Islet cell antibodies

In 1974 islet cell cytoplasmic autoantibodies (ICA) were first described by Bottazzo and colleagues in patients with polyglandular autoimmunity [1]. Later, ICA were described in type 1 diabetes mellitus (T1DM) patients in the absence of polyglandular autoimmunity [2].

Figure 1 – The detection of islet cell autoantibodies is described in the text.
Figure 1 – The detection of islet cell autoantibodies is described in the text.
The development of epifluorescence fostered the indirect immunofluorescence technique used to measure ICA. The detection of ICA requires that sera be incubated on sections of human cryocut, blood-group O pancreas [Figure 1]. Because pancreas can rapidly undergo autolysis, within 2 hours of removal from a cadaveric donor, the pancreas should be sectioned into blocks (~0.5 cm x 0.5 cm) and snap frozen in isopentane that has been cooled in a dry-ice-acetone bath. Because the tissue is not fixed with formalin, insulin is lost during incubation of-frozen tissue sections with serum, and therefore, ICA do not detect autoantibodies to insulin.

Physicians are advised to be sure that human pancreas is used for ICA testing and not non-human primate pancreas which is inferior to human pancreas [3]. Furthermore, ICA ELISAs should be avoided as such assays are not validated [4]. Following incubation with human serum, the tissue section is washed with phosphate-buffered saline. Next, the slide is flooded with a fluorescein isothiocyanate–labeled, antihuman IgG conjugate solution. Following incubation Figure 2 – Islet cell cytoplasmic autoantibodies are recognized by fluorescence of the islet cells surrounded by the non-fluorescent exocrine pancreas.
Figure 2 – Islet cell cytoplasmic autoantibodies are recognized by fluorescence of the islet cells surrounded by the non-fluorescent exocrine pancreas.
and another washing step, the slide is examined under a fluorescent microscope after application of glycerol and a cover slip. When ICA are present, the islets fluoresce [Figure 2]. ICA are expressed in terms of JDF (Juvenile Diabetes Foundation) units by serial dilution of the patient’s serum to endpoint in parallel with a standard serum. Thresholds vary between laboratories, but the cut-off for ICA positivity is commonly set between 5-10 JDF units, a level exceeded by ~3% of first degree relatives of persons with type 1 diabetes and ~0.4% of healthy individuals.

Polyclonal ICA react against multiple islet autoantigens including a sialoglycoconjugate, an insulinoma-associated autoantigen (islet antigen-2 or ICA512) and glutamic acid decarboxylase. The autoantigens are not beta-cell specific as ICA react with all cells of the islet. The measurement process is very labor intensive; however, the reagent and consumables costs are low. ICA are the most difficult islet autoantibodies to measure because ICA assays are subject to variations in the pancreatic tissue, conjugate, incubation times, humidity, and biological-scientist interpretation. Therefore accurate ICA testing requires rigorous attention to analytical procedures accompanied by a blinded-splits program, in which coded samples are measured in duplicate so that scoring is unbiased. In the Type 1 Diabetes TrialNet study, our laboratory performs an internal blinded-splits program in addition to the external blinded splits program carried out by TrialNet.

Seventy to 80% of individuals with new-onset T1DM are positive for ICA. Following diagnosis, ICA positivity declines. After 10 years, few patients remain ICA positive. Many studies have demonstrated that the presence of ICA in nondiabetic individuals predicts the development of T1DM and ICA were used for selection of high-risk subjects for early therapeutic intervention trials (DPT-1, Type 1 Diabetes TrialNet and ENDIT) Non-affected relatives of individuals with T1DM who were positive for ICA in the Diabetes Prevention Trial–1 (DPT-1) who also had low first-phase insulin responses to an intravenous glucose challenge had a 5-year risk for the development of T1DM of 60% and a projected 10-year risk near 90% [5]. Despite the discovery of new autoantibodies associated with T1DM (e.g., zinc transporter-8 autoantibodies), ICA continue to provide additional predictive information for the later development of T1DM [6].


  1. ^ Bottazzo GF, Florin-Christensen A, Doniach D. Islet-cell antibodies in diabetes mellitus with autoimmune polyendocrine deficiencies. Lancet 1974;2:1279–83.

  2. ^ Bottazzo GF, Mann JI, Thorogood M, Baum JD, Doniach D. Autoimmunity in juvenile diabetics and their families. Br Med J. 1978 Jul 15;2(6131):165-8.

  3. ^ Scherbaum WA, Trischler G, Pfeiffer EF. Non-human primate pancreas as a substrate for the detection of islet-cell antibodies in human sera. Diabetes Res Clin Pract. 1989 Jun 20;7(1):1-5.

  4. ^ ICA (Islet Cell Antibody) ELISA, Alpco Diagnostics, Salem, NH.

  5. ^ Diabetes Prevention Trial–Type 1 Diabetes Study Group. Effects of insulin in relatives of patients

  6. ^ Velluzzi F, Secci G, Sepe V, Klersy C, Shattock M, Foxon R, Songini M, Mariotti S, Locatelli M, Bottazzo GF, Loviselli A; Sardinian Autoimmunity Study Group. Prediction of type 1 diabetes in Sardinian schoolchildren using islet cell autoantibodies: 10-year follow-up of the Sardinian schoolchildren type 1 diabetes prediction study. Acta Diabetol. 2015 Apr 22. 0940-5429.


Nobody has commented on this article

Commenting is only available for registered Diapedia users. Please log in or register first.