Abstract
Even after stratification based on established risk factors, the progression to T1D is variable. Thus, we aimed to examine differences in metabolic patterns of change between NP and P to T1D, and identify non-metabolic factors contributing to the differences. Specifically, we analyzed OGTT and autoantibody (Aab) data from children <18 yrs participating in Diabetes Prevention Trial-Type 1 (DPT-1) or TrialNet Pathway to Prevention (PTP) , with "baseline" DPT-1 Risk Scores (DPTRS) at initial testing in the abnormal range of 6.50 to <7.50. No difference in AUC ratio (AUC C-peptide/AUC glucose) was present between P (n=85) and NP (n=161) in DPT-1 or between P (n=171) and NP (n=332) in PTP at baseline. However, during follow-up we found that by the first 6-month OGTT the AUC ratio had diverged, with lower AUC ratio in P both in DPT-1 (p=0.05) and PTP (p<0.001) , adjusting for baseline age, BMI, DPTRS, and AUC ratio. Moreover, in yearly change from baseline to last OGTT (before diagnosis in P) , AUC ratio decreased in P (p<0.001) and increased in NP (p<0.01) in DPT-1 and PTP. In assessments of factors relating to this divergence we observed that despite their metabolic similarity, at baseline P had higher harmonized IA-2A Aab positivity (70.8 vs. 49.1%, p<0.001) and titers (median 194 vs. 3, p<0.001) in PTP (not harmonized in DPT-1) . Based on this, we compared P with NP for yearly changes in IA-2A from baseline to the last visit, with adjustment for baseline values. P had a greater increase in IA-2A titer over time (p=0.025) . Furthermore, T1D progression was associated with yearly increases in IA-2A titer (HR: 1.006, 95%CI: 1.003, 1.009; p<0.001) . In conclusion, when NP and P have a similar degree of metabolic abnormality, IA-2A titers are greater in P. Moreover, as AUC ratio declines in P and improves in NP, IA-2A titers increase more in P than in NP. These findings suggest that IA-2A is a sensitive indicator of active β-cell pathology.
