What is being tested?

Islet cell cytoplasmic autoantibodies (ICA), insulin autoantibodies (IAA), glutamic acid decarboxylase antibodies (GADA), insulinoma-associated-2 autoantibodies (IA-2A) and Zinc Transporter 8 antibodies (ZnT8) are a group of tests that measure diabetes-related autoantibodies. These autoantibodies do not cause type 1 diabetes but serve as markers of the body’s destructive immune response against its own cells that produce insulin (e.g., the beta cells in the pancreas). When about 80-90% of the beta cells are destroyed by the immune system, symptoms of diabetes such as frequent urination, thirst, weight loss, and poor wound healing occur. Without sufficient insulin action, hyperglycaemia results. If the symptoms are not detected and hyperglycaemia is not treated, a diabetic medical crisis can occur that can develop over a few weeks or even a few days.

The ICA test measures a group of islet cell autoantibodies targeted against a variety of islet cell proteins. It is a semi-quantitative test performed by indirect immunofluorescence. GADA, IA-2A and ZnT8 are antibodies against three of the specific islet cell antigens. The only antigen believed to be highly specific to beta cells is insulin, and antibodies to insulin are abbreviated IAA. The IAA test does not differentiate whether the body’s immune system is making autoantibodies against insulin or if the immune system is making antibodies against insulin that has been injected (either human or animal) in the treatment of any type of diabetes. At present commercial assays for insulin antibodies are not accurate enough for clinical use. They are not recommended in the clinical investigation of autoimmune diabetes.

About 10% of all cases of diabetes are type 1 (autoimmune) in origin. Of these, about 75% are diagnosed in patients younger than 20 years old. Type 1 diabetes was previously known as juvenile or insulin-dependent diabetes but has been re-characterised to reflect beta cell destruction. Islet autoantibodies can be detected in the blood stream months to years before the development of type 1 diabetes. While nondiabetic individuals with islet autoantibodies are at high risk for the development of type 1 diabetes, not every person with islet autoantibodies will develop type 1 diabetes. When autoimmune type 1 diabetes is present, one or more of the diabetes autoantibodies will be present in about 95% of patients at the time of initial diagnosis.

How is it used?

In general, autoantibody testing is not required to make a diagnosis of autoimmune type 1 diabetes. In the general population, there is no benefit to autoantibody screening, but it can be useful in patients at high risk of diabetes (e.g. siblings of known type 1 diabetics and offspring of diabetic parents).

Diabetes-related (islet) autoantibody testing is primarily ordered to help distinguish between autoimmune type 1 diabetes and diabetes due to other causes (e.g., diabetes resulting from obesity and insulin resistance). If ICA, GADA, ZnT8 or IA-2A are present in an individual with diabetes, the diagnosis is highly likely to be that of type 1 diabetes. IAA testing must be performed before insulin therapy is initiated, and is difficult as commercial assays are less than optimal for these antibodies. At present they are only recommended in research settings.

It is generally recommended that testing be done for the GADA/IA-2A combination since this approach is more cost-effective. The autoantibody profile is usually different between children and adults. IAA is usually the first marker to appear in young children at risk of diabetes. As the disease evolves, this may disappear and ICA, GADA and IA-2A become more important. IA-2A is less commonly positive at the onset of type 1 diabetes than either GADA or ICA. Whereas about 50% of children with new-onset type 1 diabetes will be IAA positive, IAA positivity is not common in adults. The presence of ICA antibodies in anyone has the greatest predictive value in the development of type 1 diabetes. ZnT8 may occasionally be the sole antibody present in type 1 diabetes.

In research settings where investigators want to predict the development of type 1 diabetes, islet autoantibody testing can be carried out. The more islet autoantibodies that a nondiabetic person has in their blood stream, the higher their risk for later developing type 1 diabetes.

When is it requested?

A combination of these autoantibodies may be ordered when a patient is newly diagnosed with diabetes and the doctor suspects that the condition may be due to an autoimmune process. One or more of the autoantibodies may be ordered on the siblings of a patient diagnosed with type 1 diabetes or on the offspring of diabetic parents. This may be done initially and then again at intervals recommended by the doctor in a research setting.

What does the result mean?

Normally, nondiabetic individuals in the general population will not have any of these islet autoantibodies. However, when islet autoantibodies are detected in individuals in the general population or siblings of affected patients, there may be an increased risk for type 1 diabetes since false positives are known to occur. Many of these islet autoantibody-positive individuals will never develop diabetes. When type 1 diabetes does not develop, the level of the islet autoantibody in the blood is usually low and the islet autoantibody may be transient.

Some patients who do have type 1 diabetes will never develop detectible amounts of islet autoantibodies, although this is rare. The majority of people (95% or more) with new-onset type 1 diabetes will have at least one islet autoantibody. Therefore, if one or more islet autoantibody (e.g., ICA, ZnT8, GADA, IA-2A, and/or IAA) are present in a patient with symptoms of diabetes, the diagnosis of type 1 diabetes is often considered confirmed.

In nondiabetic individuals who are positive for one or more islet autoantibodies, there is an increased risk for type 1 diabetes as mentioned above. The more islet autoantibodies that are present, the higher is the individual’s risk for developing type 1 diabetes. If a non-diabetic individual with one or more islet autoantibodies has a low insulin response to the intravenous injection of glucose, their risk for type 1 diabetes can be very high. In first degree relatives of patients with type 1 diabetes who have ICA and have a low insulin response to the intravenous injection of glucose, the 5-year risk of developing type 1 diabetes is approximately 60%. Because there are no effective therapies to prevent type 1 diabetes, general population screening for islet autoantibodies or testing first degree relatives of patients with type 1 diabetes is not recommended.

Is there anything else I should know?

It is up to the doctor and patient to decide together which islet autoantibodies to test for at any given time. Because GADA, ZnT8 and IA-2A assays are automated, these tests are generally more available than ICA testing, which is labour-intensive and requires considerable expertise in interpretation.

Islet autoantibodies may also be seen in patients with other autoimmune endocrine disorders such as Hashimoto thyroiditis or autoimmune Addison disease.

Common questions

• Do ICA, GADA, ZnT8 and IA-2A destroy the beta cells?

They are associated with beta cell destruction and reflect an ongoing autoimmune process, but they are not thought to cause the damage.

• Does early detection of beta call destruction allow its prevention?

Not currently. What it does do is allow for diabetes to be addressed as soon as symptoms such as frequent urination, weight loss, and hyperglycaemia appear. This can help establish diabetic glucose control and can in turn help minimise the occurrence of complications such as the kidney and eye damage that are seen with diabetes. At present, strategies are being devised that may prevent Type 1 diabetes, through the use of monoclonal antibodies immune modulation in those at high risk.