New Genetic Risk Score May Help Detect Type 1 Diabetes

Miriam E. Tucker

January 23, 2019

A newly improved genetic risk score could become a cost-effective method for screening newborns for type 1 diabetes risk or identifying new-onset type 1 diabetes among adults, research suggests.

The findings, from a study by Seth A. Sharp, of the Institute of Biomedical and Clinical Science, University of Exeter Medical School, United Kingdom, and colleagues, were published online in Diabetes Care.

Currently, autoantibody testing of infants to screen for the risk of developing type 1 diabetes later in life is reserved for clinical trials and is not deemed to be cost-effective for use in the general population. In adults with new-onset diabetes for whom the type of diabetes isn't clear from clinical characteristics, there are tests that can be performed — including c-peptide and autoantibody screening — but sometimes neither is sufficient for definitive classification, the authors say.

In contrast, the newly devised score was highly discriminative for type 1 diabetes, particularly when onset is early. It significantly improves upon a previous type 1 diabetes genetic score developed by the Exeter group in its ability to distinguish individuals with type 1 diabetes from those with type 2 diabetes and from control persons.

The new score captures about 80% of the genetic risk for type 1 diabetes.

"Genetics accounts for about half of the disease risk, so this is really only 40% of the overall etiology, but it's enough to be quite predictive of type 1 diabetes," coinvestigator William A. Hagopian, MD, of Pacific Northwest Diabetes Research Institute, Seattle, Washington, told Medscape Medical News.

Hagopian anticipates that the test will become incorporated into routine clinical practice soon.

"I don't think it's ready for prime time yet, but it could be relatively soon, in a year or two. This will become part of the clinician's toolbox, and I believe there will be less misdiagnosis, so fewer type 2 [diabetes patients] will be put on insulin or type 1 patients put on pills."

"Highly Discriminative for Type 1 Diabetes"

Most of the risk for type 1 diabetes is explained by variation at a few very strongly associated human leukocyte antigen (HLA) loci.

The study group analyzed variants associated with type 1 diabetes in the HLA region and across the genome in 6481 case patients and 9247 control patients from the Type 1 Diabetes Genetics Consortium in an effort to more completely incorporate HLA alleles, their interactions, and recently discovered non-HLA loci into an improved type 1 diabetes genetic risk screening tool, dubbed type 1 diabetes genetic risk score 2 (T1DGRS2). The tool was then validated in the UK Biobank population.

"We captured nearly everything that was known from the genome-wide association studies and the background HLA literature," Hagopian explained.

The team then conducted simulations to see how their improved score compared to current genetic methods of diagnosis and screening.

It was highly discriminative for all type 1 diabetes cases (area under the curve [AUC], 0.92; P < .0001 vs older scores) and even more so for type 1 diabetes cases in the first few years of life (AUC, 0.96).

In simulated newborn screening, the score was nearly twice as predictive as HLA genotyping alone, and it was approximately 50% better than current genetic scores in predicting type 1 diabetes in the general population.

The researchers conclude that the T1DGRS2 has significant potential in clinical and research settings.

Score Is "Independent [of] and Complementary to" Other Tests

The information provided by the score is "independent [of] and complementary" to autoantibody testing, and therefore the two tests can potentially be combined in clinical settings," said Hagopian, a practicing endocrinologist at the University of Washington, Seattle.

Hagopian is also a researcher who in the past was instrumental in developing the type 1 diabetes–associated glutamic acid decarboxylase antibody test.

For newborns, a potentially cost-effective approach would be to use this new genetic risk score for population-based screening, and then only follow those found to be at high risk with periodic autoantibody testing, he suggested.

Among adults with new-onset type 1 diabetes — in whom c-peptide function may still be present, whereas autoantibodies may not be — "the genetic risk score is very predictive when added to antibodies, age at diagnosis, and body-mass index," he noted.

The team is currently working on cost-effectiveness analyses of the screening approaches in both newborns and adults, Hagopian said, adding, "We really think this will help in clinical practice. We're moving quickly."

Hagopian and Sharpe have disclosed no relevant financial relationships. Coauthor Richard A. Oram, PhD, holds a UK Medical Research Council Institutional Confidence in Concept grant to develop a 10-SNP biochip T1D genetic test in collaboration with Randox.

Diabetes Care. Published online January 17, 2019. Full text

For more diabetes and endocrinology news, follow us on Twitter and Facebook.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.
Post as: