Though much progress has been made in understanding the role of two major high-risk breast cancer (BC) susceptibility genes, BRCA1 and BRCA2, it remains unclear what causes the observed variation in risk between mutation carriers. This marked variability in individual cancer risk both between and within BRCA1 and BRCA2 mutation carrier families may be partly explained by modifier genes that influence mutation penetrance. Defining these modifiers should help refine individual cancer risk estimates and is also expected to be an efficient method to identify further BC susceptibility alleles in general. This approach is predicated on the concept that variants in genes that are low to moderate penetrance predisposition genes are likely to have a larger risk modification effect in BRCA1/2 mutation carriers. Association studies are usually used to assess the influence of variants in biologically plausible candidate loci on the penetrance of BRCA1/2 mutations (i.e., differences in age of onset or tissue-specificity of disease). Several such modifier loci, including the genes AIB1 and AR involved in hormone metabolism, and the RAD51 gene acting in DNA repair, have been proposed in the literature. A consortium of laboratories (CIMBA) has recently confirmed the RAD51 135 G/C variant as a BC risk modifier in BRCA2 mutation carriers, though not in BRCA1 carriers. This review describes molecular epidemiological efforts to evaluate the potential influence of polymorphic variants in candidate modifier genes on the risk of BC conferred by the BRCA1 and BRCA2 genes.