ORIGINAL RESEARCH article
Front. Oncol.
Sec. Neuro-Oncology and Neurosurgical Oncology
Volume 15 - 2025 | doi: 10.3389/fonc.2025.1597925
Evaluating Treatment Strategies and Machine Learning Based Treatment Recommendation System for Elderly Patients with High Grade Gliomas
Provisionally accepted
- First Affiliated Hospital of Chongqing Medical University, Chongqing, China
Select one of your emails
You have multiple emails registered with Frontiers:
Notify me on publication
Please enter your email address:
If you already have an account, please login
You don't have a Frontiers account ? You can register here
Background: When selecting treatment strategies, elderly high-grade glioma (eHGG) patients face challenges due to aging, comorbidities, surgical complications, and limited tolerance for intensive treatments. This study aims to evaluate the benefit of treatment strategies and develop a treatment recommendation system for eHGG patients.Methods: By propensity score matching and survival analysis, we compared the prognosis of treatment strategies, including surgery versus none, adjuvant therapies versus none, and gross total resection (GTR) versus subtotal resection (STR), among patients aged 65 and older with high-grade gliomas. A machine learning model, random survival forest, was developed to provide predictions on prognosis. The machine learning model was then used to create a personalized treatment recommendation system. An independent validation cohort was obtained from the First Affiliated Hospital of Chongqing Medical University to validate the machine learning model and the treatment recommendation system. The timedependent AUC (tdAUC), C-index, and integrated Brier score (IBS) in the testing sets were obtained.Results: Compared to the surgery-alone group, patients who received surgery plus adjuvant therapy had significantly better overall survival. Surgery plus adjuvant therapy improved survival compared to adjuvant therapy alone. Additionally, GTR combined with adjuvant therapy showed superior overall survival compared to STR with adjuvant therapy. Subgroup analysis indicated that patients with GBM, tumor size >3 cm, localized stage, white race, Grade IV tumors, and those aged 65-72 had better survival outcomes with GTR and adjuvant therapy. The C-index, tdAUC, and 1-IBS values for the external testing cohort were 0.813, 0.876, and 0.893. We successfully developed a web-based treatment recommendation system at https://gliomas.shinyapps.io/EHGG/. This system allows users to input patient-specific features and obtain individualized treatment recommendations and detailed survival probabilities.Aggressive treatment, including GTR and adjuvant therapy, can enhance survival outcomes in elderly patients with high-grade gliomas. The machine learning-based personalized treatment recommendation system presents a promising reference tool for treatment decisions.
Keywords: Surgery, Gross total resection, Subtotal resection, adjuvant therapy, Treatment Recommendation
Received: 21 Mar 2025; Accepted: 09 Jul 2025.
Copyright: © 2025 Xiang, Fu and Yang. This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) or licensor are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms.
* Correspondence: Xuelian Yang, First Affiliated Hospital of Chongqing Medical University, Chongqing, China
Disclaimer: All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article or claim that may be made by its manufacturer is not guaranteed or endorsed by the publisher.