Twinship and risk of postmenopausal breast cancer

Background: Intrauterine exposure to high levels of endogenous estrogens ha s been hypothesized to:increase the risk of breast cancer. Because estrogen s and other pregnancy hormones are substantially elevated in twin pregnanci es, and possibly more so in dizygotic twin pregnancies, we evaluated the as sociation between aspects of twin membership (i.e., belonging to a twin pai r) and the risk of breast cancer. Methods: In a cohort of 29 197 postmenopa usal Iowa women with no prior diagnosis of cancer (except for nonmelanoma s kin cancer), breast cancer risk factors were determined by use of a mailed questionnaire in 1986 (baseline); twin membership, sex of the twin, and zyg osity were determined by use of a follow-up questionnaire in 1992, Results: Within the cohort, 1.8% (n = 538) of the women reported being a twin; of t hese, 24% (n = 130) were monozygotic twins, 63% (n = 337) were dizygotic tw ins, and 13% (n = 71) did not-know their zygosity, From 1986 through 1996, 1230 breast cancers in the cohort were ascertained by linkage to the Iowa C ancer Registry. Compared with singletons, women who belonged to a twin pair were at elevated risk of breast cancer (multivariate-adjusted risk ratio [ RR] = 1.72; 95% confidence interval [CI] = 1.22-2.42), with adjustment for educational level, family history of breast cancer, height, body mass index , body fat distribution, age at menarche, age at first live birth, use of h ormone replacement therapy, and alcohol use. Multivariate-adjusted risk was elevated (in comparison with singletons) if the sex of the other twin was female (RR = 1.82; 95% CI = 1.20-2.75); however, this risk was limited to f emale dizygotic twins (RR = 2.14; 95% CI = 1.21-3.79), since no excess risk was evident for monozygotic twins (RR = 1.04; 95% CI = 0.43-2.50), The ris k to women with a male twin was also elevated (RR = 1.49; 95% CI = 0.80-2.7 8) in comparison with singletons, but this estimate was not statistically s ignificant, Conclusions: This cohort study lends further support to the the ory that there are important intrauterine influences on carcinogenesis of t he breast.