rs7454108 — HLA-DQB1 DQ8 tag

HLA-DQ8: The Second Celiac Risk Gene and a Type 1 Diabetes Marker

rs7454108 is a tag SNP¹¹ tag SNP
A genetic marker in perfect linkage disequilibrium (r²=1.0) with another variant, allowing it to serve as a proxy that identifies the HLA-DQ8 haplotype with extraordinary precision. Located in the intergenic region²² intergenic region
DNA sequence between genes, often containing regulatory elements between HLA-DQB1 and HLA-DQA2 on chromosome 6, this SNP's C allele serves as a genetic "flag" for the presence of HLA-DQB1*03:02, the defining allele of the DQ8 haplotype. The HLA-DQ8 molecule is a heterodimer³³ heterodimer
A protein complex made of two different subunits encoded by DQA1*03:01 and DQB1*03:02, and its presence dramatically increases risk for two major autoimmune diseases: celiac disease and type 1 diabetes.

The Mechanism

The HLA (Human Leukocyte Antigen) system is the body's primary method for distinguishing self from non-self. HLA-DQ molecules sit on the surface of antigen-presenting cells⁴⁴ antigen-presenting cells
Specialized immune cells that display protein fragments to T cells, where they present protein fragments (peptides) to T cells for immune surveillance. HLA-DQ8 has a unique structural pocket that binds and presents gluten peptides from wheat, barley, and rye with high affinity, triggering an inappropriate immune response in susceptible individuals. Studies show⁵⁵ Studies show
Johnson et al. Relationship of HLA-DQ8 and severity of celiac disease. Clin Gastroenterol Hepatol, 2004 that HLA-DQ8 molecules are particularly efficient at presenting immunodominant gliadin peptides after they've been deamidated⁶⁶ deamidated
Modified by the enzyme tissue transglutaminase, increasing immunogenicity by tissue transglutaminase.

For type 1 diabetes, HLA-DQ8 presents pancreatic β-cell autoantigens⁷⁷ pancreatic β-cell autoantigens
Self-proteins from insulin-producing cells including insulin, GAD65, and ZnT8 to autoreactive T cells. The highest risk genotype is DR3/4-DQ8 heterozygosity⁸⁸ heterozygosity
Carrying one copy of DQ2.5 (from DR3 haplotype) and one copy of DQ8 (from DR4 haplotype), which allows formation of a unique trans-encoded DQ molecule⁹⁹ trans-encoded DQ molecule
A DQ heterodimer formed by pairing alpha and beta chains from different chromosomes that further amplifies risk.

The Evidence

Mansour et al. demonstrated¹⁰¹⁰ Mansour et al. demonstrated
Effective detection of human leukocyte antigen risk alleles in celiac disease using tag single nucleotide polymorphisms. PLoS One, 2008 that rs7454108 identifies HLA-DQ8 carriers with 99.1% sensitivity and 99.6% specificity in European populations. This tag SNP is so reliable that it forms the basis of 23andMe's FDA-cleared¹¹¹¹ 23andMe's FDA-cleared
The first direct-to-consumer genetic health risk report cleared by the FDA celiac disease genetic risk report, analyzing rs7454108 alongside rs2187668 (the DQ2.5 tag) to identify the ~95% of celiac patients carrying permissive HLA genotypes.

A 2023 meta-analysis¹²¹² A 2023 meta-analysis
Meta-analysis and systematic review of HLA DQ2/DQ8 in adults with celiac disease. Int J Mol Sci, 2023 found HLA-DQ2 and/or DQ8 in over 95% of celiac disease patients across diverse populations, with HLA-DQ8 alone accounting for 5-10% of cases. In European populations, approximately 21% carry at least one copy of DQ8, but only 1% develop celiac disease, illustrating that HLA-DQ8 is necessary but not sufficient. Studies in siblings¹³¹³ Studies in siblings
Frequency of celiac disease and distribution of HLA-DQ2/DQ8 haplotypes among siblings of children with celiac disease. World J Clin Pediatr, 2022 show 10.7% prevalence among siblings of celiac patients—22.7 times the general population rate—with 100% of affected siblings carrying DQ2 and/or DQ8.

For type 1 diabetes, the evidence is even more dramatic. Barker et al.'s validation study¹⁴¹⁴ Barker et al.'s validation study
Two single nucleotide polymorphisms identify the highest-risk diabetes HLA genotype. Diabetes, 2008 in over 5,000 subjects from the Type 1 Diabetes Genetics Consortium found rs7454108 C allele present in 98.9% of individuals carrying DQB1*0302. The landmark PNAS study¹⁵¹⁵ The landmark PNAS study
Extreme genetic risk for type 1A diabetes. PNAS, 2006 by Aly et al. revealed that DR3/4-DQ8 siblings sharing both HLA haplotypes identical by descent¹⁶¹⁶ identical by descent
Inherited from the same parental chromosomes as their diabetic sibling with their diabetic proband had an 85% risk of developing islet autoantibodies by age 15, compared to 20% in those not sharing both haplotypes. Subsequent research¹⁷¹⁷ Subsequent research
Definition of high-risk type 1 diabetes HLA-DR and HLA-DQ types using only three single nucleotide polymorphisms. Diabetes, 2013 confirmed HLA-DRB1*04:01-DQB1*03:02 (DR4-DQ8) carries an odds ratio of 6.18 to 8.39 for type 1 diabetes.

Practical Implications

This SNP's primary clinical utility is in ruling out disease rather than predicting it. A TT genotype (no DQ8 copies) combined with absence of DQ2.5 makes celiac disease highly unlikely—approximately 98-99% negative predictive value. This can spare individuals from unnecessary small bowel biopsies¹⁸¹⁸ small bowel biopsies
Gold standard diagnostic procedure requiring endoscopy and tissue sampling when celiac disease is being considered. However, carrying one or two C alleles does not mean you will develop these conditions—it simply means you're genetically eligible.

For celiac disease, environmental factors matter enormously: gluten exposure timing in infancy, gut microbiome composition, and viral infections¹⁹¹⁹ viral infections
Particularly enteroviruses, which may trigger loss of oral tolerance to gluten may all influence whether genetic risk translates to active disease. For type 1 diabetes, additional non-HLA genes (insulin gene VNTR, PTPN22, CTLA4) and environmental triggers work in concert with HLA risk.

If you carry HLA-DQ8 and have first-degree relatives with celiac disease or type 1 diabetes, or if you experience unexplained symptoms²⁰²⁰ symptoms
Chronic diarrhea, bloating, iron-deficiency anemia, fatigue, weight loss for celiac; excessive thirst, frequent urination, unexplained weight loss for type 1 diabetes, genetic testing combined with serological screening (tissue transglutaminase antibodies for celiac, islet autoantibodies for type 1 diabetes) is warranted.

Interactions

The most clinically significant interaction is between rs7454108 (HLA-DQ8 tag) and rs2187668 (HLA-DQ2.5 tag). Individuals who are compound heterozygous²¹²¹ compound heterozygous
Carrying one copy each of DQ2.5 and DQ8 face heightened risk for both celiac disease and type 1 diabetes compared to carrying either haplotype alone. For celiac, this DQ2.5/DQ8 combination is second only to DQ2.5 homozygosity in risk magnitude. For type 1 diabetes, the DR3/4-DQ8 genotype (tagged by rs2187668 heterozygous + rs7454108 heterozygous) represents the highest genetic risk, accounting for 30-50% of childhood-onset cases. The trans-encoded DQ molecule²²²² trans-encoded DQ molecule
DQA1*05:01 from DR3 paired with DQB1*03:02 from DR4 formed in DR3/4-DQ8 individuals may have unique peptide-binding properties that amplify autoimmune risk beyond the sum of individual haplotypes.

A compound implication should be created for individuals carrying both DQ2.5 (rs2187668 CT or TT) and DQ8 (rs7454108 CT or CC), as the combined genotype warrants earlier and more intensive screening for both celiac disease and type 1 diabetes, particularly in individuals with affected family members or suggestive symptoms. The recommendation would be periodic serological screening and heightened clinical vigilance, rather than the "unlikely to develop disease" reassurance appropriate for individuals lacking both risk haplotypes.