Background: Most patients with small-cell lung cancer (SCLC) present with extensive-stage (ES) disease and have a poor prognosis despite achieving high initial response rates to platinum-based doublet chemotherapy. This study evaluated whether adding hydroxychloroquine (HCQ) to chemotherapy could improve outcomes. Methods: This was a randomised multicentre phase II trial. Eligible patients had untreated ES-SCLC, a performance status 0-2 and measurable disease. Patients were randomly assigned (1:1 ratio) to HCQ (400 mg orally twice daily) plus carboplatin-gemcitabine or carboplatin-etoposide alone. Chemotherapy was administered for up to six cycles, with HCQ given concurrently and then as single agent for up to 30 months. Primary endpoint was PFS, aiming for a hazard ratio (HR) of 0.70. Results: 72 patients were randomised (36 HCQ+chemotherapy and 36 chemotherapy alone). Median HCQ treatment duration was 4.4 months. HCQ did not improve PFS (HR 1.12 95 %CI 0.69-1.84; p = 0.64), with a median of 5.7 months (HCQ+chemotherapy) versus 6.2 months (chemotherapy). The corresponding median OS were 8.9 and 10.2 months (HR 0.83, 95 %CI 0.48-1.45, p = 0.52). Fewer patients in the HCQ arm completed four cycles of chemotherapy due to adverse events (64 % vs. 81 %). Grade >= 3 adverse events were higher in the HCQ+chemotherapy arm (83.3 % vs. 27.8 %), primarily anaemia, neutropenia, and thrombocytopenia, partly due to the initially higher gemcitabine dose used Conclusions: Combining HCQ with platinum doublet chemotherapy did not improve PFS or OS outcomes for ESSCLC, resulting in more patients stopping chemotherapy due to increased adverse events. When considered

8507 Background: All patients with malignant pleural mesothelioma (MPM) eventually relapse following standard chemotherapy. However, there is no standard treatment option in this setting. Vinorelbine, exhibits useful clinical activity but has not been formally evaluated in a randomised clinical trial, despite its widespread off-label use worldwide. BRCA1 regulates spindle assembly checkpoint in MPM and predicts vinorelbine sensitivity in preclinical models [1,2], suggesting that BRCA1 negative patients may be chemoresistant. Methods: VIM, a Cancer Research UK funded, investigator-initiated randomised controlled phase 2 multi-centre UK trial, enrolled patients with MPM who had progressed after first-line chemotherapy. Pts were randomised 2:1 to either vinorelbine (60mg/m2 weekly Q21d escalating to 80mg/m2 from cycle 2) + active supportive care (ASC) versus ASC until disease progression, unacceptable toxicity or withdrawal of consent. The primary outcome was progression free survival (PFS) defined as the time from randomisation to any progression (based on Modified RECIST criteria for assessment of response in malignant pleural mesothelioma) or death. The trial had 90% power to detect a hazard ratio of 0.65 at the one-sided 20% significance level. Secondary endpoints were overall survival (OS), tolerability and safety. Results: Between May 2016 and Oct 2018, 154 patients were recruited from 10 UK sites and randomised to vinorelbine + ASC (n=98) or ASC alone (n=56). In the Intention-to-treat analysis, after 129 events, median PFS was 4.2 months (m) for vinorelbine + ASC compared to 2.8m for ASC alone (Hazard Ratio (HR) 0.59; 95% CI: 0.41 to 0.85; one-sided p = 0.0017). 108 deaths were reported. Median OS was 9.3m for vinorelbine + ASC compared to 9.1m for ASC alone (HR=0.79; 95% CI: 0.53 to 1.17; two-sided p = 0.24). Toxicity data and subgroup analyses including the impact of BRCA1 deficiency will be presented. Conclusion: The trial met its primary endpoint. Vinorelbine demonstrates useful clinical efficacy in relapsed MPM, supporting its off-label use, as a treatment option for patients with relapsed MPM.[1] Busacca et al, J Pathol 2012, 227(2), 200. [2] Busacca et al, Mol Cancer Res, 2021, 20(2) 379. Clinical trial information: NCT02139904.

PURPOSE:We previously demonstrated that the median survival of patients with poor prognosis non-small cell lung cancer (NSCLC) considered unfit for first-line platinum chemotherapy was <4 months. We evaluated whether VeriStrat could be used as a prognostic or predictive biomarker in this population. EXPERIMENTAL DESIGN:We conducted a randomised double-blind trial among patients with untreated advanced NSCLC considered unfit for platinum chemotherapy because of poor performance status (PS) or multiple comorbidities. All patients received active supportive care (ASC) and were treated with either oral erlotinib or placebo daily. Five hundred twenty-seven patients had plasma samples for VeriStrat classification: good (VeriStrat Good [VSG]) or poor (VeriStrat Poor [VSP]). Main end-point was overall survival. RESULTS:Fifty-five percent patients had VSG, and 83% had Eastern Cooperative Oncology Group (ECOG) 2-3 at baseline. VeriStrat was strongly associated with survival. Among patients managed with ASC only, the adjusted hazard ratio (HR) was 0.54 (p < 0.001) for VSG versus VSP. The association was consistent across patient factors: HR = 0.25 (p = 0.004) and HR = 0.56 (p < 0.001) for ECOG 0-1 and 2-3, respectively, HR = 0.49 (0070 < 0.001) for age≥75 years and HR = 0.59 (p = 0.007) for stage IV. Several ECOG 2-3 patients had long survival: 2-year survival was 8% for VSG patients who had ASC, compared with 0% for VSP. VeriStrat status did not predict benefit from erlotinib treatment because the HRs for erlotinib versus placebo were similar between VSG and VSP patients. CONCLUSIONS:VeriStrat was not a predictive marker for survival when considering first-line erlotinib for patients with NSCLC who had poor PS and were not recommended for platinum doublet therapies. However, VeriStrat was an independent prognostic marker of survival. It represents an objective measurement that could be considered alongside other patient factors to provide a more refined assessment of prognosis for this particular patient group. VSG patients could be selected for treatment trials because of better survival, while VSP patients can continue to be treated conservatively or offered trials of less toxic agents. TRIAL REGISTRATION ISRCTN NUMBER:ISRCTN02370070.

Purpose Retrospective studies indicate that expression of excision repair cross complementing group 1 (ERCC1) protein is associated with platinum resistance and survival in non-small-cell lung cancer (NSCLC). We conducted the first randomized trial, to our knowledge, to evaluate ERCC1 prospectively and to assess the superiority of nonplatinum therapy over platinum doublet therapy for ERCC1-positive NSCLC as well as noninferiority for ERCC1-negative NSCLC. Patients and Methods This trial had a marker-by-treatment interaction phase III design, with ERCC1 (8F1 antibody) status as a randomization stratification factor. Chemonaïve patients with NSCLC (stage IIIB and IV) were eligible. Patients with squamous histology were randomly assigned to cisplatin and gemcitabine or paclitaxel and gemcitabine; nonsquamous patients received cisplatin and pemetrexed or paclitaxel and pemetrexed. Primary end point was overall survival (OS). We also evaluated an antibody specific for XPF (clone 3F2). The target hazard ratio (HR) for patients with ERCC1-positive NSCLC was ≤ 0.78. Results Of patients, 648 were recruited (177 squamous, 471 nonsquamous). ERCC1-positive rates were 54.5% and 76.7% in nonsquamous and squamous patients, respectively, and the corresponding XPF-positive rates were 70.5% and 68.5%. Accrual stopped early in 2012 for squamous patients because OS for nonplatinum therapy was inferior to platinum therapy (median OS, 7.6 months [paclitaxel and gemcitabine] v 10.7 months [cisplatin and gemcitabine]; HR, 1.46; P = .02). Accrual for nonsquamous patients halted in 2013. Median OS was 8.0 (paclitaxel and pemetrexed) versus 9.6 (cisplatin and pemetrexed) months for ERCC1-positive patients (HR, 1.11; 95% CI, 0.85 to 1.44), and 10.3 (paclitaxel and pemetrexed) versus 11.6 (cisplatin and pemetrexed) months for ERCC1-negative patients (HR, 0.99; 95% CI, 0.73 to 1.33; interaction P = .64). OS HR was 1.09 (95% CI, 0.83 to 1.44) for XPF-positive patients, and 1.39 (95% CI, 0.90 to 2.15) for XPF-negative patients (interaction P = .35). Neither ERCC1 nor XPF were prognostic: among nonsquamous patients, OS HRs for positive versus negative were ERCC1, 1.11 ( P = .32), and XPF, 1.08 ( P = .55). Conclusion Superior outcomes were observed for patients with squamous histology who received platinum therapy compared with nonplatinum chemotherapy; however, selecting chemotherapy by using commercially available ERCC1 or XPF antibodies did not confer any extra survival benefit.

e14612 Background: Amphiregulin is an EGFR ligand which contributes to many of the hallmarks of cancer including angiogenesis and metastasis. Recent studies demonstrate that Amphiregulin promotes an immunosuppressive tumour microenvironment by modulating the expression of T regulatory cells. Although its expression has been described on a variety of immune cells and tumour cells, there is no published data on Amphiregulin expression by monocytes to date. Previous work by this group has demonstrated that monocytes from non-small cell lung cancer patients are phenotypically and genomically different compared to healthy volunteers. This study aims to further investigate the differing genomic profile of monocytes in non-small cell lung cancer. Methods: Gene expression profiling was performed on CD14++CD16- (classical) monocytes of newly diagnosed patients with Stage IV non-small cell lung cancer (n = 6) and age-matched healthy volunteers (n = 6) using Affymetrix Human U133 Plus 2.0 array. Validation of differentially expressed genes of interest (including Amphiregulin) at mRNA level was performed using real-time PCR with TaqMan gene expression assays on independent patient (n = 6) and healthy volunteer cohorts (n = 6). Differential protein expression of Amphiregulin from the monocytes of healthy volunteers (n = 3) and non-small cell lung cancer patients (n = 3) was validated using ELISA technology. Results: Amphiregulin is one of the most upregulated genes on classical monocytes in non-small cell lung cancer patients compared to healthy volunteers (p = 0.001). The increased expression of Amphiregulin was confirmed on qPCR validation in an independent cohort of non-small cell lung cancer patients (p < 0.05) as well as in the original group (p < 0.001). Amphiregulin was also found to be increased at protein level on CD14++CD16- monocytes of non-small cell lung cancer compared to healthy volunteers. (p = 0.37). Conclusions: Amphiregulin is overexpressed by classical monocytes in non-small cell lung cancer and this over-expression is confirmed at mRNA and protein level. The production of Amphiregulin has not been ascribed to human monocytes in non-small cell lung cancer making its overexpression in this study a unique finding.

IMPORTANCE:Preclinical studies show that arginine deprivation is synthetically lethal in argininosuccinate synthetase 1 (ASS1)-negative cancers, including mesothelioma. The role of the arginine-lowering agent pegylated arginine deiminase (ADI-PEG20) has not been evaluated in a randomized and biomarker-driven study among patients with cancer. OBJECTIVE:To assess the clinical impact of arginine depletion in patients with ASS1-deficient malignant pleural mesothelioma. DESIGN, SETTING, AND PARTICIPANTS:A multicenter phase 2 randomized clinical trial, the Arginine Deiminase and Mesothelioma (ADAM) study, was conducted between March 2, 2011, and May 21, 2013, at 8 academic cancer centers. Immunohistochemical screening of 201 patients (2011-2013) identified 68 with advanced ASS1-deficient malignant pleural mesothelioma. INTERVENTIONS:Randomization 2:1 to arginine deprivation (ADI-PEG20, 36.8 mg/m2, weekly intramuscular) plus best supportive care (BSC) or BSC alone. MAIN OUTCOMES AND MEASURES:The primary end point was progression-free survival (PFS) assessed by modified Response Evaluation Criteria in Solid Tumors (RECIST) (target hazard ratio, 0.60). Secondary end points were overall survival (OS), tumor response rate, safety, and quality of life, analyzed by intention to treat. We measured plasma arginine and citrulline levels, anti-ADI-PEG20 antibody titer, ASS1 methylation status, and metabolic response by 18F-fluorodeoxyglucose positron-emission tomography. RESULTS:Median (range) follow-up in 68 adults (median [range] age, 66 [48-83] years; 19% female) was 38 (2.5-39) months. The PFS hazard ratio was 0.56 (95% CI, 0.33-0.96), with a median of 3.2 months in the ADI-PEG20 group vs 2.0 months in the BSC group (P = .03) (absolute risk, 18% vs 0% at 6 months). Best response at 4 months (modified RECIST) was stable disease: 12 of 23 (52%) in the ADI-PEG20 group vs 2 of 9 (22%) in the BSC group (P = .23). The OS curves crossed, so life expectancy was used: 15.7 months in the ADI-PEG20 group vs 12.1 months in the BSC group (difference of 3.6 [95% CI, -1.0 to 8.1] months; P = .13). The incidence of symptomatic adverse events of grade at least 3 was 11 of 44 (25%) in the ADI-PEG20 group vs 4 of 24 (17%) in the BSC group (P = .43), the most common being immune related, nonfebrile neutropenia, gastrointestinal events, and fatigue. Differential ASS1 gene-body methylation correlated with ASS1 immunohistochemistry, and longer arginine deprivation correlated with improved PFS. CONCLUSIONS AND RELEVANCE:In this trial, arginine deprivation with ADI-PEG20 improved PFS in patients with ASS1-deficient mesothelioma. Targeting arginine is safe and warrants further clinical investigation in arginine-dependent cancers. TRIAL REGISTRATION:clinicaltrials.gov Identifier: NCT01279967.

Aim: Increased expression of excision repair cross-complementing group 1 (ERCC1) protein might be associated with platinum resistance in NSCLC, based on retrospective and meta-analyses findings. We conducted the first ever trial to prospectively evaluate ERCC1 as a predictive marker: to assess whether non-platinum therapy is superior to platinum for ERCC1 + ve, and non-inferior for ERCC1-ve tumors.

ABSTRACT Background Despite the recommendation to treat advanced NSCLC with platinum-based chemotherapy, the majority of these pts receive only active supportive care (ASC) because of poor performance or multiple co-morbidities. We previously showed that erlotinib significantly improved PFS in such patients, but with an enhanced OS and PFS effect in female pts. Here, we report mature OS data including the association of first-cycle rash (FCR). Methods 670 pts with stage IIIB/IV NSCLC (ECOG PS 2/3 or PS 0/1 with multiple co-morbidities unsuitable for chemotherapy) were randomized to receive placebo (n = 320) or erlotinib 150 mg/day (n = 350) and ASC until disease progression/toxicity. OS, PFS, adverse events, and QoL were examined. Pre-specified subgroup analyses included erlotinib-rash ≤28 days of starting treatment (FCR), and EGFR where available. Results Baseline characteristics were well balanced in these predominantly elderly NSCLC pts (median = 77 yrs, range 42-91). Among all pts, the hazard ratios (HRs) were 0.94 (95% CI 0.81-1.10, P = 0.46) for OS. Pts receiving erlotinib had a better PFS, HR = 0.83 (95% CI 0.71-0.97, P = 0.02). 59% of pts on erlotinib developed FCR. FCR was the only independent factor predictive of OS (HR = 0.17; P = 0.02) in a multivariate analysis containing rash, age, gender, histology, ECOG score and stage. There was a benefit for both OS (HR = 0.76, 95% CI 0.63-0.92, P = 0.01) and PFS (HR = 0.66, CI 0.54-0.80, P Conclusions Erlotinib prolongs PFS and OS in patients with advanced NSCLC considered unsuitable for chemotherapy, but only if they develop FCR. Disclosure All authors have declared no conflicts of interest.

PURPOSE Since treatment efficacy of cisplatin- or carboplatin-based chemotherapy in the first-line treatment of small-cell lung cancer (SCLC) remains contentious, a meta-analysis of individual patient data was performed to compare the two treatments. PATIENTS AND METHODS A systematic review identified randomized trials comparing cisplatin with carboplatin in the first-line treatment of SCLC. Individual patient data were obtained from coordinating centers of all eligible trials. The primary end point was overall survival (OS). All statistical analyses were stratified by trial. Secondary end points were progression-free survival (PFS), objective response rate (ORR), and treatment toxicity. OS and PFS curves were compared by using the log-rank test. ORR was compared by using the Mantel-Haenszel test. RESULTS Four eligible trials with 663 patients (328 assigned to cisplatin and 335 to carboplatin) were included in the analysis. Median OS was 9.6 months for cisplatin and 9.4 months for carboplatin (hazard ratio [HR], 1.08; 95% CI, 0.92 to 1.27; P = .37). There was no evidence of treatment difference between the cisplatin and carboplatin arms according to sex, stage, performance status, or age. Median PFS was 5.5 and 5.3 months for cisplatin and carboplatin, respectively (HR, 1.10; 95% CI, 0.94 to 1.29; P = .25). ORR was 67.1% and 66.0%, respectively (relative risk, 0.98; 95% CI, 0.84 to 1.16; P = .83). Toxicity profile was significantly different for each of the arms: hematologic toxicity was higher with carboplatin, and nonhematologic toxicity was higher with cisplatin. CONCLUSION Our meta-analysis of individual patient data suggests no differences in efficacy between cisplatin and carboplatin in the first-line treatment of SCLC, but there are differences in the toxicity profile.

Background: There is currently no early predictive marker of survival for patients receiving chemotherapy for malignant pleural mesothelioma (MPM). Tumour response may be predictive for overall survival (OS), though this has not been explored. We have thus undertaken a combined-analysis of OS, from a 42 day landmark, of 526 patients receiving systemic therapy for MPM. We also validate published progression-free survival rates (PFSRs) and a progression-free survival (PFS) prognostic-index model.Methods: Analyses included nine MPM clinical trials incorporating six European Organisation for Research and Treatment of Cancer (EORTC) studies. Analysis of OS from landmark (from day 42 post-treatment) was considered regarding tumour response. PFSR analysis data included six non-EORTC MPM clinical trials. Prognostic index validation was performed on one non-EORTC data-set, with available survival data.Results: Median OS, from landmark, of patients with partial response (PR) was 12.8 months, stable disease (SD), 9.4 months and progressive disease (PD), 3.4 months. Both PR and SD were associated with longer OS from landmark compared with disease progression (both p < 0.0001). PFSRs for platinum-based combination therapies were consistent with published significant clinical activity ranges. Effective separation between PFS and OS curves provided a validation of the EORTC prognostic model, based on histology, stage and performance status.Conclusion: Response to chemotherapy is associated with significantly longer OS from landmark in patients with MPM. (C) 2012 Elsevier Ltd. All rights reserved.

7022 Background: Selecting carboplatin or cisplatin in the treatment of poor prognosis and/or extensive stage SCLC remains controversial. We performed an IPD meta-analysis from published randomized trials. Methods: In June 2009, a systematic review was performed to identify randomized phase III trials comparing cisplatin- vs. carboplatin-based CT in the first-line treatment of SCLC. Primary endpoint of the meta-analysis was overall survival (OS). IPD (baseline characteristics, treatment details, toxicity and outcome measures) were obtained via collaborating with all groups identified by the systematic review. All statistical analyses were stratified by trials. OS and progression-free survival (PFS) curves were compared using the log-rank test. Response rate was compared using the Mantel-Haenszel test. Results: All four eligible trials were included in the analysis with a total of 663 patients (329 cisplatin, 334 carboplatin). Baseline characteristics were well balanced between arms. Median age was 67 years (27-86). Performance status (PS) was 0-1 in 72%, 2-3 in 28%; 68% of patients had extensive stage. With 589 deaths recorded (89%), median OS was 9.5 months with cisplatin, and 9.5 months with carboplatin (hazard ratio [HR] carboplatin vs. cisplatin 1.03, 95% confidence interval [CI] 0.88–1.22; p = 0.69). There was no evidence of treatment difference between cisplatin and carboplatin according to different gender, stage, PS or age. Response rate was 67.5% and 65.6%, with cisplatin and carboplatin, respectively (relative risk 0.96, 95% CI 0.82-1.13; p = 0.66). Median PFS was 5.3 and 5.5 months, for cisplatin and carboplatin, respectively (HR 1.01, 95% CI 0.86–1.19; p = 0.90). Toxicity profile was significantly different: hematological toxicity was higher with carboplatin, and non-hematologic toxicity higher with cisplatin. Conclusions: COCIS is the first IPD meta-analysis in the treatment of poor prognosis and/or extensive stage SCLC. We found no survival difference between cisplatin- and carboplatin-based CT but a different toxicity profile was reported.

BACKGROUND Some non-small-cell lung cancer (NSCLC) surgical series have indicated that the positive prognostic effect of female sex is limited to patients with adenocarcinoma. We carried out a retrospective analysis to investigate the role of sex and histology on efficacy, toxicity, and dose delivery after chemotherapy. PATIENT AND METHODS Individual patient data were pooled from five randomized, phase III, advanced NSCLC chemotherapy trials. Primary outcomes were response rate, overall survival (OS), toxicity, and dose delivery. A secondary analysis examined survival by sex in histological subgroups. RESULTS Of 2349 patients, 34% were women. Women had a higher response rate to chemotherapy (42% versus 40%, P = 0.01) and longer survival than men (median OS 9.6 versus 8.6 months, P = 0.002). The difference in OS remained after adjusting for age, stage, performance status, and histology (hazard ratio 0.83, 95% confidence interval 0.74-0.92, P = 0.0005). Upon further examination, longer survival in women was only seen in patients with adenocarcinoma (test for interaction P = 0.006). There were no differences in hematological toxicity or transfusions. Women experienced more grade 3-4 emesis than men (P < 0.0001) and more dose delays (P = 0.02) or dose reductions (P < 0.0001). CONCLUSION The positive prognostic effect among women is confirmed in patients receiving platinum-based chemotherapy but appears confined to those with adenocarcinoma histology.

7504 Background: Poor performance status (PS) NSCLC is an important, under-studied group of lung cancer patients, affecting 30-40% of patients. We determined the role of erlotinib plus best supportive care (BSC) in chemo-naive, poor PS advanced NSCLC patients. Methods: Chemo-naive NSCLC patients (ECOG PS 2/3 or PS 0/1 unfit for platinum chemotherapy; stage IIIB/IV) were randomized to erlotinib (150 mg/d) plus BSC (n = 350) or placebo plus BSC (n = 320). Endpoints were overall survival (OS), progression-free survival (PFS), response rate, and toxicity. The trial had 90% power to detect an increase in 1-year OS from 10% to 17.5%. Results: Between 2005 and 2009, 670 patients were randomised to either erlotinib (350) or placebo (320), from 78 UK centres. Median age was 77 yrs (range 42-91); 61% male; 38% adenocarcinoma, 39% squamous histology; the % with ECOG PS score of 0/1, 2 and 3 were 16%, 55% and 29% respectively; 35% stage IIIB and 65% stage IV disease. Baseline patient characteristics were balanced bet...

PURPOSE:Cancers rely on angiogenesis for their growth and dissemination. We hypothesized that thalidomide, an oral antiangiogenic agent, when combined with chemotherapy, and as maintenance treatment, would improve survival in patients with advanced non-small-cell lung cancer (NSCLC).PATIENTS AND METHODS:Seven hundred twenty-two patients were randomly assigned to receive placebo or thalidomide capsules 100 to 200 mg daily for up to 2 years. All patients received gemcitabine and carboplatin every 3 weeks for up to four cycles. End points were overall survival (OS), progression-free survival (PFS), response rate, grade 3/4 toxicity, and quality of life (QoL).RESULTS:The median OS rates were 8.9 months (placebo) and 8.5 months (thalidomide). The hazard ratio (HR) was 1.13 (95% CI, 0.97 to 1.32; P = .12). The 2-year survival rate was 16% and 12% in the placebo and thalidomide arms, respectively. The PFS results were consistent with those for OS. The risk of having a thrombotic event was increased by 74% in the thalidomide group: HR of 1.74 (95% CI, 1.20 to 2.52; P = .003). There were no differences in hematologic toxicities, but a slight excess of rash and neuropathy in the thalidomide group. QoL scores were similar but thalidomide was associated with less insomnia, and more constipation and peripheral neuropathy. In a retrospective analysis, patients with nonsquamous histology in the thalidomide group had a poorer survival: 2-year risk difference of 10% (95% CI, 4% to 16%; P < .001).CONCLUSION:In this large trial of patients with NSCLC, thalidomide in combination with chemotherapy did not improve survival overall, but increased the risk of thrombotic events. Unexpectedly, survival was significantly worse in patients with nonsquamous histology.

BackgroundFemale sex is a modest positive prognostic factor in non-small cell lung cancer (NSCLC), in both early and late-stage disease. Some surgical series have suggested this benefit is limited to adenocarcinoma patients. We performed a retrospective analysis to investigate the role of sex and histology on efficacy, toxicity and dose delivery.MethodsIndividual patient data from five randomized phase III NSCLC trials, investigating platinum-based first-line chemotherapy regimens, were pooled in a single database and analyzed by patient sex. Primary outcomes were tumor response rate, overall survival (OS), lung cancer specific survival (LCSS), hematological and non-hematological toxicity and dose delivery. A secondary analysis examined survival by sex in histological subgroups. ResultsOf 2349 patients included, 793 were women (34%). Baseline demographics are shown in the Table. Overall, women had a higher response rate to chemotherapy (44% versus 39%, p=0.007), and longer survival than men (median OS 9.6 versus 8.6 months, one-year survival 41% versus 35%, HR 0.86 [95% CI 0.78-0.95], p=0.002; median LCSS 9.7 versus 8.7 months, HR 0.86 [95% CI 0.78-0.95], p=0.002). In multivariate analysis female sex remained associated with longer OS (HR 0.83, 95% CI 0.74 ? 0.92, p=0.0005). In subgroup analysis, longer survival was observed in women with adenocarcinoma but not in women with non-adenocarcinoma histology (test of interaction p=0.006, see Figure).There were no differences in any hematological toxicity, or rates of blood or platelet transfusions, between sexes. Women experienced more grade 3 or 4 nausea and vomiting than men (10% versus 5%, p=0.0002). Women experienced more dose delays (39% versus 32%, p=0.02) and dose reductions (32% versus 23%, p<0.0001) than men. ConclusionsThe positive prognostic effect of female sex is confirmed in this large case series of patients receiving modern platinum-based chemotherapy, but we demonstrate that this benefit is confined to women with adenocarcinoma histology.