Abstract

Diabetes is not a single homogenous disease but composed of many diseases with hyperglycaemia as a common feature. Four factors have, historically, been used to identify this diversity: the age at onset; the severity of the disease, i.e. degree of loss of ß-cell function; the degree of insulin resistance and the presence of diabetes-associated autoantibodies (DAA). Our broad understanding of the distinction between the two major types, type 1 diabetes and type 2 diabetes, are based on these qualities, but it has become apparent that they do not precisely capture the different disease forms. Indeed, both major type of diabetes have common features, encapsulated by adult-onset autoimmune diabetes and maturity onset diabetes of the young. As a result, there has been a repositioning of our understanding of diabetes. Much of what we know about type 1 diabetes relates to childhood-onset type 1 diabetes.  This condition is induced early in life and there is a chronic and in some cases prolonged prelude before clinical diabetes develops. The genetic background to the disease is well characterised. We know that the characteristics of autoimmune diabetes vary substantially across a range of clinical, immunological and metabolic features. This landscape of autoimmune diabetes displays a disease in line with that seen in other autoimmune diseases with a similar genetic and immune background. Like them, autoimmune diabetes is predominantly: middle-aged in onset, of varying disease severity, HLA related, with the genetic load being greater the younger the age at diagnosis, with predictive autoantibodies, with titer of autoantibodies reflecting the disease severity, while many with such autoantibodies do not develop clinical disease.  Given the broad clinical phenotype there are diverse therapeutic options within autoimmune diabetes, while the term, non-autoimmune type 2 diabetes, obscures optimal management strategy because it too encompasses substantial heterogeneity. Underlying these developments is a general progression towards precision medicine with the need for precise patient characterisation, currently based on clinical phenotypes but in future augmented by laboratory-based tests.