FAMILIAL CLUSTERING OF BREAST AND PROSTATE CANCERS AND RISK OF POSTMENOPAUSAL BREAST-CANCER

Background: Previous studies have suggested that cancers of the breast and prostate cluster in families and that the presence of both diseas es in a family may be associated with increased risk of breast cancer. Purpose: Our purpose was to evaluate whether 1) prostate cancer aggre gates in families with postmenopausal breast cancer, 2) families with cancers of the breast and prostate are the same ones as families with cancers of the breast and ovary, and 3) a family history of prostate c ancer is associated with increased risk of postmenopausal breast cance r. Methods: We analyzed data from a large prospective cohort study of Iowa women that were (at baseline) aged 55-69 years in 1986. At the th ird follow-up survey in 1992, self-reported data on family history of breast, ovarian, and prostate cancers in parents and siblings were pro vided by 30 883 women. Additional information was collected to ascerta in whether the age-of-onset of breast cancer in mothers or sisters was before or after the age of 45 years. Cancer occurrence was documented using the State Health Registry of Iowa. Results: History of prostate cancer in their father or a brother was reported by 3384 (11.0%) of t he women, and a total of 4090 women (13.2%) reported breast cancer in their mother or a sister. A positive family history of both cancers wa s reported by 556 women, significantly (two-sided P<.001) greater than the 457 women expected if the family histories were independent. The aggregation of breast, prostate, and ovarian cancers was reported by 2 2 participants, greater than the 2.7 expected (two-sided P<.0001). Dur ing 6 years of follow-up, 578 breast cancers were identified in the co hort at risk. Compared with women without a family history of either c ancer, women with a family history of breast cancer had a relative ris k (RR) of 1.37 (95% confidence interval [CI] = 1.06-1.79) if the affec ted relative had onset after the age of 45 years, and an RR of 1.71 (9 5% CI = 1.13-2.61) if the affected relative had onset at or before the age of 45. A family history of prostate cancer in the absence of a fa mily history of breast cancer was associated with an RR of 1.19 (95% C I = .90-1.56). However, a family history of both breast and prostate c ancers was associated with RRs of 2.06 (95% CI = 1.23-3.45) and 2.35 ( 95% CI = .97-5.67) for breast cancer onset in relatives of greater tha n 45 and less than or equal to 45 years, respectively. Conclusions: Th ese observations are concordant with recent reports that suggest a sha red familial risk (inherited or environmental) for these hormone-depen dent malignancies.