GAD autoantibody affinity in schoolchildren from the general population

Aims/hypothesis: Subtyping GAD autoantibody (GADA) responses using affinity measurement allows the identification of GADA-positive children with a family history of type 1 diabetes who are at risk of developing diabetes. Here, we asked whether GADA affinity is a useful marker to stratify the risk of type 1 diabetes in GADA-positive schoolchildren from the general population. Methods: GADA affinity was measured by competitive binding experiments with [ ¹²⁵I]-labelled and unlabelled human 65 kDa isoform of GAD (GAD65) in sera from 97 GADA-positive children identified in the Karlsburg Type 1 Diabetes Risk Study of a general schoolchild population in north-eastern Germany. GADA epitope specificity was determined using radiobinding assays with [ ³⁵S]-labelled GAD65/67 kDa isoform of GAD (GAD67) chimeric proteins. Results: GADA affinity was high, ≥10¹⁰ l/mol, in 33 of 35 multiple islet autoantibody-positive children. In contrast, the affinity ranged widely among 62 single GADA-positive children (median 3.1 × 10⁹ l/mol; range 5.6 × 10⁶ to >4.0 × 10¹¹ l/mol; p < 0.0001). High-affinity GADA were associated with HLA-DRB103 (p = 0.02) and predominantly directed against the C-terminal and/or middle part of the GAD65 protein. At follow-up, the affinity remained relatively constant. Five of the single GADA-positive children developed additional islet autoantibodies and had high-affinity GADA. Twenty-six children progressed to type 1 diabetes; among them, 23 had GADA affinities of ≥10¹⁰ l/mol before disease onset. Conclusions/interpretation: Schoolchildren from the general population may develop heterogeneous GADA responses, and a high affinity can identify those GADA-positive children who are more likely to progress to type 1 diabetes.