Antibody dynamics reveal how fast type 1 diabetes develops in children

Antibody tests can detect the early stages of type I diabetes in affected children, even before the first symptoms of the disease occur. Researchers at the Helmholtz Zentrum München and the Paul Langerhans Institute in Dresden, both partner of the German Center for Diabetes Research, have shown that bioinformatics models can be used to obtain better predictions about the progression of the early stages of diabetes through to the clinically symptomatic disease. Their findings are reported in the medical journal ‘Diabetologia’.

Type 1 diabetes is an autoimmune disease, which is becoming increasingly prevalent - also in Germany. During the development of the disease, patients form antibodies against the insulin-producing beta cells in their pancreas. In most cases, this occurs in early childhood. Scientists from the Institute of Diabetes Research (IDF) at the Helmholtz Zentrum München have therefore developed a test, which can detect – in a single drop of blood – the presence of the antibodies that indicate an early stage of the disease. That is the case in about four out of 1.000 children between the ages of two and five years in Bavaria, according the latest results of the Fr1da study. “However, the diagnosis is complicated by the fact that we are dealing with a total of four different antibodies,” explains Dr. Peter Achenbach, Deputy Director of the IDF. “Furthermore, not all antibody types are permanently present in the pathogenesis phase. Rather, the individual antibodies behave dynamically – in other words, they can also come and go.”

Presumed high-risk group develops the disease later than expected
In the current study, the scientists wanted to focus on this complex behavior and find out whether more precise information could be gleaned from the respective antibody dynamics than before. For their analyses, they cooperated closely with a team of data modeling experts from the Scientific Computing Research Unit (ASC) headed by Dr. Wolfgang zu Castell. “Our new approach, involving the time-course analysis of antibody patterns, enables us not only to determine whether and which antibodies are present or not, but also to observe the dynamics of the different antibodies and group children with similar profiles – and then relate our findings to the development of the disease,“ says first author Dr. David Endesfelder from the ASC. The researchers examined 88 children, each of whom had developed several different antibodies and had been observed for up to 20 years as part of the BABYDIAB study.

Thanks to these new methods of analysis, the researchers were able to show, for example, that the development of clinically symptomatic diabetes was significantly delayed in some children, who – as a result of their antibody constellation – had previously been considered among those at high risk for the rapid development of the disease. “That was very surprising and shows us that with our new approach we are not merely differentiating between ones and zeros, in other words whether antibodies are present or not. We are now also beginning to gain an ever better understanding of the different facets of the developing type 1 diabetes – and at a more detailed level,” Dr. Achenbach notes.

In future, the researchers aim to further intensify their analyses so as to gain more accurate information about the development of the disease, the underlying factors and mechanisms and also make improved predictions about the potential outcomes of treatments such as oral insulin vaccination for at-risk children, which is carried out by the IDF, amongst others.

Further information
Overall, the researchers compared the profiles of 88 children in the early stages of type I diabetes with respect to the autoantibodies IAA, GADA, IA-2A and ZnT8A. The observation period was 20 years. During this time, they identified nine different clusters, each of which represented different antibody dynamics and were associated with varying rates of development of the clinical disease.

Original Publication:
Endesfelder, D. et al. (2016). A novel approach for the analysis of longitudinal profiles reveals delayed progression to type 1 diabetes in a subgroup of multiple-islet autoantibody-positive children, Diabetologia, DOI: 10.1007/s00125-016-4050-0