Addition of quantitative imaging parameters to visual analysis improves the accuracy of PSMA PET/CT for the local staging of primary prostate cancer

Prostate-specific membrane antigen (PSMA) PET/CT is recognized as the most accurate imaging modality for staging of patients with intermediate and high-risk prostate cancer (PCa). PSMA PET/CT has also shown potential in the local (T) staging of primary PCa. The purpose of this study was to explore the value of quantitative PSMA PET/CT parameters in addition to the standard visual assessment for local T-stage classification in a large single-center retrospective cohort. Sequential intermediate- and high-risk primary PCa patients who underwent staging PSMA PET/CT prior to robot-assisted radical prostatectomy were included. Visual assessment of T-stage (miT-stage) was performed alongside quantitative analysis of PSMA PET/CT parameters, including SUVmax, SUVpeak, tumor volume (PSMA-vol), total lesion PSMA expression (PSMA-TL), and tumor longest capsule contact (LCC). The pathological tumor stage derived from radical prostatectomy specimens served as the reference standard. Univariable and multivariable logistic regression analyses were performed to develop clinical risk models for predicting pT3-stage disease. A total of 223 evaluable patients with PSMA-positive primary PCa were included. Univariable analyses of individual imaging parameters yielded AUCs of 0.53–0.63 for pT3a, 0.64–0.74 for ≥ pT3b, and 0.59–0.69 for overall ≥ pT3-stage. In multivariable analyses, LCC was the sole independent predictor for pT3a stage; miT-stage, LCC, and PSMA-vol were independent predictors for ≥ pT3b-stage; and LCC together with PSMA-vol were independent predictors for overall ≥ pT3-stage. Clinical risk models incorporating these predictors achieved AUCs of 0.62 for pT3a, 0.79 for ≥ pT3b, and 0.70 for ≥ pT3-stage. Quantitative parameters derived from PSMA PET/CT scans provide additional diagnostic accuracy for detecting extraprostatic tumor extension, particularly for ≥ pT3b-stage disease, outperforming visual assessment (miT-stage) alone.