What is the earliest lesion?
Type 1 diabetes is clinically diagnosed when 70–80% of insulin producing beta cells are destroyed through a cytotoxic T-cell mediated autoimmune process. The time course of this destruction is largely unknown. Circulating autoantibodies, considered to be surrogate markers for ongoing beta cell destruction, indicate that the disease process may take many years before it leads to overt disease. Recent histopathological studies in autoantibody-positive non-diabetic organ donors give us our first insights into the pré-clinical stages of the disease process.
What do we know about islet changes before clinical diagnosis?
Very little is known about the histopathological changes in pré-diabetic patients. Although we can identify individuals who are at an increased risk for developing the disease because of the presence of circulating autoantibodies directed against beta cell antigens (insulin, IA-2, GAD, ZnT8) in combination with a genetic predisposition (HLA-DQ/DR), it has proven difficult to obtain islets of Langerhans from such individuals. Biopsies are not without risk and contain only few islets of Langerhans due to their dispersed localisation. Autopsy material is often of low quality due to rapid autolysis of the pancreas.
Organ donors provide crucial new insights
A breakthrough was provided by a large collection of pancreas biopsies taken in the process of quality control prior to the use of the organ for islet isolation and transplantation in diabetic recipients. Such biopsies and accompanying serum samples were collected for many years and carefully stored in Biobanks. They allowed the retrospective testing for autoantibody positivity and histopathological screening for islet lesions. Some 60–70 organ donors positive for one or more autoantibodies have been identified to date. Insulitis and pseudoatrophic islets, the characteristic lesions in recent onset type 1 diabetes, were found in three out of four organ donors positive for multiple autoantibodies in combination with a susceptible HLA-DQ genotype. However, the fraction of islets showing disease was small: insulitis was found in only 0.2–9% of islets and pseudoatrophic islets (ie islets devoid of beta cells) were relatively rare. Most islets (>90%) were normal and no decrease in beta cell mass was observed. These results indicate that (multiple) autoantibody positivity in non-diabetic patients point to subclinical disease, with islet changes being still relatively minor and beta cell mass not clearly decreased. They contrast with the findings in recent onset patients were 50–60% of islets are pseudo-atrophic and beta cell mass is significantly affected.
Can islets compensate for early beta cell damage?
An exciting observation found in an organ donor with subclinical disease was the presence of beta cell replication in the inflamed islets. Normally, beta cells in the adult human pancreas rarely divide. The presence of dividing beta cells under inflammatory conditions may indicate that islets have retained a capacity for regeneration that can be activated as part of a repair process.
Biobanks lead the way
Detailed molecular studies of the early stages of beta cell destruction will be necessary to identify the initiating factors that are involved and characterize the antigens towards which the cytotoxic T-cell response is directed. Large collections of human pancreas samples taken under standardized conditions and accompanied by blood samples and detailed clinical data are currently being collected to reach this aim. Initiatives include the Network for Pancreatic Organ Donors with Diabetes (www.jdrfnpod.org) and the Diabetes Biobank in Brussels (diabetesbiobank.com).