Gene-based evidence for burden of rare pathogenic variants in pharmacogenes and oncogens of Czech breast cancer patients

摘要

Panel of 509 pharmacogenes and oncogenes was selected for targeted next generation sequencing. Data mining in databases and functional impact software tools were used as an evidence for potential pathogenicity. Survey of several in silico methods revealed significant differences in resultant set of genes mostly due to number of tools used. Following the most stringent approach with fully concordant prediction in comparison to specific combination of tools and majority vote prediction, ten prioritized genes were statistically correlated with the clinical patient data. The highest burden of pathogenic variants correlating with clinical data was statistically significant in CFTR gene. Patients carrying these variants had significantly reduced disease-free-survival in comparison with patients without these variants.