Peak C-peptide (C-P) and glucose levels are known predictors of T1D, but T1D prediction by times to those peak levels has not been studied. We thus assessed T1D prediction by times to peak C-P or glucose during OGTTs in 709 Ab+ Diabetes Prevention Trial-type 1 participants with 2-hour OGTTs [mean age: 13.9±9.6 years; BMI percentile (%ile): 59.3±29.0; 56% male]. Log rank testing and Cox regression with adjustments for age, sex, BMI %ile, proband relation, HLA, and Ab number were used. Risks were higher for time to peak C-P >60 mins [HR: 1.96 (CI: 1.49-2.63), p<0.001)] and for time to peak glucose >30 mins [HR: 2.22 (1.67-2.94), p<0.001)]. With time to peak C-P and peak C-P as covariates, T1D associations were greater for time to peak C-P (X2=22.8 vs. 9.3, both p<0.001). Time to peak glucose and peak glucose associations also remained significant (both p<0.001). Findings were validated in Ab+ relatives in the TrialNet Pathway to Prevention study (Figure); those with a peak C-P >60 mins or a peak glucose >30 mins had higher risk (both p<0.001).

In conclusion, Ab+ relatives with delayed peak C-P or glucose are at higher T1D risk, independent of peak levels. Moreover, the time to peak C-P appears to be more predictive than the peak level. Times to peak C-P and glucose have potential use as predictors of T1D and as markers for progression.

H.M. Ismail: None. P. Xu: None. C. Evans-Molina: None. J.P. Palmer: None. J. Sosenko: None.

Readers may use this article as long as the work is properly cited, the use is educational and not for profit, and the work is not altered. More information is available at