Advanced machine learning strategies for predicting therapy response in preclinical glioblastoma using longitudinal MRI
Abstract
Glioblastoma (GB) is the most aggressive primary brain tumor, characterized by a poor prognosis, limited response to therapy, and high rates of recurrence. Early therapeutic response assessment is challenging due to phenomena such as pseudoresponse and pseudoprogression. This study explores the potential of advanced machine learning (ML) strategies to predict long-term therapy outcomes using longitudinal T2weighted Magnetic Resonance Imaging (MRI) data from a preclinical GL261 glioblastoma mouse model, acquired prior and during treatment. We compare two distinct approaches: a classical pipeline based on radiomic features coupled with an XGBoost classifier, and a deep learning (DL) pipeline using a fine-tuned EfficientNetB0 model. Our results demonstrate that while the radiomics approach identifies interpretable imaging biomarkers and achieves good predictive performance (AUC ≈ 0.770, 95% CI 0.703–0.832), the DL-based model outperforms it across most evaluation metrics, reaching an AUC of 0.868 (95% CI 0.810– 0.918) and a sensitivity of 0.818. The DL model shows better generalization across individual subjects, and the discriminative performance improves progressively throughout the follow-up period for both approaches. Interpretability analysis via Grad-CAM confirms that the DL model’s predictions are driven by anatomically relevant features. These findings suggest that DL, enhanced by transfer learning, has the potential to serve as a powerful non-invasive tool for the early prediction of treatment efficacy in glioblastoma, paving the way for more robust and personalized therapy monitoring.
Related
- 2 Preclinical modeling of human glioblastoma immunotherapy resistance using a STING pathway defective mouse model
- Combined HDAC and ATM inhibition with radiation in preclinical glioblastoma models
- An Orthotopic Xenograft Model for Preclinical Evaluation of CAR T Cell Therapy in Glioblastoma.
- Ferroptosis-Mediated Radiosensitization in Glioblastoma: Preclinical Insights and Translational Opportunities
- Personalized Therapeutic Strategies in Glioblastoma: Bridging Molecular Markers and In Vivo Observations for Targeted Treatment
BibTeX
@article{gonzalez2026advanced,
title = {{Advanced machine learning strategies for predicting therapy response in preclinical glioblastoma using longitudinal MRI}},
author = {Javier González and Ana Paula Candiota and Alfredo Vellido},
journal = {Scientific Reports},
year = {2026},
month = oct,
doi = {10.1038/s41598-026-73929-2},
url = {https://www.nature.com/articles/s41598-026-73929-2},
note = {CC BY 4.0},
}
RIS
TY - JOUR
TI - Advanced machine learning strategies for predicting therapy response in preclinical glioblastoma using longitudinal MRI
AU - Javier González
AU - Ana Paula Candiota
AU - Alfredo Vellido
PY - 2026
DA - 2026/10/04
JO - Scientific Reports
DO - 10.1038/s41598-026-73929-2
UR - https://www.nature.com/articles/s41598-026-73929-2
AB - Glioblastoma (GB) is the most aggressive primary brain tumor, characterized by a poor prognosis, limited response to therapy, and high rates of recurrence. Early therapeutic response assessment is challenging due to phenomena such as pseudoresponse and pseudoprogression. This study explores the potential of advanced machine learning (ML) strategies to predict long-term therapy outcomes using longitudinal T2weighted Magnetic Resonance Imaging (MRI) data from a preclinical GL261 glioblastoma mouse model, acquired prior and during treatment. We compare two distinct approaches: a classical pipeline based on radiomic features coupled with an XGBoost classifier, and a deep learning (DL) pipeline using a fine-tuned EfficientNetB0 model. Our results demonstrate that while the radiomics approach identifies interpretable imaging biomarkers and achieves good predictive performance (AUC ≈ 0.770, 95% CI 0.703–0.832), the DL-based model outperforms it across most evaluation metrics, reaching an AUC of 0.868 (95% CI 0.810– 0.918) and a sensitivity of 0.818. The DL model shows better generalization across individual subjects, and the discriminative performance improves progressively throughout the follow-up period for both approaches. Interpretability analysis via Grad-CAM confirms that the DL model’s predictions are driven by anatomically relevant features. These findings suggest that DL, enhanced by transfer learning, has the potential to serve as a powerful non-invasive tool for the early prediction of treatment efficacy in glioblastoma, paving the way for more robust and personalized therapy monitoring.
SN - 2045-2322
ER -