Individuals with Single Islet Autoantibody Positivity (SA+) at Type 1 Diabetes (T1D) Diagnosis Have Distinct Characteristics

Abstract

Multiple autoantibody positivity (MA+) usually precedes clinical T1D. However, a subset of patients who develop T1D have SA+. We hypothesized that individuals who at T1D diagnosis express SA+ compared with MA+ have different demographic, metabolic, immunologic and genetic characteristics. We studied 620 TrialNet participants who developed clinical T1D (median age=12.3 years [range=1.4-58.6], 47.5% male, 88% white). Using multivariable modeling and adjusting for potential confounders, we investigated differences between participants with SA+ vs. MA+ (of autoantibodies to GAD65 [GADA], insulin [mIAA], IA2, ZnT8 and ICA) at T1D diagnosis. We found 95 participants who had SA+ (70% GADA+) and 525 participants with MA+. In univariate analyses, individuals with SA+ vs. MA+ were older at T1D diagnosis (median 16.4 vs. 11.7 years, p<0.0001) and had higher fasting C-peptide (1.87 vs. 1.47 nmol/L, p=0.007), higher 30-0 minute C-peptide difference (1.42 vs. 1.18, p=0.009) and lower Index60 (composite measure of fasting C-peptide, 60-min glucose and 60-min C-peptide) (2.31 vs. 2.68, p=0.0005). Gender, race, ethnicity, BMI-Z-score, presence of HLA DR3-DQ2 and/or DR4-DQ8, fasting glucose, AUC glucose and AUC C-peptide were not significantly different. By multivariable analyses, SA+ at onset corresponded with older age (OR=1.05, p<0.0001), lower Index60 (OR=0.77, p=0.008), and lower likelihood of mIAA+ (OR=0.27, p<0.0001) and GADA+ (OR=0.25, p<0.0001); gender, HLA, BMI-Z-score and other autoantibody types were not significantly different. Independent factors of being SA+ differed between DR3/DR4-DQ8 heterozygous participants (older age and not having mIAA+ or GADA+) (p<0.01) and those not DR3/DR4-DQ8 (lower Index60 and not having GADA+) (p<0.01). In conclusion, the number of islet autoantibodies at T1D diagnosis is associated with distinct characteristics, which may suggest heterogeneous pathogenesis with relevance for T1D prevention and treatment. Disclosure M.J. Redondo: None. J. Sosenko: None. I. Libman: Consultant; Self; Novo Nordisk A/S. J.J. McVean: Speaker's Bureau; Self; Medtronic MiniMed, Inc. M.A. Atkinson: Other Relationship; Self; Patent Issued. D.J. Becker: None. S. Geyer: None.