C-Peptide Levels in Subjects Followed Longitudinally Before and After Type 1 Diabetes Diagnosis in TrialNet

Abstract

Objective: Insulin secretion declines rapidly after diagnosis of type 1 diabetes, followed by a slower rate of change. Previous studies have demonstrated that the C-peptide decline begins before the clinical diagnosis. Changes in insulin secretion in the same individuals studied from preclinical stages through and after clinical diagnosis have not been previously reported.

Research design and methods: Antibody-positive relatives undergo sequential oral glucose tolerance testing (OGTT) as part of TrialNet's Pathway to Prevention study and continue both OGTT and mixed-meal tolerance testing (MMTT) as part of the Long-term Investigational Follow-up in TrialNet study if they develop type 1 diabetes. We analyzed glucose and C-peptide data obtained from 80 TrialNet subjects who had OGTT before and after clinical diagnosis. Separately, we compared C-peptide response to OGTT and MMTT in 127 participants after diagnosis.

Results: C-peptide did not change significantly until 6 months before the clinical diagnosis of type 1 diabetes and continued to decline postdiagnosis, and the rates of decline for the first 6 months postdiagnosis were similar to the 6 months prediagnosis. There were no significant differences in MMTT and OGTT C-peptide responses in paired tests postdiagnosis.

Conclusions: This is the first analysis of C-peptide levels in longitudinally monitored patients with type 1 diabetes studied from before diagnosis and continuing to the postdiagnosis period. These data highlight the discordant timing between accelerated β-cell dysfunction and the current glucose thresholds for clinical diagnosis. To preserve β-cell function, disease-modifying therapy should start at or before the acute decline in C-peptide.

Figure 1

A: Population mean of C-peptide AUC mean by months from type 1 diabetes diagnosis. Number of subjects contributing to the mean is indicated along the x-axis. B: Population mean of change in C-peptide AUC mean by month from type 1 diabetes diagnosis. Number of subjects contributing to the mean change appears along the x-axis. The slope is in units of change in log(Cp +1) per month.

Figure 2

The mean rate of C-peptide decline before and after type 1 diabetes diagnosis from the fitted mixed model in which age is included as a continuous variable. Four age ranges are used to illustrate the effect of age on C-peptide decline: blue = ages 6–11; burgundy = ages 11–17; green = ages 17–20; and orange = age >20 years. The number of subjects and the number of C-peptide measurements contributing to the fit are displayed along the x-axis. The frequency is based on the points within a ±3-month window of the month indicated under the tick mark.

Figure 3

Moving averages of fasting glucose (A), 2-h glucose (B), and 2-h glucose slope (C, next page) in subjects from 24 months before diagnosis to diagnosis.

Figure 3

Moving averages of fasting glucose (A), 2-h glucose (B), and 2-h glucose slope (C, next page) in subjects from 24 months before diagnosis to diagnosis.

Figure 4

Comparison of C-peptide AUC means computed from proximal MMTT and OGTT samples after type 1 diabetes diagnosis (171 pairs from 127 subjects).