Islet autoantibody phenotypes and incidence in children at increased risk for type 1 diabetes.

AIMS/HYPOTHESIS: Autoantibodies that precede type 1 diabetes frequently develop in early childhood and target distinct beta cell proteins. The aim of this study was to determine the heterogeneity of islet autoantibody development and fate. METHODS: The ages of development of insulin autoantibodies (IAA) and GAD autoantibodies (GADA), followed by multiple islet autoantibodies and progression to diabetes were examined in 2,441 children participating in two German birth cohorts. RESULTS: In 218 children who developed islet autoantibodies, the first islet autoantibody-positive sample was characterised by single IAA in 80 (37%), multiple islet autoantibodies in 68 (31%) and single GADA in 63 (29%) children. Of the children who were single antibody positive at seroconversion, 35 (44%) IAA-positive and 15 (24%) GADA-positive children developed multiple islet autoantibodies. Single persistent antibodies had heterogeneous affinities; GADA were also heterogeneous in their binding to N-terminally truncated GAD65 and in an ELISA. Progression to diabetes occurred in >50% of children within 10 years in all groups that developed multiple islet autoantibodies and in 44% of children with persistent single high-affinity IAA or persistent single GADA that were positive in both a radiobinding assay and ELISA. The earliest autoantibody development was seen in children with single IAA that progressed to multiple islet autoantibodies or in those with persistent high-affinity single IAA, with a sharp peak in incidence observed at age 9 months. The peak incidence occurred at age 2 years for children who underwent seroconversion directly to multiple islet autoantibodies and at 5 years for children who first seroconverted to GADA and subsequently developed other autoantibodies. Seroconversion to low-affinity IAA or persistent single GADA occurred at a low incidence after the age of 9 months. CONCLUSIONS/INTERPRETATION: Children of different ages have differing susceptibilities to autoimmunisation against specific beta cell autoantigens.

Early infant feeding and risk of developing islet autoimmunity and type 1 diabetes.

We investigated whether food supplementation within the first year life or age at introduction of gluten-containing foods influenced the risk of developing islet autoimmunity and type 1 diabetes. A total of 2,291 children with a family history of type 1 diabetes were prospectively followed from birth for 28,983 patient years (median 13.1 years). Dietary exposure data were collected by questionnaires, food records and by family interview. Exposure to gluten-containing foods before age 3 months, which occurred in 19 children, increased the risk of developing islet autoantibodies (n = 4), multiple islet autoantibodies (n = 4), and type 1 diabetes (n = 3) compared to exclusive breastfeeding within the first 3 months [adjusted hazard ratio (HR) 3.97 (95 % confidence interval 1.41-11.17), 5.39 (1.89-15.35), and 3.45 (1.04-11.48), respectively] and also compared to first exposure to gluten between 3.1 and 6.0 months of age [adjusted HR 3.40 (1.19-9.70), 4.25 (1.47-12.26), and 3.43 (1.01-11.66), respectively]. Children who received infant formula or other solid food within the first 3 months and children who received gluten-containing foods after age 6 months did not have an increased risk of islet autoantibodies, multiple islet autoantibodies or type 1 diabetes. Our present data affirm that compliance to infant feeding guidelines is a possible way to reduce type 1 diabetes risk in genetically susceptible children.

IA-2 autoantibody affinity in children at risk for type 1 diabetes.

Autoantibodies to insulinoma-associated protein 2 (IA-2A) are associated with increased risk for type 1 diabetes. Here we examined IA-2A affinity and epitope specificity to assess heterogeneity in response intensity in relation to pathogenesis and diabetes risk in 50 children who were prospectively followed from birth. At first IA-2A appearance, affinity ranged from 10(7) to 10(11)L/mol and was high (>1.0×10(9)L/mol) in 41 (82%) children. IA-2A affinity was not associated with epitope specificity or HLA class II haplotype. On follow-up, affinity increased or remained high, and IA-2A were commonly against epitopes within the protein tyrosine phosphatase-like IA-2 domain and the homologue protein IA-2ß. IA-2A were preceded or accompanied by other islet autoantibodies in 49 (98%) children, of which 34 progressed to diabetes. IA-2A affinity did not stratify diabetes risk. In conclusion, the IA-2A response in children is intense with rapid maturation against immunogenic epitopes and a strong association with diabetes development.