Real‐World Validation of the Clinical Distinctiveness of IDH‐Mutant Glioblastoma in the SEER Transition Era: A Population‐Based Study Integrating Machine Learning
Dewei Du, Amu Jike, Dongnan Yu, Shizheng Tan, Shuming Wang, Shun LiABSTRACT
Introduction
The 2021 WHO classification reclassified “IDH‐mutant glioblastoma (GBM)” as “Astrocytoma, IDH‐mutant, grade 4.” This study aims to provide real‐world validation of this reclassification using the specific ICD‐O‐3 code (9445/3) from the Surveillance, Epidemiology, and End Results (SEER) “Transition Era” (2018–2022) and develop a machine learning (ML)‐based prognostic model.
Methods
Patients diagnosed with IDH‐mutant GBM (9445/3) and GBM NOS (9440/3) were identified. Propensity Score Matching (PSM) and Inverse Probability of Treatment Weighting (IPTW) were employed to minimize bias. A doubly robust Cox regression model was constructed to quantify survival benefits. Nine ML algorithms were integrated to develop a prognostic signature, which was interpreted using SHAP (Shapley Additive exPlanations) analysis.
Results
Of 13,443 patients, 312 were IDH‐mutant. After matching, the IDH‐mutant group exhibited significantly superior overall survival (OS) and cancer‐specific survival (CSS) ( p < 0.001). IDH mutation emerged as a potent independent favorable prognostic factor, associated with a 65.0% lower mortality risk (HR = 0.350, p < 0.001). Subgroup analysis confirmed robust benefits from chemotherapy. The Random Forest (RF) model achieved the best performance (Test AUC = 0.698). SHAP analysis identified IDH status, chemotherapy, and age as the top predictors.
Conclusion
This study provides compelling evidence in support of the clinical rationale for the WHO 2021 reclassification. Despite a favorable prognosis, aggressive multimodal therapy was strongly associated with improved survival, though potential indication bias necessitates cautious interpretation and prospective validation. The developed ML model serves as a robust tool for personalized risk stratification.