For men with recurrent prostate cancer after surgery, a persistent clinical question has been whether the severity of disease pathology at the time of operation should guide decisions about adding hormone therapy to salvage radiotherapy. A definitive answer to that question now emerges from one of the largest pooled analyses of randomized trial data in this disease space — and it challenges a widely held assumption.
Drawing on individual patient data from five phase 3 randomized trials encompassing 4,781 men with a median follow-up of 9.1 years, investigators constructed an Adverse Feature Count (AFC) — a composite score from zero to four, incorporating grade group 4–5 disease, seminal vesicle invasion, positive surgical margins, and extracapsular extension. The AFC proved independently prognostic: men with more adverse features faced meaningfully worse overall survival and metastasis-free survival. However, when the analysis tested whether the AFC modified the magnitude of benefit from adding hormonal therapy to postoperative radiotherapy, no significant interaction was detected. The interaction hazard ratio for overall survival was 0.93 (95% CI 0.79–1.10; p=0.40) and for metastasis-free survival 0.88 (95% CI 0.77–1.01; p=0.08), with consistent findings across categorical analyses and a restricted two-variable sensitivity score.
This finding carries important nuance for clinical practice. While pathologic staging clearly stratifies prognosis — AFC is a genuine risk marker — it does not appear to identify a subgroup that disproportionately benefits from hormonal intensification. The implication is that decisions about hormone therapy addition should not be driven by pathologic burden alone. This meta-analysis benefits from the gold standard of individual patient data pooling, which provides substantially more statistical power than aggregate analyses. A key limitation acknowledged is that PSA levels at radiotherapy initiation were capped at ≤0.5 ng/ml, restricting generalizability to the early salvage setting. Still, for a condition where hormone therapy carries meaningful side-effect burden — cardiovascular risk, bone loss, metabolic effects — the finding that pathologic complexity does not predict amplified benefit is both practically and paradigmatically significant.