Abstract 2234: Incidence of clinically significant prostate cancer after negative prostate MRI - comparison to general population

Abstract

Abstract Prostate MRI has high NPV (around 90%) for clinically significant prostate cancer (csPCa, ISUP grade group ≥ 2)1. However, clinicians are still faced with significant uncertainty when considering whether to avoid or to proceed to systematic biopsies in a man with negative MRI (nMRI). When making such an informed decision, information on the risk of subsequent csPCa diagnosis after nMRI in respect to risk of csPCa in general population may be beneficial. Hence, our objective was to compare incidence rate (IR) of csPCa and overdiagnosis rate (diagnosis of ISUP grade group 1 cancer, isPCa) after nMRI (PI-RADS 1-2) to outcomes in general population, and to assess the value of age and PSA-density in stratification. We utilized institutional and national registries to retrospectively identify 1,685 50-79-year-old PCa-free men who underwent a primary nMRI at our institution, of whom 906 men had PSA-D < 0.15 ng/ml/cm3. We compared the IR of csPCa and overdiagnosis rate in relation to general population (n=230,246) during follow-up (up to four years) under clinical routine since December 2015 (PIRADS v2 era). We calculated age-standardized IR ratios (IRR) of csPCa and odds ratios (OR) of overdiagnosis with 95% confidence intervals (95% CI) as age-pooled and among 10-year interval groups (50-59, 60-69 and 70-79 years). Follow-up times in general population cohort, nMRI cohort, and nMRI and low PSA-D cohort were 885,792, 3,363, and 1,821 years and IRs of csPCa were 406, 2,201, and 989 per 100,000 person years. In reference to general population, IRR of csPCa in nMRI cohort was 4.2 (95% CI 3.3-5.4) which reduced to 1.8 (95% CI 1.1-2.9) if PSA-D was < 0.15 ng/ml/cm3. ORs of isPCa were 3.0 (95% CI 2.0-4.6) and 4.6 (95% CI 2.2-9.5), respectively. IRRs of csPCa of 50-59, 60-69 and 70-79-year-old men were 14.1 (95% CI 10.5-25.4), 3.2 (95% CI 2.1-4.6) and 3.4 (95% CI 1.6-3.9). Limitations of this study include retrospective and registry-based design. In conclusion, the risk of subsequent csPCa after nMRI in young men is significant, when compared to population risk. However, with advancing age, these risks converge. For a subgroup of men with PSA-D < 0.15 ng/ml/cm3, the relative risk of csPCa is low in all age groups and comparable to population risk. At the same time, the risk of overdiagnosis after nMRI is substantially higher than in general population. In light of our results, elderly men with nMRI and men at all age groups with nMRI and PSA-D < 0.15 ng/ml/cm3 could be considered candidates for follow-up instead of systematic biopsy.