Prevalence of pathogenic variants and variants of unknown significance in patients at high risk of breast cancer: A systematic review and meta-analysis of gene-panel data. (December 2018)

• Record Type: Journal Article
• Authors: van Marcke, C.; Collard, A.; Vikkula, M.; Duhoux, F.P.
• Is Part Of: Critical reviews in oncology/hematology. Volume 132(2018)
• Page Start: 138
• Page End: 144
• Publication Date: 2018-12
• Subjects: Hereditary breast cancer -- Genetic predisposition -- Variant of unknown significance -- Gene-panel testing -- Pathogenic variant; Oncology -- Periodicals; Hematology -- Periodicals; 616.994
• DOI: 10.1016/j.critrevonc.2018.09.009
• Languages: English
• ISSNs: 1040-8428

Abstract

Graphical abstract: Highlights: Most gene-panel studies for high-risk breast cancer only report on pathogenic variants (PV) and lack data regarding variants of unknown significance (VUS). Currently, the probability of detecting a VUS is significantly higher than the probability of detecting a PV. This difference increases with the size of the panel. Only BRCA1 and BRCA2, the genes previously tested in a single-gene approach, carried a higher probability of being detected with a PV than with a VUS. All future gene-panel studies should report VUS results. Large public databases should gather both PV as VUS results for continuous research. Abstract: Background: Gene-panels are used to assess predisposition to breast cancer by simultaneous testing of multiple susceptibility genes. This approach increases the identification of variants of unknown significance (VUS) that cannot be used in clinical decision-making. We performed a systematic review of published studies to calculate the prevalence of VUS and pathogenic variants (PV) in routinely tested breast cancer susceptibility genes in patients at high risk of breast cancer. Methods: We comprehensively searched the literature using Medline through May 23, 2017 for studies conducting gene-panel testing on germline DNA of women with familial breast cancer and reporting on both PVs and VUSs. A meta-analysis of the collected data was carried out to obtain pooled VUS and PV prevalence estimates per gene using a generalized linear mixed model with logit link for binomial distribution. Results: Of 602 publications, 4 were eligible and included 1870 patients. The panels encompassed 4–27 considered genes. Overall, the estimated probability per gene of a PV and VUS was 55% (95% confidence interval (CI) 26%–81%) and 91% (95% CI 78%–97%), respectively ( p = 0.0066). The estimated probability per patient of a PV and VUS was 8% (95% CI 1%–34%) and 23% (95% CI 7%–52%), respectively ( p = 0.0052). The ratio of VUS to PV was highest in the mismatch repair genes MLH1, MSH2, MSH6, PMS2 (18.7), CDH1 (13.4) and ATM (9.5). Amongst the 1468 patients tested for BRCA1 and BRCA2, only these two genes had a VUS to PV ratio of less than one (0.2 and 0.6, respectively). Conclusion: With the current panels, the probability of detecting a VUS is significantly higher than the probability of detecting a PV. Better classification of VUSs is therefore critical and requires gene-specific VUS-assessment in every future study of gene-panel testing in patients at high risk of breast cancer.