Two large-scale studies provide increasingly convincing evidence that lung cancer is affected by hormonal factors, an issue that has been controversial for many years. During Saturday’s Cancer Prevention Oral Abstract Session, Rowan T. Chlebowski, MD, PhD, of Harbor-UCLA Medical Center, presented results from a secondary analysis of data from the Women’s Health Initiative (WHI) showing that women who took estrogen plus progestin for relieving symptoms of menopause had a higher mortality rate from non-small cell lung cancer (NSCLC) than women who did not take the hormones (Abstract CRA1500). Christina S. Baik, MD, MPH, of Tufts Medical Center, presented results from the Nurses’ Health Study (NHS) demonstrating that the risk of lung cancer may be increased by certain hormonal and reproductive factors but not by postmenopausal hormone use (Abstract 1501). Together with previous study data, the results of these studies illustrate lung cancer as a heterogeneous disease affected by numerous hormonal factors that warrant continued examination.

Dr. Chlebowski examined the incidence of NSCLC in 16,608 women enrolled in the WHI clinical trial and determined mortality rates in those who developed disease during the trial period (5.6 years) and during follow-up (2.4 years). Women in WHI were randomly assigned to daily conjugated equine estrogen (CEE) (0.625 mg) plus medroxyprogesterone acetate (MPA) (2.5 mg) or placebo. Although previous research had suggested hormones are relevant to development of NSCLC, this is the first study to examine this relationship in a randomized, controlled clinical trial.

Overall, there was no statistically significant difference in the incidence of NSCLC in the hormone therapy group compared with the placebo group; 96 patients (0.14%) who received CEE plus MPA developed NSCLC, compared with 72 patients (0.11%) who received placebo (p = 0.12). How-ever, a variance was observed after 5 years, with more diagnoses in the CEE plus MPA group. In addition, among women who were diagnosed with NSCLC, mortality was 61% higher for those who received hormone therapy compared with those who did not (67 vs. 39 deaths, respectively; p = 0.02).

Smoking further increased the risk of developing NSCLC. Among women who were current smokers and who received CEE plus MPA, the mortality rate from NSCLC was 3.14%, compared with 2.3% of current smokers in the placebo group. According to Dr. Chlebowski, this translates to a preventable death for 1 in 100 current smokers using combined hormone therapy in the trial. Therefore, it is recommended that current smokers should discontinue smoking or should carefully consider this increased mortality risk before beginning or continuing combined hormone therapy.

“Many women entering menopause have symptoms that make them consider hormone therapy,” said Dr. Chlebowski. “The link we describe between hormone therapy with CEE plus MPA and death from non-small cell lung cancer should influence discussions between physicians and women considering hormone therapy use, especially for those with a smoking history.”

Supporting Data for Risks, Protective Effects
The fact that hormones play a role in lung cancer is further supported by data presented by Dr. Baik from the NHS. The NHS, which began in 1976 in Channing Laboratory at Harvard Medical School, is still on-going and is one of largest prospective, observational cohort studies in the United States. Although a few earlier studies have evaluated hormonal risk factors in lung cancer, the current study of NHS data is the largest study to date, examining 107,171 women with 1,729 lung cancer cases.

Notably, associations were strongest among current smokers, in small cell histology, and in squamous cell to a lesser degree. No significant associations were seen with adenocarcinoma. These results suggest that endogenous hormonal exposure during a woman’s lifetime may have a protective effect against lung cancer development.

The use of postmenopausal hormone therapy did not result in a significant association with lung cancer overall, but when analyzed separately by histology, there was a trend for increased risk in adenocarcinoma as well as a decreased risk in small cell and squamous histology.

Dr. Baik stated that the results of this study indicate lung cancer is a heterogeneous disease with complex hormonal mechanisms in carcinogenesis, which may vary by histologic subtypes, smoking status, and other factors.

Discussant Jill Siegfried, PhD, of the University of Pittsburgh, provided context for the studies through a discussion of previously published data on lung cancer and hormonal factors. She enumerated evidence for sex differences and the role of hormones in lung cancer, including evidence of high estrogen levels being associated with tumor progression and poor outcome. Dr. Siegfried also noted that estrogen-receptor beta (ER-beta) expression occurs in up to 80% of NSCLC tumors. She shared some unpublished data of her own that indicate high ER-beta expression correlates with shorter overall survival, while progesterone receptor expression appeared to have a protective effect, noting that looking at one hormone system alone could confound our understanding of the effects of hormones in lung cancer.

In contrast to data showing the negative effect of estrogen after cancer diagnosis, other data suggests that estrogen present before cancer diagnosis may offer some protection against tumor development, as in Dr. Baik’s study. Hypotheses to explain the incongruities between protective effects of estrogen before diagnosis versus harmful effects after diagnosis are just beginning to be explored in her laboratory and in others, said Dr. Siegfried, though more extensive study is needed.