributes to its aggressive behavior and that FAK may represent a novel therapeutic target in SCLC.

Method:
Two SCLC cell lines, one growing in suspension (NCI-H82) and one adherent (NCI-H446), were treated with a FAK small-molecule inhibitor, PF-573,228 (PF-228) (1 to 5µM), or stably transfected with FAK shRNA and/or FAK-related non kinase (FRNK) domain (doxycycline-inducible) using lentivirus vector. Cell proliferation, cell cycle, apoptosis, motility, and invasion were studied by standard methodologies. FAK expression/activity was evaluated by WB. Active Rac1 expression was evaluated by WB after enrichment of cell lysates in active GTP-bound Rac using Rac pull down columns.

Result:
PF-228 decreased FAK activity by decreasing phospho-FAK (Tyr397) expression in both SCLC cell lines, without modifying total FAK expression. This induced significant inhibition of cell proliferation and DNA synthesis, cell cycle arrest in G2/M phases, and increase of apoptosis dose-dependently. PF-228 also decreased motility in the adherent cell line NCI-H446. Transfection of FAK shRNA decreased total and phospho-FAK expression but this did not affect cell proliferation, DNA synthesis, and cell cycle. We hypothesized that the absence of anti-proliferative effects was due to the loss of the FAK-targeting domain (FAT), normally attached to the focal adhesion complex where it inhibits other proteins supporting proliferation (such as Rac). To test this, we restored FAT function by transfecting FRNK, a physical repressor of FAK activity, into cells stably transfected with FAK shRNA and demonstrated inhibition of cell proliferation and DNA synthesis. Expression of FRNK in SCLC cell lines not previously transfected with FAK shRNA also significantly decreased cell proliferation and DNA synthesis. Moreover, FAK shRNA transfection increased active Rac1 levels, while FRNK re-expression in the cell lines previously transfected with FAK shRNA decreased it.

Conclusion:
This work demonstrates that FAK has a dual role in SCLC: 1/ it supports proliferation, migration, invasion, and inhibits apoptosis through the kinase domain, suggesting that inhibition of FAK kinase activity may represent a suitable therapeutic target for SCLC, and 2/ it inhibits SCLC proliferation through the non-kinase C-terminal domain FRNK, which keeps Rac inactive.

Background:
Scotland has one of the highest incidences of SCLC globally (cancerresearchuk.org). Lung cancer is the leading cause of cancer related deaths in the UK (cancerresearchuk.org), with SCLC having a worse prognosis than NSCLC and fewer treatment options. Initially, SCLC has a high response rate to first line chemotherapy but rapidly becomes resistant to chemotherapy. There is little evidence to support and guide third line chemotherapy regimens. A West of Scotland (WoS) retrospective database (2007 to 2013) has been developed to gain useful insights into this aggressive disease to improve patient outcomes.

Method:
A retrospective case note review was undertaken of 1325 SCLC patients from the WoS. 58 patients were identified as receiving third line chemotherapy. Information was collected regarding radiological staging, performance status (PS), treatment history, time to progression and survival. Results were analysed using STATA version 14.

Result:
17% of SCLC patients in this WoS database received no chemotherapy, 60% one line (n= 801), 22% (n= 296) two lines, 4% (n=58) three lines and 0.004% (n= 6) four lines of chemotherapy. Of the 58 patients that received third line chemotherapy 28 had Topotecan, 17 CAV, 7 platinum +/- etoposide and 6 included a variety of other cytotoxics. For all patients receiving third line chemotherapy the median survival was 4.7 months (CI 4.1-5.3) and progression free survival was 3.0 months (CI 2.5 – 3.6). Median survival was 3.5 months (CI 2.6- 5.6) for Topotecan, 4.1 months (CI 2.7-5.0) for CAV and 9.0 months (CI 1.8-12.2) for platinum +/ etoposide. 32 (55%) patients received 3 different regimens for each line of chemotherapy, 25 patients (43%) received a platinum-based chemotherapy regimen twice and 1 patient received a platinum based regimen all three lines of chemotherapy. Only 2 patients received a clinical trial novel treatment for their third line therapy.

Conclusion:
SCLC continues to have a poor prognosis with few patients receiving three lines of chemotherapy. The patients that maintain some sensitivity to platinum based chemotherapy tend to have a marginally better response to third line therapy. Clincial trial opportunities for this group of patients were limited. Patients should be enrolled in clinical trials wherever possible.

Background:
Metastasis are responsible for the death of 90% of patients with lung cancer (LC) indicating the necessity to know the multiple signaling pathways involved. Among them, high-grade neuroendocrine lung carcinomas (NELC) invade and metastasize rapidly. Therefore, biomarkers of aggressiveness in LC remain to be determined, especially in NELC. Epithelial to mesenchymal transition (EMT) genes profile emerge promise as indicator of invasion and metastasis. The aim was to investigate the expression of EMT markers and assessed their relationship with the clinicopathological features and prognosis.

Method:
Fresh frozen tissue from SCLC (n=15) and NSCLC (ADc n=23 and SqCC n=10) and matched normal tissue samples were collected for qRT-PCR analysis carried out on StepOnePlus™ Real-Time PCR with RT[2] Profiler PCR Array System for the EMT pathway with 84 target genes (Qiagen, Dusseldorf, Germany). Gene expression was correlated with clinicopathological variables in the SCLC and NSCLC groups. Survival curves were calculated using the Kaplan-Meier method and risk factors determined by multivariate Cox regression model. Differences were regarded as statistically significant at P<0.05.

Result:
Female patients presented significant higher expression of EGFR (p=0.03), ILK (p=0.05), JAG1 (p=0.01), MMP2 (p=0.04) and SNAI2 (p=0.04) genes. Tobacco history was associated with increased expression of EGFR (p<0.01), ITGAV (p=0.05), SPP1 (p<0.01) and WNT5A (p=0.02). NSCLC presented similar levels of EMT genes evaluated. Tumors from SCLC and NSCLC in advanced N and M stage expressed significant high levels of genes related to cellular membrane [EGFR (p=0.03), ILK (p<0.01), FR11 (p=0.05), ITGAV (p=0.02), ITGB1 (p<0.01), DSP (p=0.04)], extracellular matrix [COL5A2 (p=0.04), COL1A2 (p=0.04)], cytoplasm [GSK3B (p=0.01), VPS13A (p=0.02), MAP1B (p=0.01)] and nucleus SNAI2 (p=0.04). Interestingly, SCLC tumors expressed higher levels of FR11 (p=0.02), GSK3B (p=0.04), ILK (p<0.01), ITGB1 (p=0.01), JAG1 (p<0.01) and MAP1B (p=0.01) indicating more aggressiveness than NSCLC. A mathematical model controlled for N and M stage, histologic type and the gene expression showed that patients with SCLC expressing high levels of MMP2 and SPARC presented significant high risk of death (OR 5.41 and 4.94, respectively) compared to those with lower expression. Patients with NSCLC with low levels of ILK, SPP1, COL1A2, ITGB1 presented a low risk of death (OR -7.02, -0.4, -1.3 and -3.02, respectively).

Conclusion:
Different expression of EMT genes in SCLC and NSCLC, its relationship with histologic types, advanced stage, lymph node metastasis and death suggest a possible role of these markers in their malignancy, but more importantly provide a potential biomolecular marker to predict outcome.

Background:
SCLC accounts for 15%-20% of all lung cancer, and has poor prognosis. The aims of this study were to evaluate the patient characteristics and depict prognostic factors in a series of SCLC patients treated at Marmara University Hospital (MUH) Istanbul.

Method:
Among SCLC patients who were admitted to MUH since 01 January 2010, 154 had satisfactory data to analyse. Demographic data, pathology & radiology reports, lab investigations, information regarding local & systemic therapies were noted from written & electronic patient records. Patient and tumour characteristics were reported descriptively. OS difference between subgroups were analysed with Log-rank & factors that had independent effect on survival detected with Cox regression tests. OS data were calculated with Kaplan-Meier estimator. A p value <.05 was accepted as significant unless reported otherwise.

Result:
The median follow-up time was 17 (min-max; 3-84) months. Median survival time of all patients was 10 months; 1 year, 2 & 3 years survival rates were 41%, 22%, and 12%, respectively. Median survival of patients with limited stage and extension stage SCLC were; 22,6 Ms (9,9-35,3), and 9 Ms (7,2-10,8), respectively. On univariate analysis patient with low initial serum haemoglobin (<12 gr/dl), abnormal sodium or ALT (> 40 IU/l) levels, poor ECOG PS (2-3), advanced VA stage, having brain, liver, bone or adrenal mets, and having a paraneoplastic syndrome (PNS) had worse survival estimates. Whereas only ECOG PS (p=0.007, HR 2.2 [1.2-4]), and having a PNS (p=0.04, HR 1.66 [1.02-2.7]) maintained independent prognostic effect on survival in Cox analysis.

Conclusion:
Results of our small retrospective SCLC series showed that median survival of extensive stage SCLC patients is poor (< 1 year) which underlies the need of novel anticancer therapeutics for this group of patients. Our multivariate model pointed out well known prognostic factors but not the VA stage. This may be explained with the small population size of our study.

Background:
Inadequate drug delivered to target tumors contributes to ineffective treatment. However, the delivered drug concentration is difficult to assess in patients in a timely and clinically-relevant manner. To address this barrier to PARP inhibitor(PARPi) therapy, we evaluated a radiolabeled PARP inhibitor([F18]PARPi) as a pharmacodynamic biomarker. We hypothesized that [F18]PARPi PET imaging can measure PARP inhibitor concentration and activity intratumorally, thereby, predicting therapeutic efficacy. Here, we applied this approach to patient-derived xenografts(PDX) of small cell lung cancer(SCLC).

Method:
To study [F18]PARPi PET as a biomarker of talazoparib(TAL), SCRX-Lu149 PDXs were orally gavaged with different doses of TAL. Mice were injected with [F18]PARPi and imaged with PET, with the expectation that TAL would competitively block [F18]PARPi binding to PARP. Organ retrieval and gamma counting was performed for drug and radiotracer biodistribution. Ex vivo PARP enzymatic activity was measured by ELISA of PAR levels. Differences in PET uptake and the tumor volumetric endpoint (time to reach 1000 mm[3]) were analyzed by student t-test and the log-rank test, respectively.

Result:
In PK PET imaging with 0.2 mg/kg TAL, greatest blocking of the radiotracer was noted at 1 hour after gavage with less blocking as time from dosing was extended (avg of 3 mice: 4.5, 2.2, 2.7, 3.1, and 3.4% max injected dose per gram [ID/g] for untreated, 1, 3, 6, and 24 h after drug, respectively). [F18]PARPi PET differentiated between doses of 0.1 and 0.3 mg/kg TAL at 3 h after dosing (3.9 vs 2.1% ID/g or 13% vs 53% relative blocking, respectively; p=0.003). No differences were noted in heart, lung, esophagus, muscle, or bone. PET uptake correlated with ex vivo enzymatic inhibition/PAR levels (p=0.0009). PET measured differences in drug doses corresponded with improved outcomes for PDXs treated with 0.3 mg/kg in combination with radiotherapy(RT; median 84 days, p=0.04) but not 0.1 mg/kg + IR (66 days) compared to RT alone (52 days).

Conclusion:
[F18]PARPi PET can differentiate between multiple doses and timing of orally administered TAL and correlates with drug efficacy. This likely reflects differences in intratumoral drug level and demonstrates the potential of this PET radiotracer to assess differences in drug delivery and efficacy for patients treated with PARP inhibitors.

Background:
The impact of interstitial lung disease (ILD) on the clinical outcome in small cell lung cancer (SCLC) patients has not been fully understood. The purpose of this study is to investigate the impact of ILD on treatment and survival in SCLC.

Method:
From April 2006 to March 2016, 67 SCLC patients treated with chemotherapy at Kumamoto University hospital were evaluated. Therapeutic response, progression-free survival (PFS) and overall survival (OS) were compared between SCLC with ILD (n=16) and SCLC without ILD (n=51).

Result:
Patients’ characteristics including age, sex, smoking, serum LDH, stage (limited/ extensive disease) between two groups were similar, except for lower proportion of chemoradiotherapy in SCLC with ILD (7% vs. 57%, P= <0.001). SCLC with ILD had a significantly lower objective response rate than SCLC patients without ILD (50% vs. 84%, P= 0.005). SCLC patients with ILD had a significantly shorter PFS (median, 184 days vs. 290 days, P= 0.008) and OS (median, 241 days vs. 704 days, P= <0.001) than patients without ILD. In addition, multivariate analysis for PFS and OS revealed that the coexisting ILD was independent predictive factor for poor survival in SCLC patients treated with chemotherapy (PFS, hazard ratio [HR] 2.06, 95% confidence interval [CI], 1.01-4.18, P= 0.046, OS, HR 3.29, 95% CI 1.53-7.08, P= <0.002). Acute exacerbation of ILD was observed in 31% of SCLC patients with ILD.

Conclusion:
This study suggests that coexisting ILD is a predictive factor for poor PFS and OS in SCLC treated with chemotherapy.

Background:
Diffuse idiopathic pulmonary neuroendocrine cell hyperplasia (DIPNECH) is a rare, often benign pulmonary condition which is characterized by diffuse proliferation of neuroendocrine cells in respiratory epithelium. DIPNECH lesions are less than 5 mm in size and are limited to the basement membrane without any invasion. There is limited evidence regarding epidemiology, natural history of disease progression and management of this rare entity. We would like to present our single center experience.

Method:
Retrospective record based descriptive study of all DIPNECH patients managed at University of Kentucky Markey Cancer Center over past 5 years.

Result:
Our patient cohort had 8 females and one male with mean age of 64.5 years at the time of diagnosis. Dyspnea on exertion and dry cough were the commonest presenting symptoms. Three patients were under surveillance without treatment, two patients were treated with short acting somatostatin analog, another three were treated with long acting monthly somatostatin analogs and azithromycin combination, one patient got long acting somatostatin analog and everolimus combination. Two patients had concomitant well-differentiated neuroendocrine tumor of lung.

Conclusion:
DIPNECH is a rare pathology which can have profound effects on patient’s quality of life. Paroxysmal coughing spells can be difficult to treat. Our limited single center experience shows encouraging response to use of somatostatin analog, azithromycin and everolimus in the management of debilitating DIPNECH associated symptoms.

Background:
Despite rising incidence, there remains limited data guiding the prognostication and treatment of patients with bronchopulmonary carcinoid tumors, particularly atypical carcinoids. We report outcomes of a large, modern, single-institutional series of patients treated for localized or locally advanced atypical carcinoid of the lung.

Method:
We retrospectively analyzed the demographic, histologic and treatment histories of 69 patients (74% female) with median age of 65 at diagnosis (range 31-83) who were treated between 2004-2016. The Kaplan-Meier method was used for overall survival (OS) estimates and compared by log-rank. Cox proportional hazards models were used for univariate (UVA) and multivariate analyses (MVA).

Result:
Median follow-up time was 33.6 months. The majority (96%) of patients underwent surgical resection (86% R0, 9% R1, 3% R2) with common approaches being lobectomy (59%), wedge resection (13%) and pneumonectomy (9%). Three patients (4%) received definitive radiotherapy as their local treatment. Nearly half (49%) of patients had nodal involvement with a stage distribution of 39% stage I, 25% stage II and 36% stage III. Twenty-one patients received chemotherapy as part of their initial treatment, 81% of whom had stage III disease. Sixteen patients received radiotherapy (median 50.4 Gy, range 18-66 Gy) as part of their initial treatment, most of whom received postoperative radiation for N2 disease (63%). Five patients (31%) received postoperative radiotherapy due to concern of incompletely resected disease. Higher stage was significantly associated with poorer OS (p=0.04). 3-year OS for Stage I, II and III disease was 96%, 88% and 72%, respectively. Stage I disease also had a significantly lower risk of distant metastasis compared to Stage II/III disease (17% vs. 31% at 3 years p=0.04). On UVA, Stage III disease was significantly associated with poorer OS (HR 4.7, p=0.021) and risk of distant failure (HR 2.8, p=0.039). Multivariate modeling showed that older age (HR 1.05, p=0.03) and stage III status (HR 6.6, p=0.009) were predictive of poorer OS. For stage III patients treated surgically, receipt of adjuvant therapy (chemotherapy and/or radiotherapy) was not significantly associated with OS (p=0.36) or distant failure (p=0.69).

Conclusion:
This is one of the largest reported series of atypical pulmonary carcinoid patients treated with curative intent. We observed generally favorable prognosis in this cohort that was primarily treated with surgery. We did not observe a significant impact of adjuvant therapy on outcomes, but small patient numbers limit our ability to quantify their potential effect.

Background:
Chemo-radiation is considered to be the standard treatment for the management of limited disease of small cell lung cancer (SCLC). Even in this early stage, the role of surgery in SCLC is still controversial. We sought to examine the role of surgery; complete resection in terms of survival in SCLC.

Method:
A retrospective review was undertaken of patients who underwent surgery for SCLC between 2001 and 2015. Patients were staged according to the 7[th] edition of the Tumor, Node, Metastasis classification of lung cancer. Actuarial survival estimated with Kaplan Meier method and comparisons were undertaken using Cox regression hazard model. Clinicopathological factors and predictors of survival were analyzed.

Result:
We identified 49 patients who underwent complete resection. The mean follow up period was 1343 days. The mean age was 70.7 years. 40 patients were men and 9 were women. The number of patients of clinical stage was stage IA :21, IB: 15, IIA: 4, IIB: 6, IIIA:3. Operative procedure was lobectomy in 43, segmentectomy in 1, wedge resection in 5. The number of patients of pathological stage was stage IA :15, IB: 11, IIA: 14, IIB: 7, IIIA:2. Adjuvant chemotherapy was performed in 26 patients (53.1%). The 5-year overall survival (OS) rate in all patients was 58.8%. The 5-year OS was 61.3% in c-stage I, 54.5% in c-stage II, and 50% in c-stage III. The 5-year OS were 66.2% in p-stage I, 55.4% in p-stage II, and 50% in p-stage III. The 5-year OS of patients with adjuvant chemotherapy was significantly better than that of patients without adjuvant chemotherapy (77.8% vs. 39.8%, p=0.005). Multivariable Cox regression hazard model demonstrated that adjuvant chemotherapy was prognostic factor of overall survival (OS) (hazard ratio 0.255 (.095-.688), p=0.007)

Conclusion:
Surgical outcome for early stage SCLC was satisfied one. The role of surgery for this group seemed to be important. Adjuvant chemotherapy may improve prognosis and long-term survival will be expected.

Background:
The combination chemotherapy of etoposide and a platinum salt represents the standard therapy for almost 30 years in patients with ED-SCLC (extensive-disease small-cell lung cancer). While, the survival benefit of chemotherapy is not obvious. Antiangiogenic agents have been confirmed to have survival benefits for patients with NSCLC (non-small cell lung cancer). However, there is no established role for antiangiogenic therapy in SCLC

Method:
we conducted this meta-analysis to evaluate antiangiogenic agents plus chemotherapy in patients with ED-SCLC. Pub-med, EMBASE, Cochrane Library and ClinicalTrials.gov website were systematically searched for RCTs (randomized controlled trials) that compared antiangiogenic agents plus chemotherapy with chemotherapy alone in ED-SCLC.

Result:
We firstly found that antiangiogenic agents plus chemotherapy was well tolerated and could provide a statistically significant improvement in PFS (progression-free survival) for patients with ED-SCLC (HR = 0.76 [95% CI 0.65, 0.88], P = 0.0003). The results of ORR (objective response rate) (RR = 1.06 [95% CI 0.96, 1.18], P = 0.23) and OS (overall survival) (HR = 0.98 [95% CI 0.80, 1.21], P = 0.85) showed no benefit for antiangiogenic agents plus chemotherapy.

Conclusion:
Our study firstly shows that the addition of antiangiogenic agents to standard chemotherapy is well tolerated and can provide a statistically significant improvement in PFS for patients with ED-SCLC. Further, maintenance therapy with antiangiogenic agents is an effective treatment option for ED-SCLC patients who received four to six cycle of chemotherapy. Additionally, four-drug chemotherapy plus antiangiogenic agents may be better for ED-SCLC patients with PS of 1 to 2. However, whether antiangiogenic agents plus chemotherapy can influence OS for ED-SCLC needs further validation.

Background:
Patients with epidermal growth factor receptor (EGFR) mutant advanced non-small cell lung cancer (NSCLC) developed resistance to first- or second-generation EGFR-tyrosine kinase inhibitor (TKI) after 9-13 months and third-generation EGFR-TKI after 10 months, respectively. Small cell lung cancer (SCLC) transformation is a rare resistance mechanism in these patients.

Method:
Tissue re-biopsy was performed in 35 patients with EGFR mutant advanced NSCLC who failed first-line EGFR-TKI; and 4 patients with acquired T790M mutant advanced NSCLC who failed third-generation EGFR-TKI, to look for SCLC transformation

Result:
SCLC transformation was identified in 2 out of 35 (5.7%) patients who failed first-line EGFR-TKI and 1 out of 4 (25.0%) patients who failed third-generation EGFR-TKI. The first patient was a 70-year-old never-smoker male who was diagnosed with stage IV lung adenocarcinoma harboring exon 19 deletion mutation in April 2014. He had partial response (PR) to gefitinib 250 mg daily but developed symptomatic progressive disease (PD) after 26 months. Re-biopsy of his enlarging primary lung lesion showed SCLC transformation. The second patient was a 43-year-old never-smoker male who was diagnosed with stage IV lung adenocarcinoma harboring exon 19 deletion mutation in November 2015. He had PR to gefitinib 250 mg daily but developed symptomatic PD after 15 months. Re-biopsy of his rapidly enlarging primary lung lesion showed SCLC transformation. His plasma cell-free tumour DNA (cftDNA) was positive for T790M mutation. The third patient was a 62-year-old never-smoker female who was diagnosed with stage IV lung adenocarcinoma harboring exon 21 L858R mutation in November 2015. She had PR to gefitinib 250 mg daily but experienced symptomatic PD after 8 months’ of gefitinib therapy. Re-biopsy of her primary lung tumor revealed the presence of T790M mutation and her treatment was switched to osimertinib 80 mg daily. After an initial PR, she developed PR in the 12[th] month of osimertinib treatment. Biopsy from a metastatic inguinal lymph node showed SCLC. The first and second patients were given standard SCLC chemotherapy with carboplatin and etoposide but did not respond. The third patient sought treatment in another hospital.

Conclusion:
Re-biopsy is recommended in all patients with symptomatic PD while on EGFR-TKI treatment. SCLC transformation under the pressure of first, second and third-generation EGFR-TKI is an emerging challenge to the management of advanced NSCLC. Other than conventional carboplatin and etoposide chemotherapy, new treatment strategies should be explored to improve the outcome of patients who develop SCLC transformation.

Background:
Pulmonary large cell neuroendocrine carcinoma (LCNEC) represents a rare entity in lung cancer. Due to poor understanding of its biologic characters, optimal treatment strategy for patient with LCNEC remains undetermined. Recent data reveals that LCNEC can be divided into SCLC and NSCLC type based on distinct genomic signatures. It has been considered that SCLC is a central-type lung cancer and LCNEC usually locates in peripheral or midzone of lung. In the present study, we examined that whether there are significant differences between central tumors and peripheral tumors of LCNEC, in terms of clinicopathologic features, survival, and genomic profiles.

Method:
A total of 126 cases (113 cases with surgical samples) of pulmonary LCNEC were included in the present study. The tumors with invasion of the segmental and/or lobar bronchus were classified as central LCNEC and those without as peripheral LCNEC. EGFR mutations, ALK translocations, ROS1 translocations, Kras mutations, RET translocations and BRAF mutations were detected. Overall survival (OS) was determined from the date of operation until reported death or last follow-up visit. OS was analyzed by the Kaplan-Meier plots and the log-rank test was used to calculate the significance between groups. The prognostic factors for OS were analyzed using univariate and multivariate COX analyses.

Result:
Central tumors were associated with smoking history (p=0.047), higher T stage (p<0.001), N stage (p=0.001), TNM stage (p=0.014), and larger tumor size (p<0.001) compared with peripheral tumors. Although neuroendocrine marker expression of CD56, CGA, and SYN was not significantly different according tumor location, central tumors had higher expression of NSE (p=0.003). Moreover, peripheral tumors had higher incidence of EGFR mutations (18.8 vs. 0%, p=0.023) and similar incidence of Kras mutations (10.4 vs. 8.0%, p=1.000). Tumors harboring EGFR mutations were all pure LCNEC. No ALK translocations, ROS1 translocations, RET translocations and BRAF mutations were identified. The median OS was 3.71 years. TNM stage (p=0.039) and N stage (p=0.068) were associated with survival. Interestingly, central tumors had poorer survival compared with peripheral tumors, in terms of median OS (1.51 vs. 4.04 years), 1-year OS rate (54.0 vs. 83.9%), 2-year OS rate (37.0 vs. 75.9%), 3-year OS rate (31.7 vs. 59.9%). After multivariate analyses, tumor location was still an independent prognostic factor for OS (HR, 2.675, 95% CI, 1.384-5.171, p=0.003).

Conclusion:
Primary tumor location of LCNEC, divided into central and peripheral type, has distinct clinicopathologic feature, genomic characteristics and survival, which may help classify and manage patients with LCNEC.

Background:
Small cell lung cancer (SCLC) as the most aggressive tumor deserves a special attention. The aim of this research was to define the place of surgery of patients with SCLC in order to improve the results of treatment.

Method:
Clinical material for research consists of 49 patients in stage IA-IIIA with SCLC, which were radically operated in Ugra (region Russia) between 1999 and 2015. Among patients predominate males 41 (83,7%), versus females – eight (16,3%).

Result:
All patients underwent radical operations R0. All resection types were included (pneumonectomy, bilobectomy, and lobectomy). By 37 patients (75,5%) systematic nodal dissection (SND) was carried out, by 5 (10,2%) - mediastinal lymph node sampling (MLS) and by 7 patients mediastinal node dissection was not carried out. By SCLC combination treatment was used more often – 35 (71,4%). By that only in 11 cases additional adjuvant of thoracic radiotherapy was used. In 14 cases only surgical resection was used (28,6%). 5-year and 10-year overall survival (OS) rate was 47,6% and 37,2%. Median survival rate was 49 months. Five and ten-year OS rate by surgery combined with adjuvant chemotherapy was 53,8% and 39,5%, as compared to only surgical treatment – 35,6%. Median survival rate was 67 and 25 months, respectively. At I stage satisfactory results were achieved: 5-year and 10-year OS rate was 68,8% and 62,5% (р<0,05), – that corresponds with results of treatment of patients with non-small cell lung cancer with similar stage of process. Median survival rate was not achieved. At II stage 5-year and 10-year OS rate was 41,7 and 41,7%. Median survival rate was 36 months. At III stage unsatisfactory results were obtained. 5-year OS rate was 25,1%. Median survival rate was only 13 months.

Conclusion:
SCLC at I and II stages is the indication to radical treatment, mandatory including surgical resection with SND and adjuvant chemotherapy. The main method of treatment at III stage is chemotherapy or chemoradiotherapy.

Background:
The prognosis for patients with large-cell neuroendocrine carcinoma (LCNEC) of the lung is extremely poor, and optimal treatment strategies have not yet been established. To improve prognoses in patients with LCNEC, we analyzed immunohistochemical expression and gene mutations of several known molecular targets in LCNECs and compared the expression levels of these targets with those in lung adenocarcinomas.

Method:
Twenty-six patients with primary LCNEC and 40 patients with adenocarcinoma were analyzed. Topoisomerase II, epidermal growth factor receptor (EGFR)-L858R, and somatostatin receptor expression were evaluated by immunohistochemistry, and EGFR mutations were evaluated using direct DNA sequencing and the Scorpion-amplified refractory mutation system.

Result:
There was no significant difference between patients with LCNEC and adenocarcinoma in relation to age, gender, smoking status, pathological stage (8[th]), performance status, and pulmonary function. In patients with LCNEC and adenocarcinoma, positive rates of topoisomerase II, EGFR-L858R, and somatostatin were 65.4% and 15.0% (p < 0.0001), 0.0% and 20.0% (p = 0.0182), and 50.0% and 5.0% (p < 0.0001), respectively. The frequencies of EGFR mutations were 0.0% and 37.5% in LCNEC and adenocarcinoma (p = 0.0002), respectively.

Conclusion:
LCNEC showed overexpression of topoisomerase II, somatostatin, suggesting it was possible to have good response to treatment with etoposide and octreotide compared with adenocarcinoma. EGFR mutations were not observed in LCNEC. These results may indicate a favorable response to adjuvant treatments that are not typically prescribed for non-small cell lung cancer.

Background:
Pleural lavage cytology (PLC) is an independent prognostic factor for resected non-small cell lung cancer (NSCLC). However, there has been no standard method for the test and the optimal dose of saline is still unclear. We investigated the feasibility of 20 mL saline in this retrospective study.

Method:
From January 2001 to December 2011, 466 consecutive patients who underwent a curative resection for NSCLC and received both a pre- and post- PLC by using 20ml of saline were enrolled. The prognostic significance of each PLC status and other clinicopathological factors were investigated.

Result:
Patients with a positive pre-(n=24) or post-PLC(n=28) had significantly worse 5-year survival rates than those with negative results (pre-PLC positive/negative; 32.6%/69.9%, p=0.001, post-PLC positive/negative; 21.4%/71.1%, p<0.001, respectively). The post-PLC (p<0.001) was found to be an independent prognostic factor for the overall survival by a multivariate analysis, whereas the pre-PLC was not (p=0.95). In stage I disease, patients with a positive post-PLC had a significantly worse survival regardless of the pre-PLC status. These results were comparable to those of the past reports by using more saline of 50-1000mL.

Conclusion:
Using 20 mL of saline would be feasible for PLC and may be helpful to simplify the procedure.

Background:
Postoperative outcome for NSCLC patients with lymph node metastasis is unfavorable even after surgery with/without adjuvant chemotherapy. We sought to investigate postoperative prognostic factors in NSCLC patients with lymph node metastasis.

Method:
We retrospectively reviewed NSCLC patients with pathologically identified lymph node metastasis who underwent complete resection. Demographic, clinical, and pathologic factors (sex, age, smoking index, Performance Status, preoperative carcinoembryonic antigen [CEA], surgical procedure, lymphadenectomy extent, histology, tumor differentiation, tumor size, pT factor, metastatic node site, and adjuvant chemotherapy) were analyzed using the log-rank test as univariate analyses and a Cox proportional hazards regression model for multivariate analyses to identify independent predictors of favorable disease-specific survival (DSS).

Result:
Of the 146 eligible patients, 113 were male and 33 were female. The median age and preoperative CEA were 67 years and 5.3 ng/mL, respectively. Lobectomy or greater resection and segmentectomy were performed in 140 and 6 patients, respectively. Tumor histology was adenocarcinoma in 58 patients, squamous cell carcinoma in 70, and others in 18. Node metastasis was identified in hilum only in 85 patients and mediastinum in 61. Adjuvant chemotherapy was performed in 61 patients (platinum doublet in 56 and tegafur/uracil in 5, respectively) and was not performed in 85. The 5-year DSS was 58%. The CEA ≤5.3 ng/mL (HR: 0.368), without mediastinal node metastasis (HR: 0.436), and platinum doublet adjuvant chemotherapy (HR: 0.491) were identified as significant predictors of favorable DSS. The 5-year DSS in patients with CEA ≤5.3 and >5.3 ng/mL were 73% vs 41%, respectively (p<0.001). The 5-year DSS in patients who underwent platinum doublet chemotherapy or none/others were 67% vs 53%, respectively (p=0.047). Figure 1



Conclusion:
Even if NSCLC patients have lymph node metastasis, favorable postoperative prognosis may be expected in patients with low preoperative CEA. Platinum doublet adjuvant chemotherapy should be considered in patients with lymph node metastasis on pathological examination.

Background:
The world’s population is rapidly aging, and the age of patients with lung cancer will increase as well. The prognostic nutritional index (PNI), controlling nutritional status (CONUT), and the geriatric nutritional risk index (GNRI) are useful parameters for evaluating immune-nutritional status. We aimed to perform a multicenter retrospective study to investigate the correlations of these immune-nutritional parameters with postoperative comorbidities or surgical outcomes of elderly patients with non-small cell lung cancer (NSCLC).

Method:
We selected 272 consecutive patients with NSCLC aged >75 years treated from January 2005 to December 2012 and evaluate three preoperative immune-nutritional parameters as potential predictive factors of postoperative comorbidities or as prognostic factors for surgically resected elderly patients with NSCLC.

Result:
Both PNI and GNRI as well as sex and preoperative respiratory comorbidities, were significantly associated with postoperative comorbidities (P =0.0287, 0.0443, 0.0191 and 0.0177, respectively). Multivariate analyses showed that preoperative GNRI (P = 0.0161) as well as sex (P < 0.0001), preoperative serum carcino embryonic antigen levels (P = 0.0128), preoperative serum cytokeratin 19 fragment levels (P = 0.0125), pleural invasion (P = 0.0214) and lymphatic vessel invasion (P = 0.0165) significantly affected overall survival (OS). Abnormal GNRI was significantly associated with histology (P = 0.0419) and outcome (P =0.0077). In Kaplan–Meier analysis of OS by preoperative GNRI, the abnormal GNRI group had significantly shorter OS than the normal GNRI group (5-year OS, 45.15% vs. 64.10%, P = 0.0007, log-rank test). CONUT score did not have any correlation with postoperative comorbidities or surgical outcome.

Conclusion:
Preoperative GNRI is a novel preoperative predictor of postoperative comorbidities as well as a prognostic factor that may identify high-risk elderly patients with NSCLC.

Background:
Treatment of patients with lung cancer associated with interstitial pneumonia (IP) is difficult because of post-operative complications or treatment-related deaths. Indeed, post-operative acute exacerbation of IP in patients with lung cancer and IP is associated with a high mortality rate. In our institute, we have unified surgical methods and improved peri-operative management of such patients since 2013. In the current study, we retrospectively studied the clinical features, surgical methods, and peri-operative management, and analyzed the clinicopathologic features of patients with lung cancer associated with IP.

Method:
We selected patients with lung cancer associated with IP who underwent surgery from January 2004 to May 2017. A presumptive diagnosis of IP or an IP pattern on computed tomography (CT) was confirmed histologically by examination of resected specimens and/or the presence of the clinical diagnostic criteria. Acute exacerbation of IP was defined according to the guidelines of the Japanese Respiratory Society when the following criteria were fulfilled within 1 month: (1) increased respiratory distress; (2) fibrosis, newly developed ground glass opacity and infiltrative shadow on high-resolution computed tomography (HRCT); and (3) > 10 Torr decrease in PaO~2~ under the same oxygenation conditions. (4) no evidence of pulmonary infection, heart failure, pneumothorax, and pulmonary embolism. We studied the following clinicopathologic features in patients with lung cancer associated with IP: gender, age, surgical methods, pathologic stages, mortality rate, and post-operative complications, including acute exacerbation of IP.

Result:
Fifty-three patients underwent surgery. The mean age was 70.8 years (50 males and 3 females). Forty-three, 1, and 9 patients underwent lobectomies, a segmentectomy, and partial resections, respectively. Twenty-three, 17, and 13 patients were stage I, II, and III, respectively. Eight patients had post-operative acute exacerbations of IP and there were three in-hospital deaths caused by acute exacerbations. Of note, there have been no in-hospital deaths since 2013.

Conclusion:
It is possible to prevent severe post-operative complications in patients with lung cancer associated with IP with the aid of intra-operatively and optimal peri-operative management.

Background:
Introduction:Lung cancer is the most common form and cause of cancer death world-wide. Radical surgical resection, with or without adjuvant treatment, is still a prerequisite for cure.In European countries the proportion of patients who undergo surgery for this disease varies between 10 and 30% .Advances in operative and postoperative care have led to a decline in complications and mortality rates during the last two decades.In spite of different additional modes of treatment, survival is still poor.

Method:
The study comprised 388 consecutive patients referred to University Hospital of Lung Disease Tirane, from a defined population, during a 12-years period (2004-2017).There were in total 968 patients who underwent lung resection for bronchogenic carcinoma.There were 690 males and 278 females included 15-87 years with a mean age of 65.5 years.Postoperative events studied were divided into major and minor complications or death during the first 30 days after surgery.

Result:
Early mortality (within 30 days) after lung cancer surgery.During the study period an increasing number of women and of patients older than 70 years underwent surgery.The 30 day mortality rate was 3.4% (33 patients), 1.2% after single lobectomy(11 patients) and 3.1 % after pneumonectomy(22 patients).Major complications occurred in 48 patients (4.9%). Minor complications occurred in 75 patients (7.7%).Male gender, smoker, FEV1 70% of expected value, squamous cell carcinoma and pneumonectomy were risk factors predicting adverse outcome .

Conclusion:
Our results show low mortality and morbidity after lung cancer surgery.However, patients with reduced lung capacity and those undergoing pneumonectomy should be treated with great care, as they run a considerable risk of major complications or death during the first 30 days postoperatively.Older age (>70 years) does not appear to be a contraindication to lung cancer surgery, but patients in this group should undergo careful preoperative evaluation.

Background:
There is consensus for the treatment modalities of lung cancer. However, there were cases in which inadequate treatments were performed due to unavoidable reasons which could make different oncologic outcomes. So authors investigated factors affecting oncologic outcomes in patients with pathologic stage II and III lung cancer who underwent limited resection or omitted adjuvant therapy.

Method:
From January, 2010 to December, 2012, 231 patients underwent operation for non-small-cell lung cancer in our hospital. Of them, 63 patients who were diagnosed as pathologic stage II and III were enrolled in this research, and medical records of them were reviewed including demographic factors, treatment modalities, 5-year overall survival rate, 5-year recurrence and so on.

Result:
The mean age was 64.03±7.94 years old and preoperative performance status was good in all patients. The median follow-up duration was 58.7 [1.3, 88.4] months. Patients (group A) with resection more than lobectomy and adjuvant therapy were 50, patients (group B) with limited resection were 4, and patients with omitted adjuvant therapy were 13 (group C). 5-year overall survival rate of group A was 69.7% (stage IIA-84.4%, IIB-40.0%, IIIA-48.5%), but, all of group B were recurred and passed away. Of group C (stage IIA-9, IIB-1, IIIA-3), 5 were recurred and dead. Multi-variable analysis (Cox proportional hazard regression analysis) of prognostic factors for overall survival (OS) and recurrence (R) was performed, which showed limited resection (OS; hazard ratio 32.179, p=0.001 and R; hazard ratio 61.337, p=0.000) and pathologic stage (OS; hazard ratio 1.898, p=0.017 and R; hazard ratio 1.517, p=0.037) were presented as significant poor prognostic factors. And omitted adjuvant therapy didn’t significantly affect oncologic outcome, but had negative influence (OS; hazard ratio 8.102, p=0.074 and R; hazard ratio 5.528, p=0.13).

Conclusion:
This study has many weak points including small sample size, retrospective study and so on, but this study suggests that several factors affect prognosis in advanced stage non-small-cell lung cancer but, especially, the limited resection in advanced stage has been a significantly negative impact and should be considered carefully.

Background:
Positive margins after pulmonary resection for cancer is an uncommon and challenging occurrence, with no consensus currently available to guide best adjuvant treatments. Our objective was to determine the rate of positive parenchymal margins after non-pneumonectomy lung resection with an assessment of adjuvant treatment strategies and outcomes.

Method:
Ethics board approval was obtained for a retrospective analysis of prospectively collected data on all lung resections performed at the Ottawa Hospital during the period 2008-2014. Individual patient records were then examined to confirm margin status by a review of the final pathology report. Survival and disease-free intervals were analysed using log-rank statistics, with significance set at 5%.

Result:
Over the study period, 1428 patients underwent non-pneumonectomy lung resection. A total of 29 cases (2%) were identified with a positive lung parenchymal margin (PPM). A matched subset (n=662) of the remaining 1379 patients with negative parenchymal (NPM) were used as control group. Median followup for NPM was 36.7 [0-105.6] and for PPM was 29.1 [0.5-71.5]. Overall, lobectomy was the most common pulmonary resection performed; wedge resections represented the majority of the PPM (n=16, 55%) followed by lobectomy (n=9; 31%) and segmental resections (n=1; 3%). Overall survival (Figure) and disease-free survival (not shown) were significantly (p<0.0001) worse for the PPM (Hazard ratio 5.59, 95%CI [2.05-15.6]) with a median survival of 31.5 months; the control group had not reached median survival. Stage I and II NSCLC were predominant in both groups; however postive margins were more often associated with metastatic disease (24% in PPM; 0.9% in NPM). The majority of PPM went on to receive additional treatment (n=17; 68%) consisting of adjuvant chemotherapy (n=7; 28%), radiotherapy (n=4; 16%), chemoradiotherapy (n=5; 20%) and re-resection in one case. The remaining patients were observed. Recurrence was found at the staple line in 7 cases (24%); the remainder recurred at distant sites.

Conclusion:
The overall rate of PPM in this study is low (2%) as compared to reported rates of 5-15%. Wedge resection for metastases was associated with most cases of PPM; most patients received additional treatment and had distant sites of recurrence. Overall survival was significantly worse for PPM despite adjuvant therapy, which likely reflects the underlying disease.

Background:
Expression levels of thymidylate synthase (TS), dihydropyrimidine dehydrogenase (DPD), thymidine phosphorylase (TP), and orotate phosphoribosyltransferase (OPRT) may predict the clinical efficacy of 5-fluorouracil (5-FU) -based chemotherapy in patients with cancer. We investigated the differences in mRNA expression levels of these enzymes in non-small-cell lung cancer (NSCLC) and evaluated them as prognostic factors for NSCLC treated by surgical resection.

Method:
Intratumoral mRNA levels of TS, DPD, OPRT, and TP were quantified in 71 patients following a complete resection in pathological stage I and II NSCLC (adenocarcinoma or squamous cell carcinoma) using the Danenberg tumor profile (DTP) method.

Result:
TP was the only significant prognostic factor for overall survival (OS) following complete resection of stage I and II NSCLC. Median values of TP mRNA expression significantly differed between the high and low mRNA expression groups for OS. OS at 5 years was significantly better in the low TP mRNA expression group than the high TP mRNA expression group (82.5% vs. 63.6%, p < 0.001). The Cox’s proportional hazard model indicated that the pathological stage, sex, and TP expression were independent prognostic factors for OS. Univariate analysis for disease free survival (DFS) indicated that the pathological stage was the only prognostic factor for DFS. However, DFS at 5 years tended to be better in low TP mRNA expression group than in high TP mRNA expression group (88.9% vs. 67.3%, p=0.083).

Conclusion:
TP mRNA expression presents an independent prognostic factor for OS in patients with stage I and II NSCLC following complete resection.

Background:
Postoperative hemorrhage after lung cancer surgery is a potentially fatal complication. This study aimed to investigate the indications and timing of reoperation for postoperative hemorrhage after lung cancer surgery.

Method:
We identified all patients who underwent lung resection and mediastinal lymph node dissection for lung cancer between October 2001 and September 2015 at Sichuan Cancer Hospital, identifying 57 who had undergone reoperation for hemostasis. The records of these 57 patients were reviewed and analyzed.

Result:
The most common postoperative hemorrhage site was the separation surface of the original pleural adhesions (29.8%). The median time interval between the initial operation and reoperation was 12 hours (range, 2-432 hours), and most patients (77.2%) underwent reoperation within 24 hours. The overall morbidity and mortality rates of reoperation were 50.9% and 5.3%, respectively. The morbidity rates of early reoperation group (≤ 24 hours) and late reoperation group were 43.2% and 77.0%, respectively, which were significantly different (P=0.033). The mortality rates of early reoperation group and late reoperation group were 0 and 23.1%, respectively, which were also significantly different (P=0.010).

Table 1 Origin of postoperative bleeding for reoperation

Origin of bleeding	Number of patients (n=57)	Percentage
Pulmonary resection surface	1	1.8
Pulmonary artery trunk	1	1.8
Pulmonary artery branch	2	3.5
Chest wall invasion resection surface	4	7.0
Intercostal blood vessel	4	7.0
Bronchial artery	6	10.5
Hemothorax-unknown origin	11	19.3
Lymph node dissection surface	11	19.3
Pleural adhesion separation surface	17	29.8

Figure 1



Conclusion:
Once indications of reoperation for postoperative hemorrhage after lung cancer surgery are identified, reoperation within 24 hours after the initial operation can get a better short-term outcome.

Background:
Postoperative pulmonary complications such as pneumonia are significant negative predictors of short- and long-term survival after thoracic surgery. A preoperative cardio-pulmonary function is known as a predictor for postoperative pulmonary complications in patients with lung cancer. However, little is known about the relationship between preoperative exercise capacity and complication of pneumonia after lung resection. The 6-min walk distance (6MWD) measured by the 6-min walk test (6MWT) is a simple, safe, and inexpensive field test that can be used to evaluate the functional exercise capacity. We examined the association between preoperative 6MWD and development of postoperative pneumonia.

Method:
A retrospective study was conducted on patients with malignant lung tumors who were scheduled to undergo lung resection at Nagoya University Hospital from January 2014 to December 2015 (Institutional Review Board approval No. 2015-0413). Preoperative pulmonary function tests and 6MWT were assessed. A logistic regression model and receiver operating characteristic (ROC) curves were used to analyze clinical variables and compare the performance on 6MWD and percentages of predicted values of forced expiratory volume in 1 s (%FEV~1~) and diffusion capacity of the lung for carbon monoxide (%DLco).

Result:
The data from a total 321 patients including 283 with primary lung cancer and 38 with metastatic lung tumors were analyzed. Preoperative 6MWD significantly correlated with age, FEV~1~, forced vital capacity (FVC), %DLco, and serum albumin level. Pneumonia developed in 13 patients (4.0%) and the 6MWD of patients with pneumonia was significantly lower than that of patients without (413.9±89.0 vs. 495.2±93.2 m, p=0.002). Incidences of smoking history and comorbidity of COPD and interstitial lung disease were significantly higher and %FEV~1~, FEV~1~/FVC, %DLco, and serum albumin level were significantly lower in patients with pneumonia than in those without. Length of hospital stay after surgery was significantly longer in patients with pneumonia than in those without (30.4±29.6 vs. 7.4±7.0 days, p<0.001). In ROC analysis, 6MWD ≤450 m was a threshold for predicting postoperative pneumonia with 69.2% sensitivity and 71.1% specificity. A 6MWD ≤450 m, %FEV~1~ <80%, %DLco <80%, serum albumin <3.5 g/dL, and blood loss during surgery >200 g were significantly associated with development of postoperative pneumonia in a logistic model adjusted by age, sex, and primary lung cancer.

Conclusion:
Preoperative 6MWD is associated with development of pneumonia after lung resection for malignancies. Preoperative 6MWT is a useful screening tool in patients with primary and metastatic lung tumors.

Background:
Arterial embolism including cerebral infarction is the major concern after surgery, because it can leave fatal or serious results. Recently, it is reported that pulmonary vein stump thrombosis is easy to occur after pulmonary resection, particular left upper lobectomy. However, it is unknown whether pulmonary vein thrombus is actually associated with cerebral infarction.

Method:
A totally 296 patients, underwent postoperative enhanced CT after left upper lobectomy in our hospital from September 2002 to December 2013, were retrospectively evaluated. We examined the association with pulmonary vein thrombus and cerebral infarction in both groups with thrombosis and without thrombosis formation. The relationship of patients’ background, clinical stage, pulmonary vein stump thrombus and postoperative cerebral infarction were analyzed.

Result:
Figure 121 patients (7.1 %) had a pulmonary vein thrombus, however all cases were asymptomatic and cerebral infarction was not developed. On the other hand, cerebral infarction developed in 15 patients of 275 patients without pulmonary vein thrombus formation. Only clinical stage was significantly relevant to cerebral infarction in univariate analysis. Table 1. Incidence of cerebral infarction according to a pulmonary vein stump thrombus



Conclusion:
Left upper lobectomy of the lung is well known to be a high risk of pulmonary vein stump thrombosis and cerebral infarction in all surgical procedures of the lung. However, pulmonary vein thrombus was not necessarily associated with the cerebral infarction after left upper lobectomy of the lung in this study.

Background:
Search of optimal diagnosis and treatment strategies for non-small cell lung cancer (LC) pa­tients (LCP) (T1-4N0-2M0) realized.

Method:
We analyzed data of 708 consecutive LCP (age=57.5±8.3 years; tumor size=4.3±2.4 cm) radically operated (R0) and monitored in 1985-2017 (m=613, f=95; lobectomies=461, pneumonectomies=247, mediastinal lymph node dissection=708; combined procedures with resection of trachea, carina, atrium, aorta, VCS, vena azygos, pericardium, liver, diaphragm, ribs, esophagus=192; only surgery-S=563, adjuvant chemoimmunoradiotherapy-AT=145: CAV/gemzar + cisplatin + thymalin/taktivin + radiotherapy 45-50Gy; T1=269, T2=251, T3=131, T4=57; N0=460, N1=130, N2=118, M0=708; G1=178, G2=216, G3=314; squamous=394, adenocarcinoma=266, large cell=48; early LC=164, invasive LC=544. Multivariate Cox modeling, clustering, SEPATH, Monte Carlo, bootstrap and neural networks computing were used to determine any significant dependence.

Result:
Overall life span (LS) was 2196.3±1764.1 days and cumulative 5-year survival (5YS) reached 71.1%, 10 years – 63%, 20 years – 43.4%. 451 LCP lived more than 5 years (LS=3125.7±1560.3 days), 128 LCP – more than 10 years (LS=5123.1±1547.9 days). 195 LCP died because of LC (LS=560±372.1 days). AT significantly improved 5YS (58.3% vs. 34.1%) (P=0.001 by log-rank test) only for LCP with N1-2. Cox modeling displayed that 5YS of LCP significantly depended on: phase transition (PT) early-invasive LC in terms of synergetics, PT N0—N12, cell ratio factors (ratio between cancer cells- CC and blood cells subpopulations), G1-3, histology, glucose, RH, AT, blood cell circuit, prothrombin index, heparin tolerance, recalcification time, weight (P=0.000-0.030). Neural networks, genetic algorithm selection and bootstrap simulation revealed relationships between 5YS and PT early-invasive LC (rank=1), PT N0—N12 (rank=2), healthy cells/CC (3), lymphocytes/CC (4), thrombocytes/CC (5), eosinophils/CC (6), erythrocytes/CC (7), segmented neutrophils/CC (8), glucose (9), monocytes/CC (10), stick neutrophils/CC (11), leucocytes/CC (12). Correct prediction of 5YS was 100% by neural networks computing (area under ROC curve=1.0; error=0.0).

Conclusion:
5YS of LCP after radical procedures significantly depended on: 1) PT early-invasive cancer; 2) PT N0--N12; 3) cell ratio factors; 4) blood cell circuit; 5) biochemical factors; 6) hemostasis system; 7) AT; 8) LC characteristics; 9) anthropometric data. Optimal diagnosis and treatment strategies for LC are: 1) screening and early detection of LC; 2) availability of experienced thoracic and cardiothoracic surgeons because of complexity of radical procedures; 3) aggressive en block surgery and adequate lymph node dissection for completeness; 4) precise prediction; 5) adjuvant chemoimmunoradiotherapy for LCP with unfavorable prognosis.

Background:
Prognostic effect of EGFR gene mutation in disease progression which affecting recurrence pattern and recurrence dynamics after complete resection of lung adenocarcinoma has not been well established. We investigated the relationship between EGFR gene mutation and recurrence dynamics in completely resected lung adenocarcinoma.

Method:
We retrospectively review 527 patients who underwent curative surgery for lung adenocarcinoma from January 2006 to December 2009. Demographics, clinic-pathologic data, and prognosis data were obtained by review of medical records. The EGFR gene mutation was analyzed by nested polymerase chain reaction followed by bidirectional direct sequencing in case of recurrence. Adenocarcinoma patients who experienced recurrence and had data of EGFR mutation were included in the analysis. Pathologic stage IV were excluded. Patients were divided into M group (mutant EFGR gene) or W group (wild-type EGFR gene). Sites of recurrence and disease-free times (DFT) of two groups were compared.

Result:
Median follow-up duration was 72 months. Overall 5-year survival rate was 80.1%. Recurrence was detected in 153 patients and 5-year disease-free rate was 58.2%. EGFR gene sequencing was performed in 118 (77.1%) patients. There were 38 (32.2%) loco-regional recurrences and 80 (67.8%) distant metastases. Any mutation of EGFR gene was detected in 66 (59.9%) patients (M group) and 52 patients had wild-type EGFR gene (W group). Sites of recurrence in M group were loco-regional in 25 (37.9%) and distant metastasis in 41 (62.1%) patients. Site of recurrence in W group were loco-regional in 13 (25%) and distant metastasis in 39 (75%) patients. Sites of recurrence of both groups were not significantly different. (p=0.14) Median DFT of two groups were significantly different (M = 20.3 months vs W = 15.1 months, p=0.039). Visceral pleura invasion (p=0.045) and EGFR gene mutation (p=0.039) were prognostic factor for DFT in univariable analysis. In multivariable analysis, EGFR gene mutation was the only prognostic factor for DFT. (HR – 0.676, 95% CI – 0.371 ~ 0.986, p=0.042) No significant factor for DFT was identified in loco-regional recurrence. EGFR gene mutation, however, was the only significant prognostic factor for DFT of distant metastasis (HR – 0.561, 95% CI – 0.356 ~ 0.885, p=0.013) in univariable and multivariable analysis.

Conclusion:
In recurrent lung adenocarcinoma, EGFR gene mutation group showed longer DFT than wild-type EGFR group. EGFR gene mutation is a prognostic factor in lung adenocarcinoma and slow-growing nature of adenocarcinoma with EGFR gene mutation should be considered in surveillance after surgery.

Background:
The aim of this retrospective study was to review recurrence patterns of stage I non-small cell lung cancer (NSCLC) and identify prognostic factors for post-recurrence survival (PRS).

Method:
Among 940 patients with pathological stage I NSCLC who underwent curative resection between 2001 and 2009, 261 patients who had experienced a recurrence were included in this study. A total of 188 patients had adenocarcinoma (ADC), and 62 had squamous cell carcinoma (SCC). Oligo-recurrence was defined as one to three locoregional or distant recurrent lesions restricted to a single organ. Potentially curative local treatment (PCLT) included surgery, stereotactic radiotherapy (SRT), and photodynamic therapy.

Result:
The median follow-up duration was 65 months (range: 4–186 months), and the median disease-free interval (DFI) was 23 months (range: 2–95 months). The most common site of recurrence was the lung in 145 patients, followed by a mediastinal lymph node in 49 patients, pleura in 30 patients, and brain in 27 patients. Local treatment for recurrent tumors included surgery in 59 patients, SRT in 46 patients, photodynamic therapy (PDT) in 2 patients, and other radiotherapy in 41 patients. Seventy-eight patients received chemotherapy only, and thirty-five patients received conservative treatment. Among 125 patients who were evaluated for an epidermal growth factor receptor (EGFR) gene mutation study, positive results were detected in 63 patients, and 31 were treated with a EGFR-tyrosine kinase inhibitor (EGFR-TKI). The 3- and 5-year PRS rates were 49.1% and 33.8%, respectively. Age at recurrence, ADC cell-type, DFI, TKI, and PCLT were independent prognostic factors in a multivariate analysis.

Conclusion:
local treatment for recurrent tumors could be a good option for selected candidates. Local treatment seems to be a reasonable modality for treating oligo-recurrence, even an extrapulmonary recurrence. Use of an EGFR-TKI is applicable if an EGFR mutation has been detected. Further study is required to identify patients who are optimal candidates for local treatment.

Background:
Sublobar resection may be less than ideal for lung adenocarcinoma (ADC) with even minor micropapillary (MIP) or solid (SOL) component, given that they carry a higher incidence of locoregional recurrence. Rapid identification of these subtypes would help decision-making on the extent of resection for early-stage ADC.

Method:
Antibody arrays of adhesion and apoptosis molecules were applied for 8-pair ADCs with (≥5%) or without (<5%) MIP/SOL component to identify the differentially expressed proteins, which were further validated by Western blot. A semi-dry dot-blot (SDB) system that visualizes the presence of the target proteins, modifying from the dot-blot method, was used to for diagnosing MIP/SOL existence in a prospective cohort of 45 clinical stage I ADCs that received operation. Resected specimens were reviewed according to the new IASLC/ATS/ERS classification and each component was recorded in 5% increments.

Result:
Insulin-like growth factor-binding protein 2 (IGFBP2) and P-cadherin was found more frequently in the MIP/SOL positive group and thereby setting as the target proteins in the SDB system for detection. A total of 46 nodules with a mean size of 2.4±0.8 cm was enrolled, including 10 (21.7%) with MIP and 16 (34.8%) with SOL component. The specificity and sensitivity for detecting MIP/SOL existence through SDB method were 72.0% and 90.5%, respectively. The average test duration was 25.6±1.9 minutes. Interestingly, the test successfully diagnosed one of the two synchronic lesions to have SOL component (Figure 1).

Conclusion:
Detecting IGFBP2 and P-cadherin via SDB method may have a potential role in the rapid identification of MIP/SOL components in early-stage lung ADC. Figure 1. (A) Semi-dry dot-blot system identified the expression of IGFBP2 and P-cadherin by the comparison of chromogen density to reference. The positive result indicated the existence of micropapillary or solid components. (B) Synchronous lesions with or without solid component showed different test result. Figure 1

Background:
Current guidelines recommend adjuvant therapy for positive margins after surgical resection. However, there is no strong evidence for those adjuvant therapies. In this study, we analyzed series of patients with microscopically tumor-positive bronchial stump (R1) and compared it with patients with complete R0 resection.

Method:
From 2011 to 2015, 172 patients underwent surgical treatment for NSCLC at out center. Excluding three patients with pleural seeding and residual tumors at parenchymal resection margin, 10 patients were bronchial R1 resection. We performed a retrospective analysis of patients’ data during follow-up (mean 36.6 months).

Result:
All patients with R1 disease received adjuvant therapies. Mode of adjuvant therapy was not related to recurrence (P=0.683) within R1 group. Kaplan-Meier analysis showed that recurrence rate in R1 group was higher compared to R0 resection group (P=0.001) when no stratification for staging was done. However, overall survival was not different between two groups (P=0.087). Figure 1 Figure 2





Conclusion:
In this study, R1 resection group was revealed as a risk factor for tumor recurrence while it did not significantly affect overall survival. The type of adjuvant therapy had no relation to the recurrence of NSCLC. But further investigation with large cohort is necessary to find out the exact effect of adjuvant therapy and stage.

Background:
The IASLC staging committe hase rased the question of the role of skip metastases and single and multple metastases in a sigle station in the hillar and mediastinal regions in survival of lung cancer

Method:
Based on a database with 1068 operated patients we will try to establish an aswer to these questions. We have just this week received the dataset so we did not have time for making the necessary calculations before abstract deadline

Result:
The reults wil be added before the next deadline

Conclusion:
Will be added before the next deadline

Background:
The authors evaluate the 30-days mortality after open pneumonectomy for lung cancer and compare the results obtained for left-right pneumonectomy.

Method:
Between 2008 and 2011 there were radically resected 326 consecutive cases of lung cancer. There were 117 pneumonectomies – 72 on left and 45 on right, all performed through thoracotomy.

Result:
Overall 30-days mortality rate was 3,41% after pneumonectomy for lung cancer. After left pneumonectomy there was 0% mortality and after right pneumonectomy - 8,88% mortality. In 37 cases the patients received induction therapy – on those cases the mortality was 2,7% at 30 days, meaning 5% for right and 0% for left pneumonectomies. Without induction therapy the 80 patients operated per primam presented 3,75% 30-days mortality (12% for right and 0% for left pneumonectomies). These values are excellent for left pneumonectomies and similar to the literature for the right ones. The case selection is not the “cause of success” since 40,17% of cases were pathological stage III (TNM 6-th edition - valid at that time) and 29,05% were pathological stage II. Neither the limitation of the resection affected the mortality since the authors have performed 64,1% of pneumonectomies (75 cases) with intrapericardial approach, of which 21 pericardiectomies, 17 pericardioplasties and 3 atrial resections. All resection included mediastinal lymph node dissection. From best of our knowledge mediastinal lymph node dissection doesn’t affect 30-days mortality after pneumonectomy for lung cancer.

Conclusion:
A null mortality rate at 30-days after left pneumonectomy for lung cancer is possible, without case selection.

Background:
For diabetic patients with lung cancer, blood glucose levels and medications such as metformin and statins may impact survival. This retrospective study defined prognostic survival factors for diabetic patients with resected non-small cell lung cancer.

Method:
Between January 2005 and December 2013, 301 patients with type 2 diabetes mellitus who underwent curative resection for non-small cell lung cancer were identified and reviewed retrospectively. Prognostic factors for lung cancer–specific and overall survival were evaluated using the Cox proportional hazards regression model.

Result:
The median follow-up period was 48 months (interquartile range, 30–72). In a multivariate analysis for lung cancer–specific survival, older age, forced expiratory volume in 1 second (FEV1) < 80 % predicted, and advanced pathologic stage were significant negative prognostic factors; statin use was a positive prognostic factor [hazard ratio, 0.468; 95% confidential interval, 0.258–0.849]. In a multivariate analysis for overall survival, male sex, older age, FEV1 < 80 % predicted, and advanced pathologic stage were significant negative prognostic factors; proper glycemic control (hazard ratio, 0.645; 95% confidence interval, 0.436–0.952) and statin use (hazard ratio, 0.602; 95% confidence interval, 0.390–0.931) were positive prognostic factors.

Prognostic factors related to overall survival

Variable	Univariate HR (95% CI)	Multivariate HR (95% CI)
Male	2.682 (1.603–4.486)	2.628 (1.287–5.366)
Age	1.041 (1.016–1.066)	1.070 (1.042–1.099)
Smoking	1.927 (1.247–2.979)	0.988 (0.527–1.850)
FEV1% predicted <80	1.843 (1.254–2.708)	1.647 (1.073–2.530)
Obesity (BMI ≥25)	0.573 (0.387–0.848)	0.752 (0.496–1.138)
Proper glucose control	0.700 (0.487–1.006)	0.645 (0.436–0.952)
Cardiovascular comorbidity	1.315 (0.896–1.929)
Neoplastic comorbidity	1.259 (0.815–1.946)
Renal insufficiency	1.338 (0.871–2.055)
Metformin	1.131 (0.781–1.638)
Insulin	1.465 (0.886–2.422)
Statin	0.602 (0.398–0.911)	0.602 (0.390–0.931)
Non-adenocarcinoma	1.724 (1.203–2.470	0.905 (0.571–1.435)
Pneumonectomy	2.386 (1.364–4.175)	1.520 (0.806–2.864)
R1 (vs R0)	1.841 (0.452–7.492)
Stage
I	Reference
II	2.594 (1.694–3.973)	3.315 (1.837–5.983)
III/IV	3.818 (2.445–5.961)	6.515 (3.471–12.226)
Adjuvant treatment	1.545 (1.076–2.217)	0.606 (0.361–1.018)

Conclusion:
Proper glycemic control (HbA1C <7.0) is recommended for diabetic patients undergoing lung cancer surgery. Statin use was associated with improved overall survival and lung cancer–specific survival. Further studies are required to elucidate associations between type 2 diabetes mellitus and anti-neoplastic effects of statins and to evaluate statins as a novel adjuvant treatment for lung cancer.

Background:
Postoperative pulmonary function changes after lung resection and that decline is determined by the extent of lung resection. It has been known that pulmonary function after surgery recovers gradually within 3 months to 1 year and finally reaches a steady state. However, the long-term respiratory function is unknown. We investigated the changes in respiratory function after thoracic surgery in patients who performed more than segmentectomy to evaluate the long-term effect of lung resection.

Method:
A total of 99 patients who underwent lobectomy followed up serial pulmonary function at postoperative 1month, 6months, 1,2,3 and 5 years. We investigated the serial changes of forced expiratory volume per 1 seconds (FEV~1~%), diffusion capacity of carbon monoxide (DL~CO~%) by groups of minimal invasive surgery and adjuvant chemotherapy.

Result:
The reduced postoperative FEV~1~% recovered to preoperative value until postoperative 6 to 1 year and declined yearly. The value of DL~CO~% increased to preoperative value until postoperative 2 years and remained steady state. There was no significant change in at more than 2 years follow-up (Fig.1). The value of DL~CO~% was decreased significantly during adjuvant therapy (p=0.031) (Fig.2) and the patients performed VATS lobectomy showed faster recovery in FEV~1~% and DL~CO~% compared to thoracotomy group (p=0.001) (Fig.3 and 4). Figure 1



Conclusion:
The respiratory function after lung lobectomy recovered gradually over postoperative 2 years. Adjuvant chemotherapy or thoracotomy showed negative impact on the postoperative pulmonary function.

Background:
Surgery is usually not indicated for malignant pleural effusion due to its poor prognosis. However, malignant pleural effusion is first detected at thoracotomy　(malignant minor pleural effusion), and it is difficult to judge an appropriate mode of resection. Several reports have demonstrated that the prognosis of patients with malignant pleural effusion first detected at thoracotomy is relatively favorable, aggressive surgical approaches may result in long-term survival. In some cases long-term survival is expected for malignant pleural disease unless the causal tumor is highly responsive to systemic chemotherapy. The aim of this study is to describe the results of surgical intervention in NSCLC patients with malignant minor pleural effusion.

Method:
We retrospectively studied 21 surgical patients with NSCLC and malignant minor pleural effusion that is found during thoracotomy, from January 1989 till March 2017. Surgical-pathological staging was carried out according the 7th edition of the TNM classification of lung cancer. We examined the clinicopathological features, the current status of therapy, and surgical outcomes in patients with malignant minor pleural effusion.

Result:
The median survival time and 5-year survival rate in the 21 patients were 16.9 months and 19.3%. Primary tumor resection was perfomed in 20 patinets(95.2%), pneumonectomy was 1 case, lobectomy was 14 cases, segmetectomy was 1 case and wedge resection was 4 cases. Of our 21 petients, 10 had malignant pleural nodule. The 5-year survival rate for 10 patients with malignant pleural nodule was 0%, whereas the 5-year survival rate for 11 patients without malignant pleural nodule was 30.0%. But there is no significant difference in survival between with or without malignant pleural nodule. Univariate analyses revealed that women, adenocarcinoma histology were associated with significant longer survival. 6 patients(28.6%) survive more than 3 years and 3 patients(14.3%) survive more than 5 years, all patients were adenocarcinoma, cN0.

Conclusion:
Tumor resection is beneficial for the survival of several patients with NSCLC who have a minor malignant pleural effusion. Our results warrant to the possible benefits of surgery in NSCLC patients, and to justify the inclusion of surgery in multimodality treatment of NSCLC patients with minor malignant pleural effusion.

Background:
Postoperative complications after pneumonectomy for NSCLC depend on the extension of surgery, comorbidities and accurate registration of any adverse event. The aim of the study was to evaluate the short-term results after standard and extended pneumonectomy in NSCLC patients according to the TMM (Thoracic Morbidity and Mortality) grading system.

Method:
We included 216 NSCLC patients consecutively submitted to pneumonectomy at our institution from January 2011 to December 2015 in the retrospective study performed on data prospectively collected in an electronic clinical database. All patients were divided into two different groups: standard (n=142) and extended (n=74) pneumonectomy, where resection of adjacent organs was indicated. The patients undergone extended pneumonectomy were subdivided into single (n=49) and multi-organ resection (n=25) groups. Systemic mediastinal lymph node dissection was carried out in all cases. Morbidity and mortality rate was analyzed according to the definitions proposed by the ESTS and TMM classification system.

Result:
Postoperative morbidity and mortality rate after standard pneumonectomy (23,9% and 3,5%) was significantly lower than extended procedure (43,2% and 10,8%) (p=0.02). Multi-organ resection was an independent prognostic factor of unfavourable outcome: morbidity and mortality was significantly higher in the multi-organ group (48,0% and 16,0%), while in the single-organ group it was 40,8% and 8,2% respectively (p=0.01). The incidence of the BPF was the only one variable in the ESTS complications definition which differs significantly between the groups of standard, single- and multi-organ resection: 3,5%, 6,1% and 16,0% respectively (p=0.02). Major complications rate (grade IIIA and higher according to the TMM) was significantly higher in the multi-organ group (40,0%) than in the single-organ (28,6%) and standard (14,8%) group (p=0.01).

Conclusion:
TMM classification system is more accurate in grading and further analysis of postoperative complications after pneumonectomy in NSCLC patients in compare with ESTS criteria. Multi-organ resection should be carried out with caution due to unacceptable high morbidity and mortality rate.

Background:
Intrapulmonary lymph node retrieval is important for pathological staging and omitting this procedure may affect the outcome evaluation for patients of pN0. In this study, we aim to investigate the incidence and distribution pattern of intrapulmonary nodes involvement (level 13-14) in a pathological N1 cohort. Meanwhile, survival benefit of adjuvant chemotherapy for patients with N1 nodes involvement only limited in intrapulmonary levels was explored.

Method:
From January 2006 to December 2014, 1979 cases admitted to Peking University Cancer Hospital were treated with standard lung resection and systematic lymph node dissection and intrapulmonary lymph nodes retrieval, of which 160 cases of pathological N1 status entered the analysis. Surgeons collected intrapulmonary nodes after operation according to the protocol and all samples were sent for pathological examination. The data was prospectively collected and retrospectively analyzed. For those with N1 nodes involvement only limited in intrapulmonary levels, adjuvant chemotherapy was suggested by oncologists and patients may or may not follow the advice due to insufficient evidence in this condition. The outcome of those receiving adjuvant chemotherapy and those not was compared.

Result:
In this group, 104 cases (65%) reported level 13-14 nodes metastasis and 57 cases (36%) of N1 spreading only limited in level 13-14. The average levels and numbers of intrapulmonary node involvement were 1.09±0.29 and 1.54±0.85, respectively. For peripheral lung cancer, 38 of 89 cases (42.7%) showed non-tumor-located level 13-14 metastasis. Level 13-14 involvement in non-tumor-bearing segment of the right upper lobe was more frequent than that of the lower lobe (44.4% vs 13.3%, p=0.032), but this trend didn't occur in the left side(56.5% vs 45.8%, p=0.464). In 25 cases with tumor diameter ≤ 2cm, 13 cases (52%) presented non-tumor-bear segment metastasis(right side 38.5% vs left side 52.0%). Among 57 cases of pN1(13-14 only), 38 cases showed all three stations' collection and examination from level 10 to 12 in pathological reports, of which 19 had adjuvant chemotherapy and remaining 19 did not. Survival analysis in this 38 cases revealed survival benefit for patients receiving adjuvant chemotherapy (5-year OS 100% vs. 73.7%±11.6%, p = 0.043; 5-year DFS 53.9%±12.8% vs. 74.6%±8.5%, p=0.563).

Conclusion:
High incidence of intrapulmonary node metastasis and frequent spreading to non-tumor-located (sub)segments in pN1 group may indicate the clinical relevance of intrapulmonary node retrieval. The oncological outcome may be improved by adjuvant chemotherapy for those with N1 nodes involvement only limited in intrapulmonary levels.

Background:
Perioperative exercise and physiotherapy is increasingly recognised as beneficial for preparation and recovery in patients undergoing lung cancer surgery. Time and resource constraints may be a barrier to referral for rehabilitation prior to lung cancer surgery. We aimed to establish if provision of an exercise app to use at home would enable patients to exercise more frequently than attendance at classes and determine what the patient experience of using the app was.

Method:
We developed an app for an Apple iPad which utilised a Bluetooth connection to a pulse oximeter to provide patients with realtime feedback on their pulse rate and oxygen saturations during exercise. The app guides patients in doing pulmonary rehabilitaton exercises, which can be made more intense or gentler depending on baseline fitness. We conducted a prospective cohort study to test use of the app by patients at home before and after lung cancer resection. Incremental shuttle walk tests were performed before and after using the app to investigate the difference in preoperative functional capacity. Patients were asked to complete semi structured telephone interviews to comment on their experience of the app. Transcripts of interviews were analysed using content analysis to categorise and highlight the important messages from patients.

Result:
During the 14 month study 37 patients were recruited. A variety of patients participated; the age range was 33 to 84 years and FEV1 range was 45% to 124% predicted. The average number of sessions completed on the app was 4, double the amount if they were attending pulmonary rehabilitation classes prior to lung cancer surgery. All patients except one improved their incremental shuttle walk distance before surgery. Interviews yielded five key messages about the app; patients had a range of motivations for taking part in the study, it was simple to use, patients had a positive experience using it, they had tips for improving it and they thought it had had an impact upon their fitness levels. All the patients found the app convenient to help them perform exercises outside of hospital.

Conclusion:
Patients welcomed the app and immediate provision of the app eliminated any delay in accessing support to exercise. Patients were able to perform more sessions of exercise by using it at home as opposed to hospital based exercises classes. It was feasible to use the app in all types of patients attending for surgery and a multi-centre study is indicated to assess the impact upon postoperative outcomes.

Background:
Good prognosis following surgery for metachronous second primary lung cancer has been reported. However, distinguishing pulmonary metastasis from metachronous second primary lung cancer is difficult.

Method:
Patients who underwent multiple pulmonary resections for metachronous lung cancer at our institution between 2000 and 2014 were retrospectively analyzed. Metachronous lung cancer was defined as non-small cell cancer regardless of disease-free interval or histologic concordance.

Result:
The retrospective chart review identified 86 patients. The median patient age at the time of second resection was 72 years, and 53 out of 86 patients (62%) were male. The median time interval between first and second resection was 1540 days. The mean tumor size was 19 ± 8 mm, the mean tumor doubling time was 316 ± 305 days, and 71 out of 86 patients (83%) had concordant histology. Upon postoperative pathological examination, the metachronous lung cancer was diagnosed as second primary lung cancer in 72 patients (84%), and pulmonary metastasis in 4 patients (5%). In the remaining 10 patients (12%), second primary lung cancer was indistinguishable from pulmonary metastasis. In second primary lung cancer, pulmonary metastasis, and indistinguishable tumors, the 5-year overall survival rates were 82%, 50%, and 53%, respectively. Factors significantly associated with improved overall survival included: tumor doubling time >180 days; pathological diagnosis of second primary lung cancer; pathological stage IA; >2 year interval between first and second surgery; and a consolidation/tumor ratio of ≤0.5. A tumor doubling time of >180 days was significantly associated with pathological diagnosis of second primary lung cancer. Using multivariate analysis, we also found that tumor doubling time >180 days was the only independent predictor of superior overall survival.

Conclusion:
This study found that tumor doubling time is the most important preoperative predictor of survival and pathological diagnosis of second primary lung cancer in metachronous lung cancer. The classic criteria for the diagnosis of a metachronous primary lung cancer were defined by Martini and Melamed. A tumor was considered to be metachronous primary lung cancer if the histologic type was discordant, or if the disease-free interval was at least 2 years for tumors with a histologic type similar to that of the primary cancer. However, due to the rising incidence of adenocarcinoma, earlier detection by computed tomography, and later recurrence due to adjuvant chemotherapy, Martini and Melamed’s criteria should be modified. We propose that tumor doubling time of >180 days should be a new criterion among Martini and Melamed’s criteria.

Background:
The presence of lymph node metastasis in the ipsilateral remaining lobe and its clinical implication after lobectomy have not been evaluated or discussed.

Method:
We sampled station 12 node of the remaining lobe during the surgery and completely dissected all the segmental nodes if cancer cell metastasis was either confirmed or suspicious. We retrospectively reviewed such patients and analyzed clinicopathologic characteristics.

Result:
Between 2010 and 2016, we found 20 cases where a complete lymph node dissections of the remaining lobes were performed (male 19, female 1, median age 55 (29-71) years). The location of primary cancer was predominantly in the RLL (RLL 17, RML 2 and LLL 1) and the surgical procedure performed were mainly bi-lobectomy (RLL and RML 11, RLL 6, RML 2, LLL 1). The site of lymph nodes in the other lobe were predominantly located in the RUL (RUL 18, RUL and RML 1, LUL 1). Clinical N-stages were N0 in 6, N1 in 9, and N2 in 15 patients. Preoperative CT and PET-CT failed to predict presence of metastasis of other lobe lymph nodes. The pathologic N-stages were N0 in 3, N1 in 4 and N2 in 13. The metastasis in the other lobe was present in 14 patients whereas, 6 were negative. Twelve out of 14 patients with positive lymph node in the other lobe had mediastinal lymph node metastasis (85.7%). The overall survival rates were 94.4% at one year and 74.4% at 2 years. However, as many as 8 patients (40%) recurred at the time of the study. Figure 1



Conclusion:
In conclusion, we found some patients possessed occult metastasis in N1 nodes of the remaining lobe especially in lower lobe cancers and the majority had occult N2 metastasis as well. The prognostic and clinical implication of knowing lymph node status of the remaining lobe needs further investigations.

Background:
The effect of surgical waiting time on prognosis in patients with lung cancer remains unknown. The purpose of this study is to examine the impact of surgical waiting time on outcome in patients with lung cancer.

Method:
We retrospectively reviewed all patients who underwent complete surgical resection in a single center between January 2006 and May 2016. Waiting time is defined as the interval between the date of initial visit to the department of respiratory medicine or thoracic surgery and the date of surgery. The patients were divided into two groups based on surgical waiting time: group A (n=160), 0-37 days; group B (n=150), 38 days or longer. Patients who had received neoadjuvant therapy, history of previous primary lung cancer and waiting time greater than 6 months were excluded.

Result:
310 patients met inclusion criteria (65.8% men; median age: 67yrs; c-stage I / II / III 249 / 41 / 20, p-stage I / II / III 212 / 47 / 51, median waiting time: 37 days). Waiting time tends to be shorter as c-stage of disease (I / II / III 38 / 35 / 31.5 days) becomes more advanced. Bronchoscopy or CT guided biopsy before surgery was performed in the group A / B, 160 / 150 patients, respectively. Charlson Comorbidity Index (CCI) (0 / 1 / 2 / 3 / 4 / 5 / 6 / 7) was in the group A: B; 97 / 20 / 33 / 7 / 1 / 0 / 1 / 1: 72 / 27 / 37 / 11 / 2 / 0 / 0 / 1, respectively; group B was significantly associated with higher comorbidity rate. The 5-year RFP in the group A / B was 60.7 / 67.6% (p=0.19) (c-stage I / II / III 70.8 / 31.3 / 21.4%, 68.7 / 66.6 / 50%(p=0.92/0.32/0.3)), respectively. The 5-year OS in the group A / B was 75.6 / 82.6% (p=0.3) (c-stage I / II / III 83.3 / 68 / 21%, 85.7 / 79.3 / 0%(p=0.77 / 0.82 / 0.62)), respectively.

Conclusion:
Surgical waiting time from initial visit to operative intervention does not adversely affect recurrence and survival.

Background:
The prognostic significance of pathological necrosis in small lung adenocarcinoma has not been investigated. The purpose of this study is to investigate the prognostic role of pathological necrosis in patients with completely resected small lung adenocarcinoma ≦2cm.

Method:
All available tumor slides from patients with surgically resected lung adenocarcinoma ≦2cm in size (1998-2015) were retrospectively reviewed. Exclusion criteria: patients who received induction therapy and lung cancer surgery within preceding 2 years. Recurrence free probability and overall survival were assessed using the Kaplan-Meier method.

Result:
351 patients met inclusion criteria (48% women, median age 67yr (34-86 yrs), 50% never-smokers; 324 Stage IA, 27 Stage IB; 111 and 240 patients underwent sublobar resection and lobectomy, respectively). Presence of pathological necrosis was identified in 32 patients (9%). Presence of pathological necrosis was significantly associated with sex, smoking, clinical T classification in the 8[th] edition and pathological tumor size (p<0.01, p<0.001, p<0.01, p<0.001, respectively). Presence of pathological necrosis correlated with an increased risk of recurrence, compared with those without pathological necrosis (5-year RFP, 70.5%vs 93.8%; p<0.001). Presence of pathological necrosis did not affect OS (5-year OS, 80.8%vs 92.3%; p=0.21).Figure 1



Conclusion:
In patients with small lung adenocarcinoma ≦2cm, presence of pathological necrosis was significantly associated with increased risk of recurrence.

Background:
Optimal postoperative follow-up method for lung cancer patients is not determined. Investigation of early recurrence risk may help differentiating the patients who should be intensively follow up, and further lead to improve prognosis of those cases. We clinicopathologically evaluated the lung cancer patients with recurrence within a year after complete resection.

Method:
We retrospectively reviewed 63 patients with recurrent diseases after complete resection of lung cancer between 2006 and 2016 in our institution.

Result:
We found 32 patients (50.1%) who had recurrent disease within a year after resection (former group), 31 patients (49.2%) over a year after resection (latter group). Average ages of each group were 70.1 years for the former, and 68.4 years for the latter. As for histological types, the former were composed of Adenocarcinoma (Ad) 12 / non-Ad 20, and the latter composed of Ad 27/ non-Ad 4 cases. Ad was found more frequently in the latter than the former, significantly（p<0.0001）. Pathological stage (I/ II-IV) of each groups were 8 / 24 cases for the former, 18 / 13 cases for the latter. The former had more advanced stages than the latter, not significantly. Average tumor size (mm) was 45.5 for the former, 32.3 for the latter, not significantly. Pleural invasion (-/+) were 11/32 cases for the former, 11/31 for the latter, not significantly. On the other hand, Lymphatic invasion (-/+) were 24/32 cases for the former, 16/31 for the latter（p=0.0168）. Vessel invasion (-/+) were 23/32 for the former, 14/31 for the latter（p=0.0313）. The former had more Lymphatic invasion (+), Vessel invasion (+) patients than the latter, significantly. Postoperative survival times were 503.6 days for the former in average, 1704.1 days for the latter. The prognosis of the latter was better than the former（p＜0.05）. The number of deaths were 19 for the former (including 1 case with intercurrent disease death), 9 for the latter (same 2 cases). The former had more cases of death （p=0.0238）than the latter, significantly.

Conclusion:
Lung cancer patients with recurrence within a year after resection showed non-Ad histology and lymphovasclar invasion positivity more frequently, than those over a year after resection. Such patients had shorter survival time and more number of deaths than those over a year, showing poorer prognosis.

Background:
The high Venous Thromboembolism (VTE) incidence has been perceived in post-surgery lung cancer patients. However, there is lack of effective data from such area in China. To evaluate the incidence of VTE and define risk factors for primary lung cancer, we conducted a retrospective single center cohort study.

Method:
252 patients were enrolled in this study, included 131 primary lung cancer patients and 121 benign lung disease patients after lung surgery from July 2016 to March 2017. The Color Doppler Ultrasonography was performed before and after operation for VTE confirmation. Patients experienced new postoperative VTE, unexplained dyspnea, hemoptysis, chest pain, highly suspected PE and other symptoms, further CTPA examination was required. The VTE incidence was evaluated after discharge. Caprini RAM was performed; relative outcomes had been recorded for all patients. Caprini scores 0 to 4 were defined as low-risk, 5 to 8 as moderate-risk and over 9 as high-risk. Single factor analysis was constructed to define risk factors associated with VTE.

Result:
The overall VTE incidence was 12.3% (31 of 252). The VTE incidence after primary lung cancer surgery was 16.0% (21 of 131), much higher than the VTE incidence of benign lung diseases 8.3% (10 of 121), but did not reach statistical significance (P=0.061). Among 21 VTE lung cancer patients, 4.8% was SVT (1), 81.0% DVT (17), and 14.3% was DVT+PE (3); 15 patients (71.4%) have manifested intramuscular venous thrombosis. Those 21 VTE patients possessed high Caprini scores, which was significantly higher than non-VTE patients in benign lung disease group (Score: 7 versus 4, P﹤0.001). Postoperative VTE incidence was correlated with increasing Caprini scores. The VTE incidence in the low, moderate, and high-risk groups was 0%, 13.4% and 35.3% respectively. The high-risk patients group had significant higher incidence than the low and moderate groups (P=0.049). In lung cancer patients, pre-operative D-Dimer concentration from VTE patients was 2 times higher than non-VTE patients (0.42 versus 0.21, P=0.015) and 1.7 times higher than patients postoperatively at day 3 (1.24 versus 0.71, P=0.008). FDP concentration in VTE patients postoperatively at day 3 was 1.8 times higher than non-VTE patients (6.40 versus 3.60, P﹤0.001).

Conclusion:
The overall incidence of VTE after primary lung cancer surgery was 16.0%. High-risk patients had an incidence of 35.3%. High Caprini scores suggest extended chemoprophylaxis for patients after discharge. Pre-operative high D-Dimer, POD3 high D-Dimer and POD3 high FDP concentration indicate high VTE risk.

Background:
The diagnosis, staging, and therapeutic strategy for synchronous multiple primary non-small-cell lung cancer (SMP-NSCLC) remain unclear. Distinguishing SMP-NSCLC from intrapulmonary metastasis is difficult but of great importance for selecting the surgical procedure and prognoses.

Method:
Fifty-two patients diagnosed with SMP-NSCLC according to the modified Martini-Melamed criteria in the thoracic surgery department of the China-Japan Friendship Hospital from November 2004 to December 2015 were enrolled in this retrospective study. A total of 106 tumors were subjected to pathological examination. Close follow-up and survival analysis were performed.

Result:
The perioperative morbidity rate was 5.8%, with no cases of perioperative death. The overall 5-year survival rate was 40.6%, the cancer-specific 5-year survival rate was 54.5%, and the median survival time was 52 months. Older age (p=0.553), sex (p=0.600), smoking history (p=0.496), tumor distribution (p=0.461), video-assisted thoracoscopic surgery (p=0.398), and adjuvant chemotherapy (p=0.078) did not affect survival. Preoperative percentage of forced expiratory volume in the first second (p=0.022), Charlson comorbidity index (p=0.034), surgical procedure (p=0.040), and highest pT stage (p=0.022) were independent risk factors in the multivariate analysis. Different pathological subtypes were identified in 13 of 18 cases of multiple adenocarcinomas. Different gene mutation types and correlations between tumors were identified through next generation sequencing in those with the same pathological subtype.

Conclusion:
Postoperative survival rates in SMP-NSCLC were satisfactory. Non-radical resection might improve the prognosis for patients with a tolerable general condition and pulmonary function. Higher pT stage might result in poorer survival rates. Larger sample size and future study are still needed to identify the prognostic factors. Comprehensive histologic assessment and next generation sequencing could be effective methods for screening SMP-NSCLC.

Background:
Prediction of postoperative lung function is important in preoperative evaluation of patients with lung cancer. Postoperative predictive lung function includes a calculation method predicting from the number of bronchial branches in the non-occlusion area and a method using perfusion scintigraphy. We compared the lung function predicted from these two methods with the actual postoperative lung function and examined the correlation with each method for each lobe.

Method:
Forty seven patients underwent lung lobectomy with lung cancer of 2 cm or less from November 2015 to February 2017 were targeted. All patients were performed preoperative lung function test and perfusion scintigraphy using technetium-99m-labeled macroaggregate of albumin. In this study, cases with right middle lobectomy and bilobectomy were excluded. In all cases, lung function test was reexamined at the third to six month after surgery. The assessment of pulmonary perfusion was performed with planar imaging. Regional quantitation of lung perfusion was first assessed using the traditional method by drawing 2 regions-of-interest (ROIs) of equal size over each lung, dividing the lung into upper, middle, and lower lung zones. The mean of the counts in each of the 4 zones was calculated and compared.

Result:
Surgery was performed in 22 cases of right upper lobectomy, 10 cases of right lower lobectomy, 10 cases of left upper lobectomy and 5 cases of left lower lobectomy. Both calculation method and perfusion scintigraphy predicted postoperative FEV1 well in patients who underwent lobectomy (R= 0.94 vs R = 0.91) (both, p < 0.001). We studied upper lobectomy 32 cases and lower lobectomy 15 cases. Both methods showed significantly lower values compared with actual postoperative FEV1. In upper lobectomy cases, the average of the difference between the actual postoperative FEV1 and predicted postoperative FEV1 in calculation method was 193ml, and the average difference from predicted postoperative FEV in perfusion scintigraphy was 168ml. There was no significant difference between the two methods. In lower lobectomy cases, the average of the difference between the actual postoperative FEV1 and predicted postoperative FEV1 in calculation method was 263ml, and the average difference from predicted postoperative FEV1 in perfusion scintigraphy was 144ml. The FEV value of the calculation method was significantly lower than that of perfusion scintigraphy (p=0.0021).

Conclusion:
The conventional calculation method and perfusion scintigraphy method showed a strong correlation with actual postoperative FEV1. In lower lobectomy, the perfusion scintigraphy method showed the possibility of predicting more accurate postoperative FEV1.

Background:
Spread through air spaces (STAS) is a novel invasive pattern of lung cancer, which spreads within air spaces beyond the edge of the main tumor, but not necessarily accompanying stromal invasion. In the current study, we investigated the significance of STAS in patients with pathological stage I adenocarcinoma.

Method:
STAS was assessed in a total of 276 patients with resected pathological stage I adenocarcinoma. STAS was classified as either no STAS, low STAS (1-4 single cells or clusters of STAS), or high STAS (≥5 single cells or clusters of STAS) using a 20x objective and a 10x ocular lens. We evaluated the association between STAS and the clinicopathological characteristics and postoperative survivals.

Result:
Among 276 patients, 123 (44.6%), 48 (17.4%) and 105 (38.0%) were classified as having no, low and high STAS, respectively. Fisher’s exact test demonstrated that positivity for STAS was significantly associated with a larger radiological tumor diameter (P=0.008), a higher consolidation/tumor ratio (P<0.001), a higher maximum standard uptake value (P<0.001), a pathologically larger tumor size (P=0.004), the presence of pleural invasion (P=0.027) and a histologically invasive type (P<0.001), while STAS was not significantly associated with epidermal growth factor receptor mutations or programmed death ligand-1 expression (P=0.129 and P=0.872, respectively). Patients with the STAS had significantly shorter recurrence-free and overall survivals than those without (P<0.001 and P=0.002, respectively). According to a multivariate analysis, positivity for STAS remained an independent prognostic parameter for both the recurrence-free and overall survivals.

Conclusion:
STAS was associated with clincopathologically invasive features and was predictive of a worse survival. Figure 1

Background:
Over the last decade, attention and emphasis on regionalization of care for tertiary hospitals have increased in order to improve outcomes, especially in complex surgical procedures such as resection of lung cancer. On the other hand, the growing centralization of cancer services increases patient displacement and may delay access to specialized services and treatment. Impact of travel distance on patients’ outcome from lung cancer is poorly described. The objective of this study is to evaluate whether distance to treatment institution has an impact on survival from patients with lung cancer.

Method:
This is an observational, retrospective study, which included patients with stage I-III lung cancer that performed curative resection at Hospital São Lucas da PUCRS, Brazil. Data was collected from medical charts including demographic, clinical-pathological features and survival. Survival was estimated using the Kaplan-Meier method and log-rank test was used to perform multivariate analysis for prognostic factors. Chi-squared test was used to compare variables between groups.

Result:
Between January 2011 and December 2015, 234 patients with stage I-III lung cancer performed curative resection at Hospital São Lucas da PUCRS. The median age was 65 years (24-85 years) and 50.85% were male. 147 (62.82%) patients were from public health insurance and 87 (37.18%) had private coverage. The distribution of clinical stage ( AJCC 8th ed) was : 116 (49.57%) stage I, 71 (30.34%) stage II and 47 (20.09%) stage III. The median distance to institution was 19.7 (2-1086 km). Patients with public health insurance traveled higher distance than those with private coverage (p= 0.0004). The 3-year overall survival was 86.13% for stage I, 77.74% for stage II and 48.95% for stage III. The distance to institution had no impact in overall survival (p=0.85 for stage I, p= 0.63 for stage II and p= 0.46 for stage III).

Conclusion:
Our study shows that the distance between residence and treatment institution was not found to influence survival of the patients with lung cancer treated by surgery. Importantly, patients with public health insurance, which are in general poorer, have to travel longer distances for curative treatment.

Background:
Higher BMI is well known to increase the overall intraoperative & post operative surgical risk in obese patients with BMI>30 . Well –published studies have showed that these group of patients have higher affinity for surgical related complications & risks in other surgical specialities , but its impact in patients undergoing pulmonary resections is uncertain. This study looks into this aspect in our cardiothoracic unit .

Method:
A well-conducted retrospective cohort study was performed on all patients undergoing lobectomies for primary lung cancer between year 2011 to 2015 . These data was extracted using the operating theatre ORSOS database and cross referencing with the surgical unit database . Sole exclusion criteria was any patients with BMI<18 . Numerous statistical analysis of demographics & outcomes were tabulated .

Result:
Patient demographics , operation type & side with post operative surgical complications were analysed . Elements of lung function tests were also evaluated

Conclusion:
Identifying the outcomes post lobectomy in patients with BMI>30 will help plan better outstanding and resources for post operative care in this sub group in order to get them thru the post operative course .

Background:
Lung Cancer is most common cancer in the world . It has progressively become a disease of older people as radiological & clinical detection improves rapidly with advances in anatomic lung resections . As population grows older demographically , it poses various distinct treatment & management challenges . Thus , we looked into outcomes & factors associated with long-term survival following lung cancer resections in patients older than 70 years of age .

Method:
A retrospective study was conducted , all septuagenarians patients with lung cancer who underwent pulmonary resections , between years 2011 to 2015 , were reviewed . These data was cross-referenced & checked with the operating theatre ORSOS & national mortality data

Result:
80 patients in total . Male predominantly between ages 70 to 79 years old . Average age is 73.5 years . 19 patients were non-smokers . Average in-hosp stay was 9.9 days .5 patients needed post op ICU admission for hypotension & one patient had failed extubation needing prolonged ventilation . 67 patients had epidural catheter for post-op pain management which expedited their pain recovery process . Minor complications includes 6 patients with superficial drain site infection . No one suffered any strokes . No major complications noted . Survival at 30 days – 100% , 1 year – 10 out of 80 patients died - 87.5 % survival rate . At present – 54 out of 80 patients are alive – 67.5 %

Conclusion:
Strict operative patient selection thru MDM will identify groups of patients that will benefit from these surgeries .These group of patients did extremely well post operatively . Important factors associated with these successful surgeries were indicative of the better survival rates. Septuagenarians with lung cancer can be given a new lease of life with continued survivability following major pulmonary resections .

Background:
Superior vena cava (SVC) replacement is infrequently performed and technically challenging in low-volume centers. Veno-venous shunt (VVS) technique is used to reduce SVC pressure during SVC replacement and has not been well reported. This study aims to add experience on this subject and evaluate the surgical outcomes of patients who undergo SVC replacement combined with VVS in our center.

Method:
A retrospective analysis of six patients who received SVC replacement combined with VVS from September 2011 to February 2017 was performed. Clinical characteristics, pathological features, operative characteristics, postoperative outcomes and survival of six patients were reviewed.

Result:
There were four males and two females with a median age of 44 years (range, 35-69 years). There were three lung cancer patients and three thymoma patients at a stage from IIIA to IVA. Five patients underwent induction therapy. Complete resection was performed on five patients. One case underwent internal VVS, and the other five cases underwent external VVS. Prosthesis grafts were employed in five cases and autologous pericardium in one case. Three cases underwent single-vein reconstruction, and the other three cases underwent double-vein reconstruction. The median SVC clamping time was 75 minutes. There were no postoperative deaths and major complications. All follow-up patients were alive, and no thrombosis was found in all the grafts.

Table 1 Clinical and pathological characteristics of six patients

Case	Year	Age(years),sex	Tumor	Status	Treatment	Response after induction	Histology	Stage
1	2011	35,male	Thymoma	Primary	CRT+S+CT+S	PR	B2 thymoma	IVA
2	2012	46,male	Lung cancer	Primary	CT+S+CT	PR	Adenocarcinoma	pT4N2M0, IIIB
3	2015	37,female	Thymoma	Primary	CRT+S+CT	PR	B2 thymoma	III
4	2015	42,female	Thymoma	Recurrence	S+RCT	-	B1 thymoma	III
5	2016	54,male	Lung cancer	Primary	CT+S+CRT	SD	Squamous cell carcinoma	pT4N2M0, IIIB
6	2017	69,male	Lung cancer	Recurrence	CRT+S+CT	SD	Squamous cell carcinoma	pT4N0M0, IIIA

CRT, chemoradiotherapy; CT, chemotherapy; PR, partial response; RT, radiotherapy; S, surgery; SD, stable disease

Conclusion:
SVC replacement combined with VVS is technically feasible and safe. Although VVS technique is not a must, it may make SVC replacement safer in inexperienced centers. Surgery-based multidisciplinary treatment for selected patients with type T4 lung cancer and SVC involvement or thymoma and SVC involvement may achieve a favorable long-term outcome.

Background:
Sleeve lobectomy can expand the indications to radical organ-saving operations and avoiding pneumonectomy increase the number of operated ones, and also significantly improve the quality of life and rehabilitation conditions of patients, which is of special importance for persons with limited functional reserves.

Method:
One surgical team performed 180 bronchial sleeve resections for malignant neoplasms of the lung. In all cases an invagination technique was used to form an anastomosis per 1 cartilage semicircle. There were 168 (93,3%) patients with primary tumors and 12 (6,7%) with solid metastatic lesions. Non-small cell lung cancer (NSCLC) was performed in 149 (88.7%) cases, carcinoid tumors - in 19 (11.3%) cases. Ia stage was diagnosed in 7.4%, Ib stage in 6.7%, IIa in 25.5%, IIb in 16.1%, IIIa in 37.6% and IIIb - 6.7% of cases. There were 80.6% (145) males and 19.4% (35) females. The average age of patients was 56.9 (from 18 to 82 years). 21 variants of bronchoplastic reconstructions was performed, lobectomy - in145 (80,6%) patients, bilobectomy – in 24 (13,3%), segmentectomy – in 6 (3,3%), isolated resection of bronchus – in 5 (2,8%) patients. In common 63 (35%) bronchoangioplasty interventions were performed. Resection of the pulmonary artery was performed in 49 cases (40 - circular, 9 - edge), vena cava resection - in 9 patients (6 - circular, 3 - edge), segmental arteries were resected in 5 patients (3 - circular, 2 - edge). In all cases of NSCLC extended ipsilateral mediastinal lymph node dissection was performed.

Result:
Complications were noted in 35 patients (19.4%): pneumonia (6.1%), atelectasis of the lobe (0.6%), insufficiency of bronchial sutures (3.9%), granulation stenosis of bronchial anastomosis (1.1 %), PE (1.7%), chylothorax (1.1%), recurrent nerve paresis (1.1%), cardiac rhythm disturbances (1.7%), hemothorax (1.7%), AMI (0.6%). Postoperative mortality was in 3.9% cases. The cause of death in 3 patients was massive PE, in 1 - AMI, in 3 - failure of bronchial sutures. Long-term results were obtained in 128 patients. The one-year observed survival rate in patients after reconstructive resections was 90.6%, 3-year - 57.4%, 5-year - 42.4% (with median 48.8 months). One-year recurrence-free survival was 89.1%, 3-year - 63.8%, 5-year - 51.6% Median relapse-free survival was 55.2 months.

Conclusion:
Using the invagination technique for the bronchial anastomoses is characterized by the simplicity of technical execution and is accompanied by satisfactory immediate and long-term results.

Background:
The treatment strategy for N2 IIIA non-small cell lung cancer (NSCLC) is still controversial. Some believe that patients with bulky N2 are not good candidates for trimodality treatment. In addition, with regard to the survival of patients underwent lung resection with SVC reconstruction, patients with SVC involvement due to direct invasion of the main tumor have longer survival compared to those with SVC involvement due to mediastinal lymph node (LN) metastasis. We encountered a patient with bulky N2 NSCLC with SVC involvement.

Method:
A 69-year-old man complaining of cough was referred to our hospital for examination of a chest abnormal shadow. Chest CT showed a 58-mm pulmonary mass lesion in the right upper lobe and mediastinal LN swelling (#4R: 31 mm, #2R: 15 mm), which resulted in stenosis of the SVC. Transbronchial biopsy of the mass and EBUS-TBNA of the #4R LN showed squamous cell carcinoma. Since distant metastasis was not apparent, the patient was diagnosed with locally advanced IIIA lung cancer with bulky N2. After induction of concurrent chemoradiotherapy (2 cycles CDDP+VNR + 45 Gy radiotherapy), the lesion showed 9.5% reduction and was defined as stable disease according to the RECIST criteria.

Result:
Since it would be difficult to dissect the SVC and #4R LN, and this procedure would require substantial time, we approached by median sternotomy and right fourth intercostal thoracotomy and established the shunt between the left brachiocephalic vein and the right atrial appendage prior to cross-clump of the SVC. The SVC was resected because of extensive firm adhesion of the #4R LN, and reconstructed with a 12-mm reinforced polytetrafluoroethylene graft. The anastomosis was performed using a 5-0 Plorene suture. The patient underwent right upper lobectomy with mediastinal dissection and combined resection of the SVC. The operation time was 494 min and blood loss was 700 g. The patient was discharged on postoperative day 16. Pathological examination revealed the effect of chemoradiotherapy was Ef2, and viable cells were present in the #4 LN (ypN2).

Conclusion:
While the long-term outcome of this patient is unknown, we believe the trimodality treatment is an option for bulky N2 NSCLC with SVC involvement.

Background:
The aim of this study was to assess the effectiveness of Intrapleural Perfusion Thermo-Chemotherapy (IPTC) under video-assisted thoracoscopic surgery (VATS) for malignant pleural effusion (MPE) caused by lung carcinoma.

Method:
In this retrospective study, fifty-four patients with moderate or large amounts of ipsilateral malignant pleural effusion (MPE) secondary to non-small cell lung cancer (NSCLC) underwent pleural biopsy and IPTC under VATS. IPTC was performed by perfusing the pleural cavity with 43.0[o]C saline solution containing cisplatin (200 mg/m[2]) using a devised circuit through mechanical circulation for 60 minutes. Blood pressure, heart rate, oxygen saturation (spo2), and esophageal and rectal temperatures were monitored throughout the surgery. At the end of the perfusion, pleural biopsy was performed again for histological analysis.

Result:
The temperature at the pleural surface was stabilized at 43°C, and pleural effusion was controlled in all patients. KPS scores increased in 89.3% of patients. No patient developed bone marrow suppression reactions with noticeable bleeding after treatment, and no liver and kidney malfunctions were observed. Apoptosis was detected by light and electron microscopy after IPTC. CEA markedly decreased in all patients 1 month after IPTC. The median survival time was 21.7 months, with a one-year survival rate of 74.1%.

Conclusion:
Intrapleural Perfusion Thermo-Chemotherapy under VATS is a new, safe, less invasive and more effective approach for malignant pleural effusion (MPE) caused by lung carcinoma.

Background:
Photodynamic therapy (PDT) is one of the established treatment modality for non-small cell lung cancer (NSCLC). Early-stage lung cancer and superficial endobronchial lesions less than 1 cm in thickness can be e ectively treated with external light sources. Thicker lesions and peripheral lesions may be amenable to interstitial PDT, where the light is delivered intra-tumorally. The primary adverse event, phototoxicity, is expected to be minimized with the introduction of new photosensitizers that have shown promising results in phase I and II clinical studies. Moreover, the addition of PDT to standard-of-care surgery and chemotherapy(or target therapy) can improve survival and outcomes in patients with pleural dissemination. Therefore, pleural PDT with surgery has shown promise in the treatment of non-small cell lung cancer with pleural spread.

Method:
Between January 2007 and January 2017, we retrospectively reviewed the clinical characteristics, treatment course and outcome of 39 patients with pleural seeding by non-small cell lung cancer. The eligibility criteria are as follows: pathologic diagnosis of lung cancer or thymoma with pleural spread, medical feasibility for PDT and surgery. The exclusion criteria are as follows: younger than 18 years old, leukopenia, or thrombocytopenia; chronic renal insufficiency with serum creatinine > 2.5 mg/dL; significantly impaired liver function; pregnancy or lactation. Patients underwent anatomic resections in whom it was possible to remove all gross tumor. After finishing the procedure, the parietal pleura were stripped from the bony hemithorax as radical parietal pleurectomy. Debulking of all gross tumor was performed in the mediastinum, too. For thymoma patients, the radical thymothy- mectomy was performed concurrent with radical pleural pleurectomy. The goal was to remove all detectable tumor in the operation field before proceeding to the photodynamic therapy. .

Result:
Thirty-nine patients enrolled in this study. There are 18 men and 21 women included in this study. The mean patient age was 52.6 ± 11.9 years. Using Kaplan-Meier survival analysis, the 3-year survival rate and the 5-year survival rate were 69.4% and 59.5%, respectively. There is one ARDS occurred immediately after the procedure. After medical treatment, the condition gradually improved. Other minor complications included prolonged air-leakage (five patients) and skin redness (six patients). The complications were successfully treated using medication. There was no procedure-related mortality.

Conclusion:
Photodynamic therapy and surgery for pleural dissemination in patients with non-small cell lung cancer is feasible and associated with a good soutcome.

Background:
Non-small cell lung cancer (NSCLC) with dissemination are classified as Stage IV lung cancer and usually considered inoperable. However, dissemination detected primarily during thoracotomy has been reported to show better prognosis compared with dissemination diagnosed clinically. Here, we present a case series of surgically treated stage IV lung cancer treated with intraoperative hyper-thermotherapy with distilled water.

Method:
Clinical records of 912 consecutive NSCLC patients who underwent surgery intended for complete resection between 2006 and 2016 were retrospectively reviewed. 5 M1a cases and 1 M1b (posterior mediastinum) case were treated with intraoperative hyper-thermotherapy with distilled water. Thoracic cavity was filled with distilled water heated to 43 degrees Celsius, following segmentectomy / lobectomy and lymphadenectomy.

Result:
All cases were pathologically T2 adenocarcinomas. 3 had pN2 status, and 3 had pN0 status. 3 had EGFR mutation, with adjuvant EGFR-TKI therapy. Recurrence was confirmed in 2 cases; 1) pT2aN2M1a, tumor size 3.5 cm with no EGFR mutation, 2) pT2aN0M1b tumor size 2.5 cm with no EGFR mutation. The median observation period was 14 months (8-31) with 2year recurrence free survival rate of 53.3% and 2year overall survival rate of 66.6%. 2 cases showed a significant decrease of serum CEA post operatively. No pleural effusion was confirmed post operatively among the surviving cases. Figure 1



Conclusion:
Stage IV lung cancer cases with dissemination detected during thoracotomy have tolerable survival rates when combining intraoperative hyper-thermotherapy with distilled water post segmentectomy / lobectomy and lymphadenectomy. Multidisciplinary treatment, including surgery, for Stage IV lung cancer requires further consideration as it has potential for better prognosis.

Background:
Resections of the tracheal carinal have a special place in thoracic surgery, which is determined by the anatomical arrangements in this area, technical difficulties associated with lung ventilation and airway reconstruction.

Method:
108 operations with carinal sleeve resection were perfomed from 1998 to 2016. Histopathologically : squamous cell carcinoma (80), adenocarcinoma (15), dimorphic cancer (2),, carcinoid (4), other forms (7). As for the stages of NSCLC the patients were distributed: IIА -7 (6,7%), IIB – 4 (3,8%), IIIA – 59 (56,2%), IIIB – 33 (31,4%), IV – 2 (1,9%).LN metastases were detected in 64 (61%) patients: N1 lesions in 17 patients, N2 in 45 patients, N3 in 2 patients. Direct invasion of metastatic LN into carina was detected in 13 patients (12%). All patients underwent mediastinal LN dissection, in 8 cases bilateral. Tracheobronchial anastomoses were formed by single nodal sutures using invagination technique,then were perfomed obligatory myoplasty of the anastomosis area.Pneumonectomy with carinal sleeve resection was performed in 82 patients (on the right 77, on the left 5). In 26 cases (24%), organ-preserving interventions were performed: isolated carinal resection or lob(bilob)ectomy with a double-sleeve reconstruction of the carina. In total, 7 different variants of reconstructions of the tracheal carina were performed. In 61 patients (56,5%), resection of other extra-pulmonary structures as simultaneously performed. 66 patients (61%) received combined treatment.

Result:
Postoperative complications occurred in 33% of patients (36), mortality was 10.2% (11). The causes of in-hospital mortality were tracheobronchial anastomotic leakage (5), pneumonia of the single lung (2), arrosive bleeding (1), PE (1), ADHF (1), ARDS (1). Mortality after organ-preserving resections was significantly lower in comparison with carinal pneumonectomy (7.5% vs 12.1%, p <0.05). Preservation of lung parenchyma ensured significantly better quality of life and early rehabilitation in comparison with patients who underwent carinal pneumonectomy. The overall 5-year survival rate was 28.5%, median survival was 24.2 + 2.3 months. In multivariate analysis, the most important prognostic factor was the status of the regional lymph nodes (N). In stages N0-N1, 5-year survival rate significantly prevailed over that in N2-N3 stages (35.1% vs 14.4%, p <0.05). Combined treatment compared with only surgical treatment allowed to improve 5-year results of treatment (38.5% vs. 18.1%, p <0.1).

Conclusion:
Surgery with carinal sleeve resection provide satisfactory results in the treatment of locally advanced lung cancer . Perfection of the technique of carinal "double-sleeve" resections with preservation of the lung parenchyma provides better results in comparison with carinal pneumonectomy.

Background:
T4 NSCLC is sometimes considered as an unresectable disease due to the surgical difficulties and expected poor outcome. Especially, pulmonary resections accompanied with carinal resection and resections under extracorporeal circulation (ECC) have huge surgical invasiveness and risks of postoperative life-threatening complications. On the other hand, in practice, such surgery plays an important role to control disease in highly selected T4 cases. We reviewed our 23-year experience to clarify the benefit and risk of radical extended surgery in T4 locally advanced NSCLC.

Method:
All patients underwent curative-intent radical extended surgery such as pulmonary resection accompanied with carinal resection or resection under ECC for cT4 NSCLC (UICC ver.7) at Fukuoka University, between 1993 and 2016. All relevant data were retrieved from our institutional database and analyzed retrospectively.

Result:
27 patients were enrolled with a mean age of 57.9±11.2 (35-74) at surgery. 25 patients (92.6%) were male, and only 2 were female. 17 patients underwent pulmonary resection with carinal resection (CR group), 11 underwent pulmonary resection under ECC (ECC group). 1 case required both CR and ECC procedure. As an extent of pulmonary resection, pneumonectomy was performed in 24 patients, lobectomy in 2, and bilobectomy in 1. In CR group, all of 17 patients underwent carinal pneumonectomy (Rt; 12, Lt; 5). In ECC group, reason for ECC use were as follows; resection and reconstruction for descending aorta 6, left atrium 3, aortic arch 1, and pulmonary artery trunk 1. As ECC, cardiopulmonary bypass was used in 5 patients, extracorporeal membrane oxygenation in 6. In this cohort, multimodality approach was indicated only in 10 patients (37%). R0 resection was achieved in 22 patients (81.5%). Regarding pathologic N status, N0 was in 10 patients, N1 in 4, N2 in 12, and N3 in1. Overall survival (OS) in the whole cohort was 62.6% at 1-year, 58.1% at 3-year, 43.6% at 5-year, respectively. OS in CR group was 63.6% at 1-year, 42.4% at 5-year, OS in ECC group was 64.2% at 1-year, 48.1% at 5-year. 30-day mortality was observed in 3 patients (11.1%), in-hospital mortality was observed in 7 (25.9%). No significant difference in surgical and survival outcome was observed between two groups.

Conclusion:
Our long-term results of extended surgery in T4 NSCLC were feasible. Outcome of carinal resection was equal to resection under ECC. However, surgical and short-term outcome could be improved. Individualized approach combined with other modality therapy will be a next step to better short- and long-term outcome.

Background:
VATS technique has been increasingly used worldwide for the management of lung cancer[1]. VATS lobectomy has been shown to be superior to traditional open lobectomy with shorter length of stay, fewer perioperative complications and improved quality of life[1]. There remain concerns regarding the use of VATS for larger oncological resections including pneumonectomy and sleeve resections. We add more confirmatory data to several retrospective cohort studies demonstrating the safety of VATS pneumonectomy in selected patients[1,2].

Method:
With ethics approval, a retrospective cohort study was performed at a single-centre in Melbourne, Australia. It included all patients who had undergone a pneumonectomy between 1999 and 2017. The primary outcome was overall survival. Secondary outcomes included: 30-day and 90-day mortality, disease free survival and length of stay.

Result:
79 patients underwent pneumonectomy between 1999 and 2017. 27 patients underwent pneumonectomy via VATS approach. There were 76 patients with NSCLC, two with carcinoid and one with melanoma. There was no difference in the tumour size between the two cohorts (VATS median 47mm versus open median 50mm, p=0.12). There was no significance difference in node positive disease between the two cohorts (p=0.14). The 30-day and 90-day mortality rate was 3.8% and 5.1% respectively, with all events occurring in the open cohort. Median overall survival for all patients was 22 months, with a median disease-free survival of 14 months. There was no statistically significant difference in overall survival depending on operative access (median survival VATS 86 months versus open 26.2 months, p=0.12). There was no difference in disease-free survival from NSCLC between the two groups (VATS median 86 months versus open median DFS 15.9 months, p=0.21). The length of stay was shorter in the VATS cohort (7 days versus 8 days, p=0.008). The number of lung cancer cases performed as VATS at our institution has increased from 41% between 2002 and 2011 to 84% since 2014. The rate of VATS pneumonectomy has increased from 18% between 2002 and 2011 to 58.6% since 2014, however this increase lagged three years behind less major resections (i.e. lobectomy).

Conclusion:
In concordance with other recent retrospective cohort studies, our study demonstrates both the safety of VATS pneumonectomy and the oncological efficacy in appropriately selected patients. References: 1. Sahai RK, Nwogu CE, Yendamuri S et al. Is thorascopic pneumonectomy safe? Ann Thorac Surg. 2009;88:1086-1092. 2. Nagai S, Imanishi N, Matsuoka et al. Video-assisted thorascopic pneumonectomy: retrospective outcome analysis of 47 consecutive patients. Ann Thorac Surg. 2014;97(6):1908-1913.

Background:
To compare the effects of neoadjuvant/induction chemotherapy or chemoradiation on morbidity, mortality, and long-term survival in patients with locally advanced NSCLC undergoing pneumonectomy.

Method:
All pneumonectomies following neoadjuvant treatment performed for NSCLC between 2000 and 2016 were retrospectively reviewed. The study included 162 patients (28 females; median patient age, 55.4 years [range, 31–73]). Neoadjuvant treatment consisted of chemotherapy in 115 patients (71%, group I) and chemoradiation in 47 patients (29%, group II). Chemotherapy was cisplatin-based, and 2–6 cycles of treatment were completed. Radiotherapy was administered sequentially (dose, 45–60 Gy). Surgery was performed 3–6 weeks after neoadjuvant treatment. Both groups were assessed for 90-day mortality, morbidity, and long-term survival.

Result:
Right pneumonectomy was performed in 60 (37%) patients, and the procedure was completed in a standard manner in 64.2% of the patients. Morbidity was observed in 27.7% of the patients (27,8% in group I; 27.6% in group II,p=0.98). The incidence of bronchopleural fistula was 4.3% (4.2% in group I; 4.3% group II). The 90-days mortality rate was 3.1% (5 patients in group I, 0 in group II,p=0.17). The mortality rates for right and left pneumonectomy were 3.3 (2/60 patients) and 3% (3/102 patients), respectively (p=0.61). The 5-year survival rates were 46.2% in group I and 54.2% in group II, (P = 0.16).

Conclusion:
Pneumonectomy after neoadjuvant chemotherapy or chemoradiation appears to be safe with an acceptable morbidity, mortality, and long-term survival. Chemoradiation did not improve long-term survival compared to chemotherapy despite comparable 90-day mortality and postoperative morbidity.

Background:
There are few treatment options with curative intent for locoregional recurrence or residual tumor of locally advanced lung cancer after definitive chemoradiation therapy. Lung resection; salvage surgery is likely to be one of the options for local control in this situation. However, perioperative complications and survival benefit of salvage surgery are not well-reported.

Method:
Patients who underwent lung resection after definitive chemoradiation therapy for the treatment of non-small cell lung cancer were selected. Frequency and content of perioperative complications, 5-y overall survival rate and disease free survival rate were retrospectively analyzed.

Result:
A total of 13 patients treated between January 2001 and December 2016 were eligible for evaluation. (12 men and 1 women, mean age 54 years, Median follow-up was 39.7 months.) The indication for surgery was primary tumor regrowth (69%) or tumor persistence (31%). The prior median radiation therapy dose was 60Gy (range 60-77Gy). The indication of for surgery were primary tumor regrowth (8 patients)or tumor persistence(5 patients). All patients underwent an anatomical resection, surgical procedure included lobectomy in 10 patients, pneumonectomy in 2 patients, bilobectomy in 1 patients. 2 patients underwent a bronchoplasty. Median estimated blood loss was 247ml, and median operative duration was 278 min. Compared with anatomical resection we usually perform, salvage surgery needs longer operative duration. Postoperative complications occurred in 4 patients(31%) without perioperative death within 90 days : arrhythmia, delayed pulmonary fistula, acute exacerbation of interstitial pneumonia and empyema. the 5-y overall survival and 5-y recurrence free survival rate were 73.3% and 55.0%,respectively.

Conclusion:
Salvage surgery for locoregional recurrence or residual tumor after definitive chemoradiotherapy was acceptable in safety. It should be considered as a treatment option for selected patients. However, the technique of salvage surgery is complicated, it needs an adequate experience.

Background:
In non-small cell lung cancer (NSCLC), pulmonary resection for stage IV patients was not recommended in standard therapy. However, various new treatments for advanced or recurrence NSCLC patients such as molecular-targeted therapy for driver oncogenes or immune checkpoint therapy have improved the survival of those patients in these days. Therefore, we need to review the role of surgery for advanced NSCLC patients again.

Method:
Clinical records of 334 patient diagnosed stage IV in Shiga University of Medical Science between 2006 and 2015 were reviewed and clinicopatholocgial features and overall survival were analyzed retrospectively.

Result:
32 patients underwent surgery in this period, and the patients included 25 men and 7 women, with median age of 70.0 years (52-82 years). There were 22 adenocarcinomas, 8 squamous cell carcinomas and 2 other histological subtypes. Surgery included 19 pulmonary resections, 3 spinal fixation surgery, 7 intracranial surgery and 3 other surgery. Median overall survival following surgery or non-surgery was 15.8 month and 9.9 months, respectively (P=0.002). Moreover, in 19 cases of pulmonary resection, there were 7 patients with pleural disseminations (IVa) and 12 patients with distant metastases (IVb). In case with disseminations (IVa), median overall survival was also significantly longer in patients with pulmonary resection than in those of non-operative 31 patients in the same period (42.2 months vs 8.3 months, P=0.038).

Conclusion:
Although further study in larger sets of patients would be warranted, surgery should not be excluded from treatment modality for stage IV patients, especially IVa cases if we could selected appropriately.

Background:
Lung cancer with ALK or EGFR activation inevitably acquires resistance to respective TKIs despite an initial good response. Relapses with only a limited number of regions, so-called oligo-recurrences, occur in a subset of such patients. Here, we present two cases of lung cancer treated with surgery and continued TKI therapy after acquiring resistance to EGFR or ALK TKI.

Method:
Retrospecive review of patient charts.

Result:
Case1: A 46-year-old man was diagnosed as having ALK-positive adenocarcinoma with pleural dissemination by exploratory thoracotomy. After 2.5 years’ treatment with alectinib, the primary tumor in the left lower lobe gradually progressed. Left S6 segmentectomy was performed. Genetic analyses of resected specimens revealed ALK G1202R resistant mutation. Alectinib treatment was resumed after surgery and the patient is free of disease 1.5 year after surgery. Case2: A 65-year-old woman presented with lung cancer with ureteral metastasis. Genetic analyses of resected ureteral tumor revealed EGFR L858R point mutation. Gefitinib was initiated and partial response was observed. After 1 year treatment with gefitinib, right middle lobectomy was performed to resect the remaining tumor. Gefitinib treatment was continued and recurrence-free survival of 2 years was achieved.

Conclusion:
These two patients appear to benefit from surgery and continued TKI therapy after acquiring resistance to EGFR or ALK TKI. It may be one of the treatment strategy in selected patients.

Background:
Platelet-Derived growth factor beta (PDGFR-β) is a functional regulator of mesenchymal cells. It is reported that PDGFR-β is expressed by cancer associated fibroblasts, and PDGFR signaling supports cancer cells. Stromal expression of PDGFR-β is reported to be associated with poor prognosis in prostate, breast, pancreas, and gastric cancers. However, the significance of stromal PDGFR-β expression in non-small cell lung cancer (NSCLC) in patients undergoing preoperative chemo- or chemoradio-therapy had not been undetermined.

Method:
Seventy-two patients with NSCLC undergoing preoperative chemo- or chemoradio-therapy between 1996 and 2006 were assessed for expression of stromal PDGFR-β by immunohistochemistry using resected specimens. After cancer cells and stromal tissues were identified by HE staining, stromal PDGFR-β expression was defined as positive when it was observed in >5% of the stromal area. Relationships between stromal PDGFR-β expression and disease-free survival (DFS) and disease-specific survival (DSS) were analyzed.

Result:
The mean age of the 72 patients was 59.7 years. Sixty-one (85%) were male and 11 (15%) female. Forty patients (56%) underwent preoperative chemoradiotherapy and 32 patients (44%) underwent preoperative chemothearapy. Indications for preoperative chemotherapy were N2 disease in 51 (71%), T3 or T4 disease in 20 (28%), and other reasons in the remaining patient (1%). Regimens for preoperative chemotherapy were cisplatin (CDDP)-based in 34 patients (47%) and carboplatin (CBDCA)-based in 37 (52%). Type of resection were pneumonectomy in 7 (10%), bilobectomy in 6 (8%), lobectomy in 57 (79%), and sublobar resection in 2 (3%) patients. Complete resection were achieved in 59 patients (82%). The pathologic stage (7[th] ed.) was IA in 11 (15%), IB in 12 (17%), IIA in 6 (8%), IIB in14 (19%), IIIA in 19 (27%), IIIB in 5 (7%), and IV in 5 (7%) patients. The histological type were adenocarcinoma in 32 (44%), squamous cell carcinoma in 35 (49%), and others in 5 (7%) patients. On immunohitological examination, stromal cells expressed PDGFR-β in 47 cases (65%). Five-year DFS and DSS in the stromal PDGFR-β-positive group was significantly worse than in the negative group (29.1%-vs-64.6%, p=0.01 and 40.2%-vs-66.1%, p=0.02, respectively).

Conclusion:
Stromal PDGFR-β expression is negatively associated with DFS and DSS in patients with NSCLC undergoing preoperative chemo- or chemoradio-therapy.

Background:
Although preoperative serum tumor marker levels, such as carcinoembryonic antigen (CEA) and squamous cell carcinoma antigen (SCC) are often evaluated in non-small cell lung cancer patients, the implication of these levels are still unknown. This study examined the predictive effect of preoperative tumor marker levels on pathological characteristics of lung adenocarcinoma.

Method:
We retrospectively reviewed patients with lung adenocarcinoma who underwent macroscopic complete resection. The pathological metastasis and/or involvement was defined that positive pleural effusion or lavage cytology, pleural involvement, pulmonary metastasis, lymph node metastasis, and/or lymphovascular involvement were identified on pathological examination. To identify predictors for the pathological metastasis and/or involvement, tumor markers (CEA, SCC, Sialyl Lewis[x]-1 [SLX], cytokeratin-19 fragments [CYFRA], neuron-specific enolase [NSE], and pro-gastrin-releasing peptide [ProGRP]), and demographic and clinical factors were analyzed by a univariate analysis and multivariate logistic regression analysis. For the significant tumor markers, optimal cutoff points were determined with a receiver operating characteristic analysis.

Result:
Of the 263 eligible patients, 138 were male and 125 were female. The median age was 70 years. The median preoperative CEA, SCC, SLX, CYFRA, NSE, and ProGRP levels were 3.7 ng/ml，0.8 ng/ml，28 U/ml，1.8 ng/ml，8.4 ng/ml，and 46.9 pg/ml, respectively. According to the 7[th] edition of the TNM classification, 186 patients (71%) had c-stage IA disease, 48 (18%) had c-stage IB disease, 26 (10%) had c-stage II disease, and 3 (1%) had c-stage III disease. Positive pleural effusion, positive pleural lavage cytology, pleural involvement, pulmonary metastasis, lymph node metastasis, lymphatic permeation, and vascular invasion were identified in 3 (1%), 4 (2%), 48 (18%), 9 (3%), 28 (11%), 20 (8%), and 35 patients (13%), respectively, and in total, 83 patients (32%) developed the pathological metastasis and/or involvement. The univariate analysis identified CEA, smoking index, size, solid size, and c-stage as significant predictors. A multivariate analysis revealed CEA (OR: 1.113, p=0.005) and solid size (OR: 1.052, p<0.001) as significant predictors. The optimal cutoff point was determined as 6.0 ng/ml for the preoperative CEA, and 35 of the 63 patients (56%) with ≥6.0 ng/ml of CEA developed the pathological metastasis and/or involvement whereas 48 of the 200 patients (24%) with <6.0 ng/ml of CEA developed pathological metastasis and/or involvement.

Conclusion:
Our results suggested the predictive effect of the high preoperative CEA level on pathological metastasis and involvement in patients with lung adenocarcinoma, and thus, we may consider preoperative CEA to decide surgical procedure for these patients, such as the extent of pulmonary resection and lymphadenectomy.

Background:
Reconstruction of the airways following tracheal resection, tracheo-oesophageal complications and lung-sparing oncological surgery is challenging. Airway reconstruction materials ideally must possess balanced properties of rigidity, flexibility and have an intact surface of epithelium to maintain a functional airway. A novel approach is the use of bio-scaffolds which allow regeneration of respiratory epithelium and eliminate the need for immunosuppressive treatment. Initial experiences with decellularised porcine small intestine submucosa extracellular matrix (CorMatrix®) as a scaffold for vascular, cardiac and pericardial tissue repair and reconstruction has been encouraging. The properties of extracellular matrix are potentially suitable for airway reconstruction.

Method:
We initially used the bio-scaffold for repairing tracheal defects and tracheo-oesophageal fistulae (n=6). We then extended its use to reconstruct distal airways to facilitate lung-sparing oncologic resections of intermediate grade lung tumours (n=3). These patients would otherwise have required pneumonectomy (Case 1 and 2) or bi-lobectomy in a patient with borderline lung function (Case 3). We describe our experience using extracellular matrix for airway reconstruction in lung-sparing oncological surgery.

Result:
Three patients underwent complex sleeve resections with varying degrees of bronchial airway reconstruction using extracellular matrix. The lung tumours were: (1) muco-epidermoid tumour of the left upper lobe and left main bronchus; (2) typical carcinoid of the right main bronchus; (3) typical carcinoid of the intermediate bronchus and right middle lobe. All patients underwent frozen section pathology examination at the time of surgery and formal histology confirmed pT1N0 and R0 resection status. There were no perioperative complications. All patients underwent bronchoscopy prior to discharge and serial bronchoscopies were carried out 6 weeks, 3 months and at >1 year. All patients had complete integration of the scaffold by epithelialisation and the patch was not evident at 3 months. All segmental bronchi were widely patent. At 18 months there was no tumour recurrence.

Conclusion:
The use of decellularised porcine intestinal submucosa extracellular matrix (CorMatrix®) as a bio-scaffold is safe and shows remarkable epithelialisation and integrity. It is an option for airway reconstruction to facilitate lung-sparing surgery without compromising oncologic margins.

Background:
Patients with lung cancer who harbor multiple pulmonary sites of disease have been challenging to classify. Although the International Association for the Study of Lung Cancer Staging and Prognostic Factors Committee propose to tailor TNM classification of multiple pulmonary sites of lung cancer to reflect the unique aspects of four different patterns of presentation, tough challenges faced by clinicians are still not easily overcome.

Method:
Surgical tumor and normal tissue specimens were collected from six patients who were diagnosed with pathologically confirmed multiple lung cancers, with each tumor in the separate lobe, and treated at Beijing Cancer Hospital, Peking University, Beijing, China. Whole-exome sequencing was used to depict the genomic profiles of each tumor, and the average sequencing depth was 123× per sample (range, 84× to 154×; s.d., 19×).

Result:
In this study, we analyzed genomic profiles of 12 tumors from 6 patients with multiple lung cancers. Eight tumors from 4 patients demonstrated distinct genomic profiles, suggesting all were independent primary tumors, which were consistent with comprehensive histopathological assessment. Noteworthy common genomic characteristics were seen in 4 tumors from 2 patients. Compared with TCGA lung cancer cohort, one out of 6 patients carried significantly higher somatic nonsynonymous mutational burden, which were also discrepant between two separate lesions. Figure 1



Conclusion:
The current findings suggest that the tailor TNM classification of multiple pulmonary sites of lung cancer still encounter real-world challenges. A deeper understanding of the spatial and temporal dynamics of the carcinogenesis and evolution of lung cancer will be required to address these challenges.

Background:
Bronchoplasty is regarded as an alternative surgical procedure to pneumonectomy for non-small cell lung cancer (NSCLC). Among the bronchoplasties, wedge bronchoplasty is associated with a better blood supply than sleeve bronchoplasty; however, the surgical outcomes for these procedures have not been evaluated. The aim of the current study was to compare the surgical outcomes of wedge and sleeve bronchoplasties.

Method:
This was a retrospective analysis of 36 patients with NSCLC who underwent bronchoplasties from January 2000 to June 2016. The indication for wedge or sleeve bronchoplasty was based on bronchoscopic and radiologic findings. In addition, we selected the procedure based on the intra-operative findings, including frozen section diagnosis. There were 17 patients (47%) who underwent wedge bronchoplasties and 19 patients (53%) who underwent sleeve bronchoplasties.

Result:
There were 29 males (81%) and 7 females (19%), with a median age of 69 years (range, 43-82 years). The median duration of follow-up was 56 months. In both groups, right upper lobectomies dominated the other lobes. There were 9 patients (53%) in the wedge bronchoplasty group and 10 patients (53%) in the sleeve bronchoplasty group. Concomitant vascular reconstruction was performed in 2 patients (12%) in the wedge bronchoplasty group and 2 patients (11%) in the sleeve bronchoplasty group (p=1.00). The covering of bronchial anastomosis was performed in 9 patients each in the wedge and sleeve bronchoplasty groups (53% and 47%, respectively; p=0.70). There were no severe complications related to anastomoses, such as bronchopleural fistulas and bronchial stenosis. There were no operative mortalities in either group. Post-operative recurrences developed in 6 patients (35%) in the wedge bronchoplasty group and 8 patients (42%) in the sleeve bronchoplasty group (p=0.52). Mediastinal lymph node recurrences were most common; specifically, there were 3 (18%) in the wedge bronchoplasty group and 4 (21%) in the sleeve bronchoplasty group (p=1.00). There were no hilar lymph node or surgical margin recurrences in either group. The 5-year overall survival was 87% in the wedge bronchoplasty group and 60% in the sleeve bronchoplasty group (p=0.07). The 5-year recurrence-free survival was 60% in the wedge bronchoplasty group and 53% in the sleeve bronchoplasty group (p=0.43).

Conclusion:
Surgical outcomes following wedge bronchoplasty were not inferior to sleeve bronchoplasty; however, additional cases are needed to establish the safety and survival benefit of wedge bronchoplasty in patients with NSCLC.

Background:
Lymph node metastasis is a significant prognostic factor in patients with resected esophageal cancer. Although the role of postoperative radiotherapy has not been confirmed, a large number of studies have shown that postoperative radiotherapy can improve the local control and survival, especially for stage III and lymph node positive patients. This study retrospectively analyzed the prognostic factors of postoperative radiotherapy for locally advanced esophageal squamous cell carcinoma (ESCC), and evaluated the prognostic value of different status of lymph node metastasis.

Method:
Data from 121 patients with locally advanced ESCC who underwent radical resection and received postoperative radiotherapy from 2006 to 2013 were reviewed retrospectively. OS and DFS were estimated using Kaplan-Meier. Univariate analysis and multivariate analysis were performed to investigate prognostic factors by the Log-rank test and the Cox regression model. The effects of different status of lymph node metastasis on OS and recurrence patterns were compared.

Result:
The median DFS of all patients was 22.57 months, and median OS was 32.90 months. Multivariate analysis showed that Karnofsky Performance Status, length of tumor and positive lymph nodes ratio (LNR) were independent prognostic factors of DFS and OS. For patients with lymph nodes metastasis, LNR has better prognostic value for OS (AUC=0.673, P =0.04) compared with the number of positive lymph nodes（AUC=0.584 P=0.31）. The median OS of patients with LNR≤ 15% and LNR>15% were 33.43 and 19.20 months, P =0.04. Patients without lymph node skip metastasis (NSM) had better OS than that with NSM, but the difference was not statistically significant. Among the patients with both LNR>15% and NSM, OS was significantly worse than other lymph node positive patients, with median OS of 14.33 vs. 32.50 months, P=0.02. The analysis of the treatment failure patterns showed that more distant metastases were observed in patients with LNR>15%, while more local and regional recurrences were observed with LNR≤ 15%.

Conclusion:
The status of lymph node metastasis was related to the prognosis of postoperative radiotherapy for locally advanced ESCC. LNR has better prognostic value for OS, treatment failure patterns varied in different LNR. The patients with both high LNR and NSM have significantly poorer prognosis.

Background:
To compare the perioperative outcome of minimally invasive (MIE) esophagectomy performed with a single- or a multi-incision in treating esophageal cancer.

Method:
Patients with esophageal cancer who underwent MIE from 2006 to 2016 were evaluated. A 3–4-cm incision was created in both the thoracoscopic and the laparoscopic phases during the single-incision MIE procedures. A propensity-matched comparison was made between the two groups of patients.

Result:
We analyzed a total of 48 pairs of patients with propensity-matched from the cohort of 360 patients undergoing MIE during 2006–2015. There is no statistical difference in terms of postoperative ICU and hospital stay, number of dissected lymph nodes and presence of major surgical complications (anastomotic leakage and pulmonary complications) between the two groups of patients. The pain score one week after surgery was significantly lower in the single-incision group (p < 0.05). There was no surgical mortality in the single-incision MIE group.

Conclusion:
Minimally invasive esophagectomy performed with a single-incision approach is feasible for treating patients with esophageal cancer, with a comparable perioperative outcome with that of multi-incision approaches. The postoperative pain one week after surgery was significantly reduced in patients undergoing single-incision MIE.

Background:
c-Met is overexpressed in cancer cells and plays a crucial role in apoptosis evasion. BPI-9016M, a small-molecule inhibitor of c-Met, could enhance the cytotoxicity of various DNA-damaging agents and promote the cell apoptosis. Here, we evaluated the radiosensitizaion potential of BPI-9016M in Eca109 human esophageal squamous cell carcinoma (ESCC) cell line.

Method:
Cell viability was determined by CCK8 assay. The radiosensitization effect of BPI-9016M was evaluated by clonogenic survival and progression of tumor xenograft. Cell apoptosis were determined by flow cytometric analysis and TUNAL. Cell apoptosis regulators were detected by western blot analysis. Radiation-induced DNA double strand break (DSB) and homologous recombination repair (HRR) were detected by the activation of ATR-Chk 1/ATM-Chk2 pathways.

Result:
BPI-9016M induced radiosensitization in Eca109 cell of ESCC cell line, associated with 1) down-regulating mutation P53 and Bcl-2; 2) decreases phosphorylated ATR and ATM focus formation, and the expression of γ-H2AX; 3) up-regulates the rate of cell apoptosis protein cleaved-Caspase(Figure 1). The combination of BPI-9016M with irradiation delayed the growth of ESCC tumor xenograft to a greater extent compared with either treatment modality alone (P < 0.05). Figure 1. BPI-9016M enhanced radiation induced apoptosis and inhibited ATM- and ATR-dependent DNA damage homologous recombination repair (HRR) pathways analyzed by western blot.

Conclusion:
Our findings suggest that the enhanced apoptosis and the inhibition of HRR contribute to radiosensitization by c-Met inhibitor BPI-9016M in ESCC cell.

Abstract:
Endoscopic Staging of Lung Cancer Kazuhiro Yasufuku During the management of patients with lung cancer, accurate lymph node staging is important not only to determine the prognosis but also to decide the most suitable treatment plan. Non-invasive staging such as computed tomography (CT) and positron emission tomography (PET) indicate size and metabolic activity, respectively. However imaging alone is inaccurate and therefore tissue sampling is the preferred and most reliable. Surgical staging by mediastinoscopy has been the gold standard for mediastinal lymph node staging but requires general anesthesia and complications cannot be ignored. Endoscopic ultrasound techniques provide a minimally invasive alternative for surgical staging and have become available for oncologists around the world. The current available endoscopic ultrasound techniques for mediastinal staging include transesophageal endoscopic ultrasound guided fine needle aspiration (EUS-FNA) and endobronchial ultrasound guided transbronchial needle aspiration (EBUS-TBNA). Both procedures are performed in an outpatient setting under local anesthesia. EUS-FNA is a sensitive and safe method of evaluating the inferior mediastinal nodes (stations 7, 8, and 9) and some parts of the anterior mediastinal nodes if the lymph nodes are accessible from the esophagus. However, in spite of the strength of EUS-FNA for evaluating the inferior mediastinal nodes, its ability to evaluate lesions anterior to the trachea is limited. On the other hand, EBUS-TBNA has reach to the paratracheal and subcarinal (stations 2R, 2L, 4R, 4L, 7), as well as the N1 lymph nodes (stations 10, 11, 12). In experienced hands, EBUS can be used through the esophagus for a EUS-like approach to sample inferior mediastinal lymph nodes. With the transvascular approach, AP window lymph nodes (station 5) can be sampled by EUS-FNA and/or EBUS-TBNA. Specialized centers have reported the sampling of station 6 via EUS-FNA. Thus, EUS-FNA and EBUS-TBNA are complementary methods for lymph node staging in lung cancer and most of the mediastinum and the hilum can be evaluated with these endoscopic procedures beyond the reach of mediastinoscopy. Based on the current evidence, EBUS-TBNA and EUS-FNA presents a minimally invasive endoscopic procedure of choice for mediastinal staging of NSCLC with discrete N2 or N3 lymph node enlargement, provided negative results are confirmed by surgical staging. When combined the techniques offer safe and accurate assessment of mediastinum, with accuracy surpassing that of the pervious gold standard – cervical mediastinoscopy. EBUS-TBNA and/or EUS-FNA can also be repeated with ease and have been used for mediastinal restaging in patients who underwent neoadjuvant therapy in preparation for definitive surgical intervention. New size needles are now available for sampling of the lymph nodes during EBUS-TBNA including 25-gauge and 19-gauge needles. Smaller needles may provide greater reach with good quality samples, whereas larger 19-gauge needle may provide bigger tissue for histological evaluation of the lymph nodes samples. There are limitations of using cytological samples obtained during EBUS-TBNA or EUS-FNA for PD-L1 expression. The use of the 19-gauge needle may solve this problem.

Abstract:
In the past several years, X-ray fluoroscopy had been commonly employed to determine the lung field during transbronchial biopsy (TBB); however, precise localization of a PPL has not always been possible leading to low diagnostic yield. For ground glass opacities (GGOs), the value of X-ray fluoroscopy even becomes less. The value of virtual X-ray fluoroscopy and CT fluoroscopy has potential but remains to be known. The advent of endobronchial ultrasound (EBUS) has dramatically increased precise bronchoscopic confirmation of the location of a PPL before sampling. In particular, the radial probe EBUS is used to indicate that a lesion has been reached. For solid peripheral pulmonary lesions (PPLs), Kurimoto et al have described three major types of echogenicities that might differentiate between benignity and malignancy. For GGOs, we have observed constant radial-EBUS patterns that we called blizzard and mixed blizzard signs. Currently, several acquired resistance mechanisms and rare driver oncogenes are identified in non-small cell lung cancer (NSCLC) relapses. Re-biopsy increases valuable information to guide treatment strategies, but the utility and feasibility of bronchoscopic re-biopsy especially endobronchial ultrasound (EBUS) guided re-biopsy has not been investigated. We recently reported the utility of bronchoscopic (EBUS-guided) re-biopsy for detecting the mutation in NSCLC. Re-biopsy by both EBUS-TBNA and EBUS-GS were useful and safe sampling procedures for mutation analysis of EGFR-TKI resistant NSCLC. Another alternative approach, specifically liquid biopsy, now present as a crucial point in the field. Liquid biopsy has grown in importance because the genetic profile of tumors can affect how well they respond to a certain treatment. A recent paper showed that the concordance between re-biopsy and liquid biopsy, including plasma DNA and circulating tumor cell, was 57–60%. The usefulness of monitoring T790M status in liquid biopsy was already reported. Although liquid biopsy has the potential to detect new mutations after chemotherapy, several reports have demonstrated some difficulties in detecting tumor-derived mutations in plasma. Therefore, liquid biopsy and re-biopsy may be considered to be complementary methods of mutation analysis. I would like to share our data and actual cases in this talk. Figure 1Figure 2

Abstract:
Introduction: Malignant airway obstruction can result from primary airway tumors, extension of adjacent primary tumors, or metastatic tumors. Partial or complete airway obstruction can deteriorate functional status of patients and result in impending respiratory failure. Malignant airway obstruction is considered to be one of the most distressing causes of morbidity and mortality in lung cancer patients. Bronchoscopic intervention can provide immediate relief from suffocation, improve general condition, and provide a bridge, allowing time for additional treatment such as surgery, radiation, or chemotherapy in patients suffering from malignant airway obstruction. Indication: Any patients who suffer from respiratory distress due to central airway obstruction are indicated for bronchoscopic intervention. However, patients should tolerate the morbidity of intervention, the length of the airway obstruction less than 4cm, and the duration of obstruction less than 2 month due to the technical limitation. Method: Due to it is safe from massive hemoptysis and respiratory failure, most experienced bronchoscopists prefer rigid bronchoscopy under general anesthesia, using intravenous propofol injection. After the induction of anesthesia, the patients are intubated with a rigid bronchoscope tube and a flexible bronchoscope is introduced through the rigid bronchoscope tube, and the narrowed central airway was evaluated. In every case, the obstructed airway is dilated gently using an 10 mm rigid bronchoscope tube initially and then progressively larger bronchoscope tubes until an adequate airway caliber was established. When indicated, a controlled radial expansion balloon is used to enlarge the airway sufficiently to allow bronchoscopic dilatation. Any intraluminal mass is removed mechanically using rigid bronchoscopic forceps or a snare. Frequently, a neodymium-yttrium aluminum garnet (Nd-YAG) or diode laser is used to ablate the residual endobronchial tumor or to cauterize the tumor bed after most of the tumor had been excised. After mechanical dilatation, the airway is maintained by inserting a silicone stent (Dumon-style stent) in patients whose airway is not maintained due to extrinsic compression or malacia. The silicone stents are inserted through the rigid bronchoscope using a standard Dumon technique. Outcome: In experienced center, the overall success rate is more than 90% after the emergency bronchoscopic intervention. A successful outcome is accompanied by subjective improvement in the symptoms and radiographic findings. After stabilizing the airway with the bronchoscopic treatment, favorable outcome is expected if additional definitive therapy can be applied, such as surgery, radiation, or chemotherapy. Nowadays, bronchoscopic intervention can achieve prolonged survival with sustained significant improvement of quality of life. Complications: Tracheal perforation, massive bleeding, respiratory failure and cardiac arrhythmia can develop after bronchoscopic intervention. However, the overall complication rate is not over 5% in experienced center. Conclusion: Bronchoscopic intervention in patients with malignant airway obstruction is helpful for the palliation the airway, allowing the multimodality therapeutic approach and prolonging the life of the patients. References 1. Jeon K, Kim H, Yu CM, et al. Rigid bronchoscopic intervention in patients with respiratory failure caused by malignant central airway obstruction. J Thorac Oncol 2006;1:319-323. 2. Cavaliere S, Venuta F, Foccoli P, et al. Endoscopic treatment of malignant airway obstructions in 2,008 patients. Chest 1996;110:1536-1542. 3. Han CC, Prasetyo D, Wright GM. Endobronchial palliation using Nd:YAG laser is associated with improved survival when combined with multimodal adjuvant treatments. J Thorac Oncol 2007;2:59-64. 4. Chhajed PN, Eberhardt R, Dienemann H, et al. Therapeutic bronchoscopy interventions before surgical resection of lung cancer. Ann Thorac Surg 2006;81:1839-1843. 5. Stratakos G, Gerovasili V, Dimitropoulos C, et al. J Cancer. 2016;25: 794-802.

Abstract:
World-wide, aside from men in France, Spain and the Netherlands, peripherally located adenocarcinomas have now overtaken squamous cell carcinoma as the predominant lung cancer cell type. With the implementation of lung cancer screening programs using low dose CT and increasing use of CT imaging for clinical investigations, a large number of people are found to have lung nodules. In contrast to symptomatic lung cancer, the size of screening CT detected or incidental lung nodules suspicious of malignancy is much smaller. Over 75% of screening CT detected lung cancers are ≤20 mm with 20% to 47% of the lung cancers found in the first screening CT and 33% to 62% of lung cancers found in annual repeat screening CT are ≤10 mm.[1-4] Because of the small size of these lesions, currently only 20% to 34% of screening CT detected lung cancers are diagnosed by endoscopy. The diagnostic yield of bronchoscopic biopsies is modest.[1,2] In the real world setting, even with advanced bronchoscopic methods such as navigation bronchoscopy and radial ultrasound, the diagnostic yield of peripheral lung lesions is less than 60%.[5,6] Several factors account for the suboptimal diagnostic yield. The diameter of the airways leading to the lesion may be smaller than the 1.4 mm diameter radial EBUS probe. The lesion may be eccentric rather than perpendicular to the biopsy forceps. Removable of the imaging probe from a guide sheath and re-insertion of biopsy forceps or needle may cause displacement or migration of the guide sheath to a different airway. To improve the diagnostic accuracy, other methods are being developed for endoscopic detection and biopsy of peripheral lung lesions ≤20 mm. Bronchoscopic transparenchymal approach to access peripheral lung nodules from more central airways and real time flouroscopic transbronchial guidance systems are under evaluation.[7,8] Flexible 21G peripheral needles are becoming commercially available for transbronchial aspiration or core biopsy. Small optical imaging probes < 0.5 mm that can be inserted within a 21G needle to confirm abnormal pathology in real time using optical frequency domain imaging[9,10] or diffuse reflectance spectroscopy before taking a biopsy. These newer endoscopic approaches hold promise to improve diagnostic accuracy while maintaining the advantage of lower complication rates such as pneumothorax and bleeding compares to CT guided transthoracic lung biopsy. References 1. National Lung Screening Research Team, Church TR, Black WC, et al. Results of initial low-dose computed tomographic screening for lung cancer. N Engl J Med. 2013 May 23;368(21):1980-91. 2. Aberle DR, DeMello S, Berg CD, et al. Results of the two incidence screenings in the National Lung Screening Trial. N Engl J Med 2013; 369(10):920-31. 3. McWilliams A, Tammemagi MC, Mayo et al. Probability of cancer in pulmonary nodules detected on first screening CT. N Engl J Med 2013;369:910-9. 4. Horeweg N, van Rosmalen J, Heuvelmans MA, et al. Lung cancer probability in patients with CT-detected pulmonary nodules: a prespecified analysis of data from the NELSON trial of low-dose CT screening. Lancet Oncol. 2014 Nov;15(12):1332-41. 5. Ost DE, Ernst A, Lei X, et al. AQuIRE Bronchoscopy Registry. Diagnostic yield and complications of bronchoscopy for peripheral lung lesions. Results of the AQuIRE Registry. Am J Respir Crit Care Med. 2016 Jan 1;193(1):68-77. 6. Ali MS, Trick W, Mba BI, et al. Radial endobronchial ultrasound for the diagnosis of peripheral pulmonary lesions: A systematic review and meta-analysis. Respirology. 2017; 22(3):443-453. 7. Herth FJ, Li S, Jiayuan Sun J, Nader D. Bronchoscopic TransParenchymal Nodule Access: Evaluation of safety and feasibility of Archimedes System. Am J Respir Crit Care Med 2017;195:A7597. 8. Stoy SP, Whitten PE, Al-Zubaidi A, Hogarth K. Bronchoscopic peripheral lung nodule navigation by a novel Live fluoroscopic overlay guidance technology. Am J Respir Crit Care Med 2017;195:A2865. 9. Tan KM, Shishkov M, Chee A, et al. Flexible transbronchial optical frequency domain imaging smart needle for biopsy guidance. Biomed Opt Express 2012; 3::1947-1954. 10. Pahlevaninezhad H, Lee AM, A. R, et al. Endoscopic Doppler optical coherence tomography and autofluorescence imaging of peripheral pulmonary nodules and vasculature. Biomedical Optics Express. 2015; 6(10):4191-9.

Abstract:
Recent advances in molecular biology have revealed that lung cancer is not a single disease and that there are subsets of non-small cell lung cancer (NSCLC) with specific genetic alterations that are critical to the growth and survival of cancer cells. Alterations of the EGFR, ALK and ROS1 gene, which are present in a mutually exclusionary fashion, are representative driver oncogene mutations. Targeted drugs against each driver oncogene usually result in dramatic tumor shrinkage and prolongation of progression free survival (PFS) compared with conventional platinum doublet chemotherapy. However, there is only a weak association between WHO pathologic classification 2015 and type of driver oncogenes. Therefore, it is of utmost importance to identify who are likely to benefit from targeted drugs by performing molecular tests for each lung cancer patient who is a candidate for drug therapy. A list of driver oncogenes is further expanding; BRAF, RET, MET, HER2, NTRK1 are being recognized as new drivers that can be exploited in the clinic. It is getting more practical to screen these molecular alterations by use of next generation sequencing technology, rather than to detect each gene alterations one by one using different platforms. We have also known that not all the tumors with mutations of the same gene behave similarly. For example, while deletional mutation in exon 19 and L858R in exon 21 are two representative mutations that sensitize cancer cells to EGFR-tyrosine kinase inhibitors (TKI), G719X in exon 18 has an intermediate sensitivity and insertional mutation in exon 20 or de novo T790M are known to be resistant. It has been shown that there is a heterogeneity in efficacy of EGF-TKIs depending on the class of mutation. For example, afatinib is active among other EGFR-TKIs for exon 18 mutations. Furthermore, a certain molecular context is known to be associated with primary resistance even within lung cancers with the same EGFR mutations. For example, it is reported that mutations in the PI3K/AKT/mTOR pathway (AKT1, PIK3CA, STK11, PTEN) or TP53 mutations are more frequent in non-responders and are associated with shorter PFS. This context dependence may present in other driver oncogenes, too. Acquired resistance is almost inevitable in the treatment of lung cancer with targeted drug. Mechanisms of this resistance has been extensively studied and now we know there are at least 3 types of mechanisms; i.e., 1) target modification by the secondary mutation that alters the affinity between the drug and the target relative to the affinity between ATP and the target (e.g., T790M in EGFR, L1196M in ALK), 2) accessory pathway activation that bypass the inhibitory effect of the drug(e.g., Met amplification in EGFR), and histologic transformation, such as small cell lung cancer transformation and epithelial-mesenchymal transition. We are now able to use the newer generation of TKIs to treat some of the resistance due to the secondary mutation of the target gene. Osimertinib has recently been shown to prolong PFS of patients who acquired resistance to EGFR-TKI through T790M mutation compared with platinum-pemetrexed in the AURA 3 trial. Therefore, detection of this mutation which accounts for about 50~60 % of the acquired resistance against EGFR-TKI is important. However, re-biopsy is sometimes more challenging compared with that in the first-line setting, and therefore detection of T790M in cell-free DNA in plasma has been rapidly developed and is now approved in regulatory authorities in several countries. There is another issue which should be taken into consideration when treating patients with acquired resistance. When there are multiple metastatic lesions, resistance mechanisms may vary from one tumor to another. Hence, it can happen that while one tumor shrinks but others increase in size. It may be reasonable and thus beneficial for patients when treatment is planned according to most prevalent mechanism of resistance in the plasma as a sum of total resistant mechanism. In this talk, I would like to overview recent advances of molecular diagnosis in targeted therapy of lung cancer and also like to discuss future perspectives in this field.

Abstract not provided

Abstract:
ALK positive NSCLC represents 2-7% of advanced NSCLC. Three ALK TKIs have shown positive first line trials against either platinum-doublet chemotherapy (ceritinib and crizotinib) or against crizotinib (alectinib). At least three other first line trials against crizotinib are ongoing (brigatinib, ensartinib and lorlatinib). Activity of different ALK TKIs post crizotinib are characterized by comparable response rates but differing toxicity profiles, durations of benefit and extent of CNS activity. With changes in the first line standard, data post non-crizotinib ALK TKIs continues to emerge with attendant caution re the applicability of both biological and clinical data currently available for clinical decision making. Advances in our understanding of CNS trial endpoints has also helped facilitate cross trial comparisons of CNS activity of these different agents. Chemotherapeutic, radiotherapeutic and immunotherapeutic options other than ALK TKIs have all generated different clinical datasets - defining both some reasonable clinical options and some clearly in need of additional research.

Abstract:
As a result of recent advances, systematic genomic testing for patients with non-small cell lung cancer (NSCLC) is the new standard of care in clinical decision-making, due to the identification of driver molecular alterations that have triggered the development of new molecules targeting these specific alterations in cancer cells. Several studies have enabled to conclude that both EGFR-mutant and ALK-positive NSCLC constitute two defined subgroups of oncogene-driven tumors with potentially effective targeted therapy. Furthermore, approximately 15-20% of NSCLC diagnosed in Europe and North America bear EGFR mutations or ALK rearrangements, enhancing the significance of the development of drugs capable of interfering with their intracellular effects. Based on these results, the identification of other activating mutations has been pursued in hopes of improving survival in NSCLC by specifically treating these genomic alterations. These potential therapeutic targets include ROS1, BRAF, RET, HER2, MET exon 14 skipping mutations and NTRK, among others. Here, we seek to review the characteristics of emerging targets that enable interaction with molecules that specifically target these receptors in lung adenocarcinomas, as well as the results of preliminary studies that assess the efficacy of these new strategies applied to NSCLC. ROS1 ROS1 rearrangement characterizes a small subset (1%–2%) of NSCLC and is associated with slight/never smoking patients and adenocarcinoma histology. Crizotinib was shown to harbor relevant activity in ROS1-rearranged NSCLC. A number of agents including ceritinib, lorlatinib and entrectinib are now been developed in order to overcome the resistance to crizotinib. At present, ROS1-rearranged patients represents a clearly defined NSCLC molecular subgroup with highly active therapeutic options. BRAF BRAF mutations occur in 2%–4% of patients with NSCLC, with the most common resulting in a glutamate substitution for valine at codon 600 (V600E). Non‐V600E BRAF mutations make up the remaining BRAF mutations and may be either activating (i.e., G469A/V, K601E, L597R) or inactivating (i.e., D594G, G466V). Efforts targeting BRAF‐mutant NSCLC to date have almost exclusively focused on patients with V600E‐mutant disease. Direct inhibition of mutant BRAF and/or the downstream mitogen-activated protein kinase kinase has led to good outcomes survival in patients with BRAF-mutant metastatic NSCLC. Dabrafenib plus trametinib achieved 63·2% response rate (RR) in BRAF(V600E)-mutant NSCLC. RET RET fusions are detected in 1-2% of lung adenocarcinomas and a number of genes, such as KIF5B, CCDC6, NCO4 and TRIMM33, can act as fusion partners. In a global registry of RET positive NSCLC patients, 41 received a RET inhibitor achieving a median progression-free survival of 2.9 months and a median overall survival of 6.8 months. Response rate was 34% for those patients receiving cabozantinib and 27% for those receiving vandetanib. Overall RET inhibitors strategies seem active in a subgroup of patients with RET-rearranged NSCLC. However, RR is lower to that observed in EGFR-mutated/ALK-positive patients. HER2 HER2 mutations are identified in about 2-4% of NSCLC and are critical for lung carcinogenesis. A number of series shows the chemosensitivity of HER2-driven NSCLC, and the potential interest of HER2-targeted agents. In a recent study, NSCLC patients with HER2 mutations were treated with T-DM1 and achieved a 44% RR. MET Approximately 2-3% of NSCLCs harbor activating mutations of the MET proto-oncogene that cause exon 14 skipping (METex14) and accumulation of c-Met lacking a juxtamembrane domain. Recently, the clinical activity of anti-Met-targeted therapy was demonstrated in patients harboring MET exon 14 skipping lung cancer. MET seems a relevant target in NSCLC and a number of clinical trials with MET inhibitors in this population are now ongoing. NTRK TRK rearrangements represent the molecular driver of a subset of solid tumors, including 1-2% of NSCLCs. Preliminary data indicate that molecularly selected NSCLC patients harboring NTRK fusions derive an unprecedented clinical benefit from Trk-directed targeted therapies. There are two different targeted agents, entrectinib and larotrectinib, that are in phase II testing for any patients who have solid tumors with NTRK rearrangement, including NSCLC patients. Both drugs have achieved dramatic responses, regardless of histology in earlier phase I studies. In a study presented at ASCO 2017, larotrectinib has demonstrated consistent and durable antitumor activity in TRK fusion cancers, across a wide range of ages and tumor types. REFERENCES Shaw AT, Ou SH, Bang YJ, et al. Crizotinib in ROS-1 rearranged non-small-cell lung cancer: N Engl J Med. 2014; 371:1963-1971. Planchard D, Besse B, Groen HJ, et al. Dabrafenib plus trametininb in patients with previously treated BRAF (V600E)-mutant metastatic non-small cell lung cancer: an opne-label, multicentre phase 2 trial. Lancet Oncol. 2016;17:984-993. Gautschi O, Milia J, Filleron T, et al. Targeting RET in Patients With RET-Rearranged Lung Cancers: Results From the Global, Multicenter RET Registry. J Clin Oncol. 2017;35:1403-1410. Mazières J, Barlesi F, Filleron T, et al. Lung cancer patients with HER2 mutations treated with chemotherapy and HER2-targeted drugs: results from the European EUHER2 cohort. Ann Oncol. 2016;27:281-286. Lu X, Peled N, Greer J, et al. MET exon 14 mutation encodes an actionable therapeutic target in lung adenocarcinoma. Cancer Res. 2017 May 18. [Epub ahead of print]. Drilon A, Nagasubramanian R, Blake JF, et al. A Next-Generation TRK Kinase Inhibitor Overcomes Acquired Resistance to Prior TRK Kinase Inhibition in Patients with TRK Fusion-Positive Solid Tumors. Cancer Discov. 2017 Jun 3. [Epub ahead of print]. Hyman DM, Laetsch TW, Kummar S, et al. The efficacy of larotrectinib (LOXO-101), a selective tropomyosin receptor kinase (TRK) inhibitor, in adult and pediatric TRK fusion cancers. J Clin Oncol. 2017;35 (suppl; abstr LBA2501). Riely GL. What, When, and How of Biomarker Testing in Non-Small Cell Lung Cancer. J Natl Compr Canc Netw. 2017;15:686-688. Jordan EJ, Kim HR, Arcila ME, et al. Prospective Comprehensive Molecular Characterization of Lung Adenocarcinomas for Efficient Patient Matching to Approved and Emerging Therapies. Cancer Discov. 2017;7:596-609.

Abstract not provided

Abstract not provided