Deep learning-based radiomics and machine learning for prognostic assessment in IDH-wildtype glioblastoma after maximal safe surgical resection: a multicenter study.

in International journal of surgery (London, England) by Jianpeng Liu, Shufan Jiang, Yanfei Wu, Ruoyao Zou, Yifang Bao, Na Wang, Jiaqi Tu, Ji Xiong, Ying Liu, Yuxin Li

TLDR

  • A radiomics-based machine learning model was developed using magnetic resonance imaging to predict overall survival in IDH-wildtype glioblastoma patients after maximal safe surgical resection, achieving robust performance and providing a non-invasive tool for personalized prognostic assessment.

Abstract

Glioblastoma (GBM) is a highly aggressive brain tumor with poor prognosis. This study aimed to construct and validate a radiomics-based machine learning model for predicting overall survival (OS) in IDH-wildtype GBM after maximal safe surgical resection using magnetic resonance imaging. A total of 582 patients were retrospectively enrolled, comprising 301 in the training cohort, 128 in the internal validation cohort, and 153 in the external validation cohort. Volumes of interest (VOIs) from contrast-enhanced T1-weighted imaging (CE-T1WI) were segmented into three regions: contrast-enhancing tumor, necrotic non-enhancing core, and peritumoral edema using an ResNet-based segmentation network. A total of 4,227 radiomic features were extracted and filtered using LASSO-Cox regression to identify signatures. The prognostic model was constructed using the Mime prediction framework, categorizing patients into high- and low-risk groups based on the median OS. Model performance was assessed using the concordance index (CI) and Kaplan-Meier survival analysis. Independent prognostic factors were identified through multivariable Cox regression analysis, and a nomogram was developed for individualized risk assessment. The Step Cox [backward] + RSF model achieved CIs of 0.89, 0.81, and 0.76 in the training, internal and external validation cohorts. Log-rank tests demonstrated significant survival differences between high- and low-risk groups across all cohorts (P < 0.05). Multivariate Cox analysis identified age (HR: 1.022; 95% CI: 0.979, 1.009, P < 0.05), KPS score (HR: 0.970, 95% CI: 0.960, 0.978, P < 0.05), rad-scores of the necrotic non-enhancing core (HR: 8.164; 95% CI: 2.439, 27.331, P < 0.05), and peritumoral edema (HR: 3.748; 95% CI: 1.212, 11.594, P < 0.05) as independent predictors of OS. A nomogram integrating these predictors provided individualized risk assessment. This deep learning segmentation-based radiomics model demonstrated robust performance in predicting OS in GBM after maximal safe surgical resection. By incorporating radiomic signatures and advanced machine learning algorithms, it offers a non-invasive tool for personalized prognostic assessment and supports clinical decision-making.

Overview

  • The study aimed to develop and validate a radiomics-based machine learning model for predicting overall survival in IDH-wildtype glioblastoma patients after maximal safe surgical resection using magnetic resonance imaging.
  • The model was constructed using a ResNet-based segmentation network to segment volumes of interest from contrast-enhanced T1-weighted imaging and extract 4,227 radiomic features, which were then filtered and categorized patients into high- and low-risk groups based on the median overall survival.
  • The primary objective of the study was to develop a non-invasive tool for personalized prognostic assessment and clinical decision-making in GBM patients.

Comparative Analysis & Findings

  • The Step Cox [backward] + RSF model achieved concordance indices of 0.89, 0.81, and 0.76 in the training, internal, and external validation cohorts, indicating robust performance in predicting overall survival.
  • Log-rank tests demonstrated significant survival differences between high- and low-risk groups across all cohorts (P < 0.05), suggesting the model's ability to accurately classify patients according to their predicted risk of death.
  • Multivariate Cox analysis identified age, KPS score, rad-scores of the necrotic non-enhancing core, and peritumoral edema as independent predictors of overall survival, providing insight into the biological significance of these features in GBM progression.

Implications and Future Directions

  • The radiomics-based model offers a non-invasive tool for personalized prognostic assessment and supports clinical decision-making in GBM patients, potentially improving treatment outcomes and patient quality of life.
  • Future studies should investigate the integration of the model with other prognostic biomarkers and therapies to enhance its accuracy and clinical utility.
  • The development of predictive models for other brain tumors and cancer types could further expand the potential implications of this study.
Read Full Article