Pathology update to the Manchester Scoring System based on testing in over 4000 families. Issue 10 (10th May 2017)

• Record Type: Journal Article
• Authors: Evans, D Gareth; Harkness, Elaine F; Plaskocinska, Inga; Wallace, Andrew J; Clancy, Tara; Woodward, Emma R; Howell, Tony A; Tischkowitz, Marc; Lalloo, Fiona
• Is Part Of: Journal of medical genetics. Volume 54:Issue 10(2017)
• Page Start: 674
• Page End: 681
• Publication Date: 2017-05-10
• Subjects: BRCA1 -- BRCA2 -- Manchester Scoring System -- pathology -- high grade serous ovarian cancer -- triple negative breast cancer; Medical genetics -- Periodicals; 616.042
• DOI: 10.1136/jmedgenet-2017-104584
• Languages: English
• ISSNs: 1468-6244

Abstract

Abstract : Background: While the requirement for thresholds for testing for mutations in BRCA1/2 is being questioned, they are likely to remain for individuals unaffected by a relevant cancer. It is still useful to provide pretesting likelihoods, but models need to take into account tumour pathology. Methods: The Manchester Scoring System (MSS) is a well-used, simple, paper-based model for assessing carrier probability that already incorporates pathology data. We have used mutation testing data from 4115 unrelated samples from affected non-Jewish individuals alongside tumour pathology to further refine the scoring system. Results: Adding additional points for high-grade serous ovarian cancer <60 (HGSOC=+2) and adding grade score to those with triple-negative breast cancer, while reducing the score for those with HER2+ breast cancer (−6), resulted in significantly improved sensitivity and minor improvements in specificity to the MSS. Sporadic HGSOC <60 years thus reached a score of 15–19 points within the 10% grouping consistent with the 15/113–13.2% that were identified with a BRCA1/2 pathogenic variant. Validation in a population series of ovarian cancer from Cambridge showed high sensitivity at the 10% threshold 15/17 (88.2%). Conclusions: The new pathology-adjusted Manchester score MSS3 appears to provide an effective and simple-to-use estimate of the 10% and 20% thresholds for BRCA1/2 likelihood. For unaffected individuals, the 20-point (20%) threshold in their affected first-degree relative can be used to determine eligibility at the 10% threshold.