ELSIcon2022 • Paper • May 27, 2022
Janis Geary, Kara Hapke, Robert Cook Deegan
Genetic tests to estimate individual risk of various hereditary cancers have proliferated due to improved sequencing technology and in response to interest among clinicians and those at risk. The genes included on germline cancer panels vary substantially from lab to lab. As the market expands, so does the need for data to improve variant interpretations and standards to guide clinical care. Historical underrepresentation of non-Europeans in research and in access to clinical genetic testing can lead to inequities in data required to interpret cancer risks associated with gene variants.
We describe characteristics of hereditary cancer gene variants available through commonly used gene variant databases. We compared hereditary cancer genetic test panels across 17 major companies, and identified 13 genes included in at least one panel from each company. We then downloaded population-based data on the variants of these genes available on the Genome Aggregation Database (gnomAD). We used descriptive statistics to compare population frequency of variants and clinical significance across population groups. We found that pathogenic variant frequency of the 13 genes is not consistent among population groups (e.g.: of the 150 pathogenic/likely pathogenic BRCA1 variants, 86% were found in only 1/8 gnomAD population groups, and 0% in >4), highlighting the need to collect variant data from diverse groups to ensure variants can be interpreted effectively and equitably.
Our analysis provides a snapshot of publicly-available population level data on hereditary cancer gene variants, and demonstrates how crucial it is to ensure non-European populations have data to enable effective variant classifications.