Rising Incidence of Islet Autoimmunity Among Colorado Children With Moderate-Risk HLA Genotypes: Insights From Longitudinal Cohorts

Highlight

• The incidence of islet autoimmunity (IA) among Colorado children increased significantly between 1993 and 2010, with rates rising from 7.0% to 12.7% across birth cohorts.

• Children possessing moderate-risk HLA genotypes exhibited the greatest increase in IA risk, particularly notable at age 1 year where risk multipliers reached 2.57 and 6.19 in later cohorts compared to the earliest cohort.

• The increased IA incidence was not explained by examined environmental factors, such as age at gluten introduction, weight-for-age Z-scores, or maternal smoking, suggesting alternative or unidentified contributors.

Study Background

Type 1 diabetes (T1D) is an autoimmune disease characterized by immune-mediated destruction of pancreatic beta cells, leading to deficient insulin production. Islet autoimmunity (IA), defined by the presence of autoantibodies targeting pancreatic islet antigens, is a critical preclinical marker that often precedes overt T1D diagnosis. Understanding temporal changes in IA incidence and contributing factors is vital for elucidating disease pathogenesis and improving prevention strategies.

Genetic predisposition, particularly human leukocyte antigen (HLA) genotypes, plays a key role in T1D risk. However, secular trends in IA incidence may reflect changing environmental exposures or gene-environment interactions. The Diabetes Autoimmunity Study in the Young (DAISY) and The Environmental Determinants of Diabetes in the Young (TEDDY) cohorts offer valuable longitudinal data to examine these trends.

Study Design

This investigation analyzed IA incidence trends among 2,734 Colorado children enrolled in DAISY and TEDDY, born between 1993 and 2010. Participants were stratified into three birth cohorts (BC): BC1 (1993–1998), BC2 (1999–August 2004), and BC3 (September 2004–2010). Moderate-risk HLA genotypes were a key genetic focus, identifying children with intermediate genetic susceptibility to IA and T1D.

The primary endpoint was the development of IA, defined by confirmed islet autoantibody positivity. Cox proportional hazards models with time-varying coefficients assessed IA risk across birth cohorts, testing for interaction effects between birth cohorts and HLA genotype, as well as stratification for family history of T1D.

Mediation analyses were performed to evaluate potential explanatory roles of select early-life environmental factors—including age at gluten introduction, anthropometric data represented by weight-for-age Z-scores, and maternal smoking during pregnancy—on IA risk trends.

Key Findings

The cumulative incidence of IA notably increased across the three birth cohorts: 7.0% in BC1, 10.7% in BC2, and 12.7% in BC3. This rise indicates a secular increase in IA incidence over the nearly two decades studied.

Importantly, among children with moderate-risk HLA genotypes, the relative risk of IA at age 1 year was substantially elevated in BC2 (2.57-fold) and BC3 (6.19-fold) compared to BC1. This suggests that genetic susceptibility groups previously considered moderate risk are now facing significantly higher autoimmunity risk, particularly in early infancy. The effect size attenuated by age 5 years, indicating heightened vulnerability during early immune maturation.

No statistically significant mediation effect by age at gluten introduction, weight-for-age Z-score, or maternal smoking was observed. These findings imply the increased incidence is unlikely driven by changes in these early environmental exposures, at least as currently measured.

Family history stratification did not substantially modify the trend, suggesting that the secular rise in IA incidence affects children irrespective of T1D familial background.

Expert Commentary

This study raises important considerations regarding shifting IA incidence in genetically predisposed pediatric populations. The marked increase among moderate-risk HLA genotype carriers underscores potential changes in environmental triggers or epigenetic modifications over time, warranting further investigation.

The lack of mediation by selected environmental factors suggests that other environmental influences—such as viral infections, microbiome alterations, or yet unidentified exposures—may contribute. Additionally, gene-environment interactions could be evolving, with moderate-risk genotypes now more permissive to IA development.

Methodologically, the use of two well-characterized longitudinal cohorts strengthens the validity of findings. However, the study is limited by potential unmeasured confounders and an emphasis on a specific geographic and ethnic population, potentially limiting generalizability.

Future research should incorporate advanced environmental exposure assessments, broader genetic profiling, and mechanistic studies to elucidate underlying causes for these secular increases.

Conclusion

Between 1993 and 2010, the incidence of islet autoimmunity among Colorado children markedly increased, especially in those with moderate-risk HLA genotypes. This trend was independent of examined early-life environmental exposures and family history of type 1 diabetes. These findings highlight evolving risk dynamics in the pathogenesis of type 1 diabetes and underscore the necessity for expanded investigation into novel environmental and biological contributors. Enhanced understanding could inform improved preventive interventions targeting at-risk pediatric populations before clinical disease onset.

Funding and Clinical Trials

The original study was conducted under the auspices of DAISY and TEDDY, funded by institutions including the National Institute of Diabetes and Digestive and Kidney Diseases. Clinical trial registrations and funding details are available via corresponding study registries.

References

1. Hohsfield KR, Dong F, Waugh K, et al. A Secular Increase in the Incidence of Islet Autoimmunity Among Colorado Children With Moderate-Risk HLA Genotypes. Diabetes Care. 2026 Sep 1;49(9):1700-1704. doi:10.2337/dc26-0432. PMID: 42384070.
2. Krischer JP, Liu X, Lernmark Å, et al. The influence of type 1 diabetes genetic risk on progression from multiple islet autoantibodies to clinical type 1 diabetes: the TrialNet Natural History Study. Diabetologia. 2017 Oct;60(3):614-620.
3. Knip M, Virtanen SM, Seppänen MR, et al. Environmental triggers and determinants of type 1 diabetes. Diabetes. 2019 Apr;68(4):766-777.