Prospective evaluation of the Kaiser Permanente prostate cancer risk calculator in biopsy-naïve men with mild PSA elevations.

Abstract

290 Background: To assess the impact on biopsy outcomes in biopsy-naïve men with mild PSA elevations (<10 ng/ml) when provided risk estimates using a novel prostate cancer risk calculator. Methods: This prospective study was conducted in urology departments in Kaiser Permanente Northern California in men with PSA elevations that were <10 ng/ml between March 2021 and March 2023. Patient variables (age, race, family history of prostate cancer, body mass index, and PSA level) were entered into our validated calculator. Risk estimates for high-grade cancer (Gleason score ≥7) and overall cancer were shared with the patient who accepted or declined the biopsy. We compared age, race, family history for prostate cancer, BMI, PSA level, predicted high-grade and overall cancer risks between those that accepted and those that declined biopsy. The outcomes of those who underwent biopsy were compared to historical data of biopsy-naïve men with PSA elevations <10 ng/ml who underwent biopsy in the three years prior to the Covid pandemic between January 2016 and December 2019 (n=5141). Results: After 24 months, the calculator was used to estimate risk for 1962 biopsy-naïve men with mild PSA elevations. 1455 (74%) accepted the biopsy. Those with a family history for prostate cancer were slightly more likely to accept biopsy. Age (continuous and categorical), and BMI (continuous and categorical) were not associated with decision for biopsy. Similarly, race/ethnicity when analyzed as four categories (White, Black, Asian, Hispanic) or as two categories (Black vs. all others) was not associated with decision for biopsy. The men who accepted the biopsy had higher predicted median high-grade cancer risks (24% [IQR 17-31%] vs. 20% [IQR 14-29%]; p<0.0001) and higher predicted median overall cancer risks (48% [IQR 40-56%] vs 45% [IQR 37-53%], respectively; p<0.0001) than those who declined the biopsy. Compared to the historical group, use of the risk calculator resulted in a biopsy population enriched with high-grade cancer: 29% vs 25%; similar rates of low-grade cancer: 24% vs 24%; and fewer negative biopsies: 47% vs. 51%, respectively; overall p=0.013. Conclusions: In biopsy-naïve men with mild PSA elevations, use of our risk calculator resulted in a biopsy population that had more high-grade cancer and fewer negative biopsies compared to a historical group. Only family history and higher predicted cancer risks by the calculator were associated with biopsy acceptance.