Purpose: The goals of this study were to quantify the frequency of post-treatment prostate-specific antigen (PSA)-level bouncing following three-dimensional conformal radiation therapy (3D-CRT) for prostate cancer and to identify any relationships that may exist between bouncing activity and biochemical control (bNED). Methods: Between May 1989 and July 1995, 306 patients were treated with 3D-CRT alone. All patients had 6 or more post-treatment PSA levels and at least 5 years of PSA follow-up. The median total follow-up and total dose to the center of prostate was 79 months and 74 Gy, respectively. A bounce was defined by a minimum rise in PSA of 0.4 ng/mL over a 6-month period, followed by a drop in PSA of any magnitude. Estimates of bNED control rates were made using Kaplan-Meier methodology and comparisons were made using the log-rank test. Multivariate analysis of bNED control predictors was accomplished using a stepwise Cox proportional hazards model. Results: Nearly one third of the patients experienced at least one bounce. Bouncers were found to present with higher pretreatment PSA levels and were treated with lower dose levels to the center of prostate. Five-year bNED control estimates for nonbouncers vs. bouncers were 69% and 52%, respectively (p = 0.0024). After controlling for dose and pretreatment PSA level, total number of bounces emerged as a significant predictor of bNED control (p = 0.02). Conclusions: pouncing PSA levels occur in approximately one third of the patients treated with 3D-CRT alone, with bouncing occurring at a constant rate from 2 to 5 years post-treatment. Bouncing is associated with lower radiation dose levels, higher pretreatment PSA levels, and decreased bNED control. Nearly half of the bouncers are bNED controlled; thus, clinicians should not use bouncing as a sole indicator of relapse. (C) 2001 Elsevier Science Inc.
