Caris Life Sciences Validates AI Model for Temozolomide Response in Glioblastoma

Caris Life Sciences announced in a press release the peer-reviewed validation of an AI-based predictive signature designed to identify glioblastoma patients likely to benefit from temozolomide therapy. The study, published in *Neuro-Oncology Advances*, evaluated the model using data from more than 5,800 patients.

The Caris AI Insights in Glioblastoma model predicts O6-methylguanine-DNA methyltransferase (MGMT) promoter methylation status from next generation sequencing data, a biomarker associated with response to temozolomide. In validation testing, the AI signature showed high concordance with pyrosequencing-based MGMT assessment and improved discrimination of survival outcomes across MGMT-defined subgroups.

The model was trained on a dataset of 5,841 patients and prospectively evaluated in over 3,400 additional cases. Higher model scores correlated with longer overall survival in temozolomide-treated patients, suggesting the approach can complement existing methods to refine therapy selection for glioblastoma.

The AI signature is available through the Caris Molecular Tumor Board Report and can be requested with no additional tissue when ordering the MI Cancer Seek assay.