ESPE Abstracts

Background: Detection of type 1 diabetes (T1D) at the pre-clinical stage is possible by detecting islet autoantibodies (IA) years before the appearance of symptomatic diabetes. An efficient screening program based on these antibodies will identify children at risk of developing diabetes during childhood. The antibody detection Israeli research (ADIR) is a general population screening program in Israel searching for children with multiple IA who are at risk of developing T1D. IA are measured in capillary blood samples using the novel ultrasensitive antibody detection by agglutination-PCR (ADAP) technology by Enable Biosciences. The ADAP assay reliably detects the three cardinal IA namely, GAD (GADA), IA-2 (IA-2A), and insulin (IAA) in 1mL of serum with higher sensitivity than the currently used radio-binding assays (RBA). Nevertheless, the reliability and accuracy of the assay were not tested in venous or capillary whole blood (WB), which are a more desirable sample matrix for general population screening programs.

Objective: To assess the accuracy and reliability of the ADAP assay in venous and capillary WB.

Methods: Fifty children with T1D and fifty healthy controls participated in the study. Venous and capillary blood samples were drawn from participants with T1D, while only venous blood was drawn from the controls. The ADAP assay was performed in 4mL of WB (capillary or venous), while RBA was performed in serum as customary. The assays were compared in their ability to detect GADA, IA-2A, and IAA.

Results: The area under the curve using the receiver operating characteristic (ROC) curvs was comparable between the ADAP assay in WB and standard RBA for all 3 IA indicating similar clinical sensitivity and specificity [ GADA 0.946 (95%CI: 0.900-0.991) vs 0.949 (95%CI: 0.906-0.992), P=0.873; IA-2A 0.747 (95%CI: 0.649-0.844) vs. 0.666 (95% CI: 0.587-0.744), P=0.106; IAA 1.000 (95%CI: 1.000-1.000) vs. 1.000 (95%CI: 1.000-1.000), P=1.000]. IA levels in venous and capillary WB using ADAP were comparable for GADA (difference=0.3±0.6, P=0.565), IA2-A (difference=0.04±0.6, P=0.401), and IAA (difference=0.14±0.8, P=0.901). The correlation between the levels of IA in venous and capillary WB using ADAP was R2= 0.958 (P<0.01), R2= 0.943 (P<0.01), and R2= 0.711 (P<0.01) for GADA, IA-2A and IAA, respectively.

Conclusions: The ADAP assay is reliable in detecting IA in venous and capillary WB samples with comparable performance to standard RBA. These findings open avenues for widespread use of the ADAP assay in future general population screening programs to detect children at risk of developing T1D.