Does Multiparametric Magnetic Resonance of Prostate Outperform Risk Calculators in Predicting Prostate Cancer in Biopsy Naïve Patients?

Abstract

BackgroundEuropean Association of Urology (EAU) guidelines recommend using risk-calculators (RCs), imaging or additional biomarkers in asymptomatic men at risk of prostate cancer (PCa).ObjectivesTo compare the performance of mpMRI, a RC we recently developed and two commonly used RC not including mpMRI in predicting the risk of PCa, as well as the added value of mpMRI to each RC.Design, Setting, and ParticipantsSingle-center retrospective study evaluating 221 biopsy-naïve patients who underwent prebiopsy mpMRI.Outcome Measurements and Statistical AnalysisPatients’ probabilities of any PCa and clinically significant PCa (csPC, defined as Gleason-Score ≥3 + 4) were computed according to mpMRI, European Randomized Study of Screening for Prostate Cancer RC (ERSPC-RC), the Prostate Biopsy Collaborative Group RC (PBCG-RC) and the Foggia Prostate Cancer RC (FPC-RC). Logistic regression, AUC, and Decision curve analysis (DCA) were used to assess the accuracy of tested models.Results and LimitationThe FPC-RC outperformed mpMRI in diagnosing both any PCa (AUC 0.76 vs 0.69) and csPCa (AUC 0.80 vs 0.75). Conversely mpMRI showed a higher accuracy in predicting any PCa compared to the PBCG-RC and the ERSPC-RC but similar performances in predicting csPCa. At multivariable analysis predicting csPCa and any PCa, the addition of mpMRI findings improved the accuracy of each calculator. DCA showed that the FPC-RC provided a greater net benefit than mpMRI and the other RCs. The addition of mpMRI findings improved the net benefit provided by each calculator.ConclusionsmpMRI was outperformed by the novel FPC-RC and showed similar performances compared to the PBCG and ERSPC RCs in predicting csPCa. The addition of mpMRI findings improved the diagnostic accuracy of each of these calculators