Yu H, Shi Z, Lin X, et al. - In a clinical trial cohort, experts contrasted the performance of three polygenic risk score (PRS) methods for prostate cancer risk evaluation, including genetic risk score (GRS), pruning and thresholding (P + T), and linkage disequilibrium prediction (LDpred). A training process was needed to recognize the best P + T model (397 single nucleotide polymorphisms [SNPs]) and the LDpred model (3,011,362 SNPs). On the contrary, GRS was directly calculated on the basis of 110 established risk-related SNPs. In the testing population, for broad-sense validity, higher deciles were significantly related to greater perceived risk. Therefore, the performance of GRS was superior to P + T and LDpred. Moreover, for genetic testing at the individual level, fewer and well-established SNPs of GRS also make it more achievable and interpretable.