IPD Meta-Analysis of 5 RCTs: Pathological Features Predict Prostate Cancer Outcomes but Not ADT Benefit After Post-Prostatectomy Radiation

{"Investigators pooled individual data from five randomized trials comparing postoperative radiation alone versus radiation plus androgen deprivation therapy. An adverse feature score (0-4) summed high-grade disease, seminal vesicle invasion, extracapsular extension, and positive margins. Higher scores correlated with worse overall survival and metastasis-free survival, confirming prognostic value, but the score did not modify the hazard ratio for hormone therapy addition on either endpoint.","This result underscores the classic distinction between prognostic and predictive biomarkers. Traditional pathology captures tumor burden and aggressiveness yet fails to reflect androgen-receptor dependence or radiosensitization potential that would determine therapy response. The consistency of findings even in the low-PSA subgroup reinforces that risk stratification alone cannot guide selective use of hormone therapy and its attendant toxicities.","Future studies must therefore prioritize molecular classifiers, such as Decipher or AR-V7 assays, tested via similar individual-patient-data interaction analyses within the same trial cohorts. Until such predictive tools are validated, clinicians should weigh absolute baseline risk against hormone therapy side-effect profiles rather than assuming greater benefit for higher-risk pathology patients.","Evidence quality is high for the null interaction result because the analysis used randomized trial data with pre-specified endpoints; however, power for subgroup interactions and generalizability to contemporary dose-escalated or PSMA-guided radiation remain limited."}