A prospective study of estradiol and breast cancer in Japanese women

Kabuto M, Akiba S, Stevens RG, Neriishi K, Land CE
Cancer Epidemiol Biomark Prev 9:575-9, 2000

Summary

Elevated circulating estrogen levels have been suspected for several decades to play a critical role in the etiology of breast cancer in women. This has been based on circumstantial evidence such as the observation that early age at menarche and late age at menopause are associated with increased risk; both factors would lead to a higher lifetime estrogen exposure. However, persuasive direct evidence has been lacking until recently. In the last few years, several cohort studies from the United States and Britain have shown significantly elevated estradiol among breast cancer cases compared to non-cases prior to development of disease. We conducted a nested case-control study of serum estradiol prior to disease and later risk of breast cancer in the cohort of Japanese atomic bomb survivors. In the LSS cohort of the RERF in Hiroshima and Nagasaki, 72 women diagnosed with breast cancer between 1973 and 1983 were identified; 150 women without breast cancer were identified and matched to the cases on age, date of blood collection, exposure, radiation dose, and city. Stored serum samples from all subjects were assayed for SHBG, total E2, bioavailable E2, DHEA-s, and prolactin. Conditional logistic regression for matched case-control comparisons was conducted. The odds ratio per unit log change in bioavailable E2 was 2.2 (95% CI, 1.02-5.3) for all subjects, and 2.3 (0.55-6.8) and 2.1 (0.55-9.7), respectively, based only on pre-menopausal or post-menopausal serum. The estimated odds ratios in each quintile of bioavailable E2 level, using the lowest quintile as referent, were 1.00, 1.89, 1.43, 3.45, and 3.37 (p for trend = 0.035). These results support the hypothesis that elevated bioavailable estradiol increases risk of breast cancer in women, and are consistent with several other recent cohort studies from other parts of the world.