A widely used system for variant assessments based on lines of evidence supporting a variant’s pathogenicity or benignness was developed through a collaboration between the American College of Medical Genetics and Genomics (ACMG) and the Association for Molecular Pathology (AMP). Ten years after the ACMG/AMP classification system was introduced and adapted, variants of uncertain significance (VUS) continue to make up the vast majority of missense variant records in databases such as ClinVar. These findings were published in the Journal of Plos Genetics.
It includes genome and exome sequencing that can identify coding sequence variants affecting a wide range of diseases ranging from rare Mendelian to common cancers. Mendelian disorders (‘oncoprotein duality’) are generally due to germline dysregulation of some proto-oncogenes and tumor suppressor genes (TSGs). Collectively, rare missense variants are common in every human genome, and, in particular, interpreting the clinical impact of these variants is difficult. Rare missense variants can be better classified by using available but underused genomic databases to identify other lines of evidence for pathogenicity. Cancer driver mutations (also known as oncogenic mutations) are genetic changes founded on these alterations that are associated with the initiation and progression of cancer.
The recent acceleration of tumor sequencing initiatives, such as The Cancer Genome Atlas (TCGA) and the International Cancer Genome Consortium (ICGC), has greatly enhanced the identification of oncogenic mutations.
In the context of modifying the existing PM1 pathogenic evidence criterion to apply to germline variants in genes implicated in cancer predisposition, Walsh and colleagues first proposed that if the variant is not already found in a germline hotspot, it should be evaluated based on its overlap with cancer mutations from Cancer Hotspots. This thesis explores the idea of oncoprotein variant duality and determines how much of the germline variant classification can be anticipated from the observation that the same tumor mutation drives cancer in the same tumor type. The logic underlying this study approach is that cancer driver mutations have functional consequences at the protein level, thus we would expect such functional consequences regardless of whether the variant was somatic/mosaic/tissue-specific or constitutional/germline.
The mutations were annotated and filtered to provide a list of 2,447 missense mutations (“CH mutations”) occurring across 216 genes. These genes include proto-oncogenes (41%), tumor suppressor genes (36%), or genes that can act (15%), according to the genome census for cancers. The team believed that missense mutations that are cancer drivers in proto-oncogenes and tumor suppressor genes would have either loss of function or gain of function mechanism. Online Mendelian Inheritance in Man (OMIM) database Mendelian disease associations for these genes showed that 20% are associated with hereditary cancer predisposition syndromes.
The 216 genes included 154 for which modes of inheritance were known for cancer and Mendelian disease in OMIM. The actual cancer expression of these 154 genes was concordant with a Mendelian disease mechanism in 107 (69%), discordant in 26 (17%), and semi-concordant in 21 (14%), meaning that genes could serve as both proto-on ones and tumor suppressor genes or as Mendelian diseases where variants have both gains of function and loss of function.
These germline variants overlapping with cancer driver mutations, may offer insights into their mechanisms, loss of function in tumor suppressor genes, or gain of function in proto-oncogenes and give functional context for Mendelian diseases. This is also illustrated by the disease associations in OMIM for these genes, with 38% being hereditary cancer predisposition syndromes (e.g., VHL associated with von Hippel-Lindau syndrome) and 62% not known to represent cancer as the predominant feature.Â
This study has a few limitations, it primarily focused on a subpopulation of cancer passenger mutations from the Cancer Hotspots dataset that was last updated in 2017. However, the overlap of the additional highly recurrent missense mutations in COSMIC in 2024 with ClinVar germline variants was small, consistent with the idea that Cancer Hotspots remains a nearly comprehensive list of statistically recurrent cancer driver mutations.
Reference: Haque B, Cheerie D, Pan A, et al. Leveraging cancer mutation data to inform the pathogenicity classification of germline missense variants. PLoS Genet. 2025;21(1):e1011540. doi:10.1371/journal.pgen.1011540
