Beyond Stages: Predicting Individual Time Dependent Risk for Type 1 Diabetes

Area of Science:

• Immunology
• Endocrinology
• Diabetes Research

Background:

• Multiple autoantibodies (AABs) are nearly universal predictors of clinical type 1 diabetes.
• Stage 1 diabetes is defined by multiple AABs and normal glucose tolerance; Stage 2 by abnormal glucose tolerance.
• Progression rates within these stages are highly variable, hindering clinical management.

Purpose of the Study:

• To develop adaptable, personalized risk calculators for type 1 diabetes progression.
• To improve clinical implementation by accounting for heterogeneous disease trajectories.

Main Methods:

• Developed three risk prediction models (small, medium, large) using TrialNet Pathway to Prevention data.
• Models incorporated autoantibody status, glucose levels, and oral glucose tolerance test predictors.
• Included variables like fasting glucose, hemoglobin A1c, age, and C-peptide levels.

Main Results:

• Models significantly enhanced personalized risk assessment compared to current staging.
• Risk predictions aligned with expected trends: lower risk with age, higher risk with elevated glucose and lower insulin secretion.
• Incorporating glucose and insulin secretion data refined risk probabilities within stages, especially for Stage 2.

Conclusions:

• Heterogeneity in type 1 diabetes progression is not explained by single risk factors.
• Model-based probabilities offer a personalized risk calculator for informed clinical monitoring decisions.
• These tools aid in determining optimal timing and methods for disease progression surveillance.