Toward a PSMA PET–mpMRI Pathway for Biopsy Decision-Making in Men with Suspected Prostate Cancer: Interim Results from the Prospective BIOPSTAGE Trial

Monica Celli¹Roberta Gunelli²Fabio Ferroni¹Federica Matteucci^1*Matteo Costantini²Domenico Barone¹Lorenzo Fantini¹Eugenia Fragalà²Irene Marini¹Valentina Di Iorio¹Flavia Foca¹Manuela Monti¹Paola Caroli¹

• ¹Istituto Scientifico Romagnolo per lo Studio e la Cura dei Tumori (IRST) IRCCS, Meldola, Italy
• ²Azienda Unita Sanitaria Locale della Romagna, Ravenna, Italy

Introduction: The BIOPSTAGE trial investigates the added diagnostic value of combining prostate-specific membrane antigen positron emission tomography (PSMA PET) with multiparametric MRI (mpMRI) for non-invasive detection of clinically significant prostate cancer (csPCa). This interim analysis aimed to retrospectively optimize semiquantitative PSMA PET thresholds and evaluate potential biopsy reduction through integrated imaging. Methods: In this prospective, single-institution study, 96 men with suspected prostate cancer underwent mpMRI and PSMA PET prior to biopsy. csPCa was defined as ISUP grade group ≥2. Receiver operating characteristic (ROC) analysis identified optimal thresholds for SUVmax (≥9.1) and SUVratio (≥3.6). Diagnostic performance metrics (sensitivity, specificity, predictive values, and biopsy burden) were evaluated for mpMRI, PSMA PET, and their combination. Logistic regression and McNemar's test assessed predictors and comparative accuracy. Results: mpMRI alone achieved a sensitivity of 93.8%, specificity of 54.6%, and an AUC of 0.74. Among the 74 lesions identified by mpMRI, 14 also met semiquantitative PET positivity criteria. An additional 11 PET-positive lesions were detected without corresponding mpMRI abnormalities. Another 60 lesions were mpMRI-positive but PET–negative and would not have been targeted in an integrated strategy. If biopsy had been guided exclusively by PSMA PET– positive lesions, 25 targets would have been sampled—representing an approximately 66.2% reduction compared to 74 lesions identified by mpMRI alone. This biopsy reduction was associated with significantly improved specificity (McNemar's p < 0.001). Integrated imaging diagnostic accuracy was: sensitivity 85.7% (95% CI: 60.1–96.0%), specificity 86.3% (95% CI: 78.0–91.8%), PPV 48.0%, NPV 97.6%. Two small csPCa lesions missed by PSMA PET were PI-RADS 4, ISUP Grade Group 2. One csPCa lesion detected exclusively by PSMA PET was not visible on mpMRI (ISUP Grade Group 3). Conclusions: Integrating semiquantitative PSMA PET with mpMRI improves lesion-level specificity and was associated with a theoretical reduction of approximately two-thirds in the number of targeted biopsy cores, while maintaining high sensitivity. While PSMA PET enhances risk stratification, mpMRI remains essential—particularly for detecting small or PSMA-negative tumors. These interim findings support feasibility of a dual-imaging biopsy triage strategy, pending validation in the full BIOPSTAGE cohort and future multicenter trials.