Immunoregulatory differences between adult onset type 1 diabetes and latent autoimmune diabetes in adults (LADA)

University dissertation from Department of Clinical Sciences, Lund University

Abstract: Autoimmune diabetes is characterized by cell mediated autoimmunity against insulin producing beta cells in pancreatic islets. Type 1 diabetes is the major cause of diabetes in children and adolescents and is believed to be mediated by a Thl driven destruction of beta cells. The much more prevalent type 2 diabetes is not considered as an autoimmune disease but diabetes specific autoantibodies (ICA or GADA) are found in 5-15% of the cases. These subjects are referred to as latent autoimmune diabetes in adults (LADA). It has been questioned if LADA is a unique entity or should be considered as type 1 diabetes. Subjects with LADA have a better preserved insulin secretion and a slower progression to beta cell failure at clinical onset compared to subjects with type 1 diabetes. Furthermore, there are adult onset type 1 diabetic patients that share the rapid progression to beta cell failure with juvenile onset diabetes. The hypothesis was therefore that type 1 diabetes and LADA have different autoimmune activity. The aim was to identify some of the immunoregulatory events that could explain differences in disease progression between the groups.

Heterozygosity of the diabetes associated high risk HLA haplotypes DR3/DQ2-DR4/DQ8 was considered as a risk factor for both adult onset type 1 diabetes and LADA. Also homozygosity for the microsatellite marker TNFa2 conferred an increased risk for both type 1 diabetes and LADA even though this was probably due to linkage disequilibrium with the HLA haplotypes. However, presence of either one of HLA DR3/DQ2 or DR4/DQ8 conferred an increased risk of type 1 diabetes but not LADA.

Autoantibodies directed against glutamic acid decarboxylase 65 (GADA) were predominantly of the IgGl subclass in both adult onset type 1 diabetes and LADA. Moreover, the IgG4 subclass was the second most common subclass in LADA but below detection limit in type 1 diabetes at clinical onset. This suggests a larger involvement of Th2 cells in LADA compared to in type 1 diabetes. The levels of all the GADA IgG subclasses decreased in type 1 diabetes three years after clinical onset while LADA remained their subclass profiles indicating a sustained difference in autoimmune response against the beta cell antigen GAD65 between the groups. The slower progression of beta cell failure in LADA might be due to a more balanced T cell response that reduces the cytotoxic Thl cell mediated response. Type 1 diabetes and LADA could thus be considered as different variations of autoimmunity.