Islet autoantibodies and type 1 diabetes: does the evidence support screening?

Type 1 diabetes (T1D) is a chronic progressive autoimmune disorder with complex polygenic susceptibility, usually associated with certain HLA alleles (IDDM1 locus). Environmental factors, which are poorly defined, also contribute to the pathogenesis. T1D is characterized by lymphocyte infiltration into the islets of Langerhans in the pancreas, leading to inflammation and selective destruction of the insulin-producing -cells, resulting in hyperglycemia (1 ). Patients with T1D fail to produce insulin and are dependent on exogenous insulin to maintain life. Although it is considerably less common than type 2 diabetes, the worldwide prevalence of T1D is increasing by approximately 3% per annum. The incidence varies widely among countries. In Americans under the age of 20 years, the prevalence of T1D rose by 23% between 2001 and 2009, and 30 000 people are diagnosed annually in the US with T1D. Progression to T1D is typically marked by the presence of islet-specific autoantibodies in the serum. In humans, autoantibodies are present months to years before disease onset, and a similar trend is seen in the nonobese diabetic (NOD) mouse model of autoimmune diabetes (2 ). The rate of T1D development varies among individuals, possibly due to non-HLA genetic factors and/or environmental factors beyond the initial trigger. Most of the current information on the pathogenesis of T1D from the initial triggers to the final effector stages of -cell destruction has been derived from animal models that mimic the human disease (3 ). T lymphocytes are central determinants for -cell destruction in T1D. Nevertheless, autoantibodies and B lymphocytes are components of some autoimmune diseases and may contribute to the pathogenesis of T1D. Multiple studies have documented the role of autoantibodies and B lymphocytes in NOD mice. Observations include a low incidence of diabetes in NOD mice that are genetically deficient in B lymphocytes and the prevention of diabetes by eliminating maternal transfer of autoantibodies from NOD mice to the progeny. A phase 2 clinical trial by the T1D TrialNet AntiCD20 Study Group revealed that a single course of rituximab, an anti-CD20 monoclonal antibody that selectively depletes B lymphocytes, slowed disease progression over 1 year in newly diagnosed T1D patients. Moreover, rituximab significantly reduced the amount of autoantibodies to insulin in the patients’ serum. However, it is not known if the effect of rituximab is mediated by reduced production of autoantibodies and/or by impairment of B-cell antigen-peptide presentation to CD4-positive T cells. Thus, the potential role of autoantibodies in the pathogenesis of T1D in humans remains unresolved. Patients with T1D commonly produce autoantibodies to islet cell cytoplasm (ICA), native insulin (insulin autoantibodies, or IAAs), the 65-kDa isoform of glutamic acid decarboxylase (GAD65), the insulinoma antigen 2 protein (IA-2), and variants of zinc transporter 8 (ZnT8). At least 1 autoantibody is present in 95% of individuals with T1D upon hyperglycemia detection (4, 5 ). ICA measured by indirect immunofluorescence in frozen sections of human pancreas are the most sensitive markers. However, the assays are cumbersome, labor intensive, and difficult to standardize and have been largely replaced by immunoassays. Most current autoantibody detection methods are based on quantitative radiobinding assays. Radiolabeled recombinant human insulin, GAD65, or IA-2 is incubated with patient’s serum, and the amount of radioactivity in the precipitated complex is proportional to the antibody concentration. Nonradioactive assays, such as ELISA-base formats, are limited for most islet cell antigens (currently available only for GAD65 and ZnT8). Cross-sectional studies in relatives of patients with T1D have documented that the development of multiple autoantibodies augments the risk for progression to T1D. The risk of developing future T1D is highest in individuals with 2 or more autoantibodies. Multiple confounding factors have prevented formulation of a 1 Washington University School of Medicine, Department of Pathology and Immunology, St. Louis, MO; 2 Department of Laboratory Medicine, NIH, Bethesda, MD. * Address correspondence to the authors at: Boris Calderon, Department of Pathology and Immunology, Washington University School of Medicine, 660 South Euclid Ave., St. Louis, MO 63110; fax 314-362-1403; e-mail [email protected]; or David B. Sacks, Department of Laboratory Medicine, National Institutes of Health, Bldg. 10, Rm. 2C306, 10 Center Dr., Bethesda, MD 20892; fax 301-402-1885; e-mail [email protected]. Received September 20, 2013; accepted November 25, 2013. Previously published online at DOI: 10.1373/clinchem.2013.212381 3 Nonstandard abbreviations: T1D, type 1 diabetes; ICA, islet cell cytoplasm; IAA, insulin autoantibody; GAD65, glutamic acid decarboxylase; IA-2, insulinoma antigen 2 protein; ZnT8, zinc transporter 8; DASP, Diabetes Antibody Standardization Program; IASP, Islet Antibody Standardization Program. Clinical Chemistry 60:3 000 – 000 (2014) Perspectives