Analysis of the exome aggregation consortium (ExAC) database suggests that the BAP1-tumor predisposition syndrome is underreported in cancer patients.

The BAP1-tumor predisposition syndrome (BAP1-TPDS) has been recently identified to predispose patients to a variety of cancers and preneoplastic lesions. About 130 unrelated probands have been identified worldwide; however, the impact of the syndrome is suspected to be much larger given the diversity of the cancer phenotype. To evaluate the frequency of germline BAP1 mutations in the general and cancer populations, we analyzed the Exome Aggregation Consortium (ExAC), a database that contains 53105 exomes of unrelated individuals unaffected by cancer (general population) and exomes of 7601 unrelated individuals affected by cancer provided by the Cancer Genome Atlas (TCGA, cancer subjects). BAP1 null variants were seen at much higher frequency in the cancer subjects (0.0526%) compared to the general population (0.00188%) with a relative risk of 27.93 and (P = 0.0011, [95% CI: 3.122-249.883], Fisher's exact test). We also studied a reported BAP1 null variant, c.1203T > G, p.T401* (rs200156887), observed commonly in the general population. Sequencing and restriction fragment polymorphism of the RT-4 cell line that contains this variant revealed that it is in fact a 3bp deletion/insertion, c.1201_1203delinsGAG, a likely benign missense alteration p.Y401E explaining the relative high frequency of this variant in the general population. In conclusion, germline null mutations in BAP1 have a significantly higher frequency in cancer patients than the general population. Given the low frequency of reported families with BAP1-TPDS, our results suggest that the syndrome is underreported especially in patients with cancer.