NetraMark Launches AI-Powered Study of Glioblastoma Clinical Data
Toronto based NetraMark Holdings has entered a collaboration with a U.S. academic medical center to apply its explainable artificial intelligence platform, NetraAI, to glioblastoma datasets.
The agreement provides NetraMark with access to longitudinal cerebrospinal fluid (CSF) proteomic data generated on the SomaLogic platform, together with related de-identified materials. The project aims to define explainable patient subgroups that could guide clinical trial design, biomarker discovery, and treatment strategies for glioblastoma, one of the most aggressive and treatment-resistant brain cancers.
NetraMark focuses on generative AI and machine learning solutions for the pharmaceutical industry. Its platform is designed to support drug developers by increasing the chances of clinical trial success through improved patient segmentation and trial enrichment strategies.
NetraAI, the company’s proprietary platform, is built to work with small and complex clinical datasets. Unlike AI methods that assign every patient to a predefined class, NetraAI separates data into explainable and unexplainable subsets. This design is meant to prevent “overfitting” by removing poorly correlated patient groups that can distort models and reduce predictive value.
From the explainable subsets, NetraAI generates “NetraPersonas”—coherent groupings of patients, variables, and statistical evidence that can be leveraged to identify treatment-relevant subpopulations, factors influencing therapy and placebo responses, and potential biomarkers of efficacy or adverse events.
According to NetraMark, glioblastoma has a median survival of approximately 15 months, with fewer than 7% of patients living beyond five years. Clinical trials in this indication fail more than 90% of the time, largely due to heterogeneous patient populations, poorly defined inclusion criteria, and a lack of predictive biomarkers. By applying NetraAI, NetraMark intends to generate decision-support tools that address these obstacles.
The specific objectives include:
- Glioblastoma vs. Non-Tumor Controls – Identify proteomic markers that distinguish glioblastoma samples from non-tumor controls.
- Glioblastoma vs. Other Brain Tumors – Differentiate glioblastoma from supratentorial brain metastases to refine glioma-specific biomarkers.
- Disease Recurrence – Compare primary and recurrent gliomas to detect molecular markers linked to recurrence and tumor evolution.
- Impact of Resection – Assess paired samples collected before and after surgery to understand molecular changes associated with resection.
- Impact of Chemoradiation – Analyze pre- and post-treatment samples to evaluate therapy response and resistance mechanisms.
- Impact of Immunotherapy – Examine lumbar CSF samples taken before and after immunotherapy to identify molecular effects of treatment.
We track collaborations like this weekly in Where Tech Meets Bio—our newsletter on startups, platforms, and deals at the intersection of biotech and digital.
Topic: AI in Bio
