Prospective biopsy-controlled validation of an AI model for predicting glioblastoma infiltration: Results from the SupraGlio trial

Santiago Cepeda; Elena Hernando-Pérez; Enrique Pérez-Riesgo; Isabel Rodríguez-Valle; Olga Esteban-Sinovas; Ignacio Arrese; María Miguel Lucero-Salaverry; Tomás Zamora; María Ángeles Torres-Nieto; Luigi Tommaso Luppino; Samuel Kuttner; Marek Wodzinski; Trinidad Escudero; Jesús Garzón; Roberto Romero-Oraá; Roberto Hornero; Lucía Núñez; Carlos Villalobos; Rosario Sarabia

doi:10.1093/neuonc/noag088

Prospective biopsy-controlled validation of an AI model for predicting glioblastoma infiltration: Results from the SupraGlio trial

Neuro Oncol. 2026 Apr 20:noag088. doi: 10.1093/neuonc/noag088. Online ahead of print.

Authors

Santiago Cepeda^{1

2}, Elena Hernando-Pérez^{3

4

5}, Enrique Pérez-Riesgo³, Isabel Rodríguez-Valle³, Olga Esteban-Sinovas^{1

2}, Ignacio Arrese^{1

2}, María Miguel Lucero-Salaverry⁶, Tomás Zamora⁶, María Ángeles Torres-Nieto⁶, Luigi Tommaso Luppino⁷, Samuel Kuttner^{8

9}, Marek Wodzinski^{10

11}, Trinidad Escudero¹², Jesús Garzón¹², Roberto Romero-Oraá^{13

14}, Roberto Hornero^{13

14

15}, Lucía Núñez^{3

16}, Carlos Villalobos³, Rosario Sarabia^{1

2}

Affiliations

¹ Department of Neurosurgery, Río Hortega University Hospital, Valladolid, Spain.
² Specialized Group in Biomedical Imaging and Computational Analysis (GEIBAC), Instituto de Investigación Biosanitaria de Valladolid (IBioVALL), Valladolid, Spain.
³ Excellence Unit Institute of Biomedicine and Molecular Genetics of Valladolid (IBGM), University of Valladolid and Spanish National Research Council (CSIC).
⁴ Department of Cell Biology, Histology and Pharmacology, School of Medicine, University of Valladolid, Valladolid, Spain.
⁵ Health Sciences School, Miguel de Cervantes European University (UEMC), Valladolid, Spain.
⁶ Department of Pathology, Río Hortega University Hospital, Valladolid, Spain.
⁷ Norwegian Computing Center, Oslo, Norway.
⁸ PET Imaging Center, University Hospital of North Norway, Tromsø, Norway.
⁹ UiT Machine Learning Group, UiT The Arctic University of Norway, Tromsø, Norway.
¹⁰ Department of Measurement and Electronics, AGH University, Kraków, Poland.
¹¹ Sano Centre for Computational Medicine, Kraków, Poland.
¹² Department of Radiology, Río Hortega University Hospital, Valladolid, Spain.
¹³ Biomedical Engineering Group, University of Valladolid, Instituto de Investigación Biosanitaria de Valladolid (IBioVALL), Valladolid, Spain.
¹⁴ Center for Biomedical Research in Network of Bioengineering, Biomaterials and Nanomedicine (CIBER-BBN), Valladolid, Spain.
¹⁵ Institute for Research in Mathematics (IMUVA), University of Valladolid, Valladolid, Spain.
¹⁶ Department of Biochemistry and Molecular Biology and Physiology, School of Medicine, University of Valladolid, Valladolid, Spain.

PMID: 42010946
DOI: 10.1093/neuonc/noag088

Abstract

Background: Glioblastoma recurrence is driven by diffuse microscopic infiltration beyond the contrast-enhancing tumour margin. GlioMap is an open-access AI model predicting voxelwise infiltration and recurrence risk from multiparametric MRI. This prospective study aimed to validate GlioMap's biological accuracy and prognostic relevance through histopathological assessment, transcriptomic profiling, and survival analysis within the SupraGlio trial (NCT05735171).

Methods: Patients with newly diagnosed glioblastoma underwent neuronavigated biopsies targeting AI-predicted high-risk (HRoR) and low-risk of recurrence (LRoR) regions beyond the contrast-enhancing tumour. Histopathological infiltration served as the ground truth, and transcriptomic profiling characterised each region's molecular phenotype. Model performance was evaluated using accuracy and area under the receiver operating characteristic (ROC) curve (AUC). Survival analyses assessed the prognostic value of postoperative HRoR volume.

Results: Fifty-eight biopsies from 27 patients were analysed. GlioMap achieved 0.81 accuracy (95% CI, 0.71-0.91) and 0.84 AUC (95% CI, 0.73-0.93) for histologically confirmed infiltration. Transcriptomic analysis of 48 samples from 16 patients revealed progressive upregulation of invasion- and angiogenesis-related genes (CD44, CHI3L1, STAT3, VEGFA) and downregulation of neuronal markers (MBP, GABRA1) from LRoR to HRoR regions and tumour core, confirming a neural-to-mesenchymal gradient. Postoperative HRoR volume >1.6 cm³ predicted shorter overall survival (P = .04) and progression-free survival (P = .008).

Conclusions: To our knowledge, this study provides the first prospective, biopsy-controlled, molecular validation of an AI model for mapping glioblastoma infiltration. By accurately identifying histologically and transcriptionally infiltrated regions, GlioMap offers a biologically grounded imaging biomarker that could guide extended resection and personalised radiotherapy planning, potentially improving tumour control and patient outcomes.

Keywords: artificial intelligence; glioblastoma; infiltration; radiomics; recurrence.

© The Author(s) 2026. Published by Oxford University Press on behalf of the Society for Neuro-Oncology. All rights reserved. For commercial re-use, please contact reprints@oup.com for reprints and translation rights for reprints. All other permissions can be obtained through our RightsLink service via the Permissions link on the article page on our site—for further information please contact journals.permissions@oup.com.

Associated data

ClinicalTrials.gov/NCT05735171