Antibodies are predictive for diagnosis of celiac disease in pediatric type 1 diabetes.

Background Children and adolescents with type 1 diabetes (T1D) have a higher risk of developing celiac disease (CD) than the general population. However, the main screening antibody, IgA anti-transglutaminase 2 (TGA-IgA), can fluctuate in T1D and there is no threshold for performing diagnostic biopsies.Objectives The study aims to define an optimal TGA-IgA cut-off for performing diagnostic biopsies for CD confirmation and to assess whether tracking TGA-IgA evolution or adding other antibodies can improve biopsy indications. Methods Retrospective longitudinal analysis of pediatric T1D individuals diagnosed at two centers in Switzerland between 2000 and 2021, from T1D diagnosis to CD diagnosis or the age of 18 years. Results We included 588 individuals with T1D, compromising 2944 TGA-IgA values. 34 (5.8%) developed CD during follow-up, of whom 50% had CD- associated symptoms at CD diagnosis. Balancing sensitivity and specificity TGA-IgA around 6.1xULN was the best cut-off for performing diagnostic biopsies. The inclusion of IgG antibodies against deamidated gliadin peptides (GLA-IgG) achieved a higher AUC of 0.79 (95% CI 0.6-1) with 80% accuracy compared to each antibody alone. CD diagnosis within two years of T1D, representing two-thirds of CD, was marked by elevated TGA-IgA at T1D diagnosis. Later CD diagnosis was associated with a more gradual increase of TGA-IgA. Conclusion Our results suggest an indication for biopsy for CD confirmation at a TGA-IgA cut-off around 6.1xULN. Including GLA-IgG can potentially increase precision. TGA-IgA determination at T1D diagnosis may help to identify individuals at risk of CD early on.