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

Abstract 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.

Authors

Institutions

Publication Details

Journal
Journal of the Endocrine Society
Published
2026-08-25
DOI
https://doi.org/10.1210/jendso/bvag198
Primary Topic
Pituitary Gland Disorders and Treatments
Type
article
Field-Weighted Citation Impact
0.00
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
article

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

Joan Gil, Jordi Ferri, I Simón, M. Ángeles Gálvez et al.
Journal of the Endocrine Society
Pituitary Gland Disorders and Treatments
article

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

Joan Gil, Jordi Ferri, I Simón, M. Ángeles Gálvez, Jose Antonio Rosado, Soledad Librizzi, Silvana Sarria‐Estrada, Roger Mateu‐Estivill, Elena Valassi, Núria Vilarrasa, Gemma Xifra, Marta Araujo‐Castro, Jesús Alejandro Alzate-Grisales, Olga Giménez‐Palop, Esteban Delgado, Guillermo Serra, Silvia Martínez-Couselo, Montserrat Marqués-Pamies, Fabiola Romero, María de la Iglesia-Vayá, Manel Puig‐Domingo, Roxana Zavala, Paz de Miguel, Cristina Carrato, Rogelio García-Centeno, Sabina Ruiz, Betina Biagetti, Felicia A. Hanzu, María Calatayud, Mireia Jordà, Cristina Álvarez-Escolá, Miguel Sampedro, Concepción Blanco, Isabel Pavón, Andreu Simo-Servat, Ignacio Bernabéu, Manel Puig-Domingo, Rocío Villar-Taibo, Paloma Puyalto, Carmen Fajardo-Montañana, Betina Biagetti, Edelmiro Menéndez, Gemma Monté‐Rubio, Anna Aulinas, Antonio Picó, Federico Vázquez, Victor Navas, Eider Pascual, Paula de Pedro, Mónica Marazuela, Cristina Lama, Raúl Luque, Ariadna Izquierdo-Gómez, Miriam López-Máñez, Susan M Webb, Fernando Guerrero, Silvana Sarria, Marta Hernández, Marta Araujo-Castro, Mireia Mora, Elena Valassi
article en

Abstract

Abstract 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.

Journal of the Endocrine Society
Hebron University (PS), Universitat Autònoma de Barcelona (ES), Instituto Cajal (ES), Universitat Internacional de Catalunya (ES), Fundación para el Fomento de la Investigación Sanitaria y Biomédica de la Comunitat Valenciana (ES), Diabetes & Endocrine Associates (US), Vall d'Hebron Hospital Universitari (ES), Hospital Universitario de La Princesa (ES), Institut d'Investigació en Ciències de la Salut Germans Trias i Pujol (ES), Hospital Universitari Germans Trias i Pujol (ES), Hospital Universitario Ramón y Cajal (ES), IPS Central (PY)
Reduced inequalities
Openalex Percentile: Top 10%
Pituitary Gland Disorders and Treatments
DOIOpen Access PDFOpenAlex
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.