343-OR: HOMA2ß Stratifies Risk for Type 1 Diabetes (T1D) Development among TrialNet Pathway to Prevention Participants

Identification of individuals at high risk for T1D is essential for the successful implementation of disease modifying interventions. Progression to T1D can be predicted by stimulated measures of β cell function, including oral glucose tolerance tests and hyperglycemic clamp studies. However, these assays are both time consuming and invasive. The homeostasis model assessment (HOMA2β) is a validated mathematical tool commonly used in type 2 diabetes cohorts to assess β cell function using fasting glucose and insulin measurements from a single time point. The utility of HOMA2β in predicting T1D progression has not been tested. To this end, HOMA2β at study entry was assessed in nondiabetic relatives of individuals with T1D who were positive for one or more β cell autoantibodies (aab) and followed longitudinally in the TrialNet Pathway to Prevention cohort. A total of 6598 subjects (mean age 17.2 ± 12.8 yrs) were screened and monitored for T1D. Of these, 40.2% (n=2652) were single aab+ (mean age 21.1 ± 14.0 yrs) and 57.5% (n=3794) were multiple aab+ (mean age 14.5 ± 11.2 yrs). During follow-up, 117 single and 1020 multiple aab+ individuals progressed to T1D. Kaplan-Meier survivor estimates by HOMA2β tertiles demonstrated that risk of T1D was inversely related to HOMA2β tertiles for multiple aab+ subjects. Multivariate Cox proportional hazards analysis was performed for the lowest tertile using the highest HOMA2β tertile as the reference group. After adjustments for age, sex, HLA, and BMI, T1D HRs were significant for the lowest HOMA2β tertile in single (HR 1.91, 95% CI 1.12, 3.24, p = 0.017), multiple (HR 2.43, 95% CI 2.03, 2.91, p<0.001), and combined single and multiple aab+ cohorts (HR 2.36, 95% CI 1.99, 2.79, p<0.001). Overall, low HOMA2β levels were able to identify a high-risk subgroup of PTP participants, suggesting that HOMA2β may have utility as a single time point measurement in stratifying individuals with established autoimmunity who are at increased risk of T1D development.