Real-World Validation of the Clinical Distinctiveness of IDH-Mutant Glioblastoma in the SEER Transition Era: A Population-Based Study Integrating Machine Learning.

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.