A simple model predicts freedom from biochemical recurrence after low-dose rate prostate brachytherapy alone

The objective of this study was to describe a simple model that predicts freedom from biochemical recurrence (FFBR) in men with prostate cancer after treatment with low-dose rate prostate brachytherapy (LDRPB) alone.

One hundred thirty-two men were treated with LDRPB alone between September 1997 and April 2001. Sixty-four percent of men had low-risk disease (prostate-specific antigen [PSA] <10, Gleason <7, and T stage or =10, Gleason > or =7, or T stage T2b). The dosimetric quantifier D90 was calculated from a computed tomography scan performed 1 month after LDRPB. The percent positive biopsies (PPB) were determined for all patients. FFBR was estimated using the product limit method. All P values are 2-sided.

The median follow-up is 65 months. The median D90 is 138 Gy (range, 47-221 Gy). Fourteen men have developed evidence of biochemical relapse at a median of 27 months (range, 6-42 months). The 5-year FFBR rate for the entire cohort is 88%. On univariate analysis, variables found to be associated with FFBR included: PSA, Gleason score, T stage, risk group, PPB, and D90. Multivariate analysis indicated that D90, PPB, and risk group were independently associated with FFBR. Patients were categorized based on the following 3 adverse prognostic factors: D90 <140 Gy, PPB > or =50%, and intermediate-risk group. Group 1 (0 factors, n = 30), group 2 (1 factor, n = 72), and group 3 (> or =2 factors, n = 30) patients had 5-year FFBR rates of 100% (+/-0%), 92% (+/-6%), and 67% (+/-18%) (P < 0.0001).

We have developed a simple, robust model based on implant quality and disease factors that predicts FFBR in men with prostate cancer treated with LDRPB alone.