Natural History After Likely Cure Versus Recurrence After Prostate Radiotherapy (RAPTOR)

Background Whether freedom from biochemical recurrence (BCR) at 5 yr after definitive radiotherapy (RT) for localized prostate cancer (PCa) represents a durable cure remains uncertain.Objective To evaluate the association between the timing of BCR after RT and the long-term risks of distant metastasis (DM) and PCa-specific mortality (PCSM), overall and according to treatment intensity, including outcomes by BCR timing (≤5 yr, >5 yr, or no BCR).Design, setting, and participants Post hoc individual patient data analysis of 3120 men with intermediate- (89%) or high-risk (11%) PCa enrolled in 11 randomized trials within the Meta-Analysis of Randomized Trials in Cancer of the Prostate consortium and treated with contemporary high-dose RT (HDRT) ± androgen deprivation therapy (ADT).Outcome measurements and statistical analysis Cumulative incidences of DM at 8 yr and PCSM at 10 yr after RT were estimated using competing-risk methods. Adjusted cumulative incidences were estimated using average marginal effects, accounting for clinicopathologic factors, treatment intensity, and progressively longer durations remaining free of BCR after RT. A priori, cumulative incidences ≤10% were considered indicative of a favorable long-term prognosis.Results and limitations Among men without BCR at 5 yr, subsequent risks of DM and PCSM were low overall. In patients with intermediate-risk PCa treated with HDRT alone, the adjusted cumulative incidences of 8-yr DM and 10-yr PCSM were 6.6% and 2.5%, respectively, indicating a favorable long-term prognosis based on the prespecified threshold. In contrast, patients with intermediate-risk PCa treated with HDRT plus short-term ADT and patients with high-risk PCa treated with HDRT plus long-term ADT had higher residual metastatic risk (8-yr adjusted cumulative DM incidence of 11% and 16%, respectively) exceeding the predefined favorable prognosis threshold despite remaining BCR-free for 5 yr. Early BCR (≤5 yr) was associated with substantially higher risks of DM and PCSM across all treatment groups. Limitations include the retrospective nature of the analyses, heterogeneity in BCR definitions across trials, absence of comorbidity data and contemporary risk subclassification, and the pre-prostate-specific membrane antigen imaging era of the constituent trials.Conclusions A 5-yr BCR-free benchmark was associated with a favorable long-term prognosis for intermediate-risk PCa treated with RT alone, but it appears insufficient for patients requiring ADT intensification. External validation in contemporary cohorts with standardized BCR definitions and modern imaging is warranted.