It may be possible to predict the development of type 1 diabetes in young children by measuring the presence of autoantibodies in the blood, new data from The Environmental Determinants of Diabetes in the Young (TEDDY) study indicate
“In the TEDDY study, we have found that autoantibodies often appear during the first few years of life,” lead researcher Åke Lernmark, MD, PhD, of Lund University, said in a press release.
The TEDDY study enrolled and prospectively followed 8,600 children from Sweden, the United States, Germany and Finland who have an increased hereditary risk for type 1 diabetes, defined as having HLA-DR high-risk genotypes (DR3/4, DR4/4, DR4/8 and DR3/3).
The new findings from the NIH-funded study, which were recently published in Diabetologia, suggest three ways of predicting development of type 1 diabetes.
- If the autoantibody first discovered attacks insulin (IAA). In the study as a whole, this occurred by age 1 year. Dr. Lernmark noted in the release that if a second autoantibody is detected, the person may develop diabetes but perhaps not for 20 years.
- If the first autoantibody targets GAD65 (GADA), a protein inside the insulin-producing cells. This was most common at age 2 years in the study.
- If both autoantibodies are first found together. According to Dr. Lernmark, 40% of these children in the TEDDY study had already developed diabetes.
In the study, participants were enrolled as infants and underwent quarterly standardized autoantibody assessments for the first 4 years of life and semi-annually thereafter.
The researchers found autoantibodies in 549 of 8,503 (6.5%) children during 34,091 person-years of follow-up. About 44% only had islet autoantibodies against insulin (IAA), with the incidence peaking within the first year of life and declining during the next 5 years. About 38% only had glutamic acid decarboxylase autoantibodies (GADA), with incidence only increasing until age 2 years and then remaining constant. About 14% of those with autoantibodies had both IAA and GADA, with incidence peaking around age 2 to 3 years.
Additionally, insulinoma antigen-2 only was found in 1.6% and other combinations were found in 3.1%, according to the data.
Hereditary risk appeared to determine which autoantibody was found, with GADA only being more common than IAA only in HLA-DR3 children but less common in HLA-DR4/8 children. Nevertheless, what causes the immune system to attack the body’s own insulin cells remains unknown, although viral infection has been suggested as a potential trigger.
“It is possible that there are two different diseases involved. Perhaps one virus triggers the autoantibodies against insulin and another one the autoantibodies against GAD65,” Dr. Lernmark said.