The distribution of BRCA1/2 germline mutations in breast/ovarian cancer (BC/OC) families varies among different populations. In the Chilean population, there are only two reports of mutation analysis of BRCA1/2, and these included a low number of BC and/or OC patients. Moreover, the prevalence of BRCA1/2 genomic rearrangements in Chilean and in other South American populations is unknown. In this article, we present the mutation-detection data corresponding to a set of 326 high-risk families analyzed by conformation-sensitive gel electrophoresis and heteroduplex analysis. To determine the contribution of BRCA1/2 LGRs in Chilean BC patients, we analyzed 56 high-risk subjects with no pathogenic BRCA1/2 point mutations. Germline BRCA1/2 point mutations were found in 23 (7.1%) of the 326 Chilean families. Families which had at least three BC and/or OC cases showed the highest frequency of mutations (15.9%). We identified 14 point pathogenic mutations. Three recurrent mutations in BRCA1 (c.187_188delAG, c.2605_2606delTT, and c.3450_3453delCAAG) and three in BRCA2 (c.4969_4970insTG, c.5374_5377delTATG, and c.6503_6504delTT) contributed to 63.6 and 66.7% of all the deleterious mutations of each gene, which may reflect the presence of region-specific founder effects. Taken together BRCA1/2 recurrent point mutations account for 65.2% (15/23) of the BRCA1/2 (+) families. No large deletions or duplications involving BRCA1/2 were identified in a subgroup of 56 index cases negative for BRCA1/2 point mutations. Our study, which is the largest conducted to date in a South American population, provides a comprehensive analysis on the type and distribution of BRCA1/2 mutations and allelic variants.

The double-strand break (DSB) DNA repair pathway has been implicated in breast cancer (BC). RAD51 and its paralogs XRCC3 and RAD51D play an important role in the repair of DSB through homologous recombination (HR). Some polymorphisms including XRCC3-Thr241Met, RAD51-135G>C, and RAD51D-E233G have been found to confer increased BC susceptibility. In order to detect novel mutations that may contribute to BC susceptibility, 150 patients belonging to 150 Chilean BRCA1/2-negative families were screened for mutations in XRCC3. No mutations were detected in the XRCC3 gene. In addition, using a case-control design we studied the XRCC3-Thr241Met, and RAD51D-E233G polymorphisms in 267 BC cases and 500 controls to evaluate their possible association with BC susceptibility. The XRCC3 Met/Met genotype was associated with an increased BC risk (P = 0.003, OR = 2.44 [95%CI 1.34-4.43]). We did not find an association between E233G polymorphism and BC risk. We also analyzed the effect of combined genotypes among RAD51-135G>C, Thr241Met, and E233G polymorphisms on BC risk. No interaction was observed between Thr241Met and 135G>C. The combined genotype Thr/Met-E/G was associated with an increased BC risk among women who (a) have a family history of BC, (b) are BRCA1/2-negative, and (c) were <50 years at onset (n = 195) (P = 0.037, OR = 10.5 [95%CI 1.16-94.5]). Our results suggested that the variability of the DNA HR repair genes XRCC3 and RAD51D may play a role in BC risk, but this role may be underlined by a mutual interaction between these genes. These findings should be confirmed in other populations.