Significant phenotypic diversity exists among T1D patients, supporting the existence of distinct disease endotypes (e.g. duration of the post-onset partial remission period) . To broaden our understanding of T1D heterogeneity and its relationship to disease progression and therapeutic responsiveness, we studied 566 new onset T1D patients (241 females, 325 males; mean age 16.5 +/- 8.2 years) that had participated in TrialNet clinical trials: TN02/MMF DZB; TN05/anti-CD20; TN08/GAD-alum; TN09/CTLA4-Ig; TN14/Canakinumab; and TN19/ATG-GCSF. Baseline samples were analyzed with a sensitive bioassay, where subject plasma is used to induce transcription in a well-controlled peripheral blood mononuclear cell population. The response was measured with a microarray, data were normalized, and transcripts exhibiting the highest median absolute deviation within each trial were identified. Weighted gene correlation network analysis was used to cluster patterns of variation into modules of co-expressed transcripts. Co-expression networks were correlated with clinical data collected over the 24-months following clinical onset. In the placebo arms of all 6 trials, we identified modules that significantly correlated with post-onset C-peptide AUC, post-onset insulin dose, Insulin-Dose Adjusted A1c, and/or the rate of C-peptide decline. This suggests that the baseline analyses identified modules that differentiated subjects in terms of post-onset beta-cell function, supporting the existence of T1D endotypes. While TN02, TN08, and TN14 did not show overall efficacy in slowing post-onset C-peptide decline, the module correlations with phenotypes related to beta-cell function were fewer in number and of lesser statistical significance in the treatment arms of all 6 trials. This suggests that there was greater beta-cell function in the treatment arms of all trials, and that varying degrees of immunomodulation had occurred even in those trials that did not show efficacy.


A.Bedrat: None. S.Jia: None. M.Roethle: None. K.Dew: None. S.M.Cabrera: None. C.Lin: None. M.Hessner: None.


National Institutes of Health (R01DK121528)

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