Type 1 diabetes is caused by cell-mediated autoimmune destruction of the insulin-producing pancreatic islet beta cells [1]. If type 1 diabetes is to be cured, it must be prevented. To be prevented, type 1 diabetes must be predicted. The presence of type 1 diabetes-associated autoantibodies (a.k.a. – “islet” autoantibodies) in non-diabetic children and adults predicts the development of type 1 diabetes. Interventions to prevent type 1 diabetes in persons with islet autoantibodies are in development and show great promise [2].

Presently type 1 diabetes is divided into 4 stages: stage 1: ≥2 islet autoantibodies with normal glucose tolerance; stage 2: ≥2 islet autoantibodies with abnormal glucose tolerance; stage 3: ≥2 islet autoantibodies with the clinical diagnosis of diabetes; and stage 4: long-standing type 1 diabetes [3]. The most important and best characterized islet autoantibodies are islet cell cytoplasmic autoantibodies (ICA), insulin autoantibodies (IAA), glutamic acid decarboxylase autoantibodies (GADA), insulinoma associated-2 autoantibodies (IA-2A), and zinc transporter-8 autoantibodies (ZnT8A).

Islet autoantibody assays with high sensitivity and high specificity are extremely desirable in predicting type 1 diabetes. Therefore, it is the goal of the Islet Autoantibody Standardization Program (IASP) to [1] improve the performance of immunoassays measuring islet autoantibodies and [2] improve the concordance of results among laboratories. Such workshops are conducted about every 18 months.

To this end IASP workshop participants are sent blinded sera from persons with stage 1 through stage 3 type 1 diabetes, and control sera. Laboratories report their results to the IASP core laboratory that holds the key to which sera were from subjects with type 1 diabetes and which sera were from control subjects. IASP then calculates each assay’s sensitivity and specificity. Sensitivity is defined as the percentage of type-1 diabetes sera that are reported as positive for the islet autoantibody. Specificity is defined as the percentage of control sera that are reported as negative for the islet autoantibody.

While specificities near or exceeding 98% are not uncommon for the better islet autoantibody assays, sensitivities for the better islet autoantibody assays vary from 60 to 80%. One hundred percent sensitivity is not expected as it has been recognized since the 1970’s that not every islet autoantibody is present in every person with new-onset type 1 diabetes. The most common islet autoantibodies at disease onset are ICA and GADA with a ~80% prevalence followed by IA-2A and ZnT8A with prevalences of ~60%. The least common islet autoantibody in new-onset persons is IAA: ~50-60% prevalence in children and <10% in adults with new-onset type 1 diabetes.

In the fall of 2019 [4], the results for the GADA autoantibody assays were reported from the 2018 IASP workshop. Thirty seven laboratories from 17 countries in North America, Europe, Asia, and Australia submitted results for 48 GADA assays that were based on 9 different assay formats. There were 50 sera from persons with type 1 diabetes and 90 sera from controls. In addition to calculating sensitivity and specificity for each of these 48 GADA assays, the IASP core laboratory calculated for each assay their “area under the receiver-operator characteristic curve” (ROC-AUC), their “adjusted sensitivity 95” (i.e., the sensitivity at a specificity of 95%), and their “partial ROC-AUC” [the AUC for a specificity of >95% (pAUC95)].

Data from the following GADA formats were reported: antibody-dependent agglutination polymerase chain reaction (ADAP), enzyme-linked immunosorbent assay (ELISA), radiobinding assay (RBA), electrochemiluminescence (ECL), luciferase immunoprecipitation system (LIPS), chemiluminescence immunoassay (CLIA), multiplex plasmonic near-infrared fluorescence (MPNIRF), Luminex (LBI) and multiplex fluorescence energy transfer (MFRET).

The better performing GADA assays used the following technologies: ADAP, ELISA, RBA, ECL, LIPS and CLIA. Traditionally the RBA assay format has been considered the “gold standard.” However in the coming years, a non-RBA format with superior accuracy and precision may prove to be a new gold standard. Specific details regarding GADA assays can be found in reference 4.

The results of the 2020 IASP workshop are to be reported in October 2020. Should readers wish to participate in future IASP workshops, they should contact the authors.


  1. Winter, W.E., Schatz, DA. Autoantibody Markers in Diabetes. Clinical Chemistry. 2011, 57: 168-75.
  2. Herold KC, Bundy BN, Long SA, et al. An Anti-CD3 Antibody, Teplizumab, in Relatives at Risk for Type 1 Diabetes [published correction appears in N Engl J Med. 2020 Feb 6;382(6):586]. N Engl J Med. 2019;381(7):603-613.
  3. Couper JJ, Haller MJ, Greenbaum CJ, Ziegler A-G, Wherrett DK, Knip M, Craig ME. ISPAD Clinical Practice Consensus Guidelines 2018Stages of type 1 diabetes in children and adolescents. https://cdn.ymaws.com/www.ispad.org/resource/dynamic/forums/20180227_035330_10305.pdf (accessed on 9-7-2020).
  4. Lampasona V, Pittman DL, Williams AJ, et al. Islet Autoantibody Standardization Program 2018 Workshop: Interlaboratory Comparison of Glutamic Acid Decarboxylase Autoantibody Assay Performance. Clin Chem. 2019;65(9):1141-1152.