Smoking and estrogen plus progestin (E+P) and lung cancer incidence and mortality.

Abstract

1524 Background: In the Women’s Health Initiative (WHI) randomized, placebo-controlled clinical trial, E+P increased deaths from lung cancer (those cancer-attributed) and after lung cancer (regardless of cause) (Lancet 2009:374:1243). To examine smoking status influence on this process, a cohort combining WHI clinical trial (CT) and observational study (OS) participants, the latter meeting criteria as in the CT, was identified to examine E+P associations with lung cancer incidence and outcome. Methods: 31,966 postmenopausal women (12,299 CT, 19,668 OS) with no prior hysterectomy and no prior hormone therapy use were classified at baseline as not hormone users or E+P users and as current or never smokers. Lung cancers were verified by medical record review. Multi-variant adjusted Cox proportional hazards regression, including pack-year use, calculated hazard ratios (HRs, 95% confidence intervals [CI]) for groups defined by smoking status and E+P use for lung cancer incidence, deaths from lung cancer and deaths after lung cancer. Results: After 12 years mean follow-up, 664 lung cancers were diagnosed with 444 deaths from lung cancer and 513 deaths after lung cancer. Analyzed from cohort entry; see Table. In nonusers of E+P, lung cancer incidence, deaths from lung cancer, and deaths after lung cancer were significantly and substantially greater in current smokers vs never smokers (p< .0001 for all comparisons). In current smokers, lung cancer incidence, deaths from lung cancer and deaths after lung cancer were significantly and substantially greater in E+P users vs non-users (p=.0021, .0005 and .0002, respectively). Conclusions: E+P use in current smokers nearly doubles their already high risk of death from and after lung cancer. Based on this risk, current smokers should not use E+P. [Table: see text]