Supervised machine learning enables non-invasive lesion characterization in primary prostate cancer with [(68)Ga]Ga-PSMA-11 PET/MRI

PURPOSE: Risk classification of primary prostate cancer in clinical routine is mainly based on prostate-specific antigen (PSA) levels, Gleason scores from biopsy samples, and tumor-nodes-metastasis (TNM) staging. This study aimed to investigate the diagnostic performance of positron emission tomography/magnetic resonance imaging (PET/MRI) in vivo models for predicting low-vs-high lesion risk (LH) as well as biochemical recurrence (BCR) and overall patient risk (OPR) with machine learning. METHODS: Fifty-two patients who underwent multi-parametric dual-tracer [(18)F]FMC and [(68)Ga]Ga-PSMA-11 PET/MRI as well as radical prostatectomy between 2014 and 2015 were included as part of a single-center pilot to a randomized prospective trial (NCT02659527). Radiomics in combination with ensemble machine learning was applied including the [(68)Ga]Ga-PSMA-11 PET, the apparent diffusion coefficient, and the transverse relaxation time-weighted MRI scans of each patient to establish a low-vs-high risk lesion prediction model (M(LH)). Furthermore, M(BCR) and M(OPR) predictive model schemes were built by combining M(LH), PSA, and clinical stage values of patients. Performance evaluation of the established models was performed with 1000-fold Monte Carlo (MC) cross-validation. Results were additionally compared to conventional [(68)Ga]Ga-PSMA-11 standardized uptake value (SUV) analyses. RESULTS: The area under the receiver operator characteristic curve (AUC) of the M(LH) model (0.86) was higher than the AUC of the [(68)Ga]Ga-PSMA-11 SUV(max) analysis (0.80). MC cross-validation revealed 89% and 91% accuracies with 0.90 and 0.94 AUCs for the M(BCR) and M(OPR) models respectively, while standard routine analysis based on PSA, biopsy Gleason score, and TNM staging resulted in 69% and 70% accuracies to predict BCR and OPR respectively. CONCLUSION: Our results demonstrate the potential to enhance risk classification in primary prostate cancer patients built on PET/MRI radiomics and machine learning without biopsy sampling. SUPPLEMENTARY INFORMATION: The online version contains supplementary material available at 10.1007/s00259-020-05140-y.