Importance of implant dosimetry for patients undergoing prostate brachytherapy

To evaluate the disease and treatment-related factors for predicting biochemical freedom from recurrence (BFR) in patients with clinically localized prostate cancer undergoing permanent prostate brachytherapy.

Between November 1992 and June 1998, 883 consecutive patients with T1-T2 prostate cancer underwent permanent prostate brachytherapy. Computed tomography-based dosimetry was performed, and the minimal dose to 90% of the prostate volume relative to the prescribed dose (D(90)) was calculated. BFR was defined as three prostate-specific antigen (PSA) rises from nadir, with patients having one or two PSA rises censored early. Follow-up was calculated by censored events. Kaplan-Meier actuarial outcome was determined, and multivariate Cox regression analysis was performed to assess the significance of the D(90), initial PSA value, Gleason score, addition of external beam radiotherapy, addition of hormonal therapy, and isotope selection.

The mean follow-up was 55 months (range 3 to 125). The 10-year BFR rate was 79.1%. Cox proportional analysis identified D(90) as a predictor of BFR (P <0.0001), along with Gleason score, initial PSA level, and clinical stage (P = 0.001, P = 0.001, and P = 0.011, respectively). The addition of external beam radiotherapy, hormonal therapy, and isotope selection did not have an impact on BFR (P = 0.128, P = 0.399, and P = 0.224, respectively).

The quality of permanent prostate brachytherapy as measured by the D(90) was the most significant predictor for BFR in this study cohort at 10 years. Furthermore, adding external beam radiotherapy and/or hormonal therapy as adjuvant therapies did not independently predict for BFR. Overall, the reported 10-year BFR rates in this study were favorable. Strategies for ensuring the best quality implant should be used and, when reporting brachytherapy outcomes, the implant quality should be noted.