📘 Type 1 diabetes-specific autoantibodies

Autoantibodies are proteins produced by your humoral immune system that mistakenly attack your own tissues. In type 1 diabetes, they specifically target proteins in pancreatic beta cells [1]. Surprisingly, however, they do not destroy the beta cells.

Their presence confirms that your diabetes is autoimmune in nature, not from other causes. The autoantibodies themselves do not destroy the beta cells. Type 1 diabetes-specific antibodies are only a sign that something else is fighting the beta cells to destroy them (the cellular immune system) [1]. They are rather witnesses to the autoimmune process, like smoke indicating the presence of fire. Their production begins months or years before diabetes symptoms appear [2].

There are four main autoantibodies tested: GAD (anti-glutamic acid decarboxylase), the most common; IA-2 (anti-tyrosine phosphatase); IAA (insulin autoantibodies), more common in young children; ZnT8 (anti-zinc transporter 8) [3]. Additionally, there is also ICA (islet cell antibodies), an older test with very low reliability, rarely used now.

Each autoantibody reflects an attack on a different protein in the beta cell. The more positive autoantibodies you have, the more certain the diagnosis of autoimmune type 1 diabetes [4]. In children and young people, all are usually tested, while in adults GAD is the most important for differentiating LADA from type 2 diabetes [3].

The presence of a single autoantibody does not mean you will develop type 1 diabetes. The risk in this case is approximately 15% in the next 15 years [5]. With two positive autoantibodies, the risk increases to approximately 70%, and with three to approximately 80% over 15 years [6]. Essentially, multiple autoantibodies mean substantially increased risk.

However, the speed of progression varies enormously [5]. Some people with autoimmunity develop diabetes in just a few months, while others in 20 years. Age and antibody titer significantly influence the speed of progression [7]. Periodic blood glucose monitoring allows early detection of stage 3 type 1 diabetes and starting treatment before ketoacidosis develops.

Once diagnosed with stage 3 type 1 diabetes (with clinical hyperglycemia), you no longer need to repeat autoantibody tests for disease monitoring [8]. Their presence or absence does not change insulin treatment and does not predict evolution. Initial testing is sufficient for possible diagnostic confirmation and documentation.

The exception is when there is not yet hyperglycemia in the diabetes range, especially for screening family members of a patient with type 1 diabetes. According to ADA/ISPAD 2024 guidelines, first-degree relatives can be tested at 6-12 months (children) or annually (adults), with increased frequency in those with multiple antibodies or positive IA-2 [8]. Children with high genetic risk can begin testing from age 2 [9]. The cost of tests often makes routine screening inaccessible.

Autoantibodies can persist for years or even decades after diagnosis, although their titer usually decreases over time [10]. GAD and IA-2 persist the longest, being detectable in 60-70% of adults after 8-10 years of disease [11]. IAA becomes uninterpretable after starting insulin treatment. ZnT8 rapidly decreases in the first 1-5 years (from ~35% to <10% after 10 years) [10].

The disappearance of autoantibodies does not mean possible cure or remission will follow. The destruction process has already been completed and very likely there is no longer sufficient target for attack. It is like after an armistice, where tanks can withdraw to a certain distance from the conflict zone, but without lasting peace accompanied by reconstruction, the destruction remains. In stage 3 of the disease, the persistent presence of autoantibodies does not mean more severe disease or worse prognosis [10].

The presence of two or more autoantibodies unequivocally confirms autoimmune type 1 diabetes and generally excludes other forms of diabetes [4]. The more autoantibodies you have, the more intense and rapid the immune attack [6]. In children, the presence of multiple autoantibodies is associated with earlier onset and more rapid loss of beta cell function [7].

For family members tested in screening, multiple autoantibodies mean almost certain progression to clinical diabetes (stage 3 of the disease) [12]. These subjects are generally included as priorities in prevention research studies. For you, after stage 3 diagnosis (with high blood glucose), the number of autoantibodies no longer influences treatment, but may be useful for genetic counseling of the family.

Yes, approximately 10% of people with typical clinical presentation of type 1 diabetes have no detectable autoantibodies [13]. This may be because the autoantibodies have disappeared, the tests do not detect all possible antibodies, or there is a non-autoimmune mechanism of beta cell destruction [14].

The absence of autoantibodies does not change the need for insulin treatment if you have the classic clinical picture: relatively sudden onset, tendency to ketoacidosis, young age, normal weight, very low C-peptide [13]. In some cases with high family burden, genetic testing for monogenic forms of diabetes (e.g. MODY) may be useful, or better yet, consultation with an experienced center [9].