Intraprostatic PSMA PET/MRI parameters and clinical markers for ISUP prediction and postoperative biochemical recurrence in patients with primary prostate cancer

PSMA PET/MRI pairs high sensitivity for PSMA-avid lesions with superior soft-tissue contrast and provides quantitative metrics that may predict clinical outcomes of prostate cancer (PCa). We investigated whether intraprostatic PET-derived metrics and PSA-based variables predict histopathological ISUP grade and biochemical recurrence (BCR) after radical prostatectomy (RPE). Men with primary PCa who underwent [⁶⁸Ga]PSMA PET/MRI before RPE were retrospectively analyzed. PET-derived tumor metrics, initial PSA (iPSA), PSA density (PSAD), and ISUP grade were assessed for association with tumor aggressiveness and BCR. Discriminatory performance was evaluated using ROC AUCs. Multivariable logistic regression retained the better-performing variable within correlated pairs (e.g., SUVpeak vs. SUVmax). Kaplan–Meier and Cox models examined recurrence-free survival. Among 101 men, all evaluated parameters differentiated low- (ISUP 1–2) from high-grade (ISUP 3–5) disease, with PET metrics showing the highest AUCs. In multivariable analysis, SUVpeak independently predicted high-grade disease (OR 1.3, 95% CI 1.1–1.5, p < 0.001). Follow-up was available for 71 patients (mean 3.2 ± 2.8 years); 36 (51%) developed BCR after 2.7 ± 2.6 years. For BCR prediction, all variables except SUVmax showed significant AUCs. In multivariable models, ISUP (OR 1.8, 95% CI 1.2–2.9, p = 0.01) and PSAD (OR 4.0, 95% CI 1.1–15.0, p = 0.039) remained significant. The combined model (ISUP + SUVpeak + PSAD) achieved an AUC of 0.788 (95% CI 0.678–0.898, p < 0.001). In Cox analysis, only high ISUP grade independently predicted time to BCR (HR 3.2, 95% CI 1.2–8.9, p = 0.02). Intraprostatic PET metrics, especially SUVpeak, effectively assess tumor aggressiveness before RPE, while PSAD adds prognostic value for recurrence. Postoperative ISUP grade remains the strongest predictor of BCR. Combined imaging and PSA-based markers may improve risk stratification and warrant validation in prospective studies.