Pharmacogenomics polygenic risk score for drug response prediction using PRS-PGx methods

• DOI: 10.1038/s41467-022-32407-9
• Publication Date: 2022-09-08T11:02:41Z
• Authors: Song Zhai; Hong Zhang; Devan V. Mehrotra; Judong Shen
• Keywords: Pharmacogenomics; Polygenic risk score; Drug response

Abstract Polygenic risk scores (PRS) have been successfully developed for the prediction of human diseases and complex traits in past years. For drug response randomized clinical trials, a common practice is to apply PRS built from disease genome-wide association study (GWAS) directly corresponding pharmacogenomics (PGx) setting. Here, we show that such an approach relies on stringent assumptions about prognostic predictive effects selected genetic variants. We propose shift PGx approaches by simultaneously modeling both further make this possible developing series PRS-PGx methods, including novel Bayesian regression (PRS-PGx-Bayes). Simulation studies methods generally outperform PRS-PGx-Bayes superior all other methods. GWAS data large cardiovascular trial (IMPROVE-IT) predict treatment related LDL cholesterol reduction. The results demonstrate substantial improvement accuracy capability capturing treatment-specific while compared with approaches.