Machine learning predicts conventional imaging metastasis-free survival (MFS) for oligometastatic castration-sensitive prostate cancer (omCSPC) using prostate-specific membrane antigen (PSMA) PET radiomics

•We are the first to investigate using radiomic imaging biomarkers from both pre-and post-treatment PSMA-PET/CT together with clinical information for predicting outcomes in omCSPC patients undergoing MDT.•Multi-zone feature extraction – We extracted features from two distinct zones: zone 1 corresponds to the GTV, and zone 2 encompasses a 5 mm expansion ring area surrounding the GTV.•Multi-institutional validation − our study benefits from patient data collected from two institutions (Johns Hopkins Hospital and Baskent University), which validated the robustness and generalizability of our findings.•Predicting 2-year MFS with AUC 0.81 is feasible from PET radiomic information derived from the GTV and its surrounding area in both pre- and post-PSMA-PET scans for omCSPS patients. This study investigated imaging biomarkers derived from PSMA-PET acquired pre- and post-metastasis-directed therapy (MDT) to predict 2-year metastasis-free survival (MFS), which provides valuable early response assessment to improve patient outcomes. An international cohort of 117 oligometastatic castration-sensitive prostate cancer (omCSPC) patients, comprising 34 from John Hopkins Hospital (JHH) and 83 from Baskent University (BU), were treated with stereotactic ablative radiation therapy (SABR) MDT with both pre- and post-MDT PSMA-PET/CT scans acquired. PET radiomic features were analyzed from a CT-PET fusion defined gross tumor volume ((GTV) or zone 1), and a 5 mm expansion ring area outside the GTV (zone 2). A total of 1748 PET radiomic features were extracted from these zones. The six most significant features selected using the Chi2 method, along with five clinical parameters (age, Gleason score, number of total lesions, untreated lesions, and pre-MDT prostate-specific antigen (PSA)) were extracted as inputs to the models. Various machine learning models, including Random Forest, Decision Tree, Support Vector Machine, and Naïve Bayesian, were employed for 2-year MFS prediction and tested using leave-one-out and cross-institution validation. Six radiomic features, including Total Energy, Entropy, and Standard Deviation from pre-PSMA-PET zone 1, Total Energy and Contrast from post-PSMA-PET zone 1, and Entropy from pre-PSMA-PET zone 2, along with five clinical parameters were selected for predicting 2-year MFS. In a leave-one-out test with all the patients, random forest achieved an accuracy of 80 % and an AUC of 0.82 in predicting 2-year MFS. In cross-institution validation, th