Genomic predictors of tumor-treating fields response in glioblastoma: STAG2 mutation is associated with lack of benefit in a real-world analysis
SNO 2026
Uriel Kim, Uloma Nwaolu, Christina Thai, Edward Williams, Metamia Ciampricotti, Adam Cohen
Abstract
Tumor-treating fields (TTFields) improve survival in glioblastoma; however, there are no validated biomarkers to guide patient selection. Identifying genomic predictors could refine treatment strategies and distinguish patients who derive meaningful benefit from those who do not. This analysis evaluates the associations between glioblastoma gene mutation status and TTField-related survival outcomes in a large real-world database. The Tempus Lens Platform (Tempus AI, Inc., Chicago, IL) was used to query the Tempus multimodal de-identified database and establish and subsequently analyze a cohort of TERT-mutated glioblastoma patients treated with temozolomide. Ten predefined genes of interest were selected based on prior studies (PIK3CA, TP53, EGFR, NF1, PTEN, CDKN2A, CDK4, RB1, MDM2, PDGFRA), and genes with pathogenic DNA variants in more than 10 patients were included. Overall survival (OS) was assessed using Kaplan-Meier methods and Cox proportional hazards models. The cohort included 890 patients (TTFields: n = 211; control: n = 679) with comparable baseline characteristics. TTFields treatment was associated with significantly improved OS compared to controls (22.4 vs. 17.6 months; p = 0.04). In this real-world cohort, no differential survival benefit with TTFields was observed for any of the prespecified genes. In contrast, there was a significant interaction between TTFields exposure and STAG2 mutation status (p = 0.02, HR = 2.68). Among control patients, OS was similar regardless of STAG2 status (14.7 months vs 18.1 months). However, among TTField-treated patients, those with STAG2-mutant tumors had significantly shorter OS than those with wild-type STAG2 (12.3 months vs 22.7 months). Common glioblastoma-associated genes do not predict benefit from TTFields. STAG2 mutation may identify a subset of patients less likely to benefit from TTFields, suggesting a potential role as a predictive biomarker pending validation in independent cohorts.
Related publications