Comparison of guidelines, BRCAPRO, and genetic counsellors estimates for the identification of BRCA1 and BRCA2 mutations in pancreatic cancer.

Abstract

e15784 Background: Germline BRCA1 and BRCA2 (BRCA) mutation carriers with pancreatic adenocarcinoma (PDAC) are eligible for precision therapy trials and their relatives should undergo genetic testing and tailored cancer prevention. We assessed the performance of strategies to identify BRCA mutation carriers in PDAC. Methods: Incident cases of PDAC were prospectively recruited for BRCA sequencing in a multidisciplinary PDAC clinic. Probands were evaluated according to the National Comprehensive Cancer Network 2017 (NCCN) and the Ontario Ministry of Health and Long-Term Care (MOHLTC) guidelines for BRCA testing. The probability of each proband carrying a BRCA mutation was estimated using BRCAPRO and by surveying genetic counsellors. Guidelines were compared across sensitivity, specificity, and positive and negative predictive values (PPV and NPV). Estimates from BRCAPRO and the genetic counsellors were compared using the area-under-the-curve (AUC) for discrimination and the Hosmer-Lemeshow test for calibration. Results: 22/484 (4.5%) of probands carried a BRCA mutation. The mutation rate was higher in probands with Ashkenazi Jewish ancestry (7/57, p=0.009) or a first-degree relative with breast cancer (8/83, p=0.036). 119 genetic counsellors responded to the survey and each proband was assessed by a mean of 5.9 genetic counsellors. The Table displays the performance of the guidelines. Discrimination was similar for the estimates from genetic counsellors and BRCAPRO (AUC 0.755 and 0.775, respectively, p=0.701). Genetic counsellors generally overestimated (p=0.008), whereas BRCAPRO severely underestimated (p<0.001), the probability that each proband carried a mutation. Conclusions: The NCCN 2017 guidelines and estimates from genetic counsellors accurately identify BRCA mutations in PDAC. The MOHLTC guidelines and BRCAPRO should be updated to account for the association between PDAC and BRCA mutations. [Table: see text]