Researchers Close In On Protein Panel For Predicting Islet Autoimmunity

The results, which published recently in Cell Reports Medicine (DOI: 10.1016/j.xcrm.2023.101093), are the culmination of a nine-year study led by Metz looking at hundreds of proteins in more than 8,000 blood samples taken from nearly 1,000 children from birth up to age 6 at sites located in the U.S. (Seattle, Colorado, and Florida) and one each in Germany, Sweden, and Finland. The children were all participants in a larger umbrella study known as TEDDY (The Environmental Determinants of Diabetes in the Young), ongoing for the past two decades. 

Metz is now issuing a “call to arms” for researchers elsewhere to start evaluating the markers in their own cohorts of type 1 diabetes patients, or to screen patients in their general population studies, noting that replication is the hallmark of science. If one or more of the proteins translate to the broader disease population, it means the community is “potentially a big step closer” to preventing this relatively common form of diabetes that typically appears in adolescence, he says. 

Follow-on studies are being planned to analyze the markers in the second phase of the TEDDY study when the same at-risk children are now older (up to 15 years of age), continues Metz. He’d also like to perform these measurements on samples from other regions in the U.S., such as the northeast, and other major cities in Europe to ensure initial findings aren’t geographically biased or specific to the TEDDY cohort for some reason. 

With further study, it’s quite possible that a smaller number of protein markers will prove to be generalizable to a larger population, he says. If every individual genetically predisposed to type 1 diabetes were tested, perhaps only a four-protein panel would be generalizable across that entire population—making the proteomic-based diagnostics approach a more manageable undertaking for laboratories around the world. 

The longer-term hope here is that some of the markers will be incorporated into the clinical setting for the routine assessment of patients for the development of autoimmunity, Metz says. Currently, there is no way to know if or when either islet autoimmunity or diabetes will occur in people who are genetically predisposed. Doctors only know that when a patient develops at least two islet autoantibodies, they have islet autoimmunity and will develop diabetes—but not what triggers autoimmunity or when diabetes will manifest. 

It can take months, even years, for symptoms of type 1 diabetes to be noticed, and they can be severe and potentially deadly. Since there is no cure, treatment is directed toward managing the amount of sugar in the blood using insulin as well modifications to diet and lifestyle. 

Fundamental research will be needed to learn if some of the proteins are involved in the biological mechanisms that are integral to the onset of diabetes, he adds. If they are directly impacting the onset of autoimmunity, they could become therapeutic targets as well as harbingers of the disease.  

Analyzing Pathways 

In the initial discovery phase of the new study, the team studied 2,252 blood samples from 184 children to pin down 376 proteins that were altered in patients who later developed islet autoimmunity or type 1 diabetes. They then performed a more in-depth validation study looking at 6,426 blood samples from 990 children. 

Some of the identified protein markers are key to important processes in the body (e.g., antigen presentation, blood clotting, inflammatory signaling, and metabolism), as well as known to be active in model studies of pancreatic cells and tissues under stress, per the team’s previous work conducted through the Human Islet Research Network (HIRN). The TEDDY and HIRN projects, both funded by the National Institute of Diabetes and Digestive and Kidney Diseases, offered the unique opportunity to compare data from clinical samples and from fundamental studies geared at identifying potential cellular mechanisms of stress, says Metz. 

As reported in a study that published several years ago in Cell Metabolism (DOI: 10.1016/j.cmet.2019.12.005), Metz and his colleagues conducted a proteomic analysis of human islets, isolated from 10 non-diabetic cadaveric donors, with and without treatment with cytokines. Since cytokines can elicit a stress response in tissues, it is a model system for the processes that might be occurring in the organs of individuals experiencing autoimmunity, he explains.  

“Although there wasn’t necessarily an exact one-to-one matchup of the proteins [between the TEDDY and HIRN study cohorts], a pathway analysis... [revealed] an overlap in the biological processes occurring in the islets that are exposed to cytokines,” says Metz. TEDDY itself was a massive clinical study of individuals genetically predisposed to developing type 1 diabetes, and thus a treasure trove of information about potential environmental triggers of the autoimmune response—including the roles of the microbiome and viral exposures as well as proteins.  

Identification of the panel of proteins predictive of the onset of islet autoimmunity up to six months in advance is one of the most exciting aspects of the latest study, he says. Since machine learning analysis wasn’t done any earlier than a half year prior to seroconversion, that predictive window could potentially be widened to find a mechanism of interest that an intervention could be designed around earlier in an individual’s lifetime.   

It is also worthy of note that the targeted proteomic assays have already been established, meaning any laboratory with similar mass spectrometry equipment could implement them, Metz says. If the panel is ultimately reduced to a handful of proteins, they could potentially be the focus of immunoassay development. 

In addition to the many changes in proteins in the human complement system seen in the latest published study, the research team recently conducted a literature review of other similar studies looking at the use of proteomics in the study of diabetes, he adds. This has further confirmed that the complement system is highly correlated to the onset of type 1 diabetes, making it worthy of further investigation to look for a possible causative role in the initiation of autoimmunity. A manuscript describing the results of the literature review is currently under peer review.