Early prostate-specific antigen (PSA) kinetics following prostate carcinoma radiotherapy: Prognostic value of a time-and-PSA threshold model

BACKGROUND. The goal of the current study was to analyze the prognostic value of early prostate-specific antigen (PSA) kinetics, with PSA assessed as reaching or failing to reach discrete threshold values at fixed time points during follow-up after external-beam radiotherapy (EBRT) for prostate carcinoma. METHODS. The authors conducted a retrospective review of PSA follow-up for 839 patients treated between May 1987 and December 2000 at the Cleveland Clinic Foundation (Cleveland, OH). They also assessed the impact on bRFS of PSA levels lower than defined threshold values at given time points during follow-up. RESULTS. During a median follow-up of 74 months (range, 24-189 months), 540 patients (64.4%) maintained bRFS, whereas 299 patients (35.6%) did not maintain bRFS. The median nadir among patients with sustained BRFS was 0.4 ng/mL, with a median time to nadir of 28.9 months. Patients who did not maintain bRFS reached a median nadir of 1.3 ng/mL at a median of 15 months (P < 0.0001 for both nadir level and time to nadir). Reaching PSA thresholds of 3.0, 2.0, 1.0, 0.5, and 0.2 ng/mL at any time during follow-up was correlated with improved bRFS (P < 0.0001, each threshold). Patients whose PSA levels crossed the appropriate thresholds within 3 and 6 months of follow-up, irrespective of the time or level of eventual nadir, exhibited significantly improved bRFS when compared with patients whose PSA levels reached those thresholds at any time during follow-up and patients whose PSA levels never reached those thresholds (all thresholds: P < 0.0001). CONCLUSIONS. Despite previous conclusions that early PSA assessment may lack prognostic value, the data obtained in the current study suggest that the kinetics of early PSA decline is predictive of long-term bRFS when assessed using a time-and-PSA threshold model. After EBRT for prostate carcinoma, PSA levels below various discrete PSA thresholds were indicative of statistically meaningful long-term outcome differences between experimental arms as early as 90 days after radiotherapy. If the time-and-PSA threshold model is shown to be predictive of prostate carcinoma mortality as well, then it may allow the scientific community to evaluate promising treatment concepts and technologies at a highly accelerated pace.