📘 Genetic inheritance and type 1 diabetes risk

HLA (Human Leukocyte Antigen) genes, located on chromosome 6, encode essential components for your immune system's ability to distinguish between self and non-self cells. In type 1 diabetes, certain HLA variants cause the immune system to mistakenly recognise the pancreatic beta cells as an infection, triggering an autoimmune attack that progressively destroys them [1].

The HLA region contributes approximately half of the total genetic risk for type 1 diabetes. The most predisposing subtypes are HLA DR3-DQ2 and DR4-DQ8. If you carry both subtypes simultaneously (DR3/DR4 heterozygote), your risk is the highest of all HLA combinations (over 15X). Approximately 90–95% of people with type 1 diabetes carry HLA-DR3 and/or HLA-DR4, compared with about 40% of the general population [1, 2].

Yes. Not all HLA variants increase risk. If you carry a protective HLA allele, your probability of developing type 1 diabetes is significantly lower, even if other risk factors are present. This means that the HLA system functions both as an accelerator and as a brake on autoimmune risk [1].

The most important protective alleles include DRB1*1501 and DQB1*0602 (DQ6). The DQB1*0602 variant is found in approximately 20% of the general population, but in less than 1% of type 1 diabetes patients. Even if you also carry a high-risk variant (on the other chromosome 6), the presence of DQB1*0602 cancels its effect and continues to significantly reduce your risk of developing type 1 diabetes [3].

If you have type 1 diabetes, your brother's or sister's lifetime risk of developing the disease is approximately 6–10%, compared with a lifetime risk in the general population of approximately 0.4%. This risk varies depending on the degree of HLA identity between you. If your sibling is HLA-identical to you, the risk rises to approximately 15–25%. If you share only one of the two HLA loci, the risk is about 5–7%, and if you share no HLA variants, the risk drops to approximately 1% or even less [4].

Concordance in dizygotic twins (each with their own placenta) for type 1 diabetes is similar to that of non-twin siblings. These figures underscore the importance of monitoring first-degree relatives of affected individuals through autoantibody screening programmes [4].

The risk for a child depends on which parent is affected and the specific HLA genotype. If the father has type 1 diabetes, the child's risk is approximately 6–9%. If the mother has type 1 diabetes, the risk is lower, approximately 2–4%. The reason for this difference is not fully understood, but it may involve maternal immune tolerance mechanisms during pregnancy [5].

If both parents have type 1 diabetes, the child's risk rises to approximately 25%, approaching the concordance rate observed in monozygotic twins (who share the same placenta). These transmission patterns confirm that type 1 diabetes is not inherited through a simple (Mendelian) pattern, but follows a complex (polygenic) model in which many genetic and environmental factors determine whether the disease ultimately manifests [5, 6].

Approximately 90% of people who develop type 1 diabetes have no known relative with the same disease. This occurs for several important reasons. Firstly, type 1 diabetes is a disease that requires the alignment of multiple risk genes in the same individual (polygenic), and each of these genes can be silently carried by family members who never develop the disease but pass them on [4].

Secondly, risk genes are common in the general population (approximately 40% carry HLA-DR3 or DR4), but only a very small fraction of carriers progress to autoimmunity and clinical disease. Thirdly, environmental triggers (viral infections, dietary factors, microbiome changes) are essential for initiating the autoimmune process in a genetically susceptible individual. Most carriers of risk genes are never exposed to the right combination of triggers during the critical window of susceptibility [7].

The genetic risk score for type 1 diabetes (T1D-GRS) is a numerical value that combines the effects of multiple genetic variants (HLA and non-HLA) into a single number. This number reflects your overall genetic susceptibility to developing type 1 diabetes. There are over 80 genetic loci associated with type 1 diabetes risk. Approximately half of the genetic risk is linked to the HLA region, while the remainder lies elsewhere (non-HLA) [8].

The T1D-GRS applied in population screening could identify over two-thirds of individuals who will develop type 1 diabetes. This approach would allow targeted autoantibody testing in the subset of individuals with the highest genetic risk, improving cost-effectiveness. The T1D-GRS could also be used for diabetes type classification and potentially in prognostic assessment [8].

Yes. Neonatal genetic screening for type 1 diabetes risk is already used in several research and population-level programmes worldwide. The approach uses a genetic risk score based on HLA and non-HLA variants to identify the subset of newborns with the highest genetic risk. These infants are then monitored through periodic autoantibody testing [9].

Such screening programmes are available in Europe (Fr1da in Germany, GPPAD – gppad.org), Australia (type1screen.org), and the United States (TrialNet – trialnet.org, ASK Health – askhealth.org, CASCADE Kids – cascadekids.org) [10].

While the HLA region contributes approximately 50% of the genetic risk, the remainder comes from multiple non-HLA loci spread throughout the entire genome, each with a small individual effect. Over 80 non-HLA loci have been confirmed through genome-wide association studies (GWAS). Variants in the insulin gene (chromosome 11) affect insulin expression in the thymus during immune system training regarding tolerance to insulin. This is the second most important genetic locus, after HLA [11].

Other important non-HLA loci for genetic risk include PTPN22 (linked to anti-tyrosine phosphatase autoantibodies), CTLA4 (directs cytotoxic T lymphocytes), IL2RA (a receptor essential for regulatory T cell function), and IFIH1 (a viral sensor with impaired function). Most non-HLA risk variants are located in regions that influence the expression of their respective genes rather than the structure of the resulting proteins [7].

If your identical (monozygotic) twin has type 1 diabetes, your lifetime risk of developing the disease is approximately 30–70%. This variability depends on the duration of follow-up and the interaction with environmental factors [12].

The fact that concordance is not 100% means that genetics alone cannot cause type 1 diabetes. If the disease were purely genetic, identical twins would always both be affected. The remaining discordance is attributed to environmental factors and epigenetic modifications [12].

Ethnicity significantly influences the risk of type 1 diabetes, primarily through differences in the frequency of high-risk and protective HLA variants across populations. Type 1 diabetes occurs most frequently in populations of Northern European descent and least frequently in East Asian and sub-Saharan African populations [13].

These differences correspond to the prevalence of high-risk HLA variants, which are more common in European populations. However, type 1 diabetes occurs in all ethnic groups worldwide, and its incidence is rising globally, including in regions with historically low incidence, such as parts of Asia and Africa [13].

If you have a relative with type 1 diabetes, genetic testing using the type 1 diabetes genetic risk score can complement, not replace, autoantibody testing. This will additionally give you a better appreciation of your risk level and identify the need for closer ongoing monitoring. If you have a first-degree relative with type 1 diabetes, programmes such as TrialNet (trialnet.org) and GPPAD (gppad.org) offer free autoantibody screening and follow-up, with the possibility of enrolment in prevention trials [9].

Autoantibody-based screening for presymptomatic type 1 diabetes should be undertaken by individuals either with a family history of type 1 diabetes or with increased genetic risk. The first step for relatives is therefore autoantibody testing, not genetic testing [14].

Type 1 diabetes shares genetic risk overlap with other autoimmune diseases. Up to one-third of people with type 1 diabetes already have or will develop at least one additional autoimmune condition during their lifetime. The most common autoimmune conditions associated with type 1 diabetes are Hashimoto's thyroiditis, Graves' disease, coeliac disease, Addison's disease, vitiligo, autoimmune hepatitis, myasthenia gravis, and pernicious anaemia [15].

This overlap occurs because HLA-DR3 and HLA-DR4 predispose to multiple autoimmune conditions. Approximately 90–95% of people with type 1 diabetes and nearly 99% of those with coeliac disease carry HLA-DQ2 and/or HLA-DQ8. The non-HLA gene PTPN22 is a risk factor for type 1 diabetes, rheumatoid arthritis, lupus, and autoimmune thyroid disease. Variants of CTLA4 increase the risk of both type 1 diabetes and Graves' disease and autoimmune thyroiditis. Monogenic polyglandular autoimmune syndromes (mutations in the AIRE gene, IPEX syndrome) can also include type 1 diabetes as one of multiple autoimmune manifestations [16].

Yes, most commonly people only carry the risk genes without developing type 1 diabetes. Approximately 30–40% of the general population carry at least one high-risk HLA variant (DR3 or DR4), but only approximately 0.4% of people develop type 1 diabetes during their lifetime. Even the highest-risk HLA combination (DR3/DR4 heterozygote) confers an absolute lifetime risk of approximately 5–7% in the absence of a positive family history [4].

This demonstrates that genetic susceptibility is necessary but not sufficient. Type 1 diabetes requires a combination of genetic predisposition, environmental triggers, and autoimmune activation (demonstrated by the appearance of islet autoantibodies), which act in a specific sequence [14].

Type 1 diabetes develops through a gene–environment interaction in which genetic susceptibility creates the foundation, while environmental factors act as triggers that initiate and/or accelerate the autoimmune destruction of pancreatic beta cells [17].

The main environmental factors potentially involved are viral infections, which can trigger autoimmunity through molecular mimicry or through direct damage to beta cells in genetically susceptible individuals. The non-HLA gene IFIH1 (which encodes a sensor for viral presence) provides a direct genetic link between viral exposure and autoimmune activation. Dietary factors have been investigated, with variable results [17].