Contemporary risk of biochemical recurrence after radical prostatectomy in the active surveillance era

Objectives To assess whether contemporary risks of biochemical recurrence (BCR) after radical prostatectomy (RP) in the AS era differ from historical estimates due to changes in tumor risk case mix and improvements in risk stratification. Materials and methods We sampled 6,682 men who underwent RP for clinically localized disease between 2000 and 2017 from the VA SEARCH database. Kaplan Meier analysis was used to calculate incidence of BCR before and after 2010 overall and within tumor risk subgroups. Multivariable Cox proportional hazard regression analysis including an interaction term between era and tumor risk was used to compare risk of BCR before and after 2010 overall and across tumor risk subgroups. Results About 3,492 (52%) and 3,190 (48%) men underwent RP before and after 2010, respectively. In a limited multivariable model excluding tumor risk, overall BCR risk was higher post-2010 vs. pre-2010 (HR: 1.15, 95%CI: 1.05–1.25; 40% vs 36% at 8 years post-RP). However, this effect was eliminated after correcting for tumor risk (HR: 0.95, 95%CI: 0.87–1.04), suggesting that differences in tumor risk between eras mediated the change. Yet, within tumor-risk subgroups, BCR risk was significantly lower for favorable intermediate-risk (HR: 0.76, 95%CI:0.60–0.96) and unfavorable intermediate-risk PC (HR: 0.78, 95%CI: 0.67–0.90), but significantly higher for high-risk PC (HR: 1.22, 95%CI: 1.07–1.38) in the post-2010 era. 8-year risks of BCR in the post-2010 era were 21% (95%CI: 16%–25%), 25% (95%CI: 20%–30%), 41% (95%CI: 37%–46%), and 60% (95%CI: 56%–64%) for low-, FIR-, UIR-, and high-risk disease, respectively. Limitations include limited long-term follow-up in the post-2010 subgroup. Conclusions Overall BCR risk has increased in the AS era, driven by a higher risk case mix and increased BCR risk among high-risk patients. Physicians should quote contemporary estimates of BCR when counseling patients.