Islet Autoimmunity in Adults With Impaired Glucose Tolerance and Recently Diagnosed, Treatment Naïve Type 2 Diabetes in the Restoring Insulin SEcretion (RISE) Study

Abstract

The presence of islet autoantibodies and islet reactive T cells (T+) in adults with established type 2 diabetes (T2D) have been shown to identify those patients with more severe β-cell dysfunction. However, at what stage in the progression toward clinical T2D does islet autoimmunity emerge as an important component influencing β-cell dysfunction? In this ancillary study to the Restoring Insulin SEcretion (RISE) Study, we investigated the prevalence of and association with β-cell dysfunction of T+ and autoantibodies to the 65 kDa glutamic acid decarboxylase antigen (GADA) in obese pre-diabetes adults with impaired glucose tolerance (IGT) and recently diagnosed treatment naïve (Ndx) T2D. We further investigated the effect of 12 months of RISE interventions (metformin or liraglutide plus metformin, or with 3 months of insulin glargine followed by 9 months of metformin or placebo) on islet autoimmune reactivity. We observed GADA(+) in 1.6% of NdxT2D and 4.6% of IGT at baseline, and in 1.6% of NdxT2D and 5.3% of IGT at 12 months, but no significant associations between GADA(+) and β-cell function. T(+) was observed in 50% of NdxT2D and 60.4% of IGT at baseline, and in 68.4% of NdxT2D and 83.9% of IGT at 12 months. T(+) NdxT2D were observed to have significantly higher fasting glucose (p = 0.004), and 2 h glucose (p = 0.0032), but significantly lower steady state C-peptide (sscpep, p = 0.007) compared to T(-) NdxT2D. T(+) IGT participants demonstrated lower but not significant (p = 0.025) acute (first phase) C-peptide response to glucose (ACPRg) compared to T(-) IGT. With metformin treatment, T(+) participants were observed to have a significantly lower Hemoglobin A1c (HbA1c, p = 0.002) and fasting C-peptide (p = 0.002) compared to T(-), whereas T(+) treated with liraglutide + metformin had significantly lower sscpep (p = 0.010) compared to T(-) participants. In the placebo group, T(+) participants demonstrated significantly lower ACPRg (p = 0.001) compared to T(-) participants. In summary, T(+) were found in a large percentage of obese pre-diabetes adults with IGT and in recently diagnosed T2D. Moreover, T(+) were significantly correlated with treatment effects and β-cell dysfunction. Our results demonstrate that T(+) are an important component in T2D.

Keywords: GADA; beta cell function; impaired glucose tolerance; islet autoantibodies; islet autoimmunity; islet reactive T-cells; pre-diabetes; type 2 diabetes.

Figure 1

Relationship of T cell positivity with month-12 β-cell function in participants with new T2D and participants with IGT. Figures show adjusted means for values at month 12 visits. All models are adjusted for age, sex, race/ethnicity, treatment group, and baseline value of dependent variable. All models except for BMI are also adjusted for baseline BMI. Models for SSCP, and ACRPg are also adjusted for baseline and month-12 insulin sensitivity (M/I). Orange symbols represent participants with new T2D and blue symbols represent participants with IGT. P-value for comparison between autoimmune negative and positive for diabetes or IGT.

Figure 2

Adjusted means of HbA1c, fasting C-peptide, SSCP, and ACPRg illustrating interactions between T cell positivity and treatment group. Participants within each treatment group are identified by colors. Participants receiving 3 months of insulin glargine followed by 9 months of metformin (blue), 12 months of liraglutide + metformin (brown), 12 months of metformin (green), or placebo (yellow). Figure shows means for values at month-12 visits. All models are adjusted for age, sex, race/ethnicity, baseline insulin sensitivity, and BMI. Positivity over 12 months: Negative results are negative at baseline and month-12 samples. Positive responses are positive at either baseline, 12 months, or baseline and month-12. P-value for comparison between autoimmune negative and positive participants.