Limitations of SUV_max-based interpretation of ¹⁸F-PSMA-1007 PET/MRI in patients with suspected prostate cancer

Objective To evaluate the diagnostic efficacy of ¹⁸F-prostate specific membrane antigen (PSMA)-1007 PET/MRI in patients with suspected prostate cancer (PCa) and explore its value in the diagnosis of PCa.

Methods A retrospective cohort analysis was conducted on clinical data and imaging data of 113 suspected PCa patients who underwent ¹⁸F-PSMA-1007 PET/MRI and puncture biopsy at the First Affiliated Hospital, Zhejiang University School of Medicine between March and December 2022. The patients' mean age was (65.4±7.7) years. Clinical indicators, biparametric magnetic resonance imaging (bpMRI) (hereafter referred to as MRI) imaging indicators, PSMA PET imaging indicators, combined imaging indicators, and comprehensive indicators of patients were obtained. Their diagnostic efficacies for PCa were compared and analyzed using area under the curve (AUC) of receiver operating characteristic curve. Binary Logistic regression model-based probability prediction was employed to classify age, serum total prostate specific antigen (tPSA), and prostate specific antigen density (PSAD) as clinical indicators; minimum apparent diffusion coefficient (ADC_min), mean apparent diffusion coefficient (ADC_avg), maximum apparent diffusion coefficient (ADC_max), and prostate imaging reporting and data system (PI-RADS) score as MRI imaging indicators; and maximum standardized uptake value (SUV_max) and molecular imaging PSMA (miPSMA) score as PET imaging indicators. MRI and PET imaging indicators were further integrated into combined imaging indicators, whereas clinical, MRI, and PET indicators were integrated into comprehensive indicators. The DeLong test was used to compare pairwise differences between indicators.

Results Among the 113 suspected PCa patients, 55(48.7%) were diagnosed with PCa by puncture biopsy, and 58(51.3%) were confirmed to have benign prostatic lesions through puncture biopsy and follow-up. The diagnostic efficacy of MRI indicators for PCa (AUC=0.943) was higher than that of PET imaging indicators (AUC=0.742) , with a statistically significant difference (Z=4.009, P<0.001), with the PI-RADS score being the most effective among MRI imaging indicators (AUC=0.911). Among PET imaging indicators, SUV_max had a higher diagnostic efficacy (AUC=0.822) than the miPSMA score (AUC=0.780), with a statistically significant difference (Z=1.972, P=0.049). The combined imaging indicators demonstrated higher diagnostic efficacy (AUC=0.946) than PET imaging indicators alone (AUC=0.742), with a statistically significant difference (Z=4.192, P<0.001). The comprehensive indicators showed higher diagnostic efficacy (AUC=0.948) than the combined imaging indicators (AUC=0.946), but the difference was not statistically significant (Z=0.477, P=0.633).

Conclusions In the interpretation of PSMA PET/MRI images for patients suspected of PCa, relying solely on the SUV_max from PET has limited diagnostic value. Combined imaging indicators can significantly improve the diagnostic efficacy of PCa.