Lung cancer deaths are significantly less common in breast cancer patients who are treated with anti-estrogens.
San Antonio, TX—Lung cancer deaths are significantly less common in breast cancer patients who are treated with anti-estrogens, said Elisabeth Rapiti, MD, University of Geneva, Switzerland, at the CTRC-AACR San Antonio Breast Cancer Symposium.
Her observational study adds to the evidence about hormones and lung cancer mortality generated from the Women’ Health Initiative (WHI). In early data from the WHI, lung cancer mortality was found to be significantly increased in women treated with postmenopausal hormone therapy, she said. Oncology & Biotech News reported on one of these earlier presentations from the ASCO Annual Meeting in May, which concluded that postmenopausal women who smoke and receive estrogen/progesterone therapy are more likely to die from lung cancer.
Based on these earlier studies, Dr Rapiti postulated that anti-estrogens would decrease the risk of lung cancer mortality. She investigated this hypothesis by examining the Geneva Cancer Registry, focusing on the 6655 breast cancer patients in the registry who were diagnosed between 1980 and 2003. Of these patients, 46% had received anti-estrogens. She followed the entire cohort for lung cancer incidence and mortality until the end of 2007. Lung cancer incidence and mortality were compared to age- and period-adjusted rates among the general population.
There were some imbalances in baseline characteristics between users and nonusers of anti-estrogens, with users being significantly older, more likely to be postmenopausal, and with a lower socioeconomic status and a more recent diagnosis (P <.001). In addition, twice as many nonusers of estrogen were categorized as having an unknown smoking history compared with users (66% vs 31%; P <.001).
Among those with a known smoking history, about 30% of women in each group (users and nonusers of anti-estrogens) were smokers, compared with 29% of women in the general population. Ex-smokers made up 14% of the group that used anti-estrogens and 13% of the nonusers, compared with 16% in the general female population. The WHI included 21,655 person years of observation in the users of anti-estrogens and 35,520 person years of observation in the nonusers.
During follow-up 40 patients developed lung cancer, comprising 12 users of anti-estrogens and 28 nonusers. The risk of women with breast cancer developing lung cancer as a secondary malignancy was 37% lower than expected among users of anti-estrogens (P = .058) and 12% higher than expected among nonusers of anti-estrogens (P = .294), though these differences failed to reach statistical significance.
In total, 18 patients died from lung cancer during follow-up. The risk of death from lung cancer after breast cancer was 87% lower than expected in the anti-estrogen users, with 2 deaths reported compared with an expected 15.3 lung cancer deaths in the estrogen users (P <.0001). Among nonusers of anti-estrogens, there were 16 deaths from lung cancer compared with an expected 21.1 deaths, which was not significant.
Because most women with an unknown smoking status had received a breast cancer diagnosis before 1990, a separate analysis was conducted of the women who received a diagnosis in 1990 or after, said Dr Rapiti. When the analysis was confined to women diagnosed with breast cancer from 1990, anti-estrogen use was associated with a 36% reduction in lung cancer incidence and a significant 85% reduction in lung cancer mortality compared to the population at large.
Further stratification by smoking status revealed that most of the protective effect of anti-estrogens on lung cancer mortality after breast cancer occurred in smokers and ex-smokers, said Dr Rapiti. In this group, the age-standardized mortality rate was 1.05 per 100,000 among users of anti-estrogens but 3.82 per 100,000 among nonusers (P =.013). In the ex-smokers, the age-adjusted mortality rate was zero per 100,000 among users of estrogen and 7.71 per 100,000 among nonusers of estrogen (P <.001). SABCS Abstract 35.