# Diagnostic value of [18F]PSMA-1007 PET/CT based on PRIMARY score combined with mpMRI in clinically significant prostate cancer

## TL;DR

This study shows that combining [18F]PSMA-1007 PET/CT with mpMRI improves detection of clinically significant prostate cancer compared to individual methods.

## Contribution

The PRIMARY score, combining PET/CT and mpMRI, is introduced as a novel diagnostic tool for prostate cancer detection.

## Key findings

- The PRIMARY score achieved an AUC of 0.910 with 83.7% sensitivity and 58.5% specificity for detecting clinically significant prostate cancer.
- A predictive model combining TPSA, SUVmax, and PRIMARY score reached an AUC of 0.968 with 91.3% sensitivity and 84.6% specificity.
- The combined model outperformed SUVmax, PI-RADS v2.1, and PRIMARY score individually in diagnostic accuracy.

## Abstract

This study aimed to assess the diagnostic efficacy of the PRIMARY score, based on the 18F-labeled prostate-specific membrane antigen (PSMA-1007) positron emission tomography (PET)/computed tomography (CT) with multiparametric magnetic resonance imaging (mpMRI) PI-RADS, in detecting clinically significant prostate cancer (csPCa).

In this retrospective cohort study, 137 patients with suspected prostate cancer (PCa) underwent [18F]PSMA-1007 PET/CT and mpMRI before transrectal ultrasound (TRUS)-guided needle biopsy was performed. Patients were categorized into csPCa and non-csPCa groups based on histopathological findings. The diagnostic performance of total prostate-specific antigen (TPSA), maximum standardized uptake value (SUVmax), the standardized Prostate Imaging Reporting and Data System (PI-RADS v2. 1) of mpMRI, and the PRIMARY score was evaluated using receiver operating characteristic (ROC) curves. The area under the curve (AUC), sensitivity, and specificity were calculated. Factors with a P-value <0.05 from the univariate analysis were included in a binary logistic regression model to develop a predictive model. Differences in the AUCs for TPSA, SUVmax, PI-RADS v2.1, the PRIMARY score, and the combined model were compared using MedCalc software. Statistical significance was set at P<0.05.

Among the 137 patients evaluated, 67.2% (92) were in csPCa and 32.8% (45) in the non-csPCa group (15 with low-grade PCa [GS 3 + 3] and 30 with benign prostatic hyperplasia or acute or chronic prostatitis). TPSA, SUVmax, PI-RADSv2.1, and the PRIMARY score significantly differed between the two groups (P<0.013). The AUCs for TPSA, SUVmax, PI-RADSv2.1, and PRIMARY score were 0.699, 0.898, 0.878, and 0.910, respectively, with corresponding diagnostic sensitivities of 53.3%, 87.0%, 90.2%, and 83.7%, and specificities of23.0%, 65. 1%, 42.6%, and 58.5%, respectively. The predictive ROC curve analysis of the model revealed an AUC of 0.968, with 91.3% sensitivity, and 84.6% specificity. MedCalc analysis showed that the AUC of the model was superior compared with that of SUVmax, PI-RADS v2.1 Score, and the PRIMARY score. The difference was statistically significant (Z= 2.273, 3.485, 2.761; P=0.023, 0.000, 0.005).

The 5-grade PRIMARY score, derived from [18F]PSMA-1007 PET/CT in conjunction with the PI-RADSv2.1 score, offers enhanced discrimination of csPCa.