Islet autoantibodies as biomarkers
المؤلف:
Holt, Richard IG, and Allan Flyvbjerg
المصدر:
Textbook of diabetes (2024)
الجزء والصفحة:
6th ed , page206-207
2025-11-11
30
Islet autoantibodies have been studied longitudinally in a large number of first- degree relatives of people with type 1 diabetes until the clinical onset of type 1 diabetes. The islet autoantibodies are not thought to be directly involved in β- cell destruction. For example, complement- mediated antibody- dependent cytotoxicity would require the autoantigen to be expressed on the β- cell surface. A subset of autoantibodies to ZnT8 may fulfil the criteria for cell sur face expression. Further studies are needed to fully establish to what extent islet autoantibodies perse contribute to pathogenesis and risk for diabetes. The demonstration of islet autoantibodies years before the onset of clinical symptoms has made it possible to identify individuals at high risk for type 1 diabetes and to initiate therapeutic intervention trials. In particular, the number of islet autoantibodies seems to affect the rate of progression to clinical onset. Children recruited and followed in three studies (DAISY, Colorado, n = 1962; DIPP, Finland, n = 8597; and BABYDIAB, Germany, n = 2818) were merged in a joint analysis. The data showed that progression to diabetes at 10- year follow- up after islet autoantibody seroconversion varied with the number of islet autoantibodies: no islet autoantibodies was 0.4% (95% confidence interval [CI], 0.2–0.6%); a single islet autoantibody was 14.5% (95% CI, 10.3–18.7%); and multiple islet autoantibodies was 69.7% (95% CI, 65.1–74.3%). These results were further corroborated in 1815 first- degree relatives who were followed until diabetes diagnosis in the TrialNet Pathway to Prevention study (Figure 1). Participants younger than 12 years had a higher risk of progressing to clinical onset regardless of whether they had two autoantibodies only (panel a) or more than two autoantibodies (panel b). These data raise the question of whether islet autoantibody screening should be used in clinical practice. The effect of age and number of autoantibodies strongly suggests that the het erogeneity in progressing to clinical onset needs to be taken into account in secondary prevention clinical trials.
Fig1. Type 1 diabetes- free cumulative incidence with 95% confidence interval (shaded area) in TrialNet participants with two (a) or more than two (b) islet autoantibodies. (a) In participants with two autoantibodies, there was a lower cumulative incidence among those older than 12 years of age (blue) compared to those younger than 12 years of age (red) (p = 0.0496). (b) In participants with more than two autoantibodies, the cumulative incidence was markedly different between those younger than 12 years (red) and those older than 12 years of age (blue) (p = 0.0008). Source: Jacobsen et al. 2020.
A first step was to use a meta- analysis to assess the evidence of an association between islet autoantibodies and the development of type 1 diabetes in a pooled population of both genetically at- risk individuals and people without a definite genetic background. In the meta- analysis, 21 prospective cohort studies with 71 482 participants, who were followed for a median of 7 years and of whom 926 developed type 1 diabetes, evaluated the role of islet autoantibodies in prediction of type 1 diabetes progression. Compared to people without autoantibodies, those positive for any type or number of islet autoantibody had a marked risk for type 1 diabetes (risk ratio [RR] 150.42 [95% CI 87.34, 259.04]). People with multiple autoantibodies had a ninefold higher risk than those with a single islet autoantibody. This meta- analysis is raising the question of the place of screening the general population for islet autoantibodies, since the likelihood of individuals with multiple islet autoantibodies eventually developing type 1 diabetes is high.
A second step was the recent discovery that the aetiology of islet autoimmunity may represent two different endotypes, dependent on whether IAA appear first compared to GADA. This raises the question of whether the first- appearing autoantibody is related to disease progression. In longitudinal sampling of IAA, GADA, and IA- 2A in a cohort of 24 662 people combined from DAISY, DIPP, DiPiS, DEW- IT, and BABYDIAB, 2172 individuals fulfilled the criteria of two or more follow- up visits and autoantibody positivity at least once, while 652 progressed to type 1 diabetes during 15 years of follow- up (Figure 2). Continuous- time hidden Markov models were used to let the data visualize the latent health state of the participants during 5 years of follow- up. Three different trajectories were dis covered from 11 latent states. TR1 represents those with multiple islet autoantibodies in the first sample. Only 40% remained diabetes free after 5 years. TR2 represents people with predominantly IAA as the first- appearing autoantibody. Diabetes- free survival was 62%. TR3 was people with GADA first who showed 88% diabetes- free survival after 5 years of follow- up. Progression rates within each trajectory could be refined by age, sex, and HLA- DR, thereby providing a clinically useful prediction of disease onset. This type of approach with the aid of further machine learning and artificial intelligence may provide the means by which islet autoantibody analyses will become meaningful for the individual.
Fig2. Three trajectories of data- driven islet autoantibodies towards stage 3 diabetes. A 11- state hidden Markov mode discovered the three trajectories for (a) diagnosed (D, 643 individuals who were diagnosed with type 1 diabetes) and (b) undiagnosed (UD, 1502 individuals who remained not diagnosed with diabetes during follow- up). The data in the table to the left describe the 11 states as the probabilities for each islet autoantibody (glutamic acid decarboxylase autoantibody [GADA], insulin autoantibody [IAA], or islet antigen- 2 autoantibody [IA- 2A]) for each state (the heat map indicate green for 0 and red for 1). Waterfall diagrams to the right show visits (dots) and the respective trajectory state over time in years (x- axis). Each person is depicted from the appearance of the first autoantibody and can then be followed from one state to the next in the waterfall. The three trajectories reveal themselves as TR1, predominantly multiple islet autoantibodies (IAb) as the first islet autoantibody (256 D, 483 UD); TR2, predominantly IAA first (273 D; 257 UD); and TR3, predominantly GADA first (11 D; 762 NP). P, progressors; NP, non- progressors. Source: Kwon et al. 2022. Licensed under CC BY 4.0.
الاكثر قراءة في الامراض المناعية
اخر الاخبار
اخبار العتبة العباسية المقدسة
