Purpose: A subgroup of prostate cancer patients who have experienced biochemical relapse after radical retropubic prostatectomy (RRP) can benefit from radiation therapy to the prostate fossa. These patients demonstrate biochemical relapse secondary to local failure in the absence of distant failure. In order to define this subgroup, we investigated the impact of pathological and biochemical variables and pre-radiation therapy biopsy of the prostate fossa on biochemical disease-free survival (bNED) and initial prostate-specific antigen response.
Methods: Sixty-two patients with localized prostate cancer who had biochemical relapse after RRP were treated with post-RRP radiation therapy localized to the prostate fossa (median dose, 6120 cGy) and were subsequently followed up for a median time of 47 months. Cox regression analyses and Kaplan-Meier estimates for bNED were used to identify prognostic variables. The Fisher's exact test was used to test the interaction of initial prostate-specific antigen response with identified prognostic variables.
Results: Cox regression analysis of bNED as a function of pathological and biochemical parameters showed that only Gleason's score was a significant predictor of bNED. On univariate analysis, seminal vesicle involvement was also found to be a significant predictor. Prostate fossa biopsy result was not significantly related to bNED. Because of the overall high rates of biochemical failure, we wished to identify a high-risk subgroup that did not have local relapse as a component of biochemical relapse after RRP. We assessed initial biochemical response following radiation therapy as a surrogate for local relapse. A complete biochemical response was observed in 50% of patients, and a partial biochemical response was observed in an additional 34%, yielding an overall biochemical response rate of 84%. When stratified by Gleason's score, seminal vesicle, pre-radiation therapy prostate-specific antigen, and biopsy result, response rates greater than 50% were seen for all subgroups.
Conclusions: Gleason's score and seminal vesicle involvement predicted bNED after post-RRP radiation therapy in our cohort. Overall biochemical response rates were high in all subgroups, suggesting that all subgroups demonstrated a high likelihood of residual local disease as a component of failure. Pre-radiation therapy biopsy was predictive of neither bNED nor overall biochemical response.