When extended metabolites profiling helps the prediction of local relapse in glioblastoma: a retrospective study using MRI and MRSI
British Journal of Radiology · 19.10.2026
DOI: 10.1093/bjr/tqag247 ↗Kurz gefasst
Fachartikel in der Zeitschrift British Journal of Radiology, online erschienen am 19.10.2026. Verfasst von Wafae Labriji und 8 weiteren Autor:innen. Beteiligte Einrichtung: Toulouse INP, Univ Toulouse, IRIT, France. Die Arbeit nennt 32 Literaturangaben. Gefördert von: ARC foundation.
Abstract (Originalsprache)
Objectives
This study investigates whether integrating postoperative MRI with advanced MR spectroscopic imaging (MRSI) improves voxel-wise prediction of future local recurrence in glioblastoma (GBM).
Methods
Fifty-eight GBM patients included in the clinical trial Spectro-Glio were retrospectively analyzed with post-operative baseline for MRI and MRSI processed to generate voxel-level features. MRSI metabolites were quantified using a Bayesian sparse model, enabling reliable estimation of both conventional and extended metabolite panels. Ground-truth relapse labels were derived from longitudinal follow-up MRI to track voxel-wise tissue class changes over time. Voxel-wise relapse prediction was performed using 3 supervised classifiers (Elastic Net, Random Forest, RBF-SVM) evaluated with leave-one-patient-out cross-validation. Feature sets combining MRI intensities, spatial descriptors, and MRSI metabolites were systematically compared.
Results
MRI-only models showed moderate discrimination (AUC, ≤0.70). Using MRSI features alone significantly improved prediction performance, with further improvements when extending the metabolite basis beyond conventional markers (AUC, 0.77-0.82; P < .001). Spatial descriptors and MRSI-based models provided the largest performance gains (AUC up to 0.95). SHAP analysis consistently identified both conventional and secondary metabolites among the most influential predictors, particularly in the absence of spatial features.
Conclusions
Postoperative MRSI and spatial descriptors significantly enhances voxel-wise prediction of GBM relapse, particularly when using an extended metabolite basis beyond conventional markers. Advances in knowledge This study supports the clinical potential of MRSI alongside spatial features to identify relapse sites not captured by anatomical MRI alone. MRSI integration of extended metabolites provides insights for voxel-wise relapse prediction, supporting complementary characterization of tissue at risk.
Autor:innen und Institutionen
- Toulouse INP, Univ Toulouse, IRIT, France.
- Toulouse INP, Univ Toulouse, IRIT, ENSEEIHT, TéSA.
- Univ Toulouse, Oncopole Claudius Regaud, IUCT-Oncopole, Toulouse, France.
- Institute of Neuroradiology, University Magna Graecia, Catanzaro, Italy.
- Univ Toulouse, INSERM, CNRS, Centre de Recherches en Cancérologie de Toulouse, Team RADOPT, Toulouse, France.
Förderung
- ARC foundation
Zitieren
APA
Labriji, W., Aissaoui, A., Tourneret, J. Y., Chaari, L., Piram, L., Sabatini, U., Lemarié, A., Cohen-Jonathan, E. M., & Ken, S. (2026). When extended metabolites profiling helps the prediction of local relapse in glioblastoma: a retrospective study using MRI and MRSI. British Journal of Radiology. https://doi.org/10.1093/bjr/tqag247
Vancouver
Labriji W, Aissaoui A, Tourneret JY, Chaari L, Piram L, Sabatini U, et al. When extended metabolites profiling helps the prediction of local relapse in glioblastoma: a retrospective study using MRI and MRSI. British Journal of Radiology. 2026. doi:10.1093/bjr/tqag247
BibTeX
@article{Labriji2026,
author = {Labriji, Wafae and Aissaoui, Amel and Tourneret, Jean-Yves and Chaari, Lotfi and Piram, Lucie and Sabatini, Umberto and Lemarié, Anthony and Cohen-Jonathan, Elizabeth Moyal and Ken, Soléakhéna},
title = {When extended metabolites profiling helps the prediction of local relapse in glioblastoma: a retrospective study using MRI and MRSI},
journal = {British Journal of Radiology},
year = {2026},
doi = {10.1093/bjr/tqag247}
}
EUVIMED