MRI-Based Radiomics Prediction of Somatostatin Receptor Ligand Response in Acromegaly: An Automated Pipeline Study

Context Radiomics may capture tumor characteristics predicting therapeutic response. Objective To develop MRI-based radiomics pipeline for predicting response to SRLs in acromegaly. Methods MRI data from 267 subjects across three datasets were used to develop an automated tumor detection and segmentation pipeline (Spanish multicenter acromegaly IGTP cohort, n = 81; two public datasets, n = 186). The three-stage pipeline comprised: (i) automated tumor detection (YOLOv8), (ii) 3D segmentation (SegResNet), and (iii) radiomic feature extraction (PyRadiomics) with supervised classification. Treatment-response radiomics included contrast-enhanced T1-weighted (CE-T1W1) MRI from 49 patients treated with SRL. SRL response was defined as ≥50% IGF-1 reduction or normalization. Eleven machine-learning classifiers were evaluated using stratified repeated k-fold cross-validation. Results Automated segmentation achieved a Dice coefficient of 0.795 using combined CE-T1WI and T2WI, decreasing to 0.765 in the acromegaly-specific cohort. The best-performing model was a three-feature logistic regression classifier, with an out-of-fold AUC of 0.798 (95% CI: 0.671-0.918), balanced accuracy of 0.793, F1-score of 0.815, and correct classification of 39/49 patients (79.6%). Performance improved as the radiomic feature set was reduced from five to two. The most consistent feature was log-sigma-3-0-mm-3D_glcm_ClusterShade, which appeared in 98% of cross-validation folds and contributed most to the model-derived discriminative signal. Conclusion In this multicenter radiomics study of SRL response prediction in acromegaly, an automated pipeline achieved good discriminative performance (AUC = 0.798). These findings highlight both the promise and the current challenges of radiomics-based treatment prediction.