Prevalence of Islet Autoantibodies in Adults Without Diabetes

Abstract

Context: Over half of all new cases of type 1 diabetes (T1D) are diagnosed in adults, yet the natural history of adult-onset T1D, particularly in nonfamilial populations, is not fully understood.

Objective: This study measured the prevalence of islet autoantibodies (IA) in adults without known diabetes and irrespective of T1D family history from Colorado (USA).

Methods: The Autoimmunity Screening for Kids study screened for IAs to insulin, glutamic acid decarboxylase (GADA), islet antigen-2, and zinc transporter 8 in 1087 adults without known diabetes [mean age 40.7 years with range 19.6-63.9 years, 63% non-Hispanic White (NHW), 10% with family history of T1D in a first-degree relative, and 78% female] from Colorado. IAs were measured using radiobinding assay and electrochemiluminescence detection methods.

Results: In total, 3.86% of adults screened positive for any IA, 0.55% screened positive for multiple IAs, and 1.75% were positive for a single IA by both detection methods. Compared to NHW, those with Hispanic race/ethnicity were more likely to screen positive for a single IA (relative risk 2.32, 95% confidence interval 1.40, 3.84, P = .001), but there was no difference in the risk of screening positive for multiple IAs when comparing across race/ethnicity. GADA was the most prevalent IA, found in 2.67% of adults.

Conclusion: IA prevalence was high in this sample of adults without known diabetes from Colorado. Further study is needed to fully characterize the risk of progression to clinical diabetes among adults who screen positive for IAs, particularly in nonfamilial populations.

Keywords: LADA; adult-onset diabetes; autoimmune diabetes; islet autoantibodies; latent autoimmune diabetes in adults; type 1 diabetes.

Figure 1.

IA screening and confirmation results. The figure depicts a flow diagram reporting the results of screening and confirmation IA testing in 1087 screened adults. The larger top black box depicts screening results, first stratified by negative or any positive result. Positive results are then stratified by degree of IA positivity [multiple islet autoantibody positive (multiple IA+), single islet autoantibody positive by both RBA and ECL detection methods (single IA+ by both methods), and single islet autoantibody positive by either RBA or ECL detection method (single IA+ by 1 method)]. The smaller bottom black box depicts results from the 24 participants who returned for confirmation testing, grouped by IA status at the time of initial screening, to show any discordance between screening and confirmation results. Abbreviations: ECL, electrochemiluminescence; IA, islet autoantibody; RBA, radiobinding assay.

Figure 2.

Risk ratios for demographic characteristics by IA positivity status. Figures depict risk ratios (dots) with 95% confidence intervals (bars) for each IA positivity status group in relation to patient characteristics: (A) multiple islet autoantibody positive (multiple IA+); (B) single islet autoantibody positive by both RBA and ECL detection methods (single IA+ by both methods); (C) single islet autoantibody positive by either RBA or ECL detection method (single IA+ by 1 method). X-axes depict risk ratios in linear scale. Abbreviations: ECL, electrochemiluminescence; FDR, first-degree relative; IA, islet autoantibody; NHW, non-Hispanic White; Non, nonrelative; RBA, radiobinding assay.