Type 1 Diabetes Autoantibodies
Prediction and Diagnosis of Autoimmune Diabetes
By Patricia W. Mueller, PhD, Peter Achenbach, MD, Vito Lampasona, Michael Schlosser, PhD, and Alistair J. K. Williams
Photo reprinted from reference 6 with permission of Elsevier.
In 2007, the economic burden of diabetes was a staggering $116 billion for direct medical costs and $58 billion for indirect costs, such as disability, work loss, and premature mortality. Today, the disease affects more than 23.6 million people in the U.S., or 7.8% of the population, with those numbers rising dramatically (1). The autoimmune forms of diabetes, which affects 2 to 4 million people, include type 1 diabetes (T1D), an estimated 5% to 10% of those with diabetes, and latent autoimmune diabetes in adults (LADA), estimated to be 5% to 10% of those diagnosed with type 2 diabetes (T2D) (2). T1D, previously called insulin-dependent diabetes mellitus or juvenile onset diabetes, occurs when the body’s immune system destroys the pancreatic β-cells that produce the hormone insulin involved in blood glucose regulation. To survive, affected individuals must take insulin.
Identifying diabetes early is critical because the disease places affected individuals at increased risk of many complications and even death. The risk of stroke or death from heart disease is 2–4 times higher than those without the disease. Diabetes is also the leading cause of new cases of blindness among adults age 20 to 74 years, and the leading cause of kidney failure, accounting for 44% of new cases in 2005. About 60% to 70% of people with diabetes have mild to severe forms of nervous system damage, and more than 60% of nontraumatic lower-limb amputations occur in people with diabetes.
This article will describe the latest advances in testing for autoantibodies predictive of T1D. Current assay standardization efforts by the Diabetes Autoantibody Standardization Program (DASP) will also be discussed.
Clinical Assessment and Trials
Traditionally, clinicians have differentiated between T1D and T2D based on phenotypic characteristics, including age at onset, abruptness of onset of hyperglycemia, ketosis-proneness, degree of obesity, prevalence of autoimmune disease, and the need for insulin replacement therapy. Studies of autoantibodies in diabetes patients now suggest that there can be some overlap between T1D and T2D, the latter of which is more commonly diagnosed in adults. A subset of adult patients diagnosed with T2D, actually have LADA. Now recognized as a slowly developing form of autoimmune diabetes found in people over 35 years old, LADA is often misdiagnosed as T2D. These patients present clinically without ketoacidosis and weight loss but progress more rapidly to insulin dependence than typical T2D patients (3). Other labels have also been applied to this subset of diabetes, including slowly progressive type 1, latent type 1, double, and type 1.5 diabetes.
Clinical trials are currently in progress to identify ways to prevent or reverse the autoimmunity of T1D. The Type 1 Diabetes TrialNet consisting of 18 clinical centers in the U.S., Canada, Europe, and Australia seeks to stop or delay the immune destruction of the insulin-producing pancreatic β cells. It will include trials of oral insulin, glutamic acid decarboxylase (GAD), Rituximab (Anti-CD20), metabolic control, as well as a nutritional prevention study of the omega-3 fatty acid, docosahexaenoic acid (DHA). The Immune Tolerance Network also conducts clinical trials for T1D, as well as other autoimmune diseases, including trials of thymoglobulin, interleukin-2 and sirolimus, and intranasal insulin.
Autoantibodies Predictive of T1D
The model proposed by Eisenbarth for T1D development has been in use for more than 20 years (4). It describes T1D as a chronic, progressive autoimmune disorder in which subjects at genetic risk experience an as yet undefined environmental insult that initiates the destruction of the insulin-producing pancreatic islet β-cells (5). Destruction of the β-cells occurs over many years and results in metabolic abnormalities such as impaired glucose tolerance and ultimately symptomatic hyperglycemia. It is this process of β-cell destruction that is accompanied by production of autoantibodies to β-cell antigens.
In 1974, researchers first identified pancreatic islet cell autoantibodies (ICAs) in T1D patients who had multi-endocrine deficiencies associated with organ-specific autoimmunity (6). In fact, this study established T1D as an autoimmune disease. Other researchers subsequently reported ICAs at a high frequency in children with newly diagnosed T1D. Now quantified in Juvenile Diabetes Foundation Units (10), the ICA levels are difficult to standardize because the assay depends on human pancreatic tissue as a substrate. Consequently labs have replaced ICA tests with tests for specific autoantibodies.
Autoantibodies in T1D
Clinicians order autoantibody tests to help distinguish between autoimmune T1D and diabetes due to other causes. Described below are the three most common autoantibody tests (see Table 1), as well as a fourth autoantibody recently shown to be useful for T1D testing.
Insulin autoantibodies. While it had long been recognized that treatment with exogenous forms of insulin could induce antibodies directed against insulin peptides, in 1983 researchers described insulin autoantibodies (IAA) in T1D patients before they received insulin therapy (7). IAAs are diverse, and in general these high-affinity autoantibodies are more predictive of T1D and share certain characteristics, including appearance at a young age, association with HLA DRB1*04, subsequent progression to multiple autoantibody positivity, binding to human insulin A chain residues 8-13, and binding to proinsulin.
IAAs are usually the first autoantibodies to appear in young children who develop T1D and can persist for many years before the appearance of T1D clinical symptoms. While they are considered one of the most important autoantibodies for predicting T1D in young children, current assays produce highly variable results, and only a few clinical laboratories consistently perform the assay with high sensitivity and specificity for the disease.
GAD65 autoantibodies. The next major autoantigen to be characterized was a 65-kDa isoform of glutamic acid decarboxylase (GAD65) (8). Cloning of the GAD65 gene enabled researchers to synthesize the protein with radioisotopically labeled amino acids in an in vitro transcription/translation reaction and to use the product as the antigen in radiobinding assays (RBAs). In the early stages of GAD65 autoimmunity, the epitopes recognized by GADA are primarily in the middle region of the protein, but later they may include regions at the middle/C-terminal region of the protein. Autoantibodies to GAD65 are particularly important in late onset autoimmunity, such as that seen in LADA patients.
IA-2 or ICA512 autoantibodies. In 1995, researchers characterized two tryptic digest fragments of islet antigens from individuals with T1D. One, a 40-kDa fragment from the intracellular portion of a tyrosine phosphatase-like protein (PTPRN gene), is now referred to as IA-2ic or ICA512ic (9). Although IA-2 autoantibodies (IA-2A) can be detected early in the course of autoimmunity, they often appear after other autoantibodies and have a higher positive-predictive value for T1D than GADA. Researchers have also characterized the other 37-kDa tryptic fragment as IA-2β or phogrin. Since almost all autoantibodies that react with IA-2β also react with IA-2, clinical labs typically do not use IA-2β autoantibodies as a first line test. However, it may be useful for identifying individuals at high risk of disease progression.
ZnT8 autoantibodies. There is growing interest in autoantibodies to a member of the zinc transporter protein family, ZnT8, for autoimmune diabetes testing. Researchers discovered ZnT8 autoantibodies by screening for highly expressed proteins in islet β-cells of the pancreas (10). ZnT8 is one of a large family of zinc transporter proteins that is associated with the membrane of secretory granules of islet β-cells. The zinc within these granules forms a complex with insulin to develop storage crystals. Human ZnT8 exists in three major polymorphic forms with amino acid differences at position 325: arginine, tryptophan, or glutamine. The arginine and tryptophan forms are the most common, and the glutamine form is rare. Sera from some T1D patients recognize ZnT8 epitopes not affected by the polymorphism at position 325. Their sera react with all polymorphic antigen forms, but some patients’ sera are specific to the polymorphic antigen. The latter sera are from T1D patients who are homozygous for either the arginine or tryptophan polymorphism. Other patients may have autoantibodies that are specific for the glutamine polymorphism, but very few T1D patients are homozygous for this polymorphism because it is not common.
Biochemical Assays and Standardization
ICAs measured by indirect immunofluorescence are still among the most sensitive markers of T1D; however, they have been largely replaced by biochemical autoantibody assays for the major specific protein antigens described above. While most labs use various forms of fluid phase RBAs for biochemical autoantibody assays, some commercial ELISAs now perform as well or better than RBAs in detecting GAD65 (11) and IA-2ic (unpublished data).
RBAs generally begin with the synthesis of 35S-methionine labeled protein antigens. For assay, the laboratorian incubates the test serum with labeled antigen and precipitates the resultant autoantibody-bound antigen with Protein-A Sepharose. Each serum sample is usually tested in duplicate on a 96-well plate. The laboratorian must then wash the plates to remove unbound labeled antigen and measure the radioactive label. For the IAA assay, commercial 125I-labeled insulin is generally used as the substrate.
Standardizing these assays is challenging. To improve comparability of measurements among laboratories, the World Health Organization (WHO) adopted a serum reference standard for GADA and IA-2A assays. These standards have assigned values of 250 units/mL for each autoantibody (12). Progress on an international standard for IAA, however, has been slow because of difficulties obtaining suitable IAA-positive sera. Standards for ZnT8 autoantibodies are also being developed.
Given the need for standardized autoantibody testing for diabetes, the Immunology of Diabetes Society (IDS) and the U.S. Centers for Disease Control and Prevention (CDC) established the Diabetes Autoantibody Standardization Program (DASP) in 2000 to improve comparability and to act as a mechanism to evaluate new autoantigens and test methodologies. The goal of the organization is to improve detection and diagnosis of autoimmune diabetes by: 1) providing technical support, training, and information about the best methods; 2) providing proficiency testing to evaluate laboratory performance; 3) supporting development of highly sensitive and specific measurement technologies; and 4) developing reference materials.
Since its inception, DASP has conducted six international workshop in which laboratories assay blinded samples from 50 patients with new onset T1D and up to 100 controls. This format provides an evaluation of the sensitivity and specificity of each test and enables DASP to assess implementation of assay methods and to document any improvement in performance. Among the major accomplishments to date, DASP demonstrated that the performance of ELISAs for measuring T1D autoantibodies can equal that of RBAs and validated ZnT8 as the fourth major T1D autoantigen. Other activities include: validation of IA-2β autoantibody assays; evaluation of the stability of the WHO GADA and IA-2 autoantibody standard; validation of affinity measurements to improve IAA test performance (13); and evaluation of standard method protocols for GADA and IA-2A.
Laboratories that have participated in multiple DASP workshops have improved the quality of their autoantibody assays. In 2009, participating labs found the majority of new-onset patient samples were positive for multiple autoantibodies, including all four major autoantigens, followed by patient samples positive for the combination of GADA, IA-2, and ZnT8 autoantibodies.
DASP is also looking at new technology to measure autoantibodies. Several DASP 2009 autoantibody assays used a new non-radioactive assay format, the luciferase immunoprecipitation system (LIPS), that looks promising.
To further improve quantitative agreement of assays for GAD65 and IA-2ic, a harmonization effort led by NIH research consortia labs and DASP committee members developed a standard method protocol (15), which is now available to all DASP participants. The National Institute of Diabetes and Digestive and Kidney Diseases research consortia laboratories have common standards, but other participating DASP laboratories standardize to the WHO standard for GADA and IA-2A. Harmonization of RBAs included the use of 35S-methionine rather than 3H-leucine as the radioisotope label for the synthesized antigen and standardization to the pTh-GAD65 plasmid for the GADA assay and the pSP64-PolyA-IA-2ic plasmid for the IA-2A assay. In the DASP 2009 workshop, the IA-2A assays showed excellent quantitative agreement among consortia laboratories and concordance for autoantibody positivity. Remaining differences in GADA measurements, however, suggest that further work is needed to bring this assay into alignment.
Guidance: Autoantibody Tests for Autoimmune Diabetes
Although autoantibody tests are not required for standard diabetes care (16), evidence suggests that they are predictive of T1D and help distinguish autoimmune T1D and LADA from non-autoimmune diabetes. Positivity to increasing numbers of autoantibodies indicates the individual’s autoimmune response is spreading and that the disease is progressing (17). The predictive ability of autoantibody tests therefore increases with the number of autoantibodies detected in an individual (Figure 1a, below) and may be influenced by the autoantibody titer, as well. There also appears to be a hierarchy of diabetes-relevance in the autoantibody response against different antigenic targets within and between islet autoantigens (17). For example, risk is relatively low in relatives with GADA or IAA alone, around 20% within 10 years. The presence of IA-2A alone, however, is associated with a risk of about 50% within 10 years, which is similar to the risk of multiple non-IA-2-autoantibodies (ICA, GADA and/or IAA) (Figure 1b, below). Among IA-2A positive relatives, risk can be further stratified according to the presence or absence of autoantibodies to IA-2B (Figure 1c, below).
Stratification of Diabetes Risk
a. Based on Numbers of Autoantibodies
b. Based on Autoantibody Combination
c. Based on IA-2A Epitopes
Stratification of diabetes risk in islet autoantibody (IAA, GADA, and/or IA-2A)-positive first-degree relatives of T1D patients (n = 180) based on autoantibody number (a), target antigen specificity (b), and IA-2A reactivity against IA-2β (c).
Reprinted with permission from The American Diabetes Association. Copyright 2004, from Diabetes 2004;53:384–392.
Clinicians now use seroconversion to autoantibody positivity as an intermediate stage for individuals at genetic risk of developing T1D. When an effective way to reintroduce tolerance in autoimmune diabetes becomes available, high-quality, high-throughput autoantibody tests will be essential for identifying individuals who can benefit from such treatment.
Future therapeutic intervention trials for disease will need to consider the characteristics of the population and assay performance in screening for the disease. These strategies will likely evolve as researchers devise new interventions and develop new tests. Two early intervention trials for which recruitment was based on ICA and/or IAA have already proven the efficacy of islet autoantibodies for T1D prediction (18, 19). Researchers have also used the presence of IAA, GADA, and IA-2A as a criterion for recruiting individuals into prevention studies and for identifying initiation of autoimmunity in natural history studies of T1D, such as the The Environmental Determinants of Diabetes in the Young (TEDDY) Study.
- CDC Diabetes Fact Sheet, 2007, (available online).
- Naik RG, Brooks-Worrell BM, Palmer J. Latent autoimmune diabetes in adults. J Clin Endocrinol Metab 2009; 94:4635–4644.
- Zimmet PZ. The pathogenesis and prevention of diabetes in adults: genes, autoimmunity, and demography. Diabetes Care 1995;18:1050–1064.
- Eisenbarth GS. Type 1 diabetes mellitus. A chronic autoimmune disease. N Engl J Med 1986;314:1360–1368.
- Taplin CE, Barker JM. Autoantibodies in type 1 diabetes. Autoimmunity 2008;41: 11–18.
- Bottazzo GF, Florin-Christensen A, Doniach D. Islet-cell antibodies in diabetes mellitus with autoimmune polyendocrine deficiencies. Lancet 1974;304:1279–1283.
- Palmer JP, Asplin CM, Clemons P, et al. Insulin antibodies in insulin-dependent diabetics before insulin treatment. Science 1983;222:1337–1339.
- Baekkeskov S, Aanstoot HJ, Christgau S, et al. Identification of the 64K autoantigen in insulin-dependent diabetes as the GABA-synthesizing enzyme glutamic acid decarboxylase. Nature 1990;347:151–156.
- Payton MA, Hawkes CJ, Christie MR. Relationship of the 37,000- and 40,000- Mr tryptic fragments of islet antigens in insulin-dependent diabetes to the protein tyrosine phosphatase-like molecule IA-2 (ICA512). J Clin Invest 1995;96:1506–1511.
- Wenzlau JM, Moua O, Sarkar SA, et al. S1C30A8 is a major target of humoral autoimmunity in type 1 diabetes and a predictive marker in prediabetes. Ann NY Acad Sci 2008;1150:256–259.
- Torn C, Mueller PW, Schlosser M, et al. Diabetes Antibody Standardization Program: evaluation of assays for autoantigens to glutamic acid decarboxylase and islet antigen-2. Diabetologia 2008;1:846–852.
- Mire-Sluis AR, Gaines Das R, Lernmark A. The World Health Organization International Collaborative Study for islet cell antibodies. Diabetologia 2000;43:1282–1292.
- Achenbach P, Schlosser M, Williams AJK, et al. Combined testing of antibody titer and affinity improves insulin autoantibody measurement: Diabetes Autoantibody Standardization Program. Clin Immunol 2007;122:85–90.
- Burbelo PD, Groot S, Dalakas MC, et al. High definition profiling of autoantibodies to glutamic acid decarboxylases GAD65/GAD67 in stiff-person syndrome. Biochem Biophys Res Commun 2008;366:1–7.
- Bonifacio E, Yu L, Williams AK, et al. Harmonization of glutamic acid decarboxylase and islet antigen-2 autoantibody assays for National Institute of Diabetes Digestive and Kidney Diseases consortia. J Clin Endocrinol Metab 2010;95:3360–3367.
- Bingley PJ. Clinical applications of diabetes antibody testing. J Clin Endocrinol Metab 2010;95:25–33.
- Achenbach P, Warncke K, Reiter J, et al. Stratification of type 1 diabetes risk on the basis of islet autoantibody characteristics. Diabetes 2004;53:384–392.
- Bingley PJ, Gale EAM. The European Nicotinamide Diabetes Intervention Trial (ENDIT) Group, Diabetologia. Progression to type 1 diabetes in islet cell antibody-positive relatives in the European Neicotinamide Diabetes Intervention Trial: the role of additional immune, genetic and metabolic markers of risk. 2006;49:881–890.
- Orban T, Sosenko JM, Cutherbertson D, et al. For the Diabetes Prevention Trial-Type 1 Study Group. Pancreatic islet autoantibodies as predictors of type 1 diabetes in the Diabetes Prevention Trial-Type1, Diabetes Care. 2009;32:2269–2274.
Patricia W. Mueller, PhD, is chief of the Molecular Risk Assessment Laboratory at the Centers for Disease Control and Prevention, Atlanta, Ga. Address all correspondence to: firstname.lastname@example.org.
Peter Achenbach, MD, is a physician and clinical scientist at the Institute of Diabetes Research, Helmholtz Center, Munich, Germany.
Vito Lampasona is a senior scientist at the Center of Genomics, Bioinformatics, and Biostatistics, San Raffaele Scientific Institute, Milan, Italy.
Michael Schlosser is a senior scientist in medical biochemistry and molecular biology, University of Greifswald, Karlsburg, Germany.
Alistair J. K. Williams is a research fellow in clinical science at North Bristol, University of Bristol, Bristol, U.K.