Outcomes of Grade Group 2 and 3 Prostate Cancer on Initial Versus Confirmatory Biopsy: Implications for Active Surveillance

Background: Active surveillance (AS) is recommended as the preferred treatment for men with low-risk disease. In order to optimize risk stratification and exclude undiagnosed higher-grade disease, most AS protocols recommend a confirmatory biopsy.Objective: We aimed to compare outcomes among men with grade group (GG) 2/3 prostate cancer on initial biopsy with those among men whose disease was initially GG1 but was upgraded to GG2/3 on confirmatory biopsy.Design, setting, and participants: We reviewed patients undergoing radical prostatectomy (RP) in two cohorts: "immediate RP group,"with GG2/3 cancer on diagnostic biopsy, and "AS group,"with GG1 cancer on initial biopsy that was upgraded to GG2/3 on confirmatory biopsy.Outcome measurements and statistical analysis: Probabilities of biochemical recurrence (BCR) and salvage therapy were determined using multivariable Cox regression models with risk adjustment. Risks of adverse pathology at RP were also compared using logistic regression.Results and limitations: The immediate RP group comprised 4009 patients and the AS group comprised 321 patients. The AS group had lower adjusted rates of adverse pathology (27% vs 35%, p = 0.003). BCR rates were lower in the AS group, although this did not reach conventional significance (hazard ratio [HR] 0.73, 95% confidence interval [CI] 0.50-1.06, p = 0.10) compared with the immediate RP group. Risk-adjusted 1- and 5yr BCR rates were 4.6% (95% CI 3.0-6.5%) and 10.4% (95% CI 6.9-14%), respectively, for the AS group compared with 6.3% (95% CI 5.6-7.0%) and 20% (95% CI 19-22%), respectively, in the immediate RP group. A nonsignificant association was observed for salvage treatment-free survival favoring the AS group (HR 0.67, 95% CI 0.42, 1.06, p = 0.087).Conclusions: We found that men with GG1 cancer who were upgraded on confirmatory biopsy tend to have less aggressive disease than men with the same grade found at initial biopsy. These results must be confirmed in larger series before recommendations can