Abstract B130: Hormone therapy as a predictor of lung cancer risk and survival among postmenopausal women enrolled in the California Teacher's Study

Abstract

Abstract Purpose: Cigarette smoking is the leading cause of lung cancer, but there exists a substantial proportion of lung cancer cases who have never smoked. Furthermore, lung cancer pathology, risk factors and prognosis differ by sex, prompting a need to investigate the influence of hormonal and reproductive factors. Most studies have shown a protective or no effect of postmenopausal hormone therapy (HT) on lung cancer risk, whereas the recent post-hoc analysis of the Women's Health Initiative showed that estrogen+progestin (E+P) decreased lung cancer survival. Given the substantial clinical implications, it is vital that the risk and survival associations be validated. Methods: We examined the associations between HT use and lung cancer risk and survival among 61,911 postmenopausal women enrolled in the California Teacher's Study. Between 1995 and 2007, 759 women (including 184 never smokers) were diagnosed with lung cancer; 425 of these died. Age-stratified, multivariable Cox proportional hazards regression was used to calculate hazard ratios (HR). Results: After adjusting for potential confounders, overall HT use was not associated with lung cancer risk. Among ever-smoking women, recent use of estrogen (E) or E+P at baseline was associated with decreased lung cancer risk (HR, 0.80; 95% confidence interval (CI), 0.66–0.96). However, the same association was not observed among never smokers (HR, 1.10; 95% CI, 0.78–1.55). Among ever-smoking women, using HT (E or E+P) for a duration of less than 5 years or 5–15 years was associated with decreases in risk (HR, 0.78, 95% CI, 0.61–0.99; HR, 0.69, 95% CI, 0.53–0.88, respectively). Additionally, recent use of E+P, but not E, was associated with decreased risk (HR, 0.76; 95% CI, 0.60–0.96; HR, 0.81, 95% CI, 0.62–1.06; respectively). After adjusting for potential confounders, any HT use (vs. no use) was associated with an increase in lung cancer survival (HR, 0.76; 95% CI, 0.60–0.95), with no difference by smoking status. The median survival time (MST) for ever users of HT was 47.7 months, compared to 37.8 months among never users (log-rank p=0.005). Among patients diagnosed with non-small cell lung cancer the difference was greater (HT ever user, MST=56.2 months vs. HT never user, MST= 39.9 months; log-rank p=0.002). Statistically significant increases in lung cancer survival were seen among recent E users (HR, 0.72; 95% CI, 0.55–0.96), but no survival difference was observed among E+P users. For E users overall, longer duration of E was associated with improved survival (0–5 years: HR, 0.81, 95% CI, 0.56–1.67; 5–15 years: HR, 0.65, 95% CI, 0.42–1.17; >15 years: HR, 0.66, 95% CI, 0.45–0.97; trend p=0.011). Conclusions: In contrast to recent findings, our results suggest that postmenopausal E+P use is associated with decreased lung cancer risk among ever smokers, while E use is associated with increased lung cancer survival overall. These results further reveal the complex nature of the role of hormonal factors in the etiology and prognosis of lung cancer. Citation Information: Cancer Prev Res 2010;3(1 Suppl):B130.