New genetic discoveries may hold clues to predict, prevent type 1 diabetes
The etiology of type 1 diabetes continues to be debated, although experts agree it is likely a mix of genetic and environmental causes. Today, more than 50 regions of the human genome are implicated in type 1 diabetes.
Within each region, researchers are identifying new genes, biological pathways and potential therapeutic targets for intervention, but a cure — or a way to prevent progression to type 1 diabetes — remains elusive.
At the same time, cases of type 1 diabetes are surging in the U.S. According to the CDC’s National Diabetes Statistics Report released in February, the number of people diagnosed with type 1 diabetes rose by nearly 30% since 2017, with the greatest increases observed among minority children.
“The big mystery in type 1 diabetes is, why does it happen?” Louis H. Philipson, MD, PhD, FACP, professor of medicine, director of the Kovler Diabetes Center at the University of Chicago School of Medicine, and former president of medicine and science for the American Diabetes Association, told Endocrine Today. “Is there an environmental cause or a genetic cause? Autoimmune diseases, as a group, tend to be not all that inheritable. That is sometimes a surprise to people.”
Several genetic polymorphisms — normal variations in DNA — are linked to increased risk for type 1 diabetes. However, these genetic variations do not cause the disease, meaning that not everyone with a diabetes-associated genetic variation will go on to develop the disease. The current view is that environmental risk factors are also needed to trigger disease development. During the past decade, research following birth cohorts, along with detailed genetic studies and data about environmental factors, has led to a change in the basic theory of the causes and pathogenesis of type 1 diabetes.
“When genetic polymorphisms associated with the disease were discovered and confirmed, some people said, this is it, the genes for type 1 diabetes have been discovered and the story ends here,” Riitta Veijola, MD, PhD, professor of pediatrics at University of Oulu, Finland, told Endocrine Today. “But we still don’t understand how the effects of those genetic variations translate to someone being more prone to develop type 1 diabetes. That is the work that has been ongoing during the last decade. We need to understand how these susceptibility genes work. Are they active or not active? Are there environmental factors that modify the activity of these genes? Which developmental phases are they involved in? We don’t know.”
Family diabetes history
The overall risk for type 1 diabetes in the general population is 0.4%, but it is higher among relatives of individuals with the disease, Maria Jose Redondo, MD, PhD, MPH, associate professor of pediatrics, diabetes and endocrinology at Baylor College of Medicine, and colleagues wrote in a review published in the September 2017 issue of Pediatric Diabetes. The human leukocyte antigen (HLA) complex on chromosome 6p21, a group of related proteins encoded by the major histocompatibility complex region, contributes about 50% of that genetic risk; the strongest association is with the genes HLA-DRB1 and HLA-DQB1, in which three amino acid alterations may account for up to 90% of the HLA risk. Other loci identified through genome-wide association studies have a slight individual effect on total genetic risk for progression to type 1 diabetes.
“The highest genetic risk is for people who have both [genes] — about 40% of people with type 1 diabetes have both haplotypes, about 1% of the general population,” Redondo told Endocrine Today. “That said, not everyone who has these haplotypes goes on to develop type 1 diabetes, and the other way around. Some have none of these high-risk haplotypes and still develop diabetes. It is interesting. We see how the incidence of type 1 diabetes is increasing. It is increasing more at the expense of those who do not have any of the high-risk genotypes.”
Twin studies and epidemiological data provide important information on the familial risk for type 1 diabetes, according to Stephen S. Rich, PhD, FAHA, director of the Center for Public Health Genomics at the University of Virginia. Early twin studies suggested that type 1 diabetes was highly heritable. Among identical twins where one twin has type 1 diabetes, up to 50% of co-twins also had type 1 diabetes, Rich said.
“This information suggested that type 1 diabetes was not completely genetic, otherwise 100% of co-twins would also have type 1 diabetes, and that type 1 diabetes was not completely due to environmental factors, otherwise 0% of the co-twins would have type 1 diabetes,” Rich told Endocrine Today. “In nonidentical twins, the risk for type 1 diabetes falls to 10%. In type 2 diabetes, the identical twin risk is higher, but that may be due to genetic factors related to obesity, lifestyle and other factors.”
Epidemiological studies provide more information regarding risk for offspring of a parent with type 1 diabetes, with studies showing risk is 10% higher for the child if the father vs. the mother has type 1 diabetes, Rich said, whereas risk for a sibling of someone with type 1 diabetes is about 10%.
“These data suggest that the genetic risk ratio for type 1 diabetes in siblings is 16-fold higher than someone in the general population; in contrast, the genetic risk ratio for type 2 diabetes in siblings is two- to fourfold higher. So, although type 1 diabetes is not common and doesn’t seem to run in families — only 10% of those who develop type 1 diabetes have a first-degree relative with type 1 diabetes — it is heritable.”
Type 1 diabetes in twins can also develop at different times, Philipson said.
“My own conception of heritability has changed over the last few decades,” Philipson said. “We thought that if one twin had it, there wasn’t more than a 50/50 chance the other twin would develop type 1. People like my friend Kevan C. Herold, MD, convinced me that it is more heritable than that; you just have to wait longer. One twin might be aged 3 or 8 years when they are diagnosed, and the second twin might be aged 60 years and get it then. The predilection to autoimmune diseases, especially diabetes, is higher than we thought in identical twins, but among first-degree relatives, it certainly is much lower than what we would say is a dominantly inherited disease.”
A family history of type 1 diabetes was associated with the development of islet autoimmunity among children participating in the ongoing The Environmental Determinants of Diabetes in the Young (TEDDY) study, whereas a family history of type 2 diabetes is associated with progression from islet autoimmunity to clinical diagnosis in the cohort, Veijola said.
In an analysis of 7,479 TEDDY children as of Jan. 31, 2016, Veijola and colleagues observed that children with a father or sibling with type 1 diabetes were more likely to develop islet autoimmunity vs. those without a father or sibling with diabetes. However, having a mother with type 1 diabetes was not a significant risk factor for islet autoimmunity.
Additionally, TEDDY children with a second-degree relative with type 2 diabetes showed significantly delayed progression from islet autoimmunity to clinical type 1 diabetes vs. children without such relatives — a finding that was consistently observed in TEDDY children across Finland, Germany, Sweden and the United States, Veijola said.
“We believe this is a true phenomenon that needs to be studied further,” Veijola said.
The finding, Veijola said, creates two new hypotheses: that type 2 diabetes-susceptibility genes may affect the progression phase of type 1 diabetes, and that families with second-degree relatives with type 2 diabetes may modify their lifestyle to prevent type 1 diabetes.
A critical application of genetics is to improve prediction so that strategies can be designed and implemented to prevent type 1 diabetes among individuals at risk, Redondo wrote in a review article published in Diabetes Care.
Redondo and colleagues have tested a previously developed and validated type 1 diabetes genetic risk score assessing 30 type 1 diabetes risk loci (HLA and non-HLA) that predicts disease risk in adults. Preliminary data from a subset of TrialNet participants suggest that the genetic risk score improves on the current predictive model of islet autoantibodies, age and metabolic factors for determining progression along the preclinical stages of type 1 diabetes. However, Redondo said such results must be validated and optimized before a type 1 diabetes genetic risk score could be used in research practice.
“We have all of these huge wells of knowledge on the genetics of type 1 diabetes, but how are you going to use it?” Redondo said. “One is for predicting diabetes, but the issue is, how do you do it? The genetic risk score allows us to put a lot more information into a single number. It is a huge contribution that lets us integrate the information and actually use it.”
Through an NIH-funded grant, Redondo and colleagues are now testing an improved version of the type 1 diabetes genetic risk score that incorporates islet autoantibody data and clinical and metabolic parameters on the entire TrialNet observational cohort — known as Pathway to Prevention — to identify the best models to predict progression overall and at each of the preclinical stages of type 1 diabetes.
“The genetic risk score improves the ability to predict even above the model we had before, even if we add HLA,” Redondo said. “Now we are embarking on the second phase where we are genotyping the whole TrialNet cohort and looking at validating findings and creating predictive models for each of those diabetes stages,” Redondo said.
Redondo said the project is expected to improve the outcomes of trials to prevent type 1 diabetes.
“This project is innovative because it seeks to shift the current practice by proposing to utilize genetics as a novel, affordable, time-independent strategy to identify individuals at risk for type 1 diabetes and select candidates for intervention trials,” Redondo said.
Although the type 1 diabetes-associated single nucleotide polymorphisms in white populations account for nearly 90% of the genetic risk, with high sensitivity and specificity, the low prevalence of type 1 diabetes makes the type 1 diabetes genetic risk score of limited utility, Rich said. However, identifying those with highest genetic risk may permit early and targeted immune monitoring to diagnose type 1 diabetes months before clinical onset of symptoms, he said.
“There are other uses of the type 1 diabetes genetic risk score being developed, including improvement in the diagnosis of type 1 diabetes compared with some monogenic forms of diabetes, such as [maturity-onset diabetes of the young], and improving when insulin could be used among individuals with type 2 diabetes,” Rich said.
Clues from monogenic diabetes
Monogenic diabetes may be “the best example today” of what precision medicine can offer and could potentially help inform precision medicine for all forms of diabetes, Philipson said.
Monogenic forms of diabetes — syndromic or non-syndromic, neonatal permanent or maturity-onset diabetes of the young (MODY) — offer the clearest opportunity in precision medicine, in that a single nucleotide change or genetic insertion or deletion can lead to diabetes, Philipson said. Genetic testing, he said, must become a standard of care in diabetes, and not just something described in the ADA Standards of Care.
“We can do much better when knowing, first, the many causes of these different types of diabetes and, second, predicting certain sequelae or other effects of those gene mutations, because they are not only in the insulin-secreting cells,” Philipson said. “It’s a big payoff, and yet we’re not at the place where insurance companies are routinely covering genetic testing. That to me is immoral. We need to move to where a genetic test is no different than getting a glucose test, sodium test or an HbA1c.”
The discoveries from research into monogenic diabetes have helped to illuminate key pathways, receptors and transcription factors that involve insulin, its synthesis and release, according to Philipson, allowing researchers to better understand metabolic control systems in humans. Further, many genes show overlap in their contributions to type 1 and type 2 diabetes, as well as monogenic diabetes, he said.
“Are genetic tests the key to precision medicine? No,” Philipson said. “They are just one facet of an individualized approach to diabetes and prediabetes. For example, in type 1 diabetes, we now have the type 1 diabetes risk score that can improve diagnosis.”
Genetics alone “does not and cannot” explain the staggering increase of diabetes around the world, Philipson said. However, a deeper knowledge of the variations in human genetics can illuminate risk factors for the disease and lead to a better understanding of the interplay between the environment and genetics that is likely behind the growing epidemic.
The gene ‘next door’
Relatively few of the regions in the genome associated with type 1 diabetes cause a change in a protein, which could then theoretically be targeted with a drug, Rich said. Researchers are instead focusing on the functional impact of genetic variants associated with type 1 diabetes and identifying the genes that they regulate.
“Most of the genetic effects appear in gene regulation — more like how much gas you give the car to go faster or slower and not like turning the ignition key on or off,” Rich said. “It is also the case that a genetic variant can ‘regulate’ more than one gene. A variant that is most associated with type 1 diabetes risk could regulate the gene ‘next door,’ but also one ‘down the block,’ ‘in another city’ or ‘in another state.’ This means that we could be misled by assuming the genetic effect in each of the many places in the genome is the closest.”
With funding from the Helmsley Charitable Trust, Philipson and colleagues are collaborating to study families in which multiple family members have type 1 diabetes or have multiple types of autoimmunity.
“Several projects are in advanced stages, and it will be illuminating to understand how those genes — which so clearly play a role in these families — play a role in whatever garden variety type 1 diabetes there is,” Philipson said. “And maybe there is no garden variety. It may be that is why genetics has not worked that well, because there are so many different causes of type 1 diabetes that to find a common pathway, you must look harder at these genes that may not show up in a scan.”
- CDC. National Diabetes Statistics Report, 2020. Available at: www.cdc.gov/diabetes/pdfs/data/statistics/national-diabetes-statistics-report.pdf. Accessed: May 29, 2020.
- Krischer J, et al. Lessons learned from The Environmental Determinants of Diabetes in the Young (TEDDY) study — Insights into early autoimmune type 1 diabetes. Presented at: American Diabetes Association Scientific Sessions; June 22-26, 2018; Orlando.
- Pociot F, et al. Lancet. 2016;doi.10.1016/S0140-6736(16)30582-7.
- Redondo MJ, et al. Diabetes Care. 2018;doi:10.2337/dc18-0087.
- Redondo MJ, et al. Pediatr Diabetes. 2017;doi:10.1111/pedi.12597.
- Rich SS. Curr Opin Endocrinol Diabetes Obes. 2017;doi:10.1097/MED.0000000000000347.
- For more information:
- Louis H. Philipson, MD, PhD, FACP, can be reached at the Kovler Diabetes Center, 900 E. 57th St., Chicago, KCBD, Room #8140, The University of Chicago, Chicago, IL 60637; email: firstname.lastname@example.org; Twitter: @lphilipson.
- Maria Jose Redondo, MD, PhD, MPH, can be reached at Texas Children’s Hospital, Department of Diabetes and Endocrinology, 6701 Fannin St., MWT 10th Floor, Houston, TX 77030; email: email@example.com; Twitter: @bcmhouston.
- Stephen S. Rich, PhD, FAHA, can be reached at the Center for Public Health Genomics, 3232 West Complex, P.O. Box 800717, Charlottesville, VA 22908; email: firstname.lastname@example.org.
- Riitta Veijola, MD, PhD, can be reached at the Department of Pediatrics, Faculty of Medicine, Aapistie 5, 90014 University of Oulu, Finland; email: email@example.com; Twitter: @RVeijola.