Large-Scale Proteomics Improve Risk Prediction for Type 2 Diabetes.

Area of Science:

• Biomarker Discovery
• Proteomics
• Epidemiology

Background:

• Type 2 diabetes (T2D) poses a significant global health challenge.
• Accurate risk prediction is crucial for timely intervention and prevention strategies.
• Existing clinical risk scores may not fully capture individual T2D susceptibility.

Purpose of the Study:

• To evaluate the added predictive value of proteomic biomarkers for 10-year T2D risk.
• To assess the performance of proteomic panels when integrated with the Cambridge Diabetes Risk Score (CDRS).

Main Methods:

• Utilized data from UK Biobank (n=21,898) for internal validation and ESTHER cohort (n=4,454) for external validation.
• Employed Olink Explore (2,085 proteins) and Olink Target 96 Inflammation panel (73 proteins) for proteomic profiling.
• Assessed incremental predictive value using C-index and net reclassification improvement.

Main Results:

• The 15-protein Olink Explore model improved the CDRS C-index by 0.029 (23.0% net reclassification) in internal validation.
• The 6-protein Olink Inflammation panel model improved the CDRS C-index by 0.016 (29.0% net reclassification) internally.
• External validation of the 6-protein model showed a C-index improvement of 0.014.

Conclusions:

• Proteomic biomarkers significantly enhance T2D risk prediction beyond the CDRS.
• The 15-protein Olink Explore model demonstrates the greatest potential for improved risk assessment.
• The successfully externally validated 6-protein Inflammation panel model offers a promising, targeted approach for T2D risk stratification.