1618-P: Insulin Secretion Variability According to the Specific Type 1 Diabetes (T1D) Risk Predictor

The risk for T1D, a major criterion for entry into disease prevention trials, can be assessed in various ways, but usually includes autoantibody (Ab) and/or metabolic measures. Given inherent variations in these biomarkers, it is possible that informative characteristics differ between populations selected by different risk measures. Thus, we assessed whether indices of insulin secretion [C-pep; first phase insulin response (FPIR)] vary between groups defined by distinctive risk measures. We used data from Diabetes Prevention Trial-Type 1 (DPT-1) participants, all being ICA positive subjects with or without biochemical Ab (bAb) positivity (GADA, IA-2A, mIAA). C-pep and FPIR values from baseline oral and intravenous glucose tolerance tests were compared between those who had 2-3 bAbs (≥2bAb) and those who had 0-1 bAb plus Index60≥0.40 (≤1bAb+Ind60). Index60 is a composite C-pep and glucose measure. These groups were chosen so that T1D risk would be similar (50% 5-year risk for both, log rank p=0.95). Ages (mean±SD) for ≥2bAb (n=289) and ≤1bAb+Ind60 (n=175) were 12.8±8.2 and 13.3±9.6 yrs, respectively (p=0.53). BMI Z-score was greater for ≥2bAb (0.42±1.12 vs. 0.15±1.09, p=0.01). C-pep levels were consistently higher for ≥2bAb than ≤1bAb+Ind60 [30-0 minute C-pep: 2.73±1.58 vs. 2.04±1.12 ng/mL, p<0.0001; area under the curve (AUC) C-pep: 3.70±1.55 vs. 3.27±1.12 ng/ml, p=0.013; AUC C-pep/AUC glucose: 0.030±0.013 vs. 0.025±0.007 (ng/ml)/(mg/dl), p<0.001]. FPIR was also higher for ≥2bAb (122.0±87.3 vs. 96.2±47.1 µU/ml, p<0.01). All comparisons remained significant with adjustments for age and BMI Z-score. Of those with ≤1bAb+Ind60, 0 bAb did not differ from 1 bAb for C-pep indices or FPIR.

In conclusion, C-pep and FPIR values can differ appreciably according to the measure that defines risk. This should be considered in choosing risk measures for prevention trials.