MUNICH — Onset of type 1 diabetes is just as likely to occur in people older than 30 years of age as in those younger, new research shows.
The data were presented September 16, 2016, here at the European Association for the Study of Diabetes (EASD) 2016 Annual Meeting by Dr Nicholas JM Thomas, of the Institute of Biomedical and Clinical Science, University of Exeter Medical School, United Kingdom.
Obtained using genetic data from the UK Biobank, the startling results refute the long-held belief that type 1 diabetes is primarily a "juvenile" condition.
Clinically, the findings are particularly relevant for primary care, where people who develop autoimmune-mediated diabetes in adulthood are often misdiagnosed as having type 2 and prescribed metformin instead of insulin.
"I think it's an eye-opener and obviously has implications for how we diagnose and manage people and also the education people receive. We very much focus on childhood and adolescence and perhaps people diagnosed later don't get the same education," Dr Thomas told Medscape Medical News in an interview.
Still, identifying these individuals in primary care is challenging for a number of reasons: The vast majority of older adults with new-onset diabetes have type 2 diabetes, and antibody tests to identify autoimmune-mediated diabetes are too expensive for routine use. Moreover, overweight/obesity is nearly universal in type 2 but also common in type 1 diabetes.
"It really is spotting a needle in a haystack," Dr Thomas remarked.
He advised that clinicians should at least be aware that adults can develop autoimmune diabetes, as either classic type 1 or the slower-onset phenomenon known as "latent autoimmune diabetes of adulthood (LADA)."
"It's knowing this does happen, and therefore just keeping an open mind when you spot someone who's not behaving like type 1 or not responding as you would anticipate when you go through the usual treatment guidelines for type 2," he said, citing the example of British Prime Minister Theresa May, who was diagnosed with type 1 diabetes at age 56 and who "progressed very rapidly."
Asked to comment, session comoderator Catharine Owen, MD, associate professor of diabetes at the Oxford Center for Diabetes, Endocrinology, and Metabolism, United Kingdom, agreed.
"I think it's absolutely crucial for people to be aware that type 1 diabetes can present at any age. Physicians shouldn't be complacent when people aren't responding to oral agents, or they're not bringing A1c down to target when they should."
Why the "Juvenile" Designation?
It's been nearly 20 years since the terms "juvenile" and "adult-onset" diabetes were officially changed to "type 1" and "type 2," yet the perception persists among the nonspecialist medical community and the lay public that adults who develop autoimmune-mediated diabetes are rare or unusual.
Indeed, Dr Thomas explained in his presentation, most studies on type 1 diabetes are conducted in children and adolescents, and very few epidemiologic data are available about type 1 diabetes later in life.
The problem, he said, is that "in childhood, almost all the cases of diabetes are type 1 diabetes, so they're easier to spot because they stand out, whereas later in life we see a dramatic increase in the prevalence of type 2 diabetes, so the cases of type 1 diabetes are drowned out and less easy" to identify.
Using medical records to study type 1 diabetes in adulthood is error-prone because clinical classification and insulin prescribing aren't systematic since there are no clinical diagnostic criteria by diabetes type.
The presence of specific autoantibodies can give a clue, but they're not measured routinely in adults and, even when they are, they have less discriminatory power than in children, resulting in a higher false-positive rate.
Novel Genetic Approach
To get around those problems, Dr Thomas and colleagues used a "robust, novel, genetic approach" using a risk score comprising 30 single nucleotide polymorphisms associated with type 1 diabetes (T1D-GRS).
They have shown type 1 diabetes is primarily restricted to patients with a high T1D-GRS (the top 50%).
They applied the type 1 genetic risk score to a cohort of 120,000 British white adults aged 40 to 70 from the UK Biobank, a network designed to study many diseases that includes people with and without diabetes and records information about age at diagnosis, whether insulin was used within a year of diagnosis, current medications, and body mass index (BMI).
As expected, when they factored in age they found that nearly all the diabetes below age 30 years was type 1 (autoimmune), and with increasing age the prevalence of type 2 diabetes rose dramatically.
However, to their surprise, they also found that the number of type 1 diabetes cases remained constant from age 30 to 60 years of age, so that roughly half of type 1 diabetes was occurring after age 30 but was "drowned out" by the vast number of type 2 cases, Dr Thomas explained.
Type 1 Diabetes After Age 30: Slightly Slimmer and Younger Than Type 2
In all, 53% (242/457) of type 1 cases are diagnosed 30 years and younger, accounting for 74% (242/326) of all diabetes cases diagnosed in this age range, whereas 47% (215/457) of cases are diagnosed from 31 to 60 years of age, accounting for just 5% (215/4335) of diabetes cases diagnosed in this age range.
In a comparison of the people genetically classified with type 2 and type 1 between 31 and 60 years of age, the type 1 group was significantly younger at diagnosis (44 vs 52 years of age, P < .0001), had lower BMI (26.2 vs 32.6 kg/m2, P < .0001), was far more likely to be on insulin within a year of diagnosis (79% vs 6%, P < .0001), and to be currently using insulin (100% vs 16%, P < .0001).
"So really, we were reassured we were capturing type 1 diabetes," Dr Thomas commented
T1D-GRS is a therefore a novel tool to investigate diabetes etiology in large cohorts without antibody measurement, he stressed.
He reiterated that type 1 diabetes is evenly distributed within the first 6 decades of life, but after age 30, the increase in type 2 diabetes makes the type 1 cases harder to recognize and treat correctly.
And Dr Owen noted that in the United Kingdom even people who develop type 1 diabetes in childhood or adolescence are often recoded in their medical record as having type 2 when they become adults.
"It's absolutely a basic misunderstanding. I think we have work to do in this area. [The new study] does highlight the issue in a nice way, although it doesn't help us identify these specific individuals."
Dr Thomas and Dr Owen have no relevant financial relationships.
European Association for the Study of Diabetes 2016 Annual Meeting; September 16, 2016; Munich, Germany. Abstract 264.
www red DiabetologNytt