Cure8 regulatory brief
This project aims to improve prediction of autoimmune disease onset and immunotherapy-related autoimmunity by combining genetics and plasma proteins.
For people with IBD, better risk models could eventually enable earlier detection or identify who might be at higher risk for immune-related toxicities from cancer immunotherapy.
Researchers in autoimmune disease and proteomics, clinicians who treat IBD or manage cancer immunotherapy side effects, and patients/researchers interested in biomarker-driven risk prediction.
This NIH-funded project plans to build predictive models that combine genetic risk (polygenic risk scores) with plasma proteomics to forecast risk of chronic autoimmune diseases and immune-related adverse events (irAE) from cancer immunotherapy.
The team will train and benchmark machine-learning models using large biobank proteomics (UK Biobank Pharma Proteomics Project) and test transferability to a prospective cohort of patients receiving immune checkpoint inhibitors.
The proposal emphasizes integrating static genetic susceptibility with dynamic protein measurements to improve prediction over either approach alone; preliminary data reportedly show improved predictive performance for inflammatory bowel disease when combining proteomic and genetic scores.
Methods include interpretable regression, nonlinear neural networks, and self-supervised proteomic embedding (masked autoencoder) to support predictions in smaller cohorts. This is a funded research project (project-record) describing planned work and aims, not study results or clinical recommendations.
It focuses on methodological development and biomarker discovery with potential future clinical application for earlier identification of high-risk individuals and defining shared vs disease-specific biomarkers.
This record describes planned and funded research rather than completed, peer-reviewed findings. Any clinical use would require further validation, replication, and regulatory review. The project leverages UK Biobank proteomics and a prospective cancer cohort to test generalizability.
Review the original publication for the complete reporting, methods, and context.
Funding disclosed by the source: National Institute of Allergy and Infectious Diseases - F31AI203561 - $50,114
This Cure8 brief is based on source text from the linked article. Cure8 is informational only and is not a substitute for professional medical advice, diagnosis, or treatment.
Genetic ancestry can influence IBD risk and disease course, and most prior studies focused on European populations.
Identifies gene and microbiome patterns linked to progression from inflammation to irreversible bowel fibrosis in Crohn’s disease, which could eventually help predict who is at higher risk and inform prevention...
A validated AI biomarker that predicts which IBD treatments are likely to work could reduce trial-and-error in treatment choices, potentially shortening time to effective therapy for people with Crohn's disease or...
