Autoantibodies in Diabetes: Useful also for Prevention?

Until diabetes can be cured or prevented, the early identification of individuals at increased will provide the opportunity for early education. In turn, this education will be the foundation to ensure good disease management and minimise future complications associated with diabetes.

Type 1 diabetes mellitus is an organ-specific autoimmune disease which leads to the destruction of pancreatic islet cells. Specific antibodies against islet-cell antigens, such as insulin (IAA), glutamic acid decarboxylase autoantibodies (GADA) and the protein tyrosine phosphatase-like molecule (IA-2A), play an important role in the pathogenesis of type 1 diabetes mellitus, all of them being linked to insulin secretion by the b -cells, and are established markers in diagnosis and in the pre-diagnostic phase. GADA is the immune marker of higher diagnostic sensitivity in adult patients, whilst in childhood diabetes IAA and IA-2A are the dominating antibodies.

GADA identifies the so-called „late-onset diabetes in adults“, LADA. LADA means type 1 diabetes only diagnosed in adulthood (in contrary to the more frequent diagnosis in childhood), having a prevalence of an estimated 0.15–0.25% in the general population, which means a similar frequency as typical type 1 diabetes. In over 80% of cases, LADA patients develop insulin dependency within a few years after the diagnosis.

The advice nowadays is to determine GADA in all patients with adult onset of diabetes mellitus to check diagnosis of potential LADA (and not the typically diagnosed type 2 in adult onset diabetes). and so predicting development of insulin dependence. In addition, because there is a high frequency of other endocrine autoimmune diseases, in this subtype of diabetes, screening for other organ-specific autoantibodies (against e.g. thyroid gland, gastric mucosa cells, adrenal gland) should be performed in every LADA patient.

Prediction of diabetes risk

There are also interesting suggestions in another recent publication which discusses the screening of non-diabetic persons for their risk of developing type 1 diabetes by looking for these autoantibodies as markers. It bases on the knowledge that islet autoimmunity can precede manifestation diabetes by years–and early diagnosis, even before clinical outbreak of the disease, could help prevent complications by prevention of unnoticed hyperglycaemic states. The prevalence of the islet autoantibodies IAA, GADA, or IA-2A in euglycaemic subjects is between 5 and 5% in relatives of type 1 diabetes patients. For the general population, determination of more than one antibody is necessary for a meaningful risk prediction, out of statistical reasons. Prevalence of multiple islet antibodies in recent population studies was measured as 0.3–0.8%.

As far as the predictive value of autoantibodies for type 1 diabetes relatives is concerned (i.e. the probability to develop diabetes if autoantibodies are present), the figures of several trials are extremely different, reaching from 1–100%, depending on kind and number of determined antibodies, time of first determination, follow-up time etc. In summary, the article under discussion says, type 1 diabetes risk can be predicted only by using a variety of autoantibody characteristics, which in turn can be used in different combinations to select subjects of defined diabetes risk for intervention trials.

It is not sure, yet, whether these results are applicable to islet autoantibody-positive subjects in the general population, who not have relatives with type 1 diabetes. Some data suggest that multiple islet autoantibody positivity bears a similar risk for development of type 1 diabetes regardless of the family history, but the numbers of individuals studied to date are too small to allow accurate risk estimations in this subgroup.

At the time being, so the final conclusion, screening for such markers in the general population should be undertaken only in defined research studies. For individuals at increased risk for type 1 diabetes, it may be offered outside of such studies, but should be limited to centres with sufficient experience in testing, providing exact risk information and psychological counselling.