Prostate biopsy status and PSA nadir level as early surrogates for treatment failure: Analysis of a prostate cancer randomized radiation dose escalation trial

Purpose: A positive biopsy after external beam radiotherapy in patients free of any evidence of treatment failure is not synonymous with eventual recurrence. Although biopsy positivity is a predictor of outcome, the utility of biopsy status as a surrogate end point, the effect of radiation dose on biopsy status, and the interrelationships of these associations to prostate-specific antigen (PSA) nadir level are not well-defined. These issues were investigated in a cohort of men with Stage T1-T3 prostate cancer who were randomized to receive between 70 Gy and 78 Gy and were prospectively biopsied at about 2 years after the completion of radiotherapy (RT). Methods and Materials: Of the 301 assessable patients in the trial, 168 underwent planned sextant or greater prostate post-RT biopsies in the absence of biochemical or clinical failure; this group constituted the study cohort. Of the 168 patients, 87 were in the 70-Gy arm and 81 in the 78-Gy arm. Biopsies were classified into four groups: negative (no tumor), atypical/suspicious cells (not diagnostic of carcinoma), carcinoma with treatment effect (CaTxEffect), and carcinoma without treatment effect (CaNoTxEffect). Any diagnosis of carcinoma in the specimen was classified as biopsy positive. Freedom from failure (FFF) included biochemical failure and/or clinical failure. Kaplan-Meier curves were calculated from the completion of RT. For those alive in the study cohort, the median follow-up was 65 months. Results: The rate of biopsy without tumor was 42%; with atypical cells, it was 28%, with CaTxEffect 21%, and with CaNoTxEffect 9%. The overall biopsy positivity rate (CaTxEffect + CaNoTxEffect) was 30%; 28% in the 70-Gy group and 32% in the 78-Gy group (p = 0.52). The distribution of PSA nadir levels was 73% less than or equal to 0.5, 20% > 0.5-1.0, 5% > 1.0 -2.0, and 1% > 2.0 ng/mL. Significantly more patients randomized to 78 Gy had a PSA nadir of less than or equal to 0.5 ng/mL (80 % vs. 67 %; p = 0.02). No relationship was found between PSA nadir level and prostate biopsy status. The 5-year FFF rate for those classified as biopsy negative was 84 % and for those biopsy positive was 60 % (p = 0.0002). Radiation dose did not significantly alter FFF rates by prostate biopsy status. Nadir PSA level correlated with FFF, although this was dependent on the inclusion of the 2 patients with a PSA nadir > 2.0 ng/mL. Conclusion: For patients free of treatment failure at the time of prostate biopsy 2 years after RT, the prognosis of no tumor cells was the same as that of atypical/suspicious cells and CaTxEffect was the same as CaNoTxEffect. The biopsy positivity rate was not altered by dose, suggesting that most of the outcome differences between the 70-Gy and 78-Gy groups were due to events occurring before prostate biopsy at 2 years and/or were not entirely dependent on biopsy status. Biopsy status,is a strong prognostic factor, but, as an early end point, it may be misleading. PSA nadir appears to have little clinical value in patients treated to doses of 2:70 Gy who are failure free 2 years after RT. (C) 2002 Elsevier Science Inc.