Smoking, estrogen, and lung cancer in women

Abstract

Mortality rates from lung cancer in women have been rising since the 1940s, and surpassed breast cancer rates in 1987. Smoking is clearly a major risk factor for lung cancer, although other factors also may influence risk among smokers. Recently, estrogen receptors have been detected in lung tumors. One study suggested that adenocarcinoma of the lung was increased among users of estrogen replacement therapy (ERT). Another report suggested reduced risk of death from lung cancer in ERT users. Given the large numbers of women taking ERT, the association between ERT and lung cancer needs to be further addressed. A case-control study design, including women from several cities in the U.S., was used to investigate the association between lung cancer (adenocarcinoma, squamous, small, and other cell types) and exposure to ERT. Over 500 cases of lung cancer and 2500 hospital-based controls with diagnoses unrelated to hormone use or smoking were utilized. Adjusted risk estimates among current and past smokers showed no association between lung cancer and long-term (>=12 months) use of ERT (OR = 0.9, 95%CI = 0.6-1.3). For small cell lung cancer, however, a significant protective effect was observed (OR = 0.4, 95%CI = 0.2-1.0). On the other hand, for adenocarcinoma, there was a suggestion of increased risk with long-term use (OR = 1.5. 95%CI = 0.9-2.4), and the difference in the risk estimates between adenocarcinoma and small cell lung cancer was significant (two-tailed p<.02). No significant associations were observed between long-term use of ERT and squamous cell lung cancer (OR = 0.8, 95%CI = 0.1-1.9), or with other types of lung cancer (OR = 0.6, 95%CI = 0.3-1.4). Based on these findings, there is no suggestion of an increased risk of lung cancer (all cell types combined) due to ERT. Further, ERT may be protective against small cell lung cancer, while perhaps leading to an increased risk of adenocarcinoma. Although numerous potential confounders were evaluated in the analyses, confounding due to diet or smoking of filtered (versus non-filtered) cigarettes could not be ruled out, as these data were unavailable. Future studies are needed on the association between ERT and lung cancer which evaluate these variables as potential confounders. Also, the significant difference observed in the risk estimates for adenocarcinoma and small cell lung cancer demonstrates the need to study each cell type of lung cancer separately.