HLA-DQB1 Test: Type 1 Diabetes, Celiac Disease, and Results

An HLA-DQB1 test identifies inherited variants in a gene that helps the immune system present protein fragments to T cells. Certain DQB1 alleles contribute strongly to susceptibility or protection for type 1 diabetes and celiac disease, and one allele, HLA-DQB106:02, is closely associated with narcolepsy type 1. The test does not diagnose any of these conditions by itself. Its meaning depends on the HLA-DQA1 and HLA-DRB1 alleles inherited alongside DQB1, the person’s ancestry, family history, symptoms, and disease-specific laboratory findings. For celiac disease, DQB102 or DQB1*03:02 may form the DQ2 or DQ8 molecules that make disease possible. For type 1 diabetes, high-risk DR-DQ haplotypes can identify susceptibility, but pancreatic autoantibodies and glucose testing show whether autoimmune diabetes is developing. Testing requires blood, saliva, or a cheek swab, needs no fasting, and remains valid throughout life.

• HLA-DQB1 results show inherited immune susceptibility, not whether type 1 diabetes or celiac disease is currently active.
• DQB102 can contribute to HLA-DQ2, while DQB103:02 can contribute to HLA-DQ8; the DQA1 partner determines the complete molecule.
• The highest type 1 diabetes risk often involves DR3-DQ2 and DR4-DQ8 haplotypes, especially when inherited together.
• DQB1*06:02 is strongly associated with narcolepsy type 1, but many carriers never develop narcolepsy.
• No fasting, gluten exposure, medication pause, or special timing is needed for the genetic sample.

Table of Contents

• What HLA-DQB1 does
• Why HLA-DQB1 testing is ordered
• Testing methods and report formats
• HLA-DQB1 results and type 1 diabetes
• HLA-DQB1 results and celiac disease
• Narcolepsy and other disease associations
• Limits and common misinterpretations
• Follow-up after testing

What HLA-DQB1 does

HLA-DQB1 is one of the highly variable human leukocyte antigen class II genes on chromosome 6. It encodes the beta chain of an HLA-DQ protein. The beta chain pairs with an alpha chain, usually encoded by HLA-DQA1, to form a molecule on antigen-presenting cells. That molecule holds small protein fragments and displays them to CD4 T cells, helping the immune system decide whether to respond.

The pairing is important. A DQB1 allele alone does not fully define the HLA-DQ molecule. For example:

• DQA105 plus DQB102 can form DQ2.5, the major celiac-associated molecule.
• DQA103 plus DQB103:02 can form DQ8, associated with both celiac disease and type 1 diabetes.
• DQA101:02 plus DQB106:02 can form DQ6.2, strongly associated with narcolepsy type 1 and often protective against type 1 diabetes.

HLA genes are usually inherited in linked groups called haplotypes. A haplotype may include HLA-DRB1, DQA1, and DQB1 alleles that tend to travel together from one parent. Clinicians therefore often speak of DR3-DQ2, DR4-DQ8, or DR15-DQ6 rather than treating each gene as isolated.

Every person normally inherits one DQB1 allele from each biological parent. A report may show two different alleles, such as DQB102:01 and DQB103:02, or the same allele twice. The asterisk separates the gene from the allele family, and the digits after the colon provide increasing resolution. DQB102 is a broad allele group; DQB102:01 is a specific allele.

The DQB1 protein does not directly damage the pancreas or intestine. Certain versions bind and present particular peptides more effectively, influencing which T-cell responses can develop. Disease still requires other genes, immune events, and environmental exposure. HLA-DQB1 is a major susceptibility factor, but it is not a single-gene cause in the usual form of type 1 diabetes or celiac disease.

Why HLA-DQB1 testing is ordered

Doctors and laboratories order HLA-DQB1 typing for several distinct purposes. The clinical question should be clear before testing because the same allele has different meaning in different settings.

For celiac disease, DQB1 typing helps determine whether a person can form DQ2 or DQ8. It is especially useful when antibody and biopsy findings conflict, when a person stopped gluten before testing, or when relatives are being evaluated. A complete celiac HLA test should include DQA1 as well as DQB1.

For type 1 diabetes research or risk programs, DQB1 may be typed with DRB1 and DQA1 to identify high-risk haplotypes. Some programs combine HLA genotype with family history, islet autoantibodies, age, and metabolic markers to estimate progression risk. Routine population diagnosis of diabetes does not require HLA testing.

For narcolepsy, HLA-DQB1*06:02 testing may support an evaluation for narcolepsy type 1. It is most useful as one piece of a sleep specialist’s workup, not as a stand-alone screen. Clinical history, overnight polysomnography, a multiple sleep latency test, and sometimes cerebrospinal fluid hypocretin measurement carry more direct diagnostic value.

HLA-DQB1 can also be included in transplant typing. In kidney and hematopoietic stem cell transplantation, DQ matching and antibodies against donor HLA-DQ may influence compatibility and risk. That application differs from autoimmune disease risk testing and usually involves broader high-resolution HLA typing.

Testing may be reasonable when:

• a close relative has celiac disease and a clinician wants to decide who needs repeated serology;
• a person is gluten-free with an uncertain celiac diagnosis;
• a child is enrolled in an evidence-based type 1 diabetes risk or autoantibody screening program;
• symptoms strongly suggest narcolepsy type 1 but the clinical findings need additional support;
• a transplant team requests DQB1 typing as part of donor-recipient matching;
• an earlier report lists an incomplete DQ result that needs clarification.

Testing is usually not helpful as a broad explanation for fatigue, digestive symptoms, high blood sugar, or daytime sleepiness without condition-specific evaluation. A positive susceptibility allele is common enough that it may be incidental.

Testing methods and report formats

The laboratory can obtain DNA from blood, saliva, or cells collected with a cheek swab. No fasting is required. Eating or avoiding gluten does not alter the genotype. Insulin, immune medicines, sleep medicines, antibiotics, supplements, pregnancy, and recent illness do not change inherited DQB1 alleles.

Methods include polymerase chain reaction with sequence-specific primers or probes, sequence-based typing, and next-generation sequencing. Older or limited assays may report low-resolution groups such as DQB102 and DQB103. High-resolution testing can report alleles such as DQB1*02:01, *03:02, or *06:02.

A useful clinical report may include:

• the two detected DQB1 alleles;
• associated DQA1 and DRB1 alleles;
• inferred haplotypes, such as DR3-DQ2 or DR4-DQ8;
• celiac-associated DQ molecule status;
• a qualitative risk category;
• assay resolution and limitations;
• whether phase—knowing which alleles occur together on one chromosome—was directly determined or inferred.

A result written only as “DQB1 positive” is incomplete because everyone normally has DQB1. The meaningful information is the allele. Likewise, “HLA-DQ positive” does not identify which DQ molecule was detected.

Direct-to-consumer services sometimes infer HLA alleles from nearby single-nucleotide variants. Imputation can be reasonably accurate for common alleles in well-represented ancestry groups but less reliable for uncommon alleles or underrepresented populations. A medical decision should rely on a validated clinical assay if the consumer result is unexpected or consequential.

HLA nomenclature can look intimidating. In DQB1*03:02, “DQB1” names the gene, “03” identifies the allele family, and “02” specifies the protein-level allele. Additional fields may show synonymous or noncoding differences. Most disease interpretations use two-field resolution, although transplant programs may require greater detail.

HLA-DQB1 results and type 1 diabetes

Type 1 diabetes develops when an autoimmune process destroys insulin-producing beta cells in the pancreas. HLA class II variation contributes a large share of inherited risk, particularly combinations across DRB1, DQA1, and DQB1. The genotype influences susceptibility but does not show whether beta-cell autoimmunity has begun.

Among many populations with European ancestry, two major risk haplotypes are:

• DR3-DQ2, commonly including DRB103:01, DQA105:01, and DQB1*02:01;
• DR4-DQ8, commonly including selected DRB104 alleles, DQA103:01, and DQB1*03:02.

A child who inherits DR3-DQ2 from one parent and DR4-DQ8 from the other often has higher genetic susceptibility than a child with either haplotype alone. Risk differs among DRB1*04 subtypes and across ancestry groups, so a DQB1 allele cannot provide a universal percentage.

Some haplotypes are protective. DQB106:02, commonly linked with DRB115:01 and DQA1*01:02, is strongly protective against classic childhood type 1 diabetes in many populations. Protection is not absolute, and it should not be used to dismiss high glucose or diabetes symptoms.

Example HLA finding	General type 1 diabetes association	What it does not mean
DR3-DQ2	Increased susceptibility	Diabetes is present or inevitable
DR4-DQ8	Increased susceptibility; strength varies by DRB1*04 subtype	Blood glucose will become abnormal
DR3-DQ2/DR4-DQ8	Often among the higher-risk combinations	A precise risk can be calculated without age, ancestry, family history, and autoantibodies
DR15-DQ6 with DQB1*06:02	Often protective	Type 1 diabetes is impossible

Disease-focused follow-up uses pancreatic islet autoantibodies, not repeated HLA tests. Common autoantibodies include insulin autoantibodies, GAD65 antibodies, IA-2 antibodies, and ZnT8 antibodies. Persistent positivity for two or more islet autoantibodies indicates early-stage type 1 diabetes even before symptoms or abnormal glucose appear. Glucose, hemoglobin A1c, oral glucose tolerance, and sometimes continuous glucose monitoring assess metabolic progression.

HLA testing does not distinguish type 1 from type 2 diabetes in every individual. Adults can have autoimmune diabetes with lower-risk HLA, and people with high-risk HLA can develop type 2 diabetes. When classification is uncertain, autoantibodies, C-peptide, clinical course, body composition, treatment response, and family history are more direct tools.

Urgent symptoms of possible diabetes include excessive thirst, frequent urination, weight loss, vomiting, abdominal pain, deep breathing, fruity-smelling breath, severe fatigue, or confusion. Genetic risk should never delay immediate glucose and ketone testing.

HLA-DQB1 results and celiac disease

Celiac disease is an immune-mediated condition triggered by gluten in genetically susceptible people. Nearly all affected people can form a celiac-compatible HLA-DQ molecule. DQB1 supplies the beta chain, while DQA1 supplies the alpha chain.

The main patterns include:

• DQ2.5: DQA105 plus DQB102. This is the most common and generally strongest celiac susceptibility molecule.
• DQ8: DQA103 plus DQB103:02. This is found in a smaller portion of people with celiac disease.
• DQ2.2: DQA102 plus DQB102:02. This usually carries lower risk than DQ2.5.
• DQ7.5: DQA105 plus DQB103:01. This can support susceptibility in a small minority of cases.

A DQB102 result may be described as “half DQ2” if the needed DQA105 partner is absent. That result is not equivalent to full DQ2.5. Conversely, DQA105 without DQB102 is the other half. The complete allele combination matters.

Two copies of DQB102 can increase risk compared with one copy. A report may identify DQ2.5 homozygosity, DQ2.5 plus DQ2.2, or DQB102 gene dose. Higher susceptibility still does not establish active disease. Many carriers remain healthy throughout life.

The main clinical value of celiac HLA testing is its negative result. If a validated test finds no compatible DQA1/DQB1 combination, celiac disease becomes very unlikely. A positive result keeps celiac disease in consideration but must be followed by disease-specific testing. The HLA-DQ2 and HLA-DQ8 test interpretation should be combined with tissue transglutaminase IgA, total IgA, selected IgG-based tests, and sometimes duodenal biopsy.

A person should generally remain on a gluten-containing diet until the diagnostic workup is complete. HLA testing stays accurate off gluten, but celiac antibodies and intestinal injury can improve, making confirmation harder. A positive DQB1 allele alone is not a reason to start a lifelong gluten-free diet.

Type 1 diabetes and celiac disease share HLA risk architecture, particularly DR3-DQ2 and DR4-DQ8. This overlap helps explain why celiac screening is recommended in many people with type 1 diabetes. It does not mean one disease automatically causes the other.

Narcolepsy and other disease associations

HLA-DQB1*06:02 has one of the strongest known HLA associations with narcolepsy type 1. Most people with narcolepsy type 1 carry the allele, often as part of the DR15-DQ6 haplotype. The association supports an immune-mediated loss of orexin-producing neurons.

The result has limited specificity. DQB1*06:02 is common in healthy populations, and most carriers never develop narcolepsy. A positive test cannot explain ordinary tiredness, insufficient sleep, sleep apnea, medication sedation, depression, or shift-work fatigue. A negative result makes typical narcolepsy type 1 less likely but does not completely exclude it, particularly in unusual cases or across diverse populations.

Narcolepsy assessment focuses on recurrent irresistible sleepiness, cataplexy—brief muscle weakness triggered by emotion—sleep paralysis, vivid dreamlike experiences near sleep, and disrupted nighttime sleep. A sleep specialist may order overnight polysomnography followed by a multiple sleep latency test. Low cerebrospinal fluid hypocretin-1 can confirm narcolepsy type 1 in the appropriate setting.

DQB1 alleles have been statistically associated with many autoimmune, infectious, and inflammatory conditions. These associations often vary by ancestry and linked HLA genes. Examples in research include autoimmune thyroid disease, multiple sclerosis, autoimmune hepatitis, and responses to particular infections. Most are not appropriate for stand-alone clinical prediction because effect sizes are modest, linkage is complex, and useful preventive actions may be unclear.

A broad HLA-DQ genetic test may provide more context than isolated DQB1 typing when the clinical question involves DQ molecule structure or multiple autoimmune associations.

Limits and common misinterpretations

HLA-DQB1 testing has several important limits:

• Susceptibility is not diagnosis. A high-risk allele can be present without autoimmunity, symptoms, or organ damage.
• The partner genes matter. DQA1 and DRB1 can change the meaning of a DQB1 result.
• Risk varies by ancestry. A risk estimate developed in one population may not transfer accurately to another.
• HLA genes are highly linked. An apparent DQB1 association may partly reflect nearby DRB1, DQA1, or other variants.
• Resolution differs. A broad DQB103 result does not distinguish DQB103:02 from other *03 alleles with different associations.
• Phase may be uncertain. Without family testing or high-resolution methods, a laboratory may infer which alleles occur together.
• Consumer data may be incomplete. Imputed alleles should be confirmed before a medical decision.

Common errors include interpreting “positive” as disease, using a protective allele to ignore symptoms, calculating a child’s future from one allele, or starting dietary or medical treatment without condition-specific evidence. Another mistake is comparing raw odds ratios from different studies as though they were personal probabilities. Odds ratios describe groups and depend on the reference genotype and population.

HLA testing cannot predict age of onset, severity, treatment response, or complications with certainty. It does not replace pancreatic autoantibodies for type 1 diabetes, celiac serology and biopsy for celiac disease, or sleep testing for narcolepsy.

An ambiguous or unexpected result may need review by a laboratory geneticist, immunogenetics specialist, or genetic counselor. Repeating the same inherited test is rarely useful unless the first assay was incomplete, low resolution, or technically uncertain.

Follow-up after testing

Start by obtaining the full report, not only a portal label. Note both DQB1 alleles, the typing resolution, DQA1 and DRB1 results, inferred haplotypes, and the laboratory’s clinical interpretation.

For type 1 diabetes susceptibility, ask whether screening for islet autoantibodies is appropriate based on age, family history, and available programs. HLA does not need to be repeated. Anyone with symptoms of high glucose needs immediate glucose testing regardless of genotype.

For celiac susceptibility, determine whether a complete celiac HLA analysis found DQ2.5, DQ8, DQ2.2, or DQ7.5. If the person is eating gluten, celiac serology usually comes next. If already gluten-free, discuss old records and the benefits and burdens of a supervised gluten challenge before changing the diet.

For possible narcolepsy, take the result to a sleep specialist. Keep a sleep schedule and symptom record, but do not use DQB1*06:02 as a self-diagnosis. Driving and safety advice may be needed when irresistible sleep episodes occur.

Family testing is most useful when it changes screening. A child’s result should not be used to restrict diet, limit activity, or create a disease label in the absence of clinical evidence. Genetic counseling can clarify inheritance, explain why siblings can receive different haplotypes, and separate relative risk from certainty.

The most useful question after any HLA-DQB1 result is which disease-specific action follows. Sometimes the answer is antibody testing or specialist evaluation; sometimes it is periodic screening; and sometimes no further testing is needed.

Using HLA risk in screening programs

HLA genotyping has often been used to enrich research cohorts for children more likely to develop islet autoimmunity. Modern screening increasingly starts with islet autoantibodies because they indicate that autoimmunity has already begun and can identify risk across HLA backgrounds. HLA remains useful for understanding inherited susceptibility and refining selected models.

A child with a high-risk DR-DQ genotype but no islet autoantibodies does not have type 1 diabetes. The genotype does not need repeating. A child with persistent multiple autoantibodies requires specialist monitoring even when glucose is normal, because stage 1 type 1 diabetes is defined by autoimmunity with normal glucose regulation.

Families should receive education about thirst, frequent urination, weight loss, fatigue, bed-wetting, vomiting, and deep breathing. Early recognition can reduce presentation with diabetic ketoacidosis. Home glucose testing or continuous monitoring should follow the screening program’s plan rather than genotype alone.

Allele names and haplotypes in the report

DQB102:01 is commonly linked with DQA105:01 and DRB103:01 on DR3-DQ2. DQB103:02 is commonly linked with DQA103:01 on DQ8 and selected DRB104 alleles. The exact DRB104 subtype changes diabetes association, illustrating why “DQB103:02 positive” is incomplete.

DQB106:02 is often part of DRB115:01-DQA101:02-DQB106:02. It is strongly protective against many forms of childhood type 1 diabetes yet strongly associated with narcolepsy type 1. These apparently opposite associations arise from different antigen-specific immune mechanisms, not from a generally strong or weak immune system.

When a report lists more than one possible haplotype, phase may be inferred from population frequencies. Family testing can resolve phase, but it is rarely necessary unless a detailed research risk model or celiac DQ molecule assignment depends on it.

For any disease association, the report should identify the population on which its risk statement is based. Allele frequencies and linked haplotypes differ, so a category developed in one ancestry group may overestimate or underestimate risk in another.

References

• The heterogeneity of type 1 diabetes: implications for understanding pathogenesis, prediction, prevention, and treatment 2025 (Review)
• Next-generation sequencing reveals additional HLA class I and II risk alleles associated with type 1 diabetes and age at onset 2024
• HLA Genotyping in Children With Celiac Disease Allows to Establish a Genomic Risk Profile for the Development of Type 1 Diabetes 2024
• The immunopathogenesis of narcolepsy type 1 2024 (Review)
• The Role of T Cells in the Pathogenesis of Narcolepsy Type 1 2024 (Review)
• Celiac Disease 2025 (Review)

Disclaimer

This article provides general information about HLA-DQB1 testing and cannot diagnose type 1 diabetes, celiac disease, narcolepsy, or another immune condition. A qualified clinician should interpret the allele with HLA-DQA1 and HLA-DRB1 findings, ancestry, symptoms, and disease-specific tests. Seek urgent medical care for symptoms of diabetic ketoacidosis, severe dehydration, confusion, or another acute illness.