Toxicity and survival by sex in patients with advanced non-small cell lung carcinoma (NSCLC) on modern Southwest Oncology Group (SWOG) trials

Abstract

7549 Background: The older lung cancer clinical trial literature suggests that women have better survival than men, possibly due to sex-related changes in drug metabolism based on estrogen levels. A detailed analysis of this issue has not been conducted in the modern chemotherapy era. Thus, the SWOG Lung Committee analyzed outcomes in recent trials, with the hypotheses that either sex-specific toxicity profiles and/or age (as a surrogate for estrogen level) may explain outcome differences. Methods: A data base of eligible patients with complete toxicity data enrolled on 6 consecutive SWOG phase II or III advanced stage NSCLC trials was formed. Treatment was platin-based, plus either gemcitabine, vinorelbine, or a taxane. Toxicities were categorized by type, number and grade with alpha=0.01 (multiple comparisons). Kaplan-Meier survival estimates and Cox multivariate models were computed (alpha=0.05), plus exploratory age by sex interaction analyses. Results: There were1,324 patients (36% women). No difference by sex was found in either maximum toxicity grade, number of toxicities, or specific toxicity types, except for marginally more severe or worse metabolic (12% vs 9%, p=0.02) and life- threatening or worse neurologic (2% vs 1%, p=0.04) toxicities in women. Median, 1- and 2-year survivals were significantly better for women (11 mos, 46%, 19%) vs men (8 mos, 35%, 13%). After adjustment for prognostic factors, women had 14% reduced risk of death (HR 0.86; 95% CI: 0.75, 0.98; p=0.02). The exploratory age by sex cutpoint analysis found women age 60 or older had better survival (11 vs 8 mos, p=0.006), whereas survival was similar by sex for those under age 60. No other age cutpoint was significant. Conclusions: Women with advanced NSCLC survive longer than men after adjustment for other prognostic factors in the modern chemotherapy era. There was no difference in toxicity profile by sex to explain this finding. The survival benefit for women appeared limited to patients age 60 and older, suggesting that estrogen levels may interact with the efficacy of current chemotherapy prescriptions or other as yet undefined factors. This finding, if validated, could be potentially exploited in designing new therapies. No significant financial relationships to disclose.