Combining prostate-specific antigen nadir and time to nadir allows for early identification of patients at highest risk for development of metastasis and death following salvage radiation therapy

Purpose: Little is known regarding the prognostic capability of prostate-specific antigen (PSA) nadir (nPSA) and time to nPSA (TnPSA) following salvage radiation therapy (SRT) for biochemical failure (BF) postradical prostatectomy (RP). We sought to assess their prognostic significance in this setting.

Methods and materials: A total of 448 patients who received SRT without androgen deprivation therapy at a single academic institution were included in this retrospective analysis. Univariate analysis and multivariate Cox proportional hazards models were used to assess BF, distant metastasis (DM), prostate cancer-specific death (PCSD), and overall survival (OS). A prognostic nomogram incorporating nPSA and TnPSA was developed and validated in randomly allocated training and validation cohorts.

Results: Median follow-up post-SRT was 64 months. Median nPSA and TnPSA were undetectable and 6.7 months, respectively. On univariate analysis, a detectable nPSA (P < .01) and TnPSA <6 months (P < .01) were predictive of all outcomes. In a training cohort, a 14-point nomogram incorporating detectable nPSA, TnPSA, Gleason score, pre-radiation therapy PSA, and seminal vesicle invasion predicted BF (hazard ratio[HR], 1.4; P < .0001), DM (HR, 1.3; P < .0001), PCSD (HR, 1.3; P < .0001), and decreased OS (HR, 1.2; P < .0001). Adding nPSA and TnPSA improved the prognostic value of the nomogram compared to using clinical predictors only. The nomogram was evaluated in a validation cohort where it was predictive of BF (c-index = 0.77), DM (0.73), and PCSD (0.69).

Conclusions: Patients with a detectable nPSA also having a TnPSA <6 months post-SRT are at high-risk for DM, PCSD, and decreased OS. These patients are unlikely to have clinically localized disease and should be considered for initiation of systemic therapies.