Search

508 Background: The VEGFR inhibitor sunitinib is a standard treatment for metastatic renal cell carcinoma (mRCC). HTN, an on-target class effect of VEGF signaling-pathway inhibitors, has been shown to correlate with clinical outcome. Studies have shown the association between genetic polymorphisms in several genes, and the development of HTN in patients treated with targeted therapies. We aimed to study the association between readily available clinicopathologic factors and the development of sunitinib induced HTN in mRCC patients. Methods: Records from mRCC patients treated with sunitinib in 9 centers across 2 countries were retrospectively reviewed. Sunitinib induced HTN was defined as systolic blood pressure ≥ 140 mmHg and/or diastolic blood pressure ≥ 90 mmHg. Analysis of the association between clinicopathologic factors and the development of HTN was performed using logistic regression. Results: Between 2004-2013, 302 patients with mRCC were treated with sunitinib. The incidence of sunitinib induced HTN was 50% (n=152). Clinicopathologic factors included in the analysis were age (median 63), gender (67% male), HENG risk (good 22%, intermediate 59%, poor 19%), smoking status (active 21%), BMI (obese=BMI ≥30, 28%; overweight=BMI 25-29.9, 37%; normal weight= BMI 1 metastatic site (82%), metastatic site (lung 72%, liver 25%, bones 40%), pre-treatment neutrophil to lymphocyte ratio (>3 in 45%), treatment line (first vs advanced), sunitinib dose reduction/treatment interruption (45%). Absence of liver metastases (OR 3.5, p=0.02), pre-treatment neutrophil to lymphocyte ratio ≤ 3 (OR 5.5, p=0.001), and BMI (overweight and normal weight vs obese, OR 2.2 and 2.3 respectively, p=0.01 both) were independently associated with the development of HTN. Conclusions: In metastatic renal cell carcinoma patients treated with sunitinib, readily available clinicopathologic factors may be used to identify patients who are prone to the development of HTN.

494 Background: Targeted txs are the tx of choice in most mRCC pts. However, HDIL2 which may produce durable responses in a small percentage of cases, is still an option in carefully selected pts. While the effect of prior HDIL2 on the outcome of targeted txs in mRCC pts is poorly defined, a recent single center report (Birkhäuser FD, Cancer J 2013) revealed an improved disease-specific survival in pts treated with prior HDIL2. We aimed to study the effect of prior HDIL2 tx on outcome of mRCC pts treated with sunitinib. Methods: Records from 302 mRCC pts treated with Su from 2004 to 2013 in 9 centers across 2 countries were retrospectively reviewed. We compared the response rate, progression free survival (PFS), and overall survival (OS), between post HDIL2 pts (n=27) and individually matched tx naïve pts (n=27). Progression free survival and overall survival were determined by Cox regression. Results: All pts had prior nephrectomy and clear cell histology. The groups were matched by age (median 61), gender (male 74%), Heng risk (favorable 37%, intermediate 59%, poor 4%), sunitinib induced hypertension (67%), sunitinib dose reduction/treatment interruption (41%), smoking status (active 7%), use of angiotensin system inhibitors (41%), the presence of more than one metastases site (96%), and pre-tx neutrophil to lymphocyte ratio (> 3 in 22%). Furthermore, they were balanced regarding the presence of lung (68%), liver (31%), and bone (43%) metastases, and the use of bisphosphonates (32%). In prior HDIL2 versus tx naïve pts, objective response was partial response/stable disease 89% (n=24) versus 74% (n=20), and progressive disease at first imaging evaluation within the first 3 months (mos) 11% (n=3) versus 26% (n=7) (p=0.29, OR 2.4). Median progression free survival was 21 versus 12 mos (HR 2.3, p=0.005), and median overall survival 25 versus 20 mos (HR 2.2, p=0.013). Conclusions: In metastatic renal cell carcinoma patients treated with sunitinib, prior high dose IL-2 therapy may improve the outcome.

256 Background: Standard treatment options for patients (pts) with metastatic castrate-resistant prostate cancer (mCRPC) include docetaxel based chemotherapy, abiraterone, and radium 223. Octogenarian pts (age 80 and older) are often considered to be unfit for chemotherapy. However, recommendations for their management is limited by the paucity of clinical trials data in this population. In countries where abiraterone in the pre-chemotherapy setting has not been approved yet, or for pts who can’t afford it, the CYP 17 inhibitor ketoconazole is used as an alternative advanced hormonal tx. We aimed to study baseline characteristics and outcome of octogenarian versus young (age 60 or younger) pts with mCRPC treated with ketoconazole. Methods: We performed an international multicenter retrospective study of pts with mCRPC, who were treated with ketoconazole at four centers across two different countries. We compared baseline characteristics and outcome of octogenarian versus young pts. The effect of very old age on prostate-specific antigen (PSA) response, progression free survival (PFS), and overall survival (OS), was tested with adjustment of other known confounding risk factors using a chi-square test and partial likelihood test from Cox model. Results: Between 2004 and 2013, 35 octogenarians (median age 83) and 33 young pts with (median age 57) mCRPC were treated with ketoconazole. The groups were balanced regarding the following baseline clinicopathologic characteristics: extent of disease (limited-axial skeleton and/or nodal versus extensive-appendicular skeleton and/or visceral), combined gleason score, pre-treatment risk category (Keizman, Oncologist 2012; based on pre-tx neutrophil to lymphocyte ratio/prostate-specific antigen doubling time, and prior response to ADT), pain intensity, ECOG performance status, alkaline phosphatase level, hemoglobin level, PSA level. In octogenarian versus young pts, PSA response (greater than or equal to 50% decline from baseline) was 40% versus 61% (OR 3.5, p=0.04), median PFS 7 versus 8 months (HR 0.91, p=0.44), and median OS 31 versus 36 months (HR 0.66, p=0.31). Conclusions: In very old vs young mCRPC patients treated with ketoconazole, PSA response was lower. Baseline clinicopathologic characteristics, PFS, and OS were not significantly different between the groups.

Background . Thymoma and thymic carcinoma with pleural spread have a high rate of locoregional recurrence and poor prognosis. Maximal debulking coupled with aggressive local treatment could offer a chance for cure. This study evaluates the early and midterm results of operation and hyperthermic pleural perfusion with cisplatinum for thymic malignancies. Methods . Fifteen patients (11 men), 20 to 67 years old (10 thymoma, 4 thymic carcinoma, 1 carcinoma in thymic cyst) underwent resection and hyperthermic pleural perfusion between 1995 to 2000. All had pleural spread proven before or intraoperatively. Six of the thymoma cases were recurrent. Current operation included resection without pleurectomy (9 patients), resection with pleurectomy (5), and extrapleural pneumonectomy (1 patient) with intraoperative hyperthermic pleural perfusion in all. Intrapleural temperature reached 40.3°C to 43°C. The total dose of cisplatinum was 150 mg or more in 14 patients. Results . Complete resection (R0) was achieved in 10 patients, subtotal (R1) in 3, and partial (R2) in 2. There was no operative mortality, no hemodynamic or respiratory disturbances during perfusion, and no hematologic, neurologic, or renal complications. Complications consisted of significant bleeding (2 patients), fever (2), and air leak (1 patient). Two patients with thymic carcinoma died after 27 and 34 months, and 1 is alive with no evidence of disease at 54 months. Two patients with thymoma died after 7 and 36 months. Eight are alive after 9 to 70 months. Four patients (all R0) are alive without local recurrence more than 60 months after operation and hyperthermic pleural perfusion. Conclusions . Operation and thermochemotherpy is feasible and safe in patients with thymic tumors. This method seems to offer excellent local control for patients with stage IV-a thymic malignancies. Midterm results suggest that operation plus hyperthermic pleural perfusion may lengthen survival in stage IV-a thymoma.

BACKGROUND Cytokeratins are constituents of the intermediate filaments of epithelial cells that are expressed in various combinations depending on the epithelial type and the degree of differentiation. The recently identified cytokeratin 20 (CK-20) was found to be expressed in colonic, gastric, and pancreatic carcinoma tumor tissues. A low rate of incidence of expression of CK-20 was found in tumor tissue from lung carcinoma but no expression was found in blood even with the sensitive reverse transcriptase-polymerase chain reaction (RT-PCR) method. The objective of the current study was to examine whether CK-20 expression in the blood can be used as a biomarker for the detection of dissemination in patients with carcinoma of the colon, stomach, and pancreas. METHODS In the current study, RT-PCR was used to determine the expression of CK-20 in the blood cells from patients with metastatic colon carcinoma (n = 22), metastatic pancreatic carcinoma (n = 28), metastatic gastric carcinoma (n = 18), metastatic lung carcinoma (n = 13), no metastatic colon carcinoma (n = 13) and no known malignant diseases (n = 22). RNA was extracted from cell pellets and analyzed by RT-PCR using primers for CK-20. RESULTS In the group of 22 patients with metastatic colon carcinoma, 14 were found to be CK-20 positive (sensitivity of 63.6% and specificity of 92.3%), 22 of the 28 pancreatic carcinoma patients showed positive CK-20 expression, and 12 of 18 patients with gastric carcinoma showed positive CK-20 expression. All patients with metastatic lung carcinoma except 1 were negative (12 of 13 patients), and 12 of 13 patients with colonic carcinoma with no known metastases also were negative. Negative CK-20 results were obtained in all 22 patients with no known malignant diseases. CONCLUSIONS The results of the current study indicate that because of its high sensitivity, RT-PCR of CK-20 is a potential biomarker for detecting metastases in blood samples from patients with carcinoma of the colon, stomach, and pancreas. Cancer 1999;86:2398–405. © 1999 American Cancer Society.

LBA8011 Background: Nintedanib (N) inhibits VEGFRs, PDGFRs, and FGFRs. LUME Lung 1 is a placebo (P) controlled phase III trial of N + docetaxel (D) in patients (pts) with locally advanced/metastatic NSCLC progressing after first-line therapy. Methods: Stage IIIB/IV or recurrent NSCLC pts (stratified by histology, ECOG PS, prior bevacizumab, and brain metastases) were randomized to N 200 mg bid + D 75 mg/m2 q21d (n=655) or P + D (n=659). 1° endpoint was centrally reviewed PFS after 713 events (2 sided stratified log-rank, α=5%, β=10%). Key 2° endpoint of OS was analyzed hierarchically after 1,121 events (2 sided, adjusted α=4.98%, β=20%), first in adenocarcinoma (adeno) pts line therapy (T

e15598 Background: Geographic differences in the outcome of pts have been described in various cancers. The VEGFR inhibitor su is a standard treatment (tx) for mRCC. The effect of geographic differences on the outcome of su tx in mRCC is poorly defined. Aims: To study the effect of geographic differences on outcome of su tx in mRCC. Methods: We performed an international multicentre retrospective study of unselected cohort of 275 mRCC pts, who were treated with su from 2004 to 2012 in 7 centers across the US and Middle East (ME; Israel). Clinicopathologic and prognostic factors, and tx outcome were compared between US (n=133) and ME (n=142) pts. Chi-square and Fisher's exact tests were used to compare categorical variables, and two-sample t-test was used to compare continuous endpoints. Progression free survival (PFS) and overall survival (OS) were determined by Cox regression. Results: Median age was 61 (US) vs 65 (ME, p=0.01). The groups were balanced regarding gender, Heng risk, past nephrectomy, RCC histology, presence of ≥ 2 metastatic sites, lung/liver/bone metastasis, use of angiotensin system inhibitors (ASI), prior cytokines/ targeted txt, su induced HTN, and su dose reduction/tx interruption secondary to side effects. The incidence of active smokers (28% vs 15%, p=0.01), bisphosphonates users (23% vs 13%, p=0.03) and pts with pre-tx neutrophil to lymphocyte ratio (NLR) ≤ 3 (63% vs 49%, p=0.04) was higher among ME pts. In US vs ME pts, objective response was partial response/stable disease 77% (n=102) vs 79% (n=112), and progressive disease at first imaging evaluation within the first 3 months (mos) 23% (n=31) vs 21% (n=30) (p=0.77, OR 1.1). Median PFS was 8 vs 12 mos (HR=1.8, p

e15597 Background: Several studies have suggested the existence of a “trial effect”, in which for a given treatment, participation in a clinical trial is associated with a better outcome of cancer patients. The VEGFR inhibitor sunitinib is a standard treatment for mRCC. The effect of clinical trial participation on the outcome of sunitinib treatment in mRCC is poorly defined. Aims: To study the effect of clinical trial participation on outcome of mRCC patients treated with sunitinib. Methods: Records from 275 mRCC patients treated with sunitinib from 2004 to 2012 in 7 centers across 2 countries were reviewed. We compared the response rate, progression free survival, and overall survival, between clinical trial participants (n=49) and a matched cohort of non participants (n=49) who who received standard therapy. Each patient participating in a clinical trial was individually matched with a non participant by clinicopathologic factors. Progression free survival and overall survival were determined by Cox regression. Results: The groups were matched by age (median 64), gender (male 67%), Heng risk (favorable 24%, intermediate 60%, poor 16%), ECOG performance status (0-1 92%), prior nephrectomy (92%), renal cell carcinoma histology (clear cell 80%), sunitinib induced hypertension (56%), and sunitinib dose reduction/treatment interruption (41%). In clinical trial participants versus non participants, objective response was partial response/stable disease 80% (n=39) versus 73% (n=36), and progressive disease at first imaging evaluation within the first 3 months (mos) 20% (n=10) versus 27% (n=13) (p=0.63, OR 1.2). Median progression free survival was 10 versus 11 mos (HR=0.96, p=0.84), and median overall survival 23 versus 24 mos (HR=0.97, p=0.89). Conclusions: In mRCC patient treated with sunitinib, the outcome of clinical trial participants was similar to matched non participants who received standard therapy.

160 Background: In patients with BRPC(M0) after local tx, the most important prognostic factor is probably the PSADT (high risk < 3, intermediate risk 3–8.9, low risk ≥9.0 months). Pre-clinical and clinical studies in several cancer types have shown that commonly prescribed medications may inhibit tumor growth. The effect of commonly prescribed medications on PSADT in pts with BRPC(M0) is poorly defined. We aimed to study the effect of commonly prescribed medications on PSADT in pts with BRPC (M0) after local tx. Methods: We reviewed the records of 156 BRPC(M0) pts enrolled in 1 prospective (Keizman, CCR 2010) and two retrospective (Keizman, prostate 2012; Mermershtain, EMUC 2011) studies, in 2 centers across 2 countries. The effect of clinicopathologic factors and the use of statins, aspirin, and angiotensin system inhibitors (ASIs; ACE-I and ARBs) on initial PSADT (from the time of first PSA relapse to the initiation of any systemic tx) was analyzed using the Mann-Whitney or Kruskal-Wallis tests and regression analyses. Results: In the whole patient cohort (n=156), median age was 62, prior local tx consisted of radical prostatectomy in128 pts and EBRT in 28, and median PSADT was 6.7 months (mos). Median PSADT in ASIs users (n=48) vs non users (n=108) was 7.85 vs 5.6 mos (p=0.011). In multivariate analysis, the use of ASIs and Gleason score were associated with median PSADT. The use of statins (n=58) or aspirin (n=72), primary tx modality, and time from primary tx to PSA relapse had no significant effect on median PSADT. PSADT risk grouping in ASIs users vs non users was low 48% vs 27%, intermediate 42% vs 49%, and high 10% vs 24% (p=0.02). In multivariate analysis, the use of ASIs, Gleason score, and prior ADT were associated with PSADT risk grouping. The use of statins or aspirin, primary tx modality, and time from primary tx to PSA relapse had no significant effect on PSADT risk grouping. Conclusions: The use of ASIs, Gleason score, and prior ADT may be associated with PSADT (median and risk grouping) of pts with BRPC (M0) after local tx. This should be investigated prospectively, and if validated, applied in clinical practice and clinical trials.

e16068 Background: Abi is a standard treatment (tx) in pts with mCRPC refractory to D. It is a potent and selective CYP 17 inhibitor, that blocks the synthesis of androgens in the testis, adrenal glands, and prostate. However, in many countries where abi has not been approved yet, keto is used as an alternative CYP 17 inhibitor. Although preclinical data suggests that keto is a less specific and potent inhibitor of CYP 17, there are limited clinical data comparing both agents. Aims: To compare the clinical effectiveness of abi vs keto in pts with mCRPC refractory to D. Methods: Records from 156 mCRPC pts treated with keto 200 - 400 mg 3x day, in 4 centers across the US and Israel, were reviewed retrospectively. 26 pts treated post D were individually matched by clinicopathologic factors to pts treated with abi (selected from a multicenter Israeli database, n=120). We compared the PSA response (decrease ≥50% from baseline), biochemical and radiological progression free survival, and overall survival between the groups. Progression free survival and overall survival were determined by Cox regression. Results: The groups were matched by Gleason score, pre-tx disease extent (limited-axial skeleton and/or nodal vs extensive- appendicular skeleton and/or visceral), ECOG PS, pre-tx risk category (favorable, intermediate, poor; Keizman, Oncologist 2012). Furthermore, they were balanced regarding median age (71 abi vs 69 keto), time from primary tx to disease relapse, time to progression on prior GnRH-a and antiandrogen, PSA response and time to progression on prior D, pre-tx pain score/alkaline phosphatase/hemoglobin/neutrophil to lymphocyte ratio/PSADT/PSA. In the groups of abi vs keto, PSA response was 46% vs 19% (OR 4.4, p=0.043), median biochemical PFS 7 vs 2 months (HR 0.65, p=0.02), median radiological PFS 6 vs 2.5 months (HR 0.63, p=0.016), median overall survival 17 vs 12 months (HR 0.53, p=0.79), and tx interruption d/t adverse events 12% vs 23% (0R 0.6, p=0.023). Conclusions: In mCRPC refractory to D, the outcome of pts treated with abiraterone was superior to ketoconazole.

LBA8011 The full, final text of this abstract will be available at abstract.asco.org at 7:30 AM (EDT) on Monday, June 3, 2013, and in the Annual Meeting Proceedings online supplement to the June 20, 2013, issue of Journal of Clinical Oncology. Onsite at the Meeting, this abstract will be printed in the Monday edition of ASCO Daily News.

453 Background: Several studies have suggested the existence of a trial effect, in which for a given treatment, participation in a clinical trial is associated with a better outcome of cancer patients. The VEGFR inhibitor sunitinib is a standard treatment for mRCC. The effect of clinical trial participation on the outcome of sunitinib treatment in mRCC is poorly defined. We aimed to study the effect of clinical trial participation on outcome of mRCC patients treated with sunitinib. Methods: Records from 275 mRCC patients treated with sunitinib from 2004 to 2012 in 7 centers across 2 countries were reviewed. We compared the response rate, progression free survival, and overall survival, between clinical trial participants (n=49) and a matched cohort of non participants (n=49) who received standard therapy. Each patient participating in a clinical trial was individually matched with a non-participant by clinicopathologic factors. Progression free survival and overall survival were determined by Cox regression. Results: The groups were matched by age (median 64), gender (male 67%), Heng risk (favorable 24%, intermediate 60%, poor 16%), ECOG performance status (0-1 92%), prior nephrectomy (92%), renal cell carcinoma histology (clear cell 80%), sunitinib induced hypertension (56%), and sunitinib dose reduction/treatment interruption (41%). In clinical trial participants vs. non participants, objective response was partial response/stable disease 80% (n=39) vs. 73% (n=36), and progressive disease at first imaging evaluation within the first 3 months (mos) 20% (n=10) vs. 27% (n=13) (p = 0.63, OR 1.2). Median progression free survival was 10 vs. 11 mos (HR=0.96, p = 0.84), and median overall survival 23 vs. 24 mos (HR=0.97, p=0.89). Conclusions: In mRCC patient treated with sunitinib, the outcome of clinical trial participants was similar to matched non participants who received standard therapy.

146 Background: Abi is a standard treatment (tx) in pts with mCRPC refractory to deocetaxel. It is a potent and selective CYP 17 inhibitor that blocks the synthesis of androgens in the testis, adrenal glands, and prostate. However, in many countries where abi has not been approved yet, keto is used as an alternative CYP 17 inhibitor. Although preclinical data suggests that keto is a less specific and potent inhibitor of CYP 17, there are limited clinical data comparing both agents. We aimed to compare the clinical effectiveness of abi vs keto in pts with mCRPC refractory to D. Methods: Records from 156 mCRPC pts treated with keto 200 - 400 mg 3x day, in 4 centers across the US and Israel, were reviewed retrospectively. 26 pts treated post D were individually matched by clinicopathologic factors to pts treated with abi (selected from a multicenter Israeli database, n=120). We compared the PSA response (decrease ≥50% from baseline), biochemical and radiological progression free survival, and overall survival between the groups. Progression free survival and overall survival were determined by Cox regression. Results: The groups were matched by Gleason score, pre-tx disease extent (limited-axial skeleton and/or nodal vs extensive- appendicular skeleton and/or visceral), ECOG PS, pre-tx risk category (favorable, intermediate, poor; Keizman, Oncologist 2012). Furthermore, they were balanced regarding median age (71 abi vs 69 keto), time from primary tx to disease relapse, time to progression on prior GnRH-a and antiandrogen, PSA response and time to progression on prior D, pre-tx pain score/alkaline phosphatase/hemoglobin/ neutrophil to lymphocyte ratio/PSADT/PSA. In the groups of abi vs keto, PSA response was 46% vs 19% (OR 4.4, p=0.043), median biochemical PFS 7 vs 2 months (HR 0.65, p=0.02), median radiological PFS 6 vs 2.5 months (HR 0.63, p=0.016), median overall survival 17 vs 12 months (HR 0.53, p=0.79), and tx interruption d/t adverse events 12% vs 23% (0R 0.6, p=0.023). Conclusions: In mCRPC refractory to D, the outcome of pts treated with abiraterone was superior to ketoconazole.

458 Background: Geographic differences in the outcome of pts have been described in various cancers. The VEGFR inhibitor su is a standard treatment (tx) for mRCC. The effect of geographic differences on the outcome of su tx in mRCC is poorly defined. We aimed to study the effect of geographic differences on outcome of su tx in mRCC. Methods: We performed an international multicenter retrospective study of unselected cohort of 275 mRCC pts, who were treated with su from 2004 to 2012 in 7 centers across the United States and Middle East (ME; Israel). Clinicopathologic and prognostic factors, and tx outcome were compared between United States (n=133) and ME (n=142) pts. Chi-square and Fisher's exact tests were used to compare categorical variables, and two-sample t-test was used to compare continuous endpoints. Progression free survival (PFS) and overall survival (OS) were determined by Cox regression. Results: Median age was 61 (United States) vs. 65 (ME, p = 0.01). The groups were balanced regarding gender, Heng risk, past nephrectomy, RCC histology, presence of ≥ 2 metastatic sites, lung/liver/bone metastasis, use of angiotensin system inhibitors (ASI), prior cytokines/ targeted txt, su induced HTN, and su dose reduction/tx interruption secondary to side effects. The incidence of active smokers (28% vs. 15%, p =0.01), bisphosphonates users (23% vs. 13%, p = 0.03) and pts with pre-tx neutrophil to lymphocyte ratio (NLR) ≤ 3 (63% vs. 49%, p = 0.04) was higher among ME pts. In United States vs. ME pts, objective response was partial response/stable disease 77% (n=102) vs. 79% (n=112), and progressive disease at first imaging evaluation within the first 3 months (mos) 23% (n=31) vs. 21% (n=30) (p = 0.77, OR 1.1). Median PFS was 8 vs. 12 mos (HR=1.8, p < 0.0001), and median OS 21 vs. 22 mos (HR=0.94, p = 0.9) in United States vs ME pts. Factors associated with PFS in multivariate analysis of the entire cohort (n=275) were geographic location (United States vs. ME), Heng risk, RCC histology, su induced HTN, ASI use, pre-tx NLR, and smoking status. Conclusions: Geographic differences in clinicopathologic factors and PFS of pts with mRCC treated with su may exist. This should be further investigated, and if validated, applied in clinical practice and clinical trials.

Background Reported response rates (RR) to 1st line gefitinib in advanced EGFR+ NSCLC range from 55 % to over 70 %. The primary goal of this retrospective analysis was to study the efficacy of gefitinib in the first line treatment of EGFR+ advanced NSCLC patients in two Israeli tertiary hospitals. Method The clinical data of patients tested positive for EGFR mutations from theShebaMedicalCenterand theMeirHospitalsince the beginning of 2010 were included in the current retrospective analysis. Results EGFR mutations were identified in 20 % of patients evaluated, with 38 patients testing positive for mutations. EGFR mutations detected were: exon 19 deletions (55 %), exon 21 L858R point mutations (26 %), exon 21 Leu861Gln (11 %), exon 18 Ala719Gly (5 %) and other (3 %). Demographic details included: median age was 69 (range 45-85); 24 females (63 %); 23 never-smokers (61 %). Histological subtypes identified as EGFR+ were: 31 adenocarcinomas (81 %), 2 squamous cell carcinomas (5 %) and 5 of non-squamous origin. 26 patients were treated with first-line gefitinib. 85 % had an ECOG performance status of 0-2. Adverse events were generally mild and similar to published reports - 9 experienced grade 1-2 rash (34 %), 6 experienced grade 1-2 diarrhea (23 %) and one patient had grade 3 hypomagnesemia. 23 of these patients have been evaluated for treatment response. The best response according to the RECIST criteria was: 3 patients achieved a partial response (13 %), 10 achieved stable disease (43 %), and 10 had progressive disease (43 %). Median time to progression was 5 months and median overall survival was 12 months. Thus, in this current retrospective study of the Israeli population, our response rate, TTP and OS is significantly lower than expected. The null-hypothesis of a minimal 50 % response rate (as seen for other western populations) is rejected with a P value of 0·01 using a two-tailed chi-square test. Conclusion Albeit the small number of patients in this current retrospective study, the low rate of radiological response suggests that in our tested population, the actual response rate is significantly lower than reported in the literature. Epidemiologic, pharmacogenetic and/or molecular mechanisms of resistance need to be explored to explain these differences.

4619 Background: Bis are used to prevent skeletal events of bone mets, and may exhibit anti tumor effects. We aimed to evaluate whether Bis can bring a RR, PFS, and OS benefit to pts with bone mets from RCC that are treated with Su. Methods: We performed an international multicenter retrospective study of pts with bone mets from RCC who were treated with Su. Pts were divided into Bis users (group 1) and nonusers (group 2). The effect of Bis on RR, PFS and OS, was tested with adjustment for known prognostic factors using a chisquare test from contingency table and partial likelihood test from Cox regression model. Results: Between 2004-2011, 244 pts with metastatic RCC were treated with Su. 92 pts had bone mets, 41 group 1 and 51 group 2. The groups were balanced regarding the following known prognostic factors: past nephrectomy, clear cell vs non clear cell histology, initial diagnosis to sunitinib treatment (tx) time, presence of ≥ 2 mets sites, presence of lung/liver mets, ECOG performance status, anemia, calcium level > 10 mg/dL, elevated alkaline phosphatase (AP), pre-tx neutrophil to lymphocyte ratio (NLR) >3, sunitinib induced HTN, and the use of angiotensin system inhibitors. They were also balanced with regard to past cytokines/targeted tx, and mean sunitinib dose/cycle. Objective response was partial response/stable disease 85% (n=35) vs 71% (n=36), and progressive disease 15% (n=6) vs 29% (n=15) (OR 3.287, p=0.07) in group 1 vs 2 respectively. Median PFS was 15 vs 5 months (HR 0.433, p=0.035), and median OS not reached with a median folloup time of 43 mos vs 12 months (HR 0.398, p=0.003), in favor of group 1. In multivariate analysis of the entire pt cohort (n=92), factors associated with PFS were Bis use (HR 0.433, p=0.035), pre-tx NLR ≤3 (HR 0.405, p=0.016), and elevated AP (HR=3.63, p=0.012). Factors associated with OS were Bis use (HR 0.32, p=0.003), elevated AP (HR 3.18, p=0.002), and Su induced HTN (HR 0.193, p< 0.001). Conclusions: Bis may improve the outcome of Su tx in RCC with bone mets. This should be investigated prospectively, and if validated applied in clinical practice and clinical trials.

e15058 Background: Obesity, smoking, hypertension (HTN) and diabetes (DM) are risk factors for RCC development. Their presence has been associated with a worse outcome of therapy (tx) in various metastatic cancers. We sought to determine their influence on the progression free survival (PFS) and overall survival (OS) of Su tx in mRCC. Methods: We performed an international multicenter retrospective study of pts with mRCC, who were treated with Su. We analyzed the pre-tx status of smoking (active vs past vs never), BMI (obese= BMI≥30 vs overweight=BMI 25-29.9 vs normal weight= BMI 10, elevated alkaline phosphatase (AP), pre-tx neutrophil to lymphocyte ratio (NLR) >3, Su induced HTN, use of angiotensin system inhibitors (ASIs), past cytokines/targeted tx, and median Su dose/cycle. PFS and OS were determined by the Kaplan-Meier method. Multivariate analyses using Cox Regression model were performed to determine their independent effect. Results: Between 2004-2011, 244 pts with mRCC were treated with Su. 51 pts were active smokers, 58 obese, 62 diabetic, and 145 had pre-tx HTN. In the entire pt cohort, median PFS was 9 months (mos) and OS 21 mos. Factors associated with PFS were active smoking (HR 2.29, p= 0.003, median PFS 4 vs 10 mos in past smokers vs 12 mos in never smokers), non clear cell histology (HR 1.7, p=0.042), pre-tx NLR >3 (HR 1.92, p3 (HR 2.5, p

4564 Background: The CYP17 inhibitor keto is active in mCRPC. The NLR, an index of systemic inflammation, is associated with prognosis in several types of cancer. We assessed the association between pre-tx NLR and outcome of mCRPC pts treated with keto. Methods: We performed an international multicenter retrospective study of pts with mCRPC, who were treated with keto. We analyzed the pre-tx NLR and previously described factors associated with keto tx outcome as prior response to hormonal tx, pre-tx PSADT, and extent of metastatic disease (limited vs extensive). Progression free survival (PFS) was determined by the Kaplan-Meier method. Multivariate analyses using Cox regression model were performed to determine their independent effect, and to form a predictive model. A survival tree analysis was used to find the best NLR cut-off value. Results: From 1999-2011, 156 mCRPR pts (median age 69) were treated with keto. 78/156 (50%) had ≥ 50% PSA decline. Overall median PFS was 8 months (mos) (range 1-144). Excluded from the analysis were 23 pts without available data on pre-tx NLR, and those with recent (≤1 mos) health event or tx (surgery, steroids, radiation) associated with a change of blood counts. 133 pts were included in the analysis. 62 (47%) had an elevated pre-tx NLR >3. Risk factors associated with PFS (table) were pre-tx NLR >3, prior response to GnRH-a

e18108 Background: The EGFR inhibitor erlotinib is a standard second line tx for mNSCLC. Statins are used in the tx of hyperlipidemia. Pre-clinical and clinical studies in several cancer types have shown that they may inhibit tumor growth. Their effect on the outcome of erlotinib as second line tx in mNSCLC is poorly defined. We aimed to study the effect of statins on the outcome of erlotinib as second line tx for mNSCLC. Methods: We performed a retrospective study of an unselected cohort of pts with mNSCLC, who were treated continuously with 150mg of oral erlotinib. Pts were divided into 2 groups: (1) statins users and (2) statins naive. The effect of statins use on objective response, progression free survival (PFS) and overall survival (OS), was tested with adjustment of other known confounding risk factors using a chisquare test and partial likelihood test from cox model. Results: Between 2005-2011, 107 pts with mNSCLC were treated with second line erlotinib. There were 51 statins users (group 1) and 56 nonusers (group 2). All users started statins before erlotinib tx initiation. The groups were balanced regarding the following known clinical prognostic factors: female gender, ECOG performance status, active smoking, anemia, adenocarcinoma histology type, EGFR mutation (positive vs negative + unknown). Objective response in group 1 vs 2 was partial response (PR) 41% vs 29% (p=0.15), stable disease (SD) 41% vs 25% (p=0. 11), and progressive disease (PD) 18% vs 46% (OR=2.5, p=0.07). Median PFS was 12 vs 3 ms (HR 0.44 in statins users, p=0.02). Median OS was 35 vs 19 ms (HR 0.63, p=0.1). Conclusions: Statins may improve the outcome of pts with mNSCLC that are treated with erlotinib as second line tx. This should be investigated prospectively, and if validated, applied in clinical practice and clinical trials.

379 Background: Bis are used to prevent skeletal events of bone mets, and may exhibit anti tumor effects. We aimed to evaluate whether Bis can bring a RR, PFS, and OS benefit to pts with bone mets from RCC that are treated with Su. Methods: We performed a multicentre retrospective study of pts with bone mets from RCC who were treated with Su. Pts were divided into Bis users (group 1) and nonusers (group 2). The effect of Bis on RR, PFS and OS, was tested with adjustment for known prognostic factors using a chisquare test from contingency table and partial likelihood test from Cox regression model. Results: Between 2004–2011, 209 pts with metastatic RCC were treated with Su. 76 pts had bone mets, 35 group 1 and 41 group 2. The groups were balanced regarding the following known prognostic factors: past nephrectomy, clear cell/non clear cell histology, time from initial diagnosis to sunitinib treatment (tx), the presence of > 2 mets sites, the presence of lung/liver mets, ECOG performance status, anemia, calcium level >10 mg/dL, elevated alkaline phosphatase, platelets count, pre-tx neutrophil to lymphocyte ratio (NLR) >3, sunitinib induced HTN, and the use of angiotensin system inhibitors. They were also balanced with regard to past cytokines/targeted tx, and mean sunitinib dose/cycle. Objective response was partial response/stable disease 86% (n=30) vs 71% (n=29), and progressive disease 14% (n=5) vs 29% (n=12) (p=0.125, OR 2.48) in group 1 vs 2 respectively. Median PFS was 15 vs 5 months (HR 2.6, p < 0.0001), and median OS 21 vs 13 months (HR 2.1, p=0.029), in favor of group 1. In multivariate analysis of the entire pt cohort (n=76), factors associated with PFS were Bis use (HR 2.2, p=0.035) and pre-tx NLR >3 (HR 0.38, p=0.009). Factors associated with OS were Bis use (HR 2.8, p=0.008), elevated alkaline phosphatase level (HR 0.287, p=0.0003), and Su induced HTN (HR 5.57, p < 0.0001). Conclusions: Bis may improve the outcome of Su tx in RCC with bone mets. Whether this is generalizable to other TKIs is not known. This should be investigated prospectively, and if validated applied in clinical practice and clinical trials.

437 Background: Obesity, smoking, hypertension (HTN) and diabetes (DM) are risk factors for RCC development. Their presence has been associated with a worse outcome of therapy (tx) in various metastatic cancers. We sought to determine their influence on the progression free survival (PFS) and overall survival (OS) of Su tx in mRCC. Methods: We performed a multicentre retrospective study of pts with mRCC, who were treated with Su. We analyzed the pre-tx status of smoking (active vs past vs never), BMI (obese=BMI≥30 vs overweight=BMI 25-29.9 vs normal weight=BMI 2 metastasis (mets) sites, lung/liver/bone mets, ECOG performance status, anemia, calcium level > 10 mg/dL, elevated alkaline phosphatase (AP), platelets count, pre-tx neutrophil to lymphocyte ratio (NLR) >3, Su induced HTN, use of angiotensin system inhibitors (ASIs), past cytokines/targeted tx, and mean Su dose/cycle. PFS and OS were determined by the Kaplan-Meier method. Multivariate analyses using Cox Regression model were performed to determine their independent effect. Results: Between 2004-2011, 209 pts with mRCC were treated with Su. 40 pts were active smoker, 51 obese, 55 diabetic, and 122 had pre-tx HTN. In the entire pt cohort, median PFS was 8 months (mos) and OS 15 mos. Factors associated with PFS were active smoking (HR 2.5, p= 0.005, median PFS 4 vs 10 mos in past smokers vs 10 mos in never smokers), non clear cell histology (HR 1.8, p=0.023), pre-tx NLR >3 (HR 0.2, p3 (HR 0.294, p Conclusions: Active smoking may decrease the PFS and OS of pts with mRCC that are treated with Su. BMI, DM, and pre-tx HTN were not found to be associated with outcome. These results should be investigated prospectively, and if validated applied in clinical practice and clinical trials.

37 Background: The CYP17 inhibitor Keto is active in mCRPC. The NLR, an index of systemic inflammation, is associated with prognosis in several types of cancer. We assessed the association between pre-tx NLR and outcome of mCRPC pts treated with keto. Methods: We performed a multicenter retrospective study of pts with mCRPC, who were treated with keto. We analyzed the pre-tx NLR and previously described factors associated with keto tx outcome as prior response to hormonal tx, pre-tx PSADT, and extent of metastatic disease (limited-axial skeleton/nodal vs extensive- appendicular skeleton/visceral). Progression free survival (PFS) was determined by the Kaplan-Meier method. Multivariate analyses using Cox regression model were performed to determine their independent effect, and to form a predictive model. A survival tree analysis was used to find the best NLR cut-off value. Results: From 1999-2011, 135 mCRPR pts were treated with keto. 67/135 (50%) had ≥ 50% PSA decline. Overall median PFS was 8 months (mos) (range 1-134 ). Excluded from the analysis were pts without available data on pre-tx NLR (n=8), and those with recent (≤1 mos) health event (surgery, n=1) or tx (steroids, n=3 or radiation, n=3) known to be associated with a change of blood counts. 120 pts were included in the analysis. 57 (48%) had an elevated pre-tx NLR >3. Risk factors associated with PFS (table) were pre-tx NLR >3, prior response to GnRH-a Conclusions: In mCRPC pts treated with keto, pre-tx NLR, prior response to hormonal tx, and pre-tx PSADT are associated with PFS, and may be used to categorize pts into risk groups. [Table: see text]

e21030 Background: Heat shock protein 90 (Hsp90) is a molecular chaperon essential for folding and stability of multiple oncogenes and an established target in cancer therapeutics (lung included). ...

8010 Background: Vandetanib is a once-daily oral inhibitor of VEGFR, EGFR and RET signaling. A phase I trial of vandetanib + pemetrexed (pem) supported further investigation of this combination (de Boer et al, Ann Oncol 2008). Methods: The primary objective was to determine whether vandetanib 100 mg/day + pem 500 mg/m2 every 21 days (max 6 cycles) prolonged progression-free survival (PFS) vs placebo + pem. Overall survival (OS), objective response rate (ORR), time to deterioration of symptoms (TDS, by Lung Cancer Symptom Scale) and safety were secondary endpoints. Efficacy and safety were assessed in females as a co-primary analysis population. Eligibility criteria included stage IIIB/IV NSCLC, PS 0–2, and previous 1st-line therapy. Results: Between Jan 07-Mar 08, 534 patients (mean age 59 yrs; 38% female; 21% squamous histology; 8% brain metastases; stage IV 84%; PS 0/1/2: 41%/53%/6%) were randomized 1:1 to receive vandetanib + pem (n=256) or placebo + pem (n=278). Baseline characteristics were similar in both arms. Median duration of follow-up was 9.0 months, with 83% patients progressed and 50% deceased. There were positive trends seen for vandetanib + pem for both PFS (hazard ratio [HR] 0.86, 97.58% CI 0.69–1.06; P=0.108) and OS (HR 0.86, 97.54% CI 0.65–1.13; P=0.219); similar advantages were observed for females. There were statistically significant advantages for ORR (19.1% vs 7.9%, P [Table: see text]

Background Bevacizumab/cisplatin/gemcitabine prolongs progression-free survival in the first-line treatment of advanced non–small-cell lung cancer (NSCLC). Pulmonary hemorrhage (PH) was reported in a phase II study of bevacizumab plus chemotherapy in NSCLC, leading to the exclusion of predominantly squamous cell carcinoma in subsequent NSCLC trials. Patients and Methods Subjects were treated on protocol BO17704, a randomized, double-blind phase III study of cisplatin/gemcitabine with or without bevacizumab (7.5 mg/kg or 15 mg/kg) for up to 6 cycles followed by bevacizumab until disease progression, for first-line treatment of advanced/recurrent nonsquamous NSCLC. Patients with previous grade ≥ 2 hemoptysis or with lesions abutting or invading major blood vessels were excluded. Pulmonary hemorrhage cases were identified by reported Adverse Event MedDRA Preferred Terms. The following preferred terms associated with PH were found in the BO17704 database: hemoptysis, respiratory tract hemorrhage, and bronchial hemorrhage. In addition, a clinical review of all serious bleeding on BO17704 reported to the Roche Databases was performed to identify possible additional cases of PH. Results Central lesions, exclusive of lymph nodes, were reported in 381 of 1043 subjects (36.5%) overall. Events in the table were as reported through the adverse event case report form (8 cases) or via clinical review of the Roche Databases* (2 cases). There was 1 fatal event in the placebo arm and 1 fatal event in the 15-mg/kg bevacizumab arm; 4 of 5 grade 3-5 events in the 7.5-mg/kg bevacizumab arm were fatal, and all grade 3-5 events in the 15-mg/kg bevacizumab arm were fatal at the time of clinical data cut-off. Of grade 3-5 PH events identified, 2 of 10 were associated with thrombocytopenia (grade 1 and 3). Grade 3/4 thrombocytopenia occurred at a rate of 23%-27% across treatment arms. Conclusion The incidence of severe PH in BO17704 (1.2% across both bevacizumab-containing arms) was lower than in E4599 (2.3%). Most PH events in BO17704 occurred in the 7.5-mg/kg bevacizumab arm, although study treatment duration was slightly longer in the 7.5-mg/kg bevacizumab arm (mean, 4.94 cycles) than in the 15-mg/kg bevacizumab arm (mean, 4.63 cycles).

Background Bevacizumab/cisplatin/gemcitabine prolongs progression-free survival in the first-line treatment of advanced non– small-cell lung cancer (NSCLC). Venous thrombosis necessitating full-dose anticoagulation (FDAC) is common in NSCLC. Because of concern about severe pulmonary hemorrhage (PH), there is limited experience with FDAC and bevacizumab in the setting of NSCLC. Herein, we report on the safety of bevacizumab therapy in 53 patients with NSCLC treated with concomitant FDAC. Patients and Methods Subjects were treated on protocol BO17704, a randomized, double-blind phase III study of cisplatin/gemcitabine with or without bevacizumab (7.5 mg/kg or 15 mg/kg) for up to 6 cycles followed by bevacizumab until disease progression, for first-line treatment of advanced/recurrent nonsquamous NSCLC. Full-dose anticoagulation was not permitted at study entry but was allowed for thrombotic events during study participation. Subjects on FDAC were identified by anticoagulant use and presence of a thrombotic adverse event after initiation of study treatment. Results Approximately 2/3 of FDAC subjects were treated with heparinoids; the remainder was treated with warfarin/warfarin derivatives. No bleeding events led to death in the FDAC population. There were no severe PH events in the FDAC population; all hemoptysis events were grade 1. Among subjects receiving FDAC, the majority of bleeding events were grade 1 epistaxis. All 5 grade 1-5 PH events in the bevacizumab 15-mg/kg arm were grade 1 hemoptysis. In the FDAC population, there was 1 grade 4 central nervous system bleed in the placebo arm and 1 grade 2 central nervous system bleed in the bevacizumab 15-mg/kg arm. Conclusion There were no cases of severe PH in the FDAC population, although there were few events overall. As expected, bleeding rates are higher in the FDAC population, regardless of treatment.

7645 Background: In patients (pts) with relapsed NSCLC, erlotinib 150 mg/d significantly prolonged survival, delayed symptom progression, and improved quality of life versus placebo (Shepherd et al, N Engl J Med 2005;353:123–32). TRUST is an open label, non- randomized trial initiated to provide erlotinib access to pts with advanced NSCLC. Methods: Eligible pts had stage IIIb/IV NSCLC, and had failed or were unsuitable for chemotherapy. Erlotinib (150 mg/d p.o.) was given until disease progression or unacceptable toxicity. Pts were monitored monthly. Results: In November 2006, data were available for 5,015 pts (ITT population) from 51 countries. Median age was 63y (range 19–95). Pt characteristics (%) were: male/female 62/38; Caucasian/Oriental/other 76/19/5; non-smoker/ex- or current-smoker 28/71 (no data 1); ECOG PS 0/1/2/3 21/53/20/6; adenocarcinoma/squamous cell/other 53/25/21; stage IIIb/IV 22/78; erlotinib 1st/2nd/3rd-line/other 14/48/37/1. Safety data were available for 4,423 pts, 55% of whom had at least one adverse event (AE). Only 5% had one or more erlotinib- related serious AEs, the most common being gastrointestinal (GI) disorders (86 pts; 63 grade [gr] 3/4). 6% of pts discontinued treatment due to erlotinib-related AEs: GI disorders in 96 pts (54 gr 3/4), skin disorders in 92 (50 gr 3/4). Unexpected erlotinib-related AEs occurred in 10% of pts (4% gr 1, 3% gr 2, 3% gr 3/4). As expected, rash was observed in 70% of pts, with the majority (84%) being of gr 1/2. 80% pts received >4 weeks of erlotinib. Among 4,405 pts, only 14% had dose reductions, mainly due to rash (83%) and diarrhea (21%). Similar safety results were seen for 2nd-line pts only. Efficacy for all and 2nd-line pts will be presented. Conclusions: These results, achieved through routine clinical use of erlotinib in unselected pts with advanced NSCLC, confirm the favorable tolerability profile seen with erlotinib in selected patients in the clinical trial setting. [Table: see text]

Background Pembrolizumab (pembro) is indicated for patients (pts) with microsatellite instability-high (MSI-H) solid tumors after 1 prior therapy, and MSI-H colorectal cancer (CRC) after prior fluoropyrimidine, oxaliplatin, and irinotecan, based in part on data showing durable clinical benefit with pembro in the phase II studies KEYNOTE (KN)164 ([NCT02460198] in 61 pts (cohort A) with MSI-H CRC) and KN158 ([NCT02628067] in19 pts with MSI-H non-CRC). Here, we report results of the antitumor activity of pembro in pts with MSI-H tumors from a pooled analysis of KN164 and KN158, with ≥18mo of additional follow-up across 28 tumor types. Methods KN164 enrolled pts with MSI-H CRC (cohort A [≥2 prior], cohort B [≥1 prior therapy]), while KN158 included pts with MSI-H non-CRC (≥1 prior therapy). MSI-H status was determined locally by IHC or PCR or centrally by PCR. Eligible pts in both received pembro 200mg Q3W. Tumor response was assessed every 9 wk. Primary endpoint was ORR by central review per RECIST v1.1. Data cutoff date was Sept 4, 2018 for KN164 and Dec 6, 2018 for KN158. Results At data cutoff, 357 pts (124 with MSI-H CRC, 233 with MSI-H non-CRC) were enrolled. Pts had median age of 59 years (range 20-87) and 350 (98%) had ≥1 prior therapy. Common MSI-H non-CRC tumor types included endometrial (n=49), gastric (n=24), cholangiocarcinoma (n=22), pancreatic (n=22), small intestinal (n=19), ovarian (n=15), brain (n=13), sarcoma (n=9), neuroendocrine (n=7), cervical and prostate (n=6) cancers. Median follow-up was 18.0mo (range 0.1-35.6). Confirmed ORR was 34% (n=121; 95% CI 29-39); 30 (8%) pts had CR. Median DOR was not reached (range 2.9 to 31.3+); 54% of pts had response duration ≥18mo. Median OS was 27.8 months mo (95% CI 21.3 to not reached), with 2-year OS rate of 52%. Median PFS was 4.0mo (95% CI 2.5-4.3) with 2-year PFS rate of 31%. Serious drug-related events occurred in 11 (9%) pts with MSI-H CRC and 18 (8%) pts with MSI-H non-CRC. The safety profile was consistent with that previously seen for pembro. A pooled safety analysis will be presented. Conclusions Pembro provides robust antitumor activity with durable responses and a manageable safety profile in pts with MSI-H cancers independent of tumor type. Clinical trial identification KN164 NCT02460198 KN158 NCT02628067. Editorial acknowledgement Luana Atherly-Henderson, Merck Sharp & Dohme, a subsidiary of Merck & Co., Inc, Kenilworth, NJ, USA. Legal entity responsible for the study Merck Sharp & Dohme Corp., a subsidiary of Merck & Co., Inc., Kenilworth, NJ, USA. Funding Merck & Co., Inc. Disclosure L.A. Diaz: Advisory / Consultancy, Research grant / Funding (institution), Travel / Accommodation / Expenses: Merck & Co., Inc.; Advisory / Consultancy: Caris; Advisory / Consultancy: Lyndra; Advisory / Consultancy: Genocea Biociences; Advisory / Consultancy: Illumina; Advisory / Consultancy: Cell Design Labs; Advisory / Consultancy, Shareholder / Stockholder / Stock options: Neophore; Leadership role, Shareholder / Stockholder / Stock options, Officer / Board of Directors: PGDx; Officer / Board of Directors: Jounce Therapeutics; Licensing / Royalties: Johns Hopkins; Licensing / Royalties: MSKCC; Spouse / Financial dependant: Amgen. D. Le: Honoraria (self), Advisory / Consultancy, Research grant / Funding (institution): Merck & Co., Inc.; Advisory / Consultancy, Research grant / Funding (institution): BMS; Research grant / Funding (institution): Aduro Biotech. M. Maio: Honoraria (institution), Advisory / Consultancy, Travel / Accommodation / Expenses: BMS; Honoraria (institution), Advisory / Consultancy, Travel / Accommodation / Expenses: MSD; Honoraria (institution), Advisory / Consultancy, Travel / Accommodation / Expenses: Roche; Honoraria (institution), Advisory / Consultancy, Travel / Accommodation / Expenses: GSK; Honoraria (institution), Advisory / Consultancy, Travel / Accommodation / Expenses: MedImmune. P.A. Ascierto: Advisory / Consultancy, Research grant / Funding (institution): BMS; Advisory / Consultancy, Research grant / Funding (institution): Roche; Advisory / Consultancy, Travel / Accommodation / Expenses: MSD; Advisory / Consultancy, Research grant / Funding (institution): Array; Advisory / Consultancy: Novartis; Advisory / Consultancy: Merck Serono; Advisory / Consultancy: Pierre Fabre, Incyte; Advisory / Consultancy: Genmab; Advisory / Consultancy: Newlink Genetics; Advisory / Consultancy: MedImmune; Advisory / Consultancy: AstraZeneca; Advisory / Consultancy: Syndax; Advisory / Consultancy: Sun Pharma; Advisory / Consultancy: Sanofi; Advisory / Consultancy: Idera; Advisory / Consultancy: Ultimovacs; Advisory / Consultancy: Sandoz; Advisory / Consultancy: Immunocore; Advisory / Consultancy: 4SC. R. Geva: Advisory / Consultancy: Bayer; Honoraria (institution), Advisory / Consultancy: MSD; Honoraria (institution), Advisory / Consultancy, Educational grant to the research unit - Novartis: Novartis; Honoraria (institution), Travel / Accommodation / Expenses: BMS; Honoraria (institution): Lilly; Honoraria (institution): Medison; Honoraria (institution), Travel / Accommodation / Expenses: Roche; Honoraria (institution): Janssen; Honoraria (institution): Takeda; Honoraria (institution), Travel / Accommodation / Expenses: Merck & Co., Inc. D. Motola-Kuba: Speaker Bureau / Expert testimony: Pfizer; Speaker Bureau / Expert testimony: Roche; Speaker Bureau / Expert testimony: Asopharma; Speaker Bureau / Expert testimony: AstraZeneca; Speaker Bureau / Expert testimony: BMS; Speaker Bureau / Expert testimony: MSD. T. Andre: Honoraria (self), Advisory / Consultancy: AstraZeneca; Honoraria (self), Advisory / Consultancy: Amgen; Honoraria (self), Advisory / Consultancy, Travel / Accommodation / Expenses: BMS; Honoraria (self), Advisory / Consultancy, Travel / Accommodation / Expenses: MSD; Honoraria (self), Advisory / Consultancy: HalioDx; Honoraria (self), Advisory / Consultancy, Travel / Accommodation / Expenses: Roche; Honoraria (self), Advisory / Consultancy: Sanofi; Honoraria (self), Advisory / Consultancy: Servier. E. Van Cutsem: Advisory / Consultancy: Astellas; Advisory / Consultancy: AstraZeneca; Advisory / Consultancy, Research grant / Funding (institution): Bayer; Advisory / Consultancy, Research grant / Funding (institution): BMS; Advisory / Consultancy, Research grant / Funding (institution): Celgene; Advisory / Consultancy, Research grant / Funding (institution): Lilly; Advisory / Consultancy, Research grant / Funding (institution): MSD; Advisory / Consultancy, Research grant / Funding (institution): Merck KGaA; Advisory / Consultancy, Research grant / Funding (institution): Novartis; Advisory / Consultancy, Research grant / Funding (institution): Roche; Advisory / Consultancy, Research grant / Funding (institution): Servier; Research grant / Funding (institution): Amgen; Research grant / Funding (institution): Boehringer Ingelheim; Research grant / Funding (institution): Ipsen. J. Delord: Advisory / Consultancy, Research grant / Funding (institution): Roche; Advisory / Consultancy, Research grant / Funding (institution): Genentech; Advisory / Consultancy, Research grant / Funding (institution): MSD; Advisory / Consultancy, Research grant / Funding (institution): BMS; Advisory / Consultancy, Research grant / Funding (institution): Novartis. D. Jager: Advisory / Consultancy: Roche; Advisory / Consultancy: BMS; Advisory / Consultancy: Bayer. T.W. Kim: Research grant / Funding (institution): Merck Serono; Research grant / Funding (institution): AstraZeneca; Research grant / Funding (institution): Pfizer. R. Guimbaud: Speaker Bureau / Expert testimony: Pierre Fabre; Speaker Bureau / Expert testimony: MSD; Travel / Accommodation / Expenses: Roche; Travel / Accommodation / Expenses: Ipsen; Travel / Accommodation / Expenses: Merck & Co., Inc.; Travel / Accommodation / Expenses: Servier; Travel / Accommodation / Expenses: Novartis. T. Yoshino: Research grant / Funding (institution): Novartis K.K.; Research grant / Funding (institution): MSD K.K.; Research grant / Funding (institution): Sumitomo Dainippon Pharma ; Research grant / Funding (institution): Chungai Pharmaceutical; Research grant / Funding (institution): Sanofi K.K.; Research grant / Funding (institution): Daiichi Sankyo Company; Research grant / Funding (institution): PAREXEL International Inc; Research grant / Funding (institution): Ono Pharmaceutical Co. M. Chen: Full / Part-time employment: Merck & Co., Inc. K. Norwood: Full / Part-time employment: Merck & Co., Inc. P. Marinello: Full / Part-time employment: Merck & Co., Inc. A. Marabelle: Advisory / Consultancy: Merck Serono; Advisory / Consultancy: eTheRNA; Advisory / Consultancy: Lytix pharma; Advisory / Consultancy: Kyowa Kirin Pharma; Honoraria (institution), Advisory / Consultancy: Bayer; Advisory / Consultancy: Novartis; Honoraria (institution), Advisory / Consultancy: BMS; Advisory / Consultancy: Symphogen; Advisory / Consultancy: Genmab; Advisory / Consultancy: Amgen; Advisory / Consultancy: Biothera; Advisory / Consultancy: Nektar; Advisory / Consultancy: GSK; Advisory / Consultancy: Oncovir; Advisory / Consultancy: Pfizer; Advisory / Consultancy: Seattle Genetics; Advisory / Consultancy: Flexus Bio; Advisory / Consultancy: Roche/Genentech; Honoraria (institution), Speaker Bureau / Expert testimony: Roche; Honoraria (institution): Pierre Fabre; Honoraria (institution): Onxeo; Honoraria (institution): EISAI; Honoraria (institution): Genticel; Honoraria (institution): Rigontec; Honoraria (institution): Daiichi Sankyo; Honoraria (institution): Imaxio; Honoraria (institution): Sanofi; Honoraria (institution): BioNTech. All other authors have declared no conflicts of interest.