Despite the advent of molecular-targeted therapies, platinum-based

Transcription

1 ORIGINAL ARTICLE Histology-Related Associations of ERCC1, RRM1, and TS Biomarkers in Patients with Non Small-Cell Lung Cancer Implications for Therapy Martin K.H. Maus, MD,* Philip C. Mack, MD, Stephanie H. Astrow, PhD, Craig L. Stephens, Gary D. Zeger, MD, Peter P. Grimminger, MD,* Jack H. Hsiang, Eric Huang, PhD, Tianhong Li, PhD, Primo N. Lara MD, Kathleen D. Danenberg, and David R. Gandara, MD Introduction: On the basis of the results of recent clinical trials, histology-based decision-making for therapy of non small-cell lung cancer has been advocated. We hypothesized associations of the biomarkers excision repair cross-complementing 1 (ERCC1), ribonucleotide reductase M1 (RRM1), and thymidylate synthase (TS) with histology as a contributing factor to reported differences in chemotherapy outcomes between squamous cell carcinoma (SCCA) and adenocarcinoma (AC) subtypes. Here, we report analysis of the Response Genetics Inc., database and implications for histologybased therapy. Methods: RNA from microdissected formalin-fixed paraffin-embedded tumors was extracted and analyzed as previously described. Specimens from 2540 individual non small-cell lung cancer patients were analyzed for one or more biomarkers, of which 1457 were categorized as AC or SCCA. Results: For each biomarker, gene expression was lower in AC compared with SCCA (<0.001), although there was a wide range between individual patients. Gene expression was higher in men versus women: ERCC1: 2.51 versus 2.22 (p = 0.005); RRM1: 1.41 versus 1.24 (p = 0.004); TS: 3.23 versus 2.83 (p < 0.001). However, SCCA was more frequent in men versus women (30%/19%; p < 0.001). When AC and SCCA were assessed separately, the statistical significance between gene expression and sex was lost (in SCCA: ERCC1, p = 0.14; RRM1, p = 0.26; TS, p = 0.11). Conclusions: This analysis represents the largest data set for gene expression of these biomarkers reported so far. Significant histologyrelated associations for ERCC1, RRM1, and TS are seen. However, *Department of General, Visceral and Tumor Surgery, University of Cologne, Germany; Response Genetics Inc., Los Angeles, California; Department of Internal Medicine, University of California Davis Comprehensive Cancer Center, University of California, Davis, Sacramento, California; Department of Pathology, Keck School of Medicine, University of Southern California, Los Angeles, California. Disclosure: Martin K.H. Maus and David R. Gandara are consultants of Response Genetics Inc.; Stephanie Astrow, Craig Stephens, Gary Zeger and Jack Hsiang are employees of Response Genetics. The other authors declare no conflict of interest. Address for correspondence: Martin K.H. Maus, MD, University of Cologne, Department of General, Visceral and Tumor Surgery, Kerpener Strasse 62, Cologne, Germany, mausman@gmx.at Copyright 2013 by the International Association for the Study of Lung Cancer ISSN: /13/ marked heterogeneity exists in individual patient tumor expression levels. Randomized phase III trials assessing the predictive value of these chemotherapy-related biomarkers are warranted. Key Words: Biomarkers in NSCLC, Personalized treatment, ERCC1, RRM1, TS, Histology, Sex. (J Thorac Oncol. 2013;8: ) Despite the advent of molecular-targeted therapies, platinum-based chemotherapy remains a standard of care for most patients with advanced stage non small-cell lung cancer (NSCLC). 1 3 Typical chemotherapy doublets include cisplatin or carboplatin matched with an agent from another drug class, chosen from a group, including paclitaxel, docetaxel, vinorelbine, gemcitabine, or pemetrexed. Until recently, the selection of which drug to match with a platinum compound in individual patients with NSCLC has been largely empiric, based on physician experience or differences in toxicity profiles. Within the last 5 years, however, the decision-making process has been altered to incorporate clinical trial data and regulatory considerations regarding histology-associated activity of pemetrexed. Specifically, pemetrexed use is now largely restricted to nonsquamous histologic subtypes, based on retrospective analyses showing inferior results in squamous cell carcinoma (SCCA) in recent phase III trials. 4 6 Proposed predictive chemotherapy biomarkers in NSCLC include the DNA repair gene excision repair cross-complementing 1 (ERCC1) for platinum compounds, ribonucleotide reductase M1 (RRM1) for platinum or gemcitabine, and thymidylate synthase (TS) for pemetrexed. We hypothesized an association between histology and gene expression levels of these biomarkers as a possible explanation for reported differences in patient outcomes from platinum-based doublet chemotherapy in SCCA versus adenocarcinoma (AC) subtypes of NSCLC. Here, we report analysis of a large tumor tissue database for these chemotherapy-related biomarkers, demonstrating that as a group, patients with AC are significantly more likely to exhibit lower expression levels than do patients with SCCA, a potential explanation for enhanced platinum-based chemotherapy efficacy, pemetrexed in particular, in NSCLC of the AC subtype. 582 Journal of Thoracic Oncology Volume 8, Number 5, May 2013

2 Journal of Thoracic Oncology Volume 8, Number 5, May 2013 Biomarkers in NSCLC PATIENTS AND METHODS Formalin-fixed paraffin-embedded (FFPE) tumor samples from NSCLC patients available from the Response Genetics Inc. (RGI) database were used for this analysis. Because this database came from a commercial genetic testing center (Response Genetics Inc.) and specimens were submitted by physicians for Clinical Laboratory Improvement Amendments certified laboratory testing, an institutional review board protocol was not required. All samples were de-identified according to Health Insurance Portability and Accountability Act (HIPAA) laboratory standards. A hematoxylin and eosin stained section of all formalinfixed paraffin-embedded specimens was evaluated by a boardcertified pathologist for tumor content. Histologic subtype was retained as delineated from the submitting center, classified as AC, SCCA, or not otherwise specified or other (NOS). Adjacent sections of the tumor were sectioned at 10 μm and stained with nuclear fast red to aid visualization for microdissection. After microdissection of tumor cells from nuclear fast red stained slides, RNA was extracted by the RGI-patented method. After isolation and lysis of the tumor cells, RNA and DNA were isolated separately from the specimen. The RNA was then reverse-transcribed to cdna followed by reversetranscriptase polymerase chain reaction (PCR). For the evaluation of expression levels of ERCC1, RRM1, and TS, cdna was analyzed by reverse-transcriptase PCR using specific primers and probes. The results were obtained as a ratio of PCR fluorescent signals of the genes in reference to beta actin. Established cutoff levels for drug sensitivity were for platinum (ERCC1 < 1.7), gemcitabine (RRM1 < 0.97), and pemetrexed (TS < 2.33). Statistical Analysis Gene expression levels of ERCC1, RRM1, and TS were log-transformed to render them normally distributed. The differences in mrna level of TS, ERCC1, epidermal growth factor receptor (EGFR), and RRM1 by sex and histology status were compared using the two-sample t test. Analysis of variance was used to test the difference in TS mrna level across the different groups. All tests were two-sided with a significance level of The primary comparisons of interest were between AC and SCCA. RESULTS Patient Characteristics Specimens from 2540 individual NSCLC patients were analyzed for one or more biomarkers. By histologic subtype, 1136 patients (45%) were categorized as AC, 457 patients (18%) as SCCA, and 947 patients (37%) were categorized as NOS. Table 1 displays patient characteristics available within the RGI database. The median age was 67 years (range, years). The male-to-female ratio was 48%:52% (1176/1274), respectively. No information regarding treatment regimens or outcomes were available from this data set. TABLE 1. Patient Demographics Histology Adenocarcinoma 1136 (45%) Squamous cell 457 (18%) carcinoma Not otherwise specified 947 (37%) Age, yr Median (range) 67 (18 96) Sex (n = 2071) Men/women 48%/52% Association of Histology with Gene Expression The focus of this analysis is comparison of gene expression between the two major histologic subtypes of NSCLC, AC versus SCCA. ERCC1, RRM1, and TS expression levels were associated with NSCLC histologic subtype and were significantly different between AC and SCCA with p value less than 0.001, although the range of values between individual patients was wide for each biomarker. Expression levels for the NOS subset were of intermediate value for each gene, ERCC1: median 2.07 ( ), RRM1: 1.35 ( ), TS: 3.06 ( ) and are not further addressed here. Comparisons of ERCC1 levels between AC (n = 1002) and SCCA (n = 436) are as follows: for AC the median expression level was 1.82 ( ), whereas for SCCA, the median was 2.48 ( ) and p value was less than Similarly, median gene expression levels of RRM1 and TS levels were significantly lower in AC than in SCCA: for RRM1: 1.06 in AC and 2.41 in SCCA (p < 0.001), and for TS: 2.5 in AC and 4.1 in SCCA (p < 0.001; Table 2). Again, the range of expression levels between individual patients was wide in all categories. The proportion of patients with median expression levels of ERCC1, RRM1, and TS below the reference cutoff levels for drug sensitivity was significantly different between AC and SCCA for each biomarker (p = 0.001; Fig. 1). For example, ERCC1 levels were categorized as low (below the reference level of 1.7) in 46% of AC, compared with 30.7% of SCCA with p value of Similar differences were observed by histology for RRM1 and TS which is as follows: RRM1 expression less than 0.97 was seen in 42% of AC compared with only 13% of SCCA; TS expression less than 2.33 was present in 45% of AC compared with 25% of SCCA. Association of Sex with Gene Expression Differences in gene expression levels for ERCC1, RRM1, and TS were also observed by sex, with expression significantly higher in male than in female patients (p < 0.001) for all three biomarkers (Table 2). However, SCCA histology was more frequent in men versus women in this series (30%/19%). When AC and SCCA were assessed separately, statistical significance between gene expression and sex was lost (i.e., in SCCA: ERCC1, p = 0.14; RRM1, p = 0.26; TS, p = 0.11). Copyright 2013 by the International Association for the Study of Lung Cancer 583

3 Maus et al. Journal of Thoracic Oncology Volume 8, Number 5, May 2013 TABLE 2. Biomarker Result by Histology and Sex Biomarker NSCLC-Total, n = 2540 NSCLC-SCCA, n = 457 NSCLC-AC, n = 1136 SCCA vs. AC, p NSCLC-Total, n = 2071 Male, n = 992 Female, n = 1079 Male vs. Female, p ERCC1 n Median < Range RRM1 n Median < <0.001 Range TS n Median < <0.001 Range Mann Whitney test Mann Whitney test NSCLC, non small-cell lung cancer; SCCA, squamous cell carcinoma; AC, adenocarcinoma; ERCC1, excision repair cross-complementing 1; RRM1, ribonucleotide reductase M1; TS, thymidylate synthase. DISCUSSION Aside from the relatively small subset of patients with cancers harboring oncogene drivers such as EGFR mutation or anaplastic lymphoma kinase (ALK) translocation, platinum-based chemotherapy remains the standard of care for advanced stage NSCLC. Recently, histology-based therapeutic decision-making for chemotherapy of advanced NSCLC has been advocated, primarily related to the chemotherapeutic agent pemetrexed. 4 8 In a large randomized phase III trial comparing cisplatin gemcitabine with cisplatin pemetrexed, overall survival results were similar. However, when the results were analyzed by histologic subtype, patients with nonsquamous histology, the majority of which had AC, fared better with cisplatin pemetrexed with p value of Conversely, survival in those with SCCA favored the cisplatin gemcitabine arm with p value of Retrospective analyses of other pemetrexed-related trials in NSCLC have shown mixed results in regard to demonstrating differences in efficacy by histology. 6,9,10 Whether a molecular basis underlies histology-specific chemotherapy selection has remained unclear. Nevertheless, both histology-based selection of chemotherapy and the prognostic and predictive value of chemotherapy-related biomarkers are being described in NSCLC guidelines. 11,12 ERCC1 (<1.7 forpla num) RRM1 (<0.97 for gemcitabine) TS ( <2.33 for pemetrexed) NSCLC-Total 43.4% NSCLC-Total 39.6% NSCLC-Total 41.3% NSCLC-AC 46% NSCLC-AC 42.2% NSCLC-AC 45.7% NSCLC-SCCA 30.7% NSCLC-SCCA 13% NSCLC-SCCA 25.9% FIGURE 1. Heterogeneity of excision repair cross-complementing 1, ribonucleotide reductase M1, and thymidylate synthase mrna expression levels in non small-cell lung cancer. 584 Copyright 2013 by the International Association for the Study of Lung Cancer

4 Journal of Thoracic Oncology Volume 8, Number 5, May 2013 Biomarkers in NSCLC Of reported chemotherapy-related biomarkers for NSCLC, ERCC1, RRM1, and TS are most prominent. 12,13 Here, we report the association of gene expression levels for these biomarkers with NSCLC histologic subtype as a potential contributing factor to differences in chemotherapeutic outcome between patients with AC and SCCA. These data may support the hypothesis of a mechanism-related basis for the current algorithm of histology-based treatment selection in NSCLC. Resistance to platinum agents is multifactorial and comprises tumor-related and host-related parameters. 14 The multitude of potential resistance mechanisms includes increased metabolic inactivation, decreased intracellular accumulation, and increased DNA repair capacity. 15 Recently, Friboulet et al. 16 reported that different isoforms of ERCC1 may add to the complexity of interpreting ERCC1 protein expression because DNA repair capacity may vary by isoform. Despite the complexity of platinum compound disposition and the multitude of potential resistance mechanisms, nucleotide excision repair gene stands out by playing a central role in determining platinum compound sensitivity. 15 By providing a rate-limiting step in removal of platinum-dna adducts, ERCC1 shifts the balance between DNA damage and DNA repair, which determines cancer cell death or survival following challenge by a platinum compound. Abnormal expression of ERCC1, most often determined by mrna or protein expression, has been reported to have both prognostic significance and predictive value for platinum-based chemotherapy in patients with NSCLC It is now 10 years since the first evidence linking mrna expression to cisplatin gemcitabine efficacy in advanced NSCLC was published. 21 Since then, a large body of data has emerged showing that although low levels of ERCC1 expression translate into an unfavorable prognosis, conversely, low ERCC1 levels are associated with improved efficacy of platinum-based chemotherapy. 17,21 In a similar fashion, expression levels of RRM1 and TS have been reported to predict efficacy of gemcitabine and pemetrexed, respectively. 7,13,22 A recent meta-analysis of RRM1 as a predictive biomarker for gemcitabine in more than 1000 NSCLC patients revealed both higher response rate (p < ) and improved survival in those patients with cancers harboring low RRM1 (p < 0.001). 23 Although the results of our study are consistent with published literature suggesting that SCCA of the lung is a more chemoresistant histologic subtype than AC, our study is limited by lack of clinical annotation for efficacy outcomes. It, therefore, remains hypothesis-generating and requires confirmation from ongoing clinical trials. 7,13,17,18,21,22,24 However, the tremendous range of gene expression levels within individual patient tumors in a given histolgoic subtype is striking for each of the three biomarkers. For example, ERCC1 levels in AC ranged from 0.04 to 24.79, suggesting that although AC as a group generally has lower levels than SCCA, the cancers of some AC patients are characterized by very high levels of nucleotide excision repair and are likely to be more resistant to platinum. Our findings may assist in explaining the mixed results of previous biomarker-related clinical trials, many of which were limited by small patient sample sizes, which were inadequate to account for histology or interpatient variation in gene expression. The standard procedure for biomarker clinical validation remains a prospectively designed biomarker-driven phase III trial, of which several variations exist, as recently reviewed by Redman et al. 25 Currently, several such trials are underway, evaluating ERCC1, RRM1, and TS, either singly or in combination. In the adjuvant chemotherapy arena, three trials of note prospectively evaluating one or more of these biomarkers are Tailored Adjuvant Post-Surgical Therapy in Early NSCLC (TASTE), Spanish Lung Cancer Adjuvant Trial (SCAT), and International Tailored Adjuvant Chemotherapy Trial (ITACA). In advanced NSCLC, the Elderly Patient Individualized Chemotherapy (EPIC) trial uses a biomarker-driven randomization to compare investigator choice of platinum doublet chemotherapy to treatment based on ERCC1, RRM1, and TS. Taken together, these studies should be sufficient to test the hypothesis of biomarker-selected chemotherapy. Of interest, our data show that AC has much lower gene expression of RRM1 than SCCA. Only 13% of SCCA demonstrates levels less than the reference cutoff of 0.97, consistent with relative resistance of SCCA of the lung to gemcitabine. Reconciling these data with the JMDB study of Scagliotti et al., 9 it is certainly possible that although gemcitabine is not an optimal drug for SCCA, it is nevertheless better than pemetrexed. Furthermore, others have made the same observation. Our analysis does not address potential chemotherapy-related biomarkers for other chemotherapy drug classes, such as antimicrotubule-directed taxanes and vinca alkaloids. Of note, a recent retrospective multitrial analysis by the Southwest Oncology Group showed no association of histology for any efficacy outcome in more than 800 patients treated with vincas or taxanes plus platinum combinations. 26 The associations between sex and mrna expression levels of ERCC1, RRM1, and TS described in our study must be interpreted within the context of associated histology. Although we found significantly higher expression levels in men for each biomarker, when controlled for histological subgroup, the statistical association for sex was lost, suggesting that, at least in part, women with NSCLC fare better with platinum-based chemotherapy because of the higher proportion of AC in women. Numerous earlier reports have showed that women with NSCLC have both a better prognosis and better chemotherapy-related outcomes but the impact of histology or other underlying factors remains to be clearly defined Finally, our findings need to be placed into context with regard to recent findings about the genomic complexity of NSCLC and inter- and intrapatient tumor heterogeneity. Histologyassociated differences in gene expression and frequency of mutations for AC and SCCA are already well recognized but have been recently emphasized by two studies using whole genome sequencing. 32,33 In comparison with most other cancer types, both AC and SCCA demonstrate marked genomic derangement, with a similar range of genomic complexity as reflected in number of mutations per megabase of DNA. Nevertheless, the frequency of oncogenic drivers, such as EGFR mutation and ALK fusion, differs dramatically between the two histologies, being found almost exclusively in AC. Of Copyright 2013 by the International Association for the Study of Lung Cancer 585

5 Maus et al. Journal of Thoracic Oncology Volume 8, Number 5, May 2013 interest, in the same data set that provided the basis for this study, we have shown that EGFR mutation associates with low ERCC1 expression, a plausible explanation for reports of high response rate to platinum-based chemotherapy in EGFRmutant NSCLC. 34 Similarly, we have demonstrated that cancers with ALK rearrangements, mostly present in AC and only sporadically in SCCA, demonstrate a significant correlation with low expression levels of TS, a mechanism-based rationale for pemetrexed activity in ALK-positive cancers In conclusion, our data demonstrate significant histology-related associations for ERCC1, RRM1, and TS mrna expression levels in a large data set. These data support the concept of decreased chemo-sensitivity of SCCA compared with AC for platinum-based chemotherapy. The predictive value of these biomarkers is being addressed in currently ongoing phase III biomarker-driven trials. ACKNOWLEDGEMENTS Supported by National Institutes of Health grant P30 CA93373 (University of California Davis Cancer Center); Response Genetics Inc.; Bonnie J. Addario Lung Cancer Foundation. REFERENCES 1. Rapp E, Pater JL, Willan A, et al. Chemotherapy can prolong survival in patients with advanced non-small-cell lung cancer report of a Canadian multicenter randomized trial. J Clin Oncol 1988;6: Cullen MH, Billingham LJ, Woodroffe CM, et al. Mitomycin, ifosfamide, and cisplatin in unresectable non-small-cell lung cancer: effects on survival and quality of life. J Clin Oncol 1999;17: Schiller JH, Harrington D, Belani CP, et al.; Eastern Cooperative Oncology Group. Comparison of four chemotherapy regimens for advanced nonsmall-cell lung cancer. N Engl J Med 2002;346: Scagliotti GV, Parikh P, von Pawel J, et al. Phase III study comparing cisplatin plus gemcitabine with cisplatin plus pemetrexed in chemotherapy-naive patients with advanced-stage non-small-cell lung cancer. J Clin Oncol 2008;26: Ciuleanu T, Brodowicz T, Zielinski C, et al. Maintenance pemetrexed plus best supportive care versus placebo plus best supportive care for nonsmall-cell lung cancer: a randomised, double-blind, phase 3 study. Lancet 2009;374: Hanna N, Shepherd FA, Fossella FV, et al. Randomized phase III trial of pemetrexed versus docetaxel in patients with non-small-cell lung cancer previously treated with chemotherapy. J Clin Oncol 2004;22: Ozasa H, Oguri T, Uemura T, et al. Significance of thymidylate synthase for resistance to pemetrexed in lung cancer. Cancer Sci 2010;101: Einhorn LH. First-line chemotherapy for non-small-cell lung cancer: is there a superior regimen based on histology? J Clin Oncol 2008;26: Scagliotti G, Hanna N, Fossella F, et al. The differential efficacy of pemetrexed according to NSCLC histology: a review of two Phase III studies. Oncologist 2009;14: Grønberg BH, Bremnes RM, Fløtten O, et al. Phase III study by the Norwegian lung cancer study group: pemetrexed plus carboplatin compared with gemcitabine plus carboplatin as first-line chemotherapy in advanced non-small-cell lung cancer. J Clin Oncol 2009;27: Ettinger DS, Akerley W, Bepler G, et al.; NCCN Non-Small Cell Lung Cancer Panel Members. Non-small cell lung cancer. J Natl Compr Canc Netw 2010;8: Febbo PG, Ladanyi M, Aldape KD, et al. NCCN Task Force report: Evaluating the clinical utility of tumor markers in oncology. J Natl Compr Canc Netw 2011;9(Suppl 5):S1 32; quiz S Aggarwal C, Somaiah N, Simon RS. Biomarkers with predictive and prognostic function in non-small cell lung cancer: ready for prime time. J Natl Compr Cancer Netw 2010;8: Gandara DR, Kawaguchi T, Crowley J, et al. Japanese-US common-arm analysis of paclitaxel plus carboplatin in advanced non-small-cell lung cancer: a model for assessing population-related pharmacogenomics. J Clin Oncol 2009;27: Chang A. Chemotherapy, chemoresistance and the changing treatment landscape for NSCLC. Lung Cancer 2011;71: Friboulet L, Thomale J, Olaussen K, et al. Functional characterization of ERCC1 isoforms in NSCLC. Cancer Research 2012;72(Suppl 1): abstr Cobo M, Isla D, Massuti B, et al. Customizing cisplatin based on quantitative excision repair cross-complementing 1 mrna expression: a phase III trial in non-small-cell lung cancer. J Clin Oncol 2007;25: Olaussen KA, Dunant A, Fouret P, et al.; IALT Bio Investigators. DNA repair by ERCC1 in non-small-cell lung cancer and cisplatin-based adjuvant chemotherapy. N Engl J Med 2006;355: Zheng Z, Chen T, Li X, et al: The DNA synthesis and repair genes RRM1 and ERCC1 in lung cancer. N Engl J Med 2007:356: Simon GR, Sharma S, Cantor A, Smith P, Bepler G. ERCC1 expression is a predictor of survival in resected patients with non-small cell lung cancer. Chest 2005;127: Lord RV, Brabender J, Gandara D, et al. Low ERCC1 expression correlates with prolonged survival after cisplatin plus gemcitabine chemotherapy in non-small cell lung cancer. Clin Cancer Res 2002;8: Bepler G, Kusmartseva I, Sharma S, et al. RRM1 modulated in vitro and in vivo efficacy of gemcitabine and platinum in non-small-cell lung cancer. J Clin Oncol 2006;24: Gong W, Zhang X, Wu J, et al. RRM1 expression and clinical outcome of gemcitabine-containing chemotherapy for advanced non-small-cell lung cancer: a meta-analysis. Lung Cancer 2012;75: Kotsakis A, Yousem S, Gadgeel M, Is Histological Subtype Significant in the Management of NSCLC. Open Lung Cancer J 2010;3, Redman MW, Crowley JJ, Herbst RS, Hirsch FR, Gandara DR. Design of a phase III clinical trial with prospective biomarker validation: SWOG S0819. Clin Cancer Res 2012;18: Chansky K, Mack PC, Crowley JJ, Lara PN, Hirsch FR, Franklin WA, Gandara DR. Chemotherapy outcomes by histological subtype of nonsmall cell lung cancer (NSCLC): analysis of the SWOG database for antimicrotubule-platinum therapy. J Thoracic Oncol 2009;4(Suppl 1). 27. Ferguson MK, Wang J, Hoffman PC, et al. Sex-associated differences in survival of patients undergoing resection for lung cancer. Ann Thorac Surg 2000;69:245 9; discussion Cerfolio RJ, Bryant AS, Scott E, et al. Women with pathologic stage I, II, and III non-small cell lung cancer have better survival than men. Chest 2006;130: Foeglé J, Hédelin G, Lebitasy MP, Purohit A, Velten M, Quoix E. Specific features of non-small cell lung cancer in women: a retrospective study of 1738 cases diagnosed in Bas-Rhin between 1982 and J Thorac Oncol 2007;2: Alexiou C, Onyeaka CV, Beggs D, et al. Do women live longer following lung resection for carcinoma? Eur J Cardiothorac Surg 2002;21: Chang JW, Asamura H, Kawachi R, Watanabe S. Gender difference in survival of resected non-small cell lung cancer: histology-related phenomenon? J Thorac Cardiovasc Surg 2009;137: Hammerman PS, et al. Comprehensive genomic characterization of squamous cell lung cancers. Nature 2012;489: Imielinski M, Berger AH, Hammerman PS, et al. Mapping the hallmarks of lung adenocarcinoma with massively parallel sequencing. Cell 2012;150: Gandara DR, Grimminger P, Mack PC, et al. Association of epidermal growth factor receptor activating mutations with low ERCC1 gene expression in non-small cell lung cancer. J Thorac Oncol 2010;5: Gandara DR, Huang E, Desai S, et al. Thymidylate synthase (TS) gene expression in patients with ALK positive non-small cell lung cancer (NSCLC): Implications for therapy. J Clin Oncol 2012;30(suppl;abstr 7582). 36. Camidge DR, Kono SA, Lu X, Okuyama S, Baron AE, Oton AB, Davies AM, Varella-Garcia M, Franklin WA, Doebele RC. Anaplastic Lymphoma Kinase (ALK) gene rearrangements in non-small cell lung cancer are associated with prolonged progression free survival on pemetrexed. J Thoracic Oncol 2011; 6(4): Shaw AT, Kim D-W, Nakagawa K, et al. Phase III study of crizotinib vs pemetrexed or docetaxel chemotherapy in patients with advanced ALK-positive NSCLC (PROFILE 1007). ESMO 37th Annual Meeting 2012;Abstract Copyright 2013 by the International Association for the Study of Lung Cancer