Predicting overall survival in glioblastoma patients using machine learning: an analysis of treatment efficacy and patient prognosis

IntroductionGlioblastoma (GBM), the most aggressive primary brain tumor, poses a significant challenge in predicting patient survival due to its heterogeneity and resistance to treatment. Accurate survival prediction is essential for optimizing treatment strategies and improving clinical outcomes.Me...

Full description

Saved in:
Bibliographic Details
Main Authors: Razvan Onciul, Felix-Mircea Brehar, Adrian Vasile Dumitru, Carla Crivoi, Razvan-Adrian Covache-Busuioc, Matei Serban, Petrinel Mugurel Radoi, Corneliu Toader
Format: Article
Language:English
Published: Frontiers Media S.A. 2025-04-01
Series:Frontiers in Oncology
Subjects:
Online Access:https://www.frontiersin.org/articles/10.3389/fonc.2025.1539845/full
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849701406472142848
author Razvan Onciul
Razvan Onciul
Felix-Mircea Brehar
Felix-Mircea Brehar
Adrian Vasile Dumitru
Adrian Vasile Dumitru
Carla Crivoi
Razvan-Adrian Covache-Busuioc
Razvan-Adrian Covache-Busuioc
Matei Serban
Matei Serban
Petrinel Mugurel Radoi
Petrinel Mugurel Radoi
Corneliu Toader
Corneliu Toader
author_facet Razvan Onciul
Razvan Onciul
Felix-Mircea Brehar
Felix-Mircea Brehar
Adrian Vasile Dumitru
Adrian Vasile Dumitru
Carla Crivoi
Razvan-Adrian Covache-Busuioc
Razvan-Adrian Covache-Busuioc
Matei Serban
Matei Serban
Petrinel Mugurel Radoi
Petrinel Mugurel Radoi
Corneliu Toader
Corneliu Toader
author_sort Razvan Onciul
collection DOAJ
description IntroductionGlioblastoma (GBM), the most aggressive primary brain tumor, poses a significant challenge in predicting patient survival due to its heterogeneity and resistance to treatment. Accurate survival prediction is essential for optimizing treatment strategies and improving clinical outcomes.MethodsThis study utilized metadata from 135 GBM patients, including demographic, clinical, and molecular variables such as age, Karnofsky Performance Status (KPS), MGMT promoter methylation, and EGFR amplification. Six machine learning models—XGBoost, Random Forests, Support Vector Machines, Artificial Neural Networks, Extra Trees Regressor, and K- Nearest Neighbors—were employed to classify patients into predefined survival categories. Data preprocessing included label encoding for categorical variables and MinMax scaling for numerical features. Model performance was assessed using ROC-AUC and accuracy metrics, with hyperparameters optimized through grid search.ResultsXGBoost demonstrated the highest predictive accuracy, achieving a mean ROC-AUC of 0.90 and an accuracy of 0.78. Ensemble models outperformed simpler classifiers, emphasizing the predictive value of metadata. The models identified key prognostic markers, including MGMT promoter methylation and KPS, as significant contributors to survival prediction.ConclusionsThe application of machine learning to GBM metadata offers a robust approach to predicting patient survival. The study highlights the potential of ML models to enhance clinical decision-making and contribute to personalized treatment strategies, with a focus on accuracy, reliability, and interpretability.
format Article
id doaj-art-b1932c2fa26f4b5b8f0fade376d99ed4
institution DOAJ
issn 2234-943X
language English
publishDate 2025-04-01
publisher Frontiers Media S.A.
record_format Article
series Frontiers in Oncology
spelling doaj-art-b1932c2fa26f4b5b8f0fade376d99ed42025-08-20T03:17:57ZengFrontiers Media S.A.Frontiers in Oncology2234-943X2025-04-011510.3389/fonc.2025.15398451539845Predicting overall survival in glioblastoma patients using machine learning: an analysis of treatment efficacy and patient prognosisRazvan Onciul0Razvan Onciul1Felix-Mircea Brehar2Felix-Mircea Brehar3Adrian Vasile Dumitru4Adrian Vasile Dumitru5Carla Crivoi6Razvan-Adrian Covache-Busuioc7Razvan-Adrian Covache-Busuioc8Matei Serban9Matei Serban10Petrinel Mugurel Radoi11Petrinel Mugurel Radoi12Corneliu Toader13Corneliu Toader14Department of Neurosurgery, “Carol Davila” University of Medicine and Pharmacy, Bucharest, RomaniaNeurosurgery Department, Emergency University Hospital, Bucharest, RomaniaDepartment of Neurosurgery, “Carol Davila” University of Medicine and Pharmacy, Bucharest, RomaniaDepartment of Neurosurgery, Clinical Emergency Hospital “Bagdasar-Arseni”, Bucharest, RomaniaDepartment of Neurosurgery, “Carol Davila” University of Medicine and Pharmacy, Bucharest, RomaniaDepartment of Pathology, University Emergency Hospital Bucharest, Carol Davila University of Medicine and Pharmacy, Bucharest, RomaniaDepartment of Computer Science, Faculty of Mathematics and Computer Science, University of Bucharest, Bucharest, RomaniaDepartment of Neurosurgery, “Carol Davila” University of Medicine and Pharmacy, Bucharest, RomaniaDepartment of Vascular Neurosurgery, National Institute of Neurology and Neurovascular Diseases, Bucharest, RomaniaDepartment of Neurosurgery, “Carol Davila” University of Medicine and Pharmacy, Bucharest, RomaniaDepartment of Vascular Neurosurgery, National Institute of Neurology and Neurovascular Diseases, Bucharest, RomaniaDepartment of Neurosurgery, “Carol Davila” University of Medicine and Pharmacy, Bucharest, RomaniaDepartment of Vascular Neurosurgery, National Institute of Neurology and Neurovascular Diseases, Bucharest, RomaniaDepartment of Neurosurgery, “Carol Davila” University of Medicine and Pharmacy, Bucharest, RomaniaDepartment of Vascular Neurosurgery, National Institute of Neurology and Neurovascular Diseases, Bucharest, RomaniaIntroductionGlioblastoma (GBM), the most aggressive primary brain tumor, poses a significant challenge in predicting patient survival due to its heterogeneity and resistance to treatment. Accurate survival prediction is essential for optimizing treatment strategies and improving clinical outcomes.MethodsThis study utilized metadata from 135 GBM patients, including demographic, clinical, and molecular variables such as age, Karnofsky Performance Status (KPS), MGMT promoter methylation, and EGFR amplification. Six machine learning models—XGBoost, Random Forests, Support Vector Machines, Artificial Neural Networks, Extra Trees Regressor, and K- Nearest Neighbors—were employed to classify patients into predefined survival categories. Data preprocessing included label encoding for categorical variables and MinMax scaling for numerical features. Model performance was assessed using ROC-AUC and accuracy metrics, with hyperparameters optimized through grid search.ResultsXGBoost demonstrated the highest predictive accuracy, achieving a mean ROC-AUC of 0.90 and an accuracy of 0.78. Ensemble models outperformed simpler classifiers, emphasizing the predictive value of metadata. The models identified key prognostic markers, including MGMT promoter methylation and KPS, as significant contributors to survival prediction.ConclusionsThe application of machine learning to GBM metadata offers a robust approach to predicting patient survival. The study highlights the potential of ML models to enhance clinical decision-making and contribute to personalized treatment strategies, with a focus on accuracy, reliability, and interpretability.https://www.frontiersin.org/articles/10.3389/fonc.2025.1539845/fullmachine learningprognostic biomarkersexplainable AIsurvival predictionclinical decision supportpersonalized medicine
spellingShingle Razvan Onciul
Razvan Onciul
Felix-Mircea Brehar
Felix-Mircea Brehar
Adrian Vasile Dumitru
Adrian Vasile Dumitru
Carla Crivoi
Razvan-Adrian Covache-Busuioc
Razvan-Adrian Covache-Busuioc
Matei Serban
Matei Serban
Petrinel Mugurel Radoi
Petrinel Mugurel Radoi
Corneliu Toader
Corneliu Toader
Predicting overall survival in glioblastoma patients using machine learning: an analysis of treatment efficacy and patient prognosis
Frontiers in Oncology
machine learning
prognostic biomarkers
explainable AI
survival prediction
clinical decision support
personalized medicine
title Predicting overall survival in glioblastoma patients using machine learning: an analysis of treatment efficacy and patient prognosis
title_full Predicting overall survival in glioblastoma patients using machine learning: an analysis of treatment efficacy and patient prognosis
title_fullStr Predicting overall survival in glioblastoma patients using machine learning: an analysis of treatment efficacy and patient prognosis
title_full_unstemmed Predicting overall survival in glioblastoma patients using machine learning: an analysis of treatment efficacy and patient prognosis
title_short Predicting overall survival in glioblastoma patients using machine learning: an analysis of treatment efficacy and patient prognosis
title_sort predicting overall survival in glioblastoma patients using machine learning an analysis of treatment efficacy and patient prognosis
topic machine learning
prognostic biomarkers
explainable AI
survival prediction
clinical decision support
personalized medicine
url https://www.frontiersin.org/articles/10.3389/fonc.2025.1539845/full
work_keys_str_mv AT razvanonciul predictingoverallsurvivalinglioblastomapatientsusingmachinelearningananalysisoftreatmentefficacyandpatientprognosis
AT razvanonciul predictingoverallsurvivalinglioblastomapatientsusingmachinelearningananalysisoftreatmentefficacyandpatientprognosis
AT felixmirceabrehar predictingoverallsurvivalinglioblastomapatientsusingmachinelearningananalysisoftreatmentefficacyandpatientprognosis
AT felixmirceabrehar predictingoverallsurvivalinglioblastomapatientsusingmachinelearningananalysisoftreatmentefficacyandpatientprognosis
AT adrianvasiledumitru predictingoverallsurvivalinglioblastomapatientsusingmachinelearningananalysisoftreatmentefficacyandpatientprognosis
AT adrianvasiledumitru predictingoverallsurvivalinglioblastomapatientsusingmachinelearningananalysisoftreatmentefficacyandpatientprognosis
AT carlacrivoi predictingoverallsurvivalinglioblastomapatientsusingmachinelearningananalysisoftreatmentefficacyandpatientprognosis
AT razvanadriancovachebusuioc predictingoverallsurvivalinglioblastomapatientsusingmachinelearningananalysisoftreatmentefficacyandpatientprognosis
AT razvanadriancovachebusuioc predictingoverallsurvivalinglioblastomapatientsusingmachinelearningananalysisoftreatmentefficacyandpatientprognosis
AT mateiserban predictingoverallsurvivalinglioblastomapatientsusingmachinelearningananalysisoftreatmentefficacyandpatientprognosis
AT mateiserban predictingoverallsurvivalinglioblastomapatientsusingmachinelearningananalysisoftreatmentefficacyandpatientprognosis
AT petrinelmugurelradoi predictingoverallsurvivalinglioblastomapatientsusingmachinelearningananalysisoftreatmentefficacyandpatientprognosis
AT petrinelmugurelradoi predictingoverallsurvivalinglioblastomapatientsusingmachinelearningananalysisoftreatmentefficacyandpatientprognosis
AT corneliutoader predictingoverallsurvivalinglioblastomapatientsusingmachinelearningananalysisoftreatmentefficacyandpatientprognosis
AT corneliutoader predictingoverallsurvivalinglioblastomapatientsusingmachinelearningananalysisoftreatmentefficacyandpatientprognosis