Approximately 15% of all breast cancer patients have a positive family history of the disease. BRCA1 and BRCA2 are two genes that explain major proportions of families with multiple cases of early-onset breast and/or ovarian cancer. Despite the high risks of breast and ovarian cancer conferred by deleterious BRCA1 and BRCA2 mutations, a strong variability in phenotype has been observed among families segregating the same mutation. This can range from early-onset breast cancer and ovarian cancer, to late-onset breast cancer without ovarian cancer. Even within a single pedigree, ages of onset of cancer can vary substantially. These observations support the idea that disease outcome in carriers is codetermined by other factors. Different risk estimates for BRCA mutations, depending on the type of population studied, also attest to this point. Risk estimates derived from families with multiple cases of early-onset breast cancer, used for linkage analysis to detect BRCA1 and BRCA2, came out substantially higher than those from population-based studies, and risks also appear to differ between populations.

Both genetic and environmental factors are thought to interact with BRCA1 and BRCA2. The influence of nongenetic factors is demonstrated by the finding that even identical carrier twins may differ in disease history. Simple chance may determine age of onset, because multiple genetic mutations are required for full tumorigenesis. Thus far we are not even sure whether modifiers of BRCA-conferred risk actually exist, but some suggestive associations have been reported, which will require independent confirmation. Rare alleles at HRAS1 were found to increase risk of ovarian cancer in BRCA1 carriers, whereas breast cancer risk has been found to be modified by rare alleles at the androgen receptor. Among Ashkenazi Jewish women, a polymorphism in the 5' UTR of the RAD51 gene increased risk of breast cancer fourfold, but only in carriers of the BRCA2-6174delT.