PSA density as a dynamic prognostic marker of biopsy grade progression during active surveillance

The safety and efficacy of active surveillance (AS) for men with prostate cancer (PCa) depend on accurate risk stratification at diagnosis and during follow-up. Initial PSA density (PSAD) has been associated with the risk of AS progression when measured at diagnosis, yet few studies have examined the impact of changes in PSAD during follow-up. We investigated whether serial PSAD measurements during AS were prognostic of subsequent biopsy grade progression.

We queried our institutional AS database to identify men with Grade Group 1 PCa, ≥ 2 prostate biopsies, and ≥ 2 PSAD measurements at least 1 year apart. The median follow-up time was 5.9 years. The primary outcome was grade progression on subsequent AS biopsy. The utility of trending PSAD during AS was evaluated using 2 models: PSAD change over time (PSAD velocity) in a cross-validated logistic regression and serial PSAD measurements in a Cox proportional hazards model with time-varying covariates.

A total of 453 patients were included, of whom 137 met the primary outcome of biopsy pathological grade progression. Among the rich multivariate model, serial PSAD measurements had the highest concordance index (0.75) as a prognostic marker of biopsy progression. Patients with a net positive PSAD velocity (increasing PSAD over AS) had an OR of 1.97 (95% CI 1.26-3.11) and those with a PSAD value ≥0.15 ng/ml² at any point during AS had a HR of 3.70 (95% CI 2.45-5.60).

Our data suggest that serial PSAD measurements and PSAD velocity are strong independent prognostic markers of biopsy grade progression in AS. As prostate volume and PSA measurements are already part of AS protocols, monitoring PSAD trends over time confers no additional cost and should be integrated into clinical practice to improve risk stratification.