Gender Differences in Non Small-Cell Lung Cancer Survival: An Analysis of 4,618 Patients Diagnosed Between 1997 and 2002

Transcription

1 ORIGINAL ARTICLES: Gender Differences in Non Small-Cell Lung Cancer Survival: An Analysis of 4,618 Patients Diagnosed Between 1997 and 2002 Antonio L. Visbal, MD, Brent A. Williams, MS, Francis C. Nichols III, MD, Randolph S. Marks, MD, James R. Jett, MD, Marie-Christine Aubry, MD, Eric S. Edell, MD, Jason A. Wampfler, BS, Julian R. Molina, MD, PhD, and Ping Yang, MD, PhD Departments of Health Sciences Research and Oncology, Divisions of General Thoracic Surgery, Pulmonary and Critical Care Medicine, and Anatomic Pathology, and Cancer Center, Mayo Clinic, Rochester, Minnesota; and Hospital Antituberculoso Santa Clara, Bogota, Colombia Background. Gender has been reported as a predictor for nonsmall cell lung cancer (NSCLC) survival. Most of the reports are limited to selected groups of patients. The magnitude of gender effect on NSCLC survival across disease stage, tumor histology, and therapies needs to be further characterized. Methods. A cohort of 4,618 patients diagnosed with NSCLC was prospectively enrolled and actively followed. Vital status of each patient was verified through multiple complementary sources. Cox proportional hazards models were developed to compare postdiagnosis survival between genders adjusting for age at diagnosis, tumor histology and grade, stage, pack-years smoked, and treatment received (resection, radiation, or chemotherapy). Results. There were 2,724 men (59%) and 1,894 women (41%), with a median age at diagnosis of 68 years in men and 66 in women (p < 0.01). More men smoked and were heavier smokers than women. Adenocarcinoma was the most frequent histology in both genders. No difference was found in stage and treatment between genders. The estimated survival in men was 51% (95% CI: 49%, 53%) and 15% (95% CI: 12%, 17%) at one and five years, respectively, and in women was 60% (95% CI: 58%, 62%) and 19% (95% CI: 16%, 22%). Men were at a significantly increased risk of mortality compared to women following a diagnosis of NSCLC (adjusted relative risk: 1.20, 95% CI: 1.11, 1.30), particularly for patients with stage III/IV disease or adenocarcinoma. Conclusions. Male gender is confirmed to be an independent unfavorable prognostic indicator for NSCLC survival. (Ann Thorac Surg 2004;78:209 15) 2004 by The Society of Thoracic Surgeons Primary lung cancer has been reported to be the number one cause of cancer death in both genders in the U.S. [1]. Remarkable changes in histology patterns and gender distribution of lung cancer patients have occurred in the past. Adenocarcinoma has become the most frequent histologic subtype in both genders, and women present with adenocarcinoma in a higher proportion than men do [2]. The ratio of men to women with lung cancer was reported as 6.8:1 in 1957 to 1960, and 1.8:1 in 1977 to 1980 [3]. Extrapolation of the incidence curves suggests that the gender ratio will reach one in the near future [1, 4], and indeed the ratio is estimated to be 1.6:1 in the United States in 2000 [5, 6]. Lung cancer death rate in women rose from 5/100,000 in 1960 to 30/100,000 in 1990, surpassing breast cancer as the main cause of cancer mortality in women [7]. The rise in lung cancer mortality in women. appears to parallel the increased Accepted for publication Nov 25, Address reprint requests to Dr Yang, Department of Health Sciences Research and Cancer Center at Mayo Clinic, 200 First St SW, Rochester, MN 55905; yang.ping@mayo.edu. prevalence of smoking [8], and previous studies have suggested an increased risk for developing lung cancer in women at any level of smoking compared to males [9 12]. The impact of gender on survival needs to be clarified in light of the recent epidemiological changes of this disease in the population. Previous studies showed that female gender was a favorable prognostic factor for nonsmall-cell lung cancer (NSCLC) in a surgical series of stage I [13, 14] and IIB [15] patients. Male gender was found to be detrimental for surgical patients more than 60 years old with NSCLC [16]. Limited number of patients and selected groups and lack of details regarding tobacco smoking in these reports, make it difficult to generalize findings to the lung cancer population. The importance of gender as a prognostic factor in lung cancer needs to be fully elucidated because of its implication in the design of experimental protocols for targeted chemoprevention, early disease screening, molecular marker based staging, and individualized treatment. The objective of this paper is to reassess gender as a 2004 by The Society of Thoracic Surgeons /04/$30.00 Published by Elsevier Inc doi: /j.athoracsur

2 210 VISBAL ET AL Ann Thorac Surg GENDER DIFFERENCE IN LUNG CANCER SURVIVAL 2004;78: prognostic factor for survival among NSCLC patients diagnosed at a medical center between 1997 and late 2002, a recent period without dramatic changes in staging and treatment. Material and Methods Study Population Between January 1, 1997 and December 31, 2002, a cohort of 4,618 patients with pathologically diagnosed and/or confirmed NSCLC at Mayo Clinic (Minnesota, USA) were prospectively enrolled in this study and actively followed. Ninety-five percent of all new lung cancer patients seen at the Mayo Clinic were identified daily from our computerized pathology reporting system with the remaining 5% referred directly from the patient-care physicians. Only patients giving informed consent for research were included in this study. All patient contact materials including interview script and follow-up questionnaires were reviewed and approved by the Mayo Foundation IRB. Methods of initial lung cancer diagnosis before pathologic confirmation include chest roentgenogram (92%) or spiral CT scan (97%) showing evidence for lung cancer. Approximately 72% of all patients had one or more symptoms that had led to the final diagnosis. Among patients who were examined by chest roentgenogram or CT scan, 33% or 5% took the procedure for lung cancer screening and 2% or 1% for nonlung cancer related symptoms, respectively. Approximately one-half of the patients who were diagnosed as a result of screening effort were asymptomatic. The effects of the screening and nonlung cancer related symptoms on patients survival are under thorough investigation in a separate project. Information abstracted from medical records for each patient included demographics (age, gender, and race), history of tobacco exposure, other diseases, lung cancer pathology, clinical staging, anatomical site, and treatment. Smoking history data were found in the patients medical record and was confirmed with a follow-up questionnaire or an interview [17]. Never smokers are defined as those who smoked less than 100 cigarettes in their lifetime. Histology classification (cell type and grade) was according to the World Health Organization Classification of Tumor 3 rd edition [18] and TNM stage as proposed in 1997 [19]. Clinical stage was assigned based upon results from chest roentgenogram, and/or CT scan, bone scan, PET scan, and magnetic resonance imaging, as available. Fiberoptic bronchoscopy was used to evaluate the primary tumor for most patients with suspected lung cancer. Surgical candidates underwent mediastinoscopy, mediastinotomy, thoracoscopy, and thoracothomy for evaluation of resectability as indicated. Patients with mediastinal lymphadenopathy (lymph node 1 cm or greater in the short axis view) underwent transbronchoscopic needle aspiration, mediastinoscopy and/or mediastinotomy for staging before a thoracotomy. If lymph nodes were found positive for malignancy, multimodality therapy was offered. Stage IV disease was histologically confirmed in the majority of patients. All pathology specimens, including the specimens from other institutions, were reviewed and histologically classified by the lung team of the Surgical Pathology Division at Mayo Clinic Rochester. Surgery was defined as a treatment when the patient had any pulmonary resection for the primary tumor, including pneumonectomy, bilobectomy, lobectomy, segmentectomy, and wedge resection, irrespective of the status of the surgical margin. In this study, if a patient was found to be nonresectable during an exploratory thoracotomy, surgery was considered as nontherapeutic and was classified as a staging procedure only. Surgical resection was the treatment of choice in early stages. For the patients stage IIIA or higher, chemotherapy and radiation therapy was given as neoadjuvant, adjuvant, or palliative according to established clinical protocols, performance status, and eligibility criteria of these patients. Approximately 12% of all patients referred to the Mayo Clinic had already had some types of treatment(s), which may include surgery (6%), chemotherapy (6%), and/or radiation therapy (5%). When the patient received any type of diagnostic procedure and therapy elsewhere, authorization for the release of medical information was requested and copies of the relevant medical records were requested and abstracted as received. All outside records were reviewed by a Mayo Clinic clinician. Patient Follow-Up All patients have been actively followed beginning at six months after diagnosis with subsequent annual follow-up by mailed questionnaires. Timely verification of the vital status of each patient to ensure virtually 100% completeness was accomplished through the multiple sources of information available to us: the Mayo Clinic registration database, next-of-kin reports, death certificates and obituary documents filed in the patient s medical records, the Mayo Clinic Tumor Registry, and the Social Security Death Index website. For living patients, the most up-to-date information was obtained from the last Mayo Clinic visit report or the last follow-up questionnaire, whichever was most recent. For deceased patients, the follow-up packet was sent to the next-of-kin to obtain proxy information regarding new diseases occurring after the initial diagnosis and cause of death, changes in smoking status, body weight, appetite, dietary supplements, and updated family history of lung cancer or other cancers. Cause of Death Information Mayo Clinic is a tertiary medical center located in Olmsted County that serves the county residents as a major primary care hospital. Although cause-of-death information was not available for every referral case, this information has been routinely and actively sought for all Olmsted County residents as a part of Rochester Epidemiology Project through death certificate, autopsy records, and physician s reports [20]. We have assessed

3 Ann Thorac Surg VISBAL ET AL 2004;78: GENDER DIFFERENCE IN LUNG CANCER SURVIVAL 211 lung-related and nonlung cancer-related death of patients who were residents of Olmsted County by gender. Statistical Methods The primary outcome used in this analysis was survival following lung cancer diagnosis. Survival was defined as the time (in days) from lung cancer diagnosis to death or the last known date the patient was reported to be alive. Patients known to be alive at last contact were censored. Survival was estimated for each gender at several time points by the Kaplan-Meier method, and survival curves were generated. Survival differences between genders were assessed for statistical significance by the log-rank test. A Cox proportional hazards model was developed to estimate the relative risk of death for men versus women. Adjusted relative risks were estimated after accounting for covariates known to be associated with survival in NSCLC patients, including age, smoking intensity, tumor histology and grade, stage, and treatment. The effect of gender on postdiagnosis survival was further assessed by estimating relative risks of death for men versus women stratifying separately on stage, histology, and treatment. Cox proportional hazards models were again developed to estimate relative risks within the strata of each individual factor, after adjusting for the other factors aforementioned. Interaction terms were created and tested to determine whether the effect of gender significantly differed across strata within a particular factor. All categorical variables were reported as percentages and compared across genders. Nominal categorical variables were compared across genders using 2 tests, and ordinal categorical variables were compared using Wilcoxon rank-sum tests. All continuous variables were reported as medians and ranges and compared between genders by Wilcoxon rank-sum tests. All analyses were done using SAS v6.12. Results Table 1. Tumor, Histology, Grade, and Disease Stage by Gender of NSCLC Patients Female (n 1889) Males (n 2721) P Value Histology Adenocarcinoma 1124, 59.5% 1311, 48.2% 0.01 Squamous 415, 22.0% 868, 31.9% Unclassified NSCLC 237, 12.5% 410, 15.1% Large Cell 70, 3.7% 81, 3.0% Adenosquamous 17, 0.9% 23, 0.8% Pleomorphic, Sacomatoid 15, 0.8% 21, 0.8% Salivary Gland Type 5, 0.3% 3, 0.1% Grade: I 28, 1.8% 18, 0.8% 0.01 II 256, 16.3% 267, 11.8% III 748, 47.6% 1104, 49.0% IV 538, 34.3% 865, 38.4% Unknown Stage: IA 305, , 12.1% 0.25 IB 201, 10.8% 311, 11.6% IIA 29, 1.6% 54, 2.0% IIB 100, 5.4% 189, 7.1% IIIA 272, 14.6% 408, 15.3% IIIB 283, 15.2% 432, 16.2% IV 675, 36.2% 956, 35.8% Unknown Among the 4,618 patients diagnosed with NSCLC during this six-year period, there were 2,724 men (59%) and 1,894 women (41%), a 1.44:1 male to female ratio. Median age at diagnosis was 68 years in men (range, 18 to 95) and 66 in women (range, 24 to 97), p less than White was the most frequent ethnic group (95.6% of women and 96.8% of men). Minority groups included Alaskan/Indian, African American, Hispanic, Asian/Pacific Islanders, and Other. Adenocarcinoma was the most frequent histopathology in both genders. Women had proportionally more adenocarcinoma, less squamous cell carcinoma, and less advanced histology grade than men did (Table 1). No significant difference in stage was observed across gender (p 0.25), though more women presented in stage IA than men (16.4% vs 12.1%) did. Regarding comorbidities, men presented with more heart disease, diabetes and nongender-specific cancer than women did (14.7% vs 7.7%, 7.2 vs 3.4%, and 26.7% vs 17.2%, respectively). The presence of any of these comorbidities in the multivariable model did not significantly alter the gender effect on survival. There were no significant differences in treatment between genders. Specifically, 51% women versus 48% men received surgical resection; 33% versus 32% received chemotherapy; and 30% versus 31% received radiation therapy. Nontherapeutic thoracotomy was performed in 84 patients with no difference between genders. When compared with women, more men were heavier smokers (Table 2). Cumulative exposure of smoking (as measured by pack-years of cigarettes) is significantly different with 30.7% of men smoking more than 60 pack years compared to 15.5% of women (p 0.01). Among the ever-smoking patients, more men were former smokers and more women were current smokers at the time of diagnosis. Survival differed significantly between genders (Fig 1). Estimated crude survival in men was 51% (95% CI: 49, 53) and 15% (12, 17) at one and five years respectively, with a median survival of 1.0 year. The estimated survival in women was 60% (58, 62) and 19% (16, 22) at one and five years, respectively with a median survival of 1.4 years. After adjusting for age at diagnosis, tumor histology and grade, stage, pack-years smoked, and treatment, men remained at significantly increased risk of mortality (relative risk: 1.20, 95% CI: 1.11, 1.30) compared to women following a diagnosis of NSCLC. All the variables in our adjusted model were significantly associated with postdiagnosis mortality except for pack years of cigarette smoking history (p 0.19).

4 212 VISBAL ET AL Ann Thorac Surg GENDER DIFFERENCE IN LUNG CANCER SURVIVAL 2004;78: Table 2. Smoking Information by Gender of NSCLC Patients Females (n 1894) Males (n 2724) P Value Smoking Status Current 535, 31.8% 640, 27.4% 0.01 Former 829, 49.2% 1516, 65.0% Never Smokers 321, 19.1% 177, 7.6% Unknown Pack Years 0 324, 20.2% 177, 8.2% , 15.1% 259, 12.0% , 25.7% 515, 23.8% , 23.6% 551, 25.4% Over , 15.5% 665, 30.7% Unknown All Smokers Age at initiation a 18 (16, 20) 17 (15, 18) 0.01 Former Smokers Years not smoking a 8 (2, 17) 12 (4, 22) 0.01 a Data were available on 1,994 patients for age at initiation and on 3,800 patients for years not smoking. We examined the gender effect on survival through multivariable Cox models stratified by disease stage, tumor histology, and initial therapy within six months of diagnosis (Table 3). We observed a significantly higher postdiagnosis risk of death among men who had stage III and IV lung cancer (Fig 2) and who had adenocarcinoma. Gender effect on survival was similar across treatment modalities. Difference in cause-of-death by gender as a potential confounder was assessed in a subset of patients who were Olmsted County residents. We have virtually complete information (96%) on cause-of-death on all Olmsted County patients (n 160) who were diagnosed during the study period. The cause of death was lung cancerrelated in 88% and 95% of men and women, respectively; however, this difference was not statistically significant. Fig 1. Kaplan-Meier survival curves of 4,618 newly diagnosed nonsmall cell lung cancer by gender, Mayo Clinic, 1997 to 2002, Minnesota. *Relative risk comparing men to women. Comment This study confirms that male gender is an unfavorable prognostic factor for NSCLC survival. Our results show an increased mortality compared to women patients in our cohort of prospectively and actively followed series of patients diagnosed over a six-year period when histologic classification [18], staging system [18, 19] and treatment options for NSCLC remained constant. The survival for men in stage IA is similar to women in stage IB, men in stage IIA is similar to women in stage IIB, and so forth. Thus, men appear to survive as the group of women in the next immediately advanced stage. After considering multiple known predictors of lung cancer survival, men remained at 20% increased risk of mortality compared to women following a diagnosis of NSCLC. Male gender has been reported as a significant independent negative prognostic factor for patients with nonsmall cell lung cancer from retrospective case series [21] or selected surgical patients [13, 16, 22]. de Perrot et al in a retrospective report of 839 men and 198 women surgically treated for NSCLC more than 20 years found a protective effect of female gender in stage I and II, and this effect was absent in stages III and IV. In this report, women were more likely asymptomatic (32 vs 20%), nonsmokers (27 vs 2%), or light smokers (31 vs 52 pack-years) than men. In a multivariable model, women demonstrated reduced risk compared to men (hazard ratio, 0.72; 95% confidence interval, 0.56 to 0.92) independent of age, presence of symptoms, smoking habits, type of operation, histology, and stage of disease. Our study reflects a group of nonselected patients, primarily Caucasians, who came to a referral center at any phase of their disease. The fact that more women were diagnosed at earlier stages raises interesting hypotheses to be studied and taken into consideration by the current screening experimental protocols for highrisk patients. Because the female gender survival advantage in early stage NSCLC might be a consequence of an earlier diagnosis, it is important to rule out if this is due to more frequent consultations and routine general medical follow-ups in women, instead of a genetic predisposition and resistance to lung cancer progression. Lung cancer patients present with more comorbid conditions than the general population [24], and men develop cardiovascular diseases ten years earlier than women. Thus, it is important to clarify whether the cause of death is cancer-specific or due to other diseases. Evidence from our Olmsted County patients suggest that gender difference in cause-of-death, ie, lung cancer related versus nonlung cancer-related, is unlikely the explanation for the observed gender difference in postdiagnosis survival. Critical issues deserve further exploration to elucidate the true difference in gender-specific survival to lung cancer: (1) Are critical genetic alterations genderspecific? (2) Are the associations of gender specific susceptibility and genetic changes a result of smoking and other environmental carcinogens exposure? (3) Are there any changes in the sex chromosome important in defin-

5 Ann Thorac Surg VISBAL ET AL 2004;78: GENDER DIFFERENCE IN LUNG CANCER SURVIVAL 213 Table 3. Stratified Analysis of Gender Difference by Stage, Histology, and Treatment Females (n 1894) Males (n 2724) P Value Stage (Adjusted for pack years, histology, and treatment) Stage IA: Relative Risk % CI... (0.85, 1.64) Stage IB: Relative Risk % CI... (0.81, 1.50) Stage IIA: Relative Risk % CI... (0.48, 1.82) Stage IIB: Relative Risk % CI... (0.80, 1.61) Stage IIIA: Relative Risk % CI... (0.99, 1.46) Stage IIIB: Relative Risk % CI... (1.02, 1.46) Stage IV: Relative Risk % CI... (1.15, 1.44) Histology (adjusted for stage, pack years, and treatment) Adenocarcinoma: Relative Risk % CI... (1.23, 1.53) Squamous: Relative Risk % CI... (0.96, 1.30) Other: Relative Risk % CI... (0.94, 1.31) Treatment (adjusted for pack years, histology, and stage) Chemotherapy: Relative Risk % CI... (0.97, 1.29) Radiation: Relative Risk % CI... (0.97, 1.29) Surgery: Relative Risk % CI... (0.92, 1.26) ing susceptibility to environmental carcinogens and/or to lung cancer? (4) Are any of the above differences important in the clinical practice for screening, prevention, and treatment choices? Our results align with findings described in the literature over the last 20 years, reflecting changes in the epidemiology of lung cancer between men and women. First, the incidence ratio of male to female was 1.44:1, confirming a narrowing ratio among genders reported by Mayo in 1988 [25] and two other studies [3, 4]. Second, we found that women are diagnosed on average two years younger than men are. Similar results were reported in a Polish series of 785 women with lung cancer diagnosed at 59.7 years and 4,619 men diagnosed at 61.9 years of age [26]. Third, the higher rate of adenocarcinoma is consistent with previous observations [2, 27, 28]. This has been attributed to the popularization of current low-tar and nicotine cigarettes in which an increase in frequency and intensity of the inhalations are required to satisfy nicotine level, exposing the bronchialveolar unit to higher amounts of smoke-born carcinogens. Controversy exists regarding differences in lung cancer risk between genders. Some authors have found an increased relative and absolute risk for female patients

6 214 VISBAL ET AL Ann Thorac Surg GENDER DIFFERENCE IN LUNG CANCER SURVIVAL 2004;78: Fig 2. Survival of stages IIIA, IIIB, and IV lung cancer patients by gender, Mayo Clinic, 1997 to 2002, Minnesota. *Adjusted for pack years of smoking, histology, and treatment. susceptible to develop lung cancer earlier in life than male patients. Although adenocarcinoma is the most common subtype in both genders, women have proportionally more adenocarcinomas than men do. Women have a better survival than their male counterparts after considering age, smoking history, tumor histology and grade, stage, comorbid conditions, treatment, and cause of death. We thank Drs Claude Deschamps, Daniel L. Miller, David E. Midthun, Alex A. Adjei, Aminah Jatoi, and Mark S. Allen for their input at various stages of this work. We also thank Susan Ernst for her technical assistance with the manuscript. This work was supported by grants CA and CA84354 from the US National Cancer Institute and by Mayo Foundation Funds. for developing lung cancer at any level of cigarette smoke exposure [9 12]. Others suggest that both genders are at equal risk [29, 30]. Recently, the presence of genetic polymorphisms in detoxification enzymes (cytochrome P450 and glutathione S transferase M1) has been reported to increase the risk for lung cancer in women [31]. Our data show that lung cancer was diagnosed earlier in women than in men, despite evidence that women started to smoke later, smoked less cigarettes per day and fewer years, and were lighter smokers. In addition, the period elapsed before lung cancer diagnosis in women after smoking cessation was shorter (Table 2). Because there appears to be a genetic predisposition for developing lung cancer, there is ground to further explore genetic differences in susceptibility for lung cancer [32] between genders. Finally, our results are important in the current immunopathology and molecular substaging for lung cancer. An immunopathology subclassification of 408 completely resected stage I NSCLCs found a differential cancerspecific survival by marker expression, histology and gender. Men with positive erb-2 and women with p53, CD4, RB, and factor VIII have a worse cancer-specific survival. Men with squamous cell carcinoma positive for erbb-2 and women with adenocarcinoma positive for p53, CD-44, and factor VIII has also a lower survival. Interestingly, interactions among molecular markers, histology, and gender showed an increased risk of dying for women with positive p53, and patients with adenocarcinoma positive for CD44 [33]. In a series of 287 prospectively collected blood samples of patients with lung cancer, cellular immunologic parameters related to age, gender, and stage with possible prognostic significance were described [34]. Therefore, future studies should consider the difference in survival by gender for NSCLC patients when designing protocols expected to impact the clinical practice for screening, prevention, and therapy. Conclusion Characteristics of our patient population reflect the current trends of lung cancer brought by changes in smoking behavior. Female patients smoke less but seem more References 1. Wingo PA, Ries LA, Giovino GA, et al. Annual report to the nation of the status of cancer, ; with a special section on lung cancer and tobacco smoking. J Natl Cancer Inst 1999;91: Levi F, Franceschi S, La Vecchia VC, Randimbison L, Te C. Lung carcinoma trends by histologic type in Vaud and Neuchatel, Switzerland, Cancer 1997;79: Andrews JL, SB Balogh K and F. B J. Lung cancer in women. Cancer 1985;55(12): Travis WD, Lubin J, Ries L, Devesa S. United States lung carcinoma incidence trends: declining for most histologic types among males, increasing among females. Cancer 1996; 77: Ries LAG, Eisner MP, Kosary CL, et al. SEER Cancer Statistics Review,-1999;2002; Jemal A, Murray T, Samuels A, Ghafoor A, Ward E, Thun M J. Cancer Statistics, CA A Cancer Journal for Clinicians 2003;53: Parker SL, Tong T, Bolden S, Wingo PA. Cancer Statistics, CA A Cancer Journal for Clinicians 1997;47: Mannino DM, Ford E, Giovino GA, Thun M. Lung cancer mortality rates in birth cohorts in the United States from 1960 to Lung Cancer 2001;31: Zang EA, Wynder EL. Difference in lung cancer risk between men, and women. Examination of the evidence. J Natl Cancer Inst 1996;88(3/4): Zang E. Re. Sex, Smoking and cancer: a reprisal. J Natl Cancer Inst 2002;94(4): Risch HA, Howe GR, Jain M, Burch JD, Holowaty EJ, Miller AB. Are female smokers at higher risk for lung cancer than male smokers? A case-control analysis by histologic type. Am J Epidemiol 1993;138: Risch HA, Miller AB. RE. Sex, smoking, and cancer: a reappraisal. J Natl Cancer Inst 2002;94(4): Alexiou C, Onyeaka CVP, Beggs D, et al. Do women live longer following lung resection for carcinoma? Eur J Cardiothorac Surg 2002;21: Ferguson MK, Wang JH, Hoffman PC, et al. Sex-associated differences in survival of patients undergoing resection for lung cancer. Ann Thorac Surg 2000;69: Burkhart HM, Allen MS, Nichols FC III, et al. Results of en bloc resection for bronchogenic carcinoma with chest wall invasion. J Thorac Cardiovasc Surg 2002;123(4): Minami H, Yoshimura M, Miyamoto Y, Matsuoka H, Tsubota N. Lung cancer in women. Sex-associated differences in survival of patients undergoing resection for lung cancer. Chest 2000;118: Yang P, Wentzlaff KA, Katzmann JA, et al. Alpha 1 - antitrypsin deficiency allele carriers in lung cancer patients. Cancer Epidemiol Biomarkers Prev 1999;8: Travis WD, Colby T V, Shimosato Y, Brambilla E, Sobin LH,

7 Ann Thorac Surg VISBAL ET AL 2004;78: GENDER DIFFERENCE IN LUNG CANCER SURVIVAL 215 Corrin B. Histological typing of lung and pleural tumours 1999: Mountain CF. Revisions in the international system for staging lung cancer. Chest 1997;111: Melton LJ III. History of the Rochester Epidemiology Project. Mayo Clin Proc 1996;71: Ferguson MK, Skosey C, Hoffman PC, Golomb HM. Sexassociated differences in presentation, and survival in patients with lung cancer. J Clin Oncol 1990;8(8): Thomas P, Doddoli C, Thirion X, et al. Stage I non-small cell lung cancer. A pragmatic approach to prognosis after complete resection. Ann Thorac Surg 2002;73(4): de Perrot M, Licker M, Bouchardy C, Usel M, Robert J, Spiliopoulos A. Sex differences in presentation, management, and prognosis of patients with non-small cell lung carcinoma. J Thorac Cardiovasc Surg 2000;119(1): Janssen-Heijnen ML, Schipper RM, Razenberg PP, Crommelin MA, Coebergh JW. Prevalence of co-morbidity in lung cancer patients and its relationship with treatment: a population-based study. Lung Cancer 1998;21: Beard CM, Jedd MB, Woolner LB, Richardson RL, Bergstralh EJ, Melton LJ III. Fifty-year trend in incidence rates of bronchogenic carcinoma by cell type in Olmsted County, Minnesota. J Natl Cancer Inst 1988;80: Radzikowska E, Glaz P. The role of sex as a prognostic factor in lung cancer. Pneumonlogia I Alergologia Polska 2000; 68(9 10): Ries LAG, Kosary CL, Hankey BF, Miller BA, Harras A, Edwards BK. SEER Cancer Statistics Review, ; Tables and Graphs: NIH Pub. No Janssen-Heijnen MLG, Coebergh JWW. Trends in incidence and prognosis of the histological subtypes of lung cancer in North America, Australia, New Zealand and Europe. Lung Cancer 2001;31: Agudo A, Ahrens W, Benhamou E, et al. Lung cancer, and cigarette smoking in women. A multicenter case-control study in Europe. Int J Cancer 2000;88: Perneger TV. Sex, smoking, and cancer: a reappraisal. J Natl Cancer Inst 2001;93(21): Dresler CM, Fratelli C, Babb J, Everley L, Evans AA, Clapper ML. Gender differences in genetic susceptibility for lung cancer. Lung Cancer 2000;30: Sellers TA, Bailey-Wilson JE, Potter JD, Rich SS, Rothschild H, Elston RC. Effect of cohort differences in smoking prevalence on models of lung cancer susceptibility. Genet Epidemiol 1992;9: D Amico TA, Aloia TA, Moore MBH, et al. Molecular biologic substaging of stage I lung cancer according to gender, and histology. Ann Thorac Surg 2000;69: Nakamura H, Kawasaki N, Hagiwara M, Saito M, Konaka C, Kato H. Cellular immunologic parameters related to age, gender, and stage in lung cancer patients. Lung Cancer 2000;28: INVITED COMMENTARY Lung cancer is the leading cause of cancer deaths in men and women in the United States. Overall, survival has gradually increased over the last 40 years, but remains at a disappointing low level around 15%. Many would argue that despite the high frequency and lethal nature of lung cancer, it has attracted far too little attention, research dollars, and screening programs compared to less common and less lethal malignancies. Even the most basic issues still lack clarity. Visbal and colleagues have conducted a careful, thorough, and prospective study to answer a basic question regarding nonsmall lung cancer (NSCLC): Is there a gender difference in survival? In their study population of mostly Caucasian (96%) patients from a major tertiary referral center, gender did have a negative impact on survival for men. Having answered the question, the real challenge is why, being able to answer this, however, is the more important question. One must be certain that there is not some inherent bias in their study such as the impact of referrals to their institution, cumulative smoke exposure, smoking habits, or health care habits of men versus women. It will be important to determine if there is a genetic, hormonal, or biologic explanation for a gender difference in survival for NSCLC. If there is a real gender difference, then finding the why may unlock some of the secrets of lung cancer that may improve our understanding and ultimately our treatment of this deadly disease. Douglas Mathisen, MD General Thoracic Surgical Unit Massachusetts General Hospital 55 Fruit St, Blake 1570 Boston, MA dmathisen@partners.org 2004 by The Society of Thoracic Surgeons /04/$30.00 Published by Elsevier Inc doi: /j.athoracsur