Cancer Predisposition Sequencing Reporter (CPSR): A flexible variant report engine for high-throughput germline screening in cancer

Abstract

The value of high-throughput germline genetic testing is increasingly recognized inclinical cancer care. Disease-associated germline variants in cancer patients areimportant for risk management and surveillance, surgical decisions and can also havemajor implications for treatment strategies since many are in DNA repair genes. Withthe increasing availability of high-throughput DNA sequencing in cancer clinics andresearch, there is thus a need to provide clinically oriented sequencing reports forgermline variants and their potential therapeutic relevance on a per-patient basis. Tomeet this need, we have developed the Cancer Predisposition Sequencing Reporter(CPSR), an open-source computational workflow that generates a structured reportof germline variants identified in known cancer predisposition genes, highlightingmarkers of therapeutic, prognostic and diagnostic relevance. A fully automated vari-ant classification procedure based on more than 30 refined American College ofMedical Genetics and Genomics (ACMG) criteria represents an integral part of theworkflow. Importantly, the set of cancer predisposition genes profiled in the reportcan be flexibly chosen from more than 40 virtual gene panels established by scientificexperts, enabling customization of the report for different screening purposes andclinical contexts. The report can be configured to also list actionable secondary vari-ant findings, as recommended by ACMG. CPSR demonstrates comparable sensitivityand specificity for the detection of pathogenic variants when compared to otheralgorithms in the field. Technically, the tool is implemented in Python/R, and is freelyavailable through Docker technology. Source code, documentation, example reportsand installation instructions are accessible via the project GitHub page: https://github.com/sigven/cpsr.