MRI-derived T stage predicts nodal positivity on PSMA PET/CT in high-risk primary prostate cancer

We evaluated whether Magnetic Resonance Imaging (MRI)-Derived T Stage independently predicts nodal disease on Prostate-Specific Membrane Antigen (PSMA) PET/CT in patients with high-risk primary prostate cancer, and developed a predictive model for clinical decision support.

MRI-derived T Stage, Prostate-Specific Antigen (PSA), Gleason score, age and PSMA node status were analysed using univariable and multivariable logistic regression in 152 men with high-risk primary prostate cancer who underwent pre-treatment multiparametric MRI and PSMA PET/CT. Model performance was assessed with area under the receiver operating characteristic (ROC) curve (AUC), calibration plots, and decision curve analysis. Subgroup analyses evaluated Gleason 3 + 4 and 4 + 3 disease separately.

MRI T3b and T4 stages were independently and significantly associated with increased risk of PSMA-positive nodal disease (p < 0.001 and p = 0.016, respectively). T3a disease was not significantly independently associated with PSMA-positive nodal disease (p = 0.086). T3b remained an independent predictor even in Gleason 4 + 3 cases (p = 0.026; odds ratio ≈ 23.57). The final model (AUC = 0.863) was well-calibrated and demonstrated clinical net benefit on decision curve analysis. In our cohort, patients with PSA < 20 ng/mL, MRI-derived T2 disease, and Gleason score < 8 (including both 3 + 4 and 4 + 3 patterns) demonstrated a 0% rate of PSMA-positive nodal disease.

MRI-derived T stage T3b disease is a strong independent predictor of PSMA-positive nodal disease. These findings support guideline-based staging criteria and support the integration of MRI-derived T Stage into PSMA PET/CT triage workflows.