Routine autoantibody measurement advised for suspected adult type 1 diabetes
medwireNews: A majority of adults with clinically diagnosed type 1 diabetes but no islet autoantibodies have a low genetic risk for the condition, suggesting misdiagnosis, report researchers.
“This would advocate for routine measurement of islet autoantibodies in all adult cases of clinically suspected type 1 diabetes,” presenter Nick Thomas (University of Exeter, UK) told delegates at the virtual 2021 Diabetes UK Professional Conference.
Thomas and team applied their previously published type 1 diabetes genetic risk score (T1DGRS), based on 30 risk variants, to 1866 people, aged between 5 and 75 years, who had clinically diagnosed type 1 diabetes.
The presenter noted that the T1DGRS cannot diagnose type 1 diabetes at an individual level, but can give an idea of the overall distribution of genetic risk in a cohort for which the diabetes subtype is unclear.
The team found that children younger than 18 years of age with a clinical diagnosis of type 1 diabetes had a similar distribution of genetic risk regardless of whether or not they had islet autoantibodies, indicating that the vast majority did indeed have type 1 diabetes, despite some not having the expected autoantibodies.
Just 7% of the autoantibody-negative group was predicted to have other forms of diabetes, and “reassuringly” this small number of incorrect diagnoses “was entirely explained by cases of monogenic diabetes,” said Thomas.
Young adults aged from 18 to 30 years with autoantibodies had a T1DGRS distribution closely matching that of the children, but those without autoantibodies had a broader distribution of genetic risk, with a lower median score, suggesting a mixture of type 1 diabetes and other forms.
In fact, the genetic risk distribution indicated that only 55% of these people with clinically diagnosed type 1 diabetes actually had the condition, said Thomas.
And for those older than 30 years, the T1DGRS again supported a type 1 diabetes diagnosis in people with autoantibodies, but the risk distribution in those without autoantibodies was very similar to that in people with type 2 diabetes, indicating a majority of incorrect diagnoses, with only 23% of people being correctly diagnosed with type 1 diabetes.
The team found just five cases of monogenic diabetes in the young adults who were autoantibody negative, and two among the older adults, “suggesting that the most likely diagnosis in these genetically discordant cases is probably type 2 diabetes,” said Thomas.
After removing cases with confirmed monogenic or probable type 2 diabetes, there was no longer any age-related differences in T1DGRS distribution, with 92% of people with genetically defined type 1 diabetes having at least one islet autoantibody.
Thomas noted, however, that age did influence the specific autoantibodies likely to be present, with glutamic acid decarboxylase antibodies becoming increasingly more likely with older age, being present in 95% of autoantibody-positive people older than 30 years.
Therefore, the presence of autoantibodies confirms a clinical type 1 diabetes diagnosis in adults, concluded Thomas.
By contrast, the absence of autoantibodies “would suggest an alternative, non-type 1 diabetes diagnosis in the majority,” he said, recommending close monitoring of C-peptide progression in those without autoantibodies.
medwireNews is an independent medical news service provided by Springer Healthcare Ltd. © 2021 Springer Healthcare Ltd, part of the Springer Nature Group