104 Molecular Subgroups of IDH Wild-Type Glioblastoma Exhibit Distinct Treatment Responses to Resection and Adjuvant Therapy, a Multi-Institutional Study

Le résumé fourni par la source

INTRODUCTION: IDH wild-type glioblastoma (GBM) remains a clinically heterogeneous disease with poor survival outcomes. Prior molecular subclassification schemes lack clinical utility. Here, we use semi-supervised consensus clustering to identify molecular subtypes with distinct prognoses and treatment responses. METHODS: Eight hundred and fifty-eight newly diagnosed, treatment-naïve IDH wild-type GBM patients with complete next-generation sequencing and MGMT methylation data were retrospectively identified from three institutions. Semi-supervised Monte Carlo consensus clustering (M3C) was applied within TERT and MGMT-defined strata. Kaplan-Meier and Cox regression analyses were used to evaluate survival. Bootstrapping with 10,000 iterations was used to identify optimal postoperative residual tumor volume thresholds. Multivariate models assessed treatment-response patterns within molecular subtypes. RESULTS: Five molecular subtypes were identified: TERT-M/MGMT-M EGFR-altered, TP53/PTEN-mutant, or CDKN2A loss; TERT-M/MGMT-U; and TERT-WT. Overall survival varied significantly between groups (p<0.0001). Residual contrast-enhancing tumor thresholds predictive of improved survival differed by group (range: 0.25–4.7 cc). Subtype-specific responses to treatment were observed on multivariate analysis: for TERT-M/MGMT-M tumors, EGFR-altered tumors benefited from temozolomide, radiation, and resection, TP53/PTEN-mutant tumors from radiation only, while CDKN2A-loss tumors did not benefit from standard therapy; TERT-M/MGMT-U tumors showed benefit from resection only; and TERT-WT tumors also showed limited response to any single modality. CONCLUSIONS: Molecular subgroups of IDH wild-type GBM show distinct prognostic and therapeutic profiles. Preoperative or intraoperative identification of these subtypes may refine surgical goals and personalize treatment strategies.

Ce résumé expose les affirmations des auteurs. BNTIC ne l’interprète pas comme une validation indépendante des résultats.