Prioritizing Germline and Somatic Variants for Precision Oncology

Abstract

Molecular sequencing has become an integral part of diagnostic and treatment algorithms for many types of cancer, yet interpreting molecular sequencing in light of evolving knowledge and relevant data types remains a significant challenge. Interpretation and discovery is particularly challenging for germline sequencing routinely collected from normal tissue of cancer patients. Here I discuss enhancements to a paired precision oncology knowledgebase and interpretation algorithm, the Molecular Oncology Almanac, to add assertions relevant to medical and radiation oncology as well as improve accessibility of crowdsourcing through development of a public API and Chrome extension. To further characterize the contribution of pathogenic germline variants to the missing heritability of chronic lymphocytic leukemia, we performed a case-control study using WES data from 962 cases. Preliminary results demonstrate enrichment of pathogenic variants in the DNA damage repair gene, CHEK2. Significant enrichment in loss-of- function variants persists after removing known low-penetrance and founder variants, suggesting robustness to ancestry correction. Final analysis with enhanced QC, relatedness, and ancestry matching methods in expanded cohorts is pending. These results demonstrate potential for lowering the barrier to incorporate molecular sequencing data into precision oncology research and clinical practice for both medical and radiation oncology, in particular germline sequencing of normal tissue.