On April 29, 2026 Caris Life Sciences (NASDAQ: CAI), a leading, patient-centric, next-generation AI TechBio company and precision medicine pioneer, reported the development and peer-reviewed validation of a new predictive signature to inform therapy selection in glioblastoma (GBM) patients. The study, published in Neuro-Oncology Advances, describes the development and evaluation of the model in a cohort of more than 5,800 GBM patients. The Caris AI Insights in Glioblastoma is featured on the Caris Molecular Tumor Board Report, an innovative profiling report that provides additional insight into tumor biology and is available upon request with no additional tissue required when ordering MI Cancer Seek.
Schedule your 30 min Free 1stOncology Demo!
Discover why more than 1,500 members use 1stOncology™ to excel in:
Early/Late Stage Pipeline Development - Target Scouting - Clinical Biomarkers - Indication Selection & Expansion - BD&L Contacts - Conference Reports - Combinatorial Drug Settings - Companion Diagnostics - Drug Repositioning - First-in-class Analysis - Competitive Analysis - Deals & Licensing
Schedule Your 30 min Free Demo!
Caris Life Sciences continues to advance precision oncology by integrating multimodal real-world datasets, available with its proprietary CodeAI platform, enabling the creation of Caris AI Insights, an engine that utilizes Whole Exome Sequencing (WES), Whole Transcriptome Sequencing (WTS), and clinical data to generate deeper biological understanding and more actionable insights for clinicians and researchers.
GBM is the most common and aggressive brain cancer, with patients typically surviving around 12 months despite treatment. Current treatments include surgery to remove the tumor, radiation and chemotherapy with temozolomide (TMZ). Unfortunately, nearly half of GBM patients do not respond to TMZ and many develop resistance that leads to recurrence. Determining whether TMZ will benefit the patient is a critical step in treatment and provides insight for developing new therapies for patients who do not respond favorably.
The novel signature, developed by Caris using multimodal molecular and clinical data, is an AI-derived model designed to infer O6-methylguanine-DNA methyltransferase (MGMT) promoter methylation status from NGS data. MGMT promoter methylation is an established biomarker associated with response to TMZ therapy in GBM patients that is commonly assessed using pyrosequencing. In the peer-reviewed validation study, the signature demonstrated high concordance with pyrosequencing-based MGMT assessment and improved discrimination of overall survival outcomes across MGMT-defined patient subgroups, while also addressing variability in MGMT classification associated with traditional testing methods. These findings show that Caris’ proprietary AI signature can complement existing testing approaches and provide additional clinical insight for isocitrate dehydrogenase (IDH)-wild type GBM patients treated with TMZ.
"The Caris AI Insights signature for GBM showcases Caris’ advanced AI capabilities in our pursuit of improving cancer patient outcomes," said Caris President David Spetzler, MS, PhD, MBA. "With peer-reviewed validation demonstrating strong concordance with traditional testing approaches and clearer prognostic stratification, we believe that this signature can complement existing testing methods to improve clinical insight for glioblastoma patients treated with TMZ."
The model was trained to predict MGMT promoter methylation status, as measured by a pyrosequencing assay, using tumor profiling data from the MI Cancer Seek NGS assay, which also has FDA-approved CDx indications. As reported in the study, the signature was developed using a clinico-genomic dataset of 5,841 patients and further evaluated in a prospective cohort of more than 3,400 cases. The model stratified patients into distinct survival groups based on the signature score, with higher scores associated with significantly longer overall survival in TMZ-treated patients.
(Press release, Caris Life Sciences, APR 29, 2026, View Source [SID1234664912])