Incorporating Lower Urinary Tract Symptoms to Predict the Risk of Positive Prostate Magnetic Resonance Imaging.

Background and objective: Many men referred from primary care entering magnetic resonance imaging (MRI)-based prostate cancer (PCa) pathways have an MRI without abnormalities, highlighting the need to improve MRI risk stratification. Optimising this process in primary care could reduce unnecessary referrals and MRIs. This study aimed to develop a prediction model incorporating the International Prostate Symptom Score (IPSS) to improve MRI risk stratification and reduce hospital referrals and MRIs in biopsy-naïve men with suspected PCa in primary care. Methods: We prospectively identified men with suspected PCa referred from primary care to a Dutch teaching hospital in 2022-2023. Standard work-up included IPSS and upfront biparametric MRI. Study outcomes included Prostate Imaging Reporting and Data System (PI-RADS) ≥ 4, the number of potentially reduced hospital referrals and subsequent MRIs, and missed PCa cases. Men with an abnormal digital rectal examination (DRE) were excluded from model development, as they have a direct indication for MRI according to current guidelines. Multivariable logistic regression identified predictors of PI-RADS ≥ 4. Model performance was assessed using the area under the curve (AUC), and clinical utility was evaluated with decision curve analysis using a predefined threshold probability of 20%. Key findings and limitations: Of 409 men, 334 without abnormal DRE were included in the model development cohort; 30% (101/334) had PI-RADS ≥ 4, of whom 59% (60/101) had significant PCa (International Society of Urological Pathology [ISUP] grade group ≥ 2). Prostate-specific antigen (PSA) and IPSS were independent predictors of PI-RADS ≥ 4. The model showed fair discrimination (AUC = 0.68; 95% confidence interval [CI] = 0.62-0.74). Decision curve analysis showed greater net benefit than 'treat none' across all thresholds and greater net benefit than 'treat all' between 17% and 30%. At a 20% threshold, 23% of MRIs could be avoided, while 3.9% of all patients would not be referred despite having a positive MRI (13% of positive MRI findings), including eight cases of missed significant PCa. Limitations include the lack of external validation. Conclusions and clinical implications: Incorporating IPSS in primary care for biopsy-naïve patients with suspected PCa improves risk stratification for MRI and offers an easily available parameter to optimise diagnostic pathways.