The following is a summary of “Comparative analysis of genetic risk scores for predicting biochemical recurrence in prostate cancer patients after radical prostatectomy,” published in the July 2024 issue of Urology by Hsieh et al.
In recent years, Genome-Wide Association Studies (GWAS) have made significant strides in identifying genetic variants associated with complex diseases; however, the majority of these variants exert minimal effects on phenotypes. To address this challenge, innovative methodologies have been developed, such as genetic risk scores (GRS), which leverage variants with low genetic impact to enhance disease risk prediction.
In this study, researchers aimed to determine the most effective GRS model for predicting disease risk, focusing specifically on prostate cancer. The study group utilized both simulated datasets and actual prostate cancer data to evaluate the predictive power of three distinct GRS models: the simple count genetic risk score (SC-GRS), the direct logistic regression genetic risk score (DL-GRS), and the explained variance weighted GRS based on directed logistic regression (EVDL-GRS).
Investigators constructed GRS models using 26 single nucleotide polymorphisms (SNPs) to assess the risk of biochemical recurrence (BCR) following radical prostatectomy. By integrating clinical variables—including age at diagnosis, body mass index, prostate-specific antigen levels, Gleason score, pathologic T stage, and surgical margins—alongside the GRS models, they significantly enhanced the predictive accuracy for BCR. Findings revealed that the DL-GRS model demonstrated superior predictive performance, evidenced by a statistical power of 0.707 in simulated data and an area under the curve (AUC) of 0.8462 in the prostate cancer cohort. Furthermore, the SNP rs455192 emerged as the most significantly associated locus with BCR, achieving a p-value of 2.496 × 10-6 in the analysis.
In conclusion, the research highlights the efficacy of genetic risk scores in predicting biochemical recurrence in patients with prostate cancer following radical prostatectomy. Among the models assessed, the DL-GRS demonstrated the greatest predictive capability, suggesting its potential utility in clinical settings for risk stratification. By combining genetic risk assessment with clinical variables, this approach not only enhances the understanding of prostate cancer progression but also paves the way for more personalized patient management strategies.
Source: bmcurol.biomedcentral.com/articles/10.1186/s12894-024-01524-6
