We recently published a new scoring system to estimate the chance of identifying mutations in the BRCA1 and BRCA2 genes (table 1).¹ A potential criticism of the original paper was that not all cases were fully screened for mutations in both BRCA1 and BRCA2. We have now extended our analyses to address this. In total, we have screened 921 samples from familial breast cancer pedigrees from the Manchester region through the whole coding sequence and intron/exons boundaries of both genes. In addition, 200 cases, with all cases with a Manchester combined score for BRCA1 and BRCA2 of 40 or above, were screened for exonic deletions/duplications by multiplex ligation dependent probe amplification (MLPA). These analyses resulted in the identification of 206 (22.5%) intragenic mutations and nine (4.5%) exonic deletions or duplications (tables 2 and 3).

Table 2 shows that the 10 point Manchester score for a 10% threshold for each gene holds up well. The results for BRCA1 are presented with and without BRCA2 positive families with male breast cancer. Similarly, the results for BRCA2 are presented with and without BRCA1 positive families with ovarian cancer. These data allow one to estimate the likelihood of identifying a mutation in the gene screened second. For example, a breast/ovarian cancer family with a score of 28 for BRCA1 and 22 for BRCA2 has a 58% chance (column 3, table 2) of harbouring a BRCA1 mutation. If BRCA1 screening in the family is negative, the chance of identifying a BRCA2 mutation is 57% (column 5, table 2). The overall chance of a BRCA2 mutation in such a family prior to testing BRCA1 is therefore 24% (57% of 42%, the remainder of 100%–58%). The overall chance of a mutation in either gene is 82% (58%+24%). A similar process could be undertaken for families with male breast cancer.

As can be seen from table 3, a combined Manchester score of 15 is the cut off for a 10% threshold for BRCA1/2 testing. The cut off for a 20% threshold is less clear. Families with scores >20 would definitely qualify and those with scores of 17 points or above could reasonably be tested. It can be assumed that virtually all families with scores of 40 or above are due to either BRCA1 or BRCA2, although not all will have identifiable BRCA1/2 abnormalities as the sensitivity of gene testing is not 100% even with full gene screening and MLPA.² Moreover, the pathogenicity of unclassified variants (which were not included in our analyses) remains unclear.