The HLA-A31:01 test checks for an inherited immune-system allele associated with a higher risk of carbamazepine hypersensitivity. Carbamazepine is used for epilepsy, trigeminal neuralgia, and some mood disorders, but it can rarely trigger serious immune reactions involving the skin and internal organs. HLA-A31:01 is linked to several reaction patterns, including a widespread maculopapular rash, drug reaction with eosinophilia and systemic symptoms (DRESS), and, less consistently, Stevens–Johnson syndrome or toxic epidermal necrolysis. A positive result does not mean a reaction will occur, and a negative result does not make carbamazepine completely risk-free. The result is most useful before the first dose, when a prescriber can compare carbamazepine with suitable alternatives. It must also be interpreted alongside ancestry, prior drug exposure, other HLA alleles—especially HLA-B*15:02—and the medical reason for treatment. Because HLA type is inherited and stable, a valid result normally does not need to be repeated.
- A positive HLA-A*31:01 result means increased carbamazepine hypersensitivity risk, not certain harm.
- A negative result lowers one genetic risk but does not exclude rash, DRESS, SJS, or TEN.
- Testing is most informative before carbamazepine is started; no fasting or medication pause is required.
- HLA-A31:01 and HLA-B15:02 answer different, partly overlapping drug-safety questions.
- Blisters, mouth or eye sores, facial swelling, fever, or a spreading rash after treatment require urgent medical assessment.
Table of Contents
- What the HLA-A*31:01 Test Measures
- Who May Benefit From Testing
- How Testing Is Performed
- What a Positive Result Means
- What a Negative Result Means
- HLA-A*31:01 Versus HLA-B*15:02
- Medication Decisions After the Result
- Symptoms, Limitations, and Next Steps
What the HLA-A*31:01 Test Measures
HLA-A*31:01 is one allele of the HLA-A gene, a highly variable human leukocyte antigen gene on chromosome 6. HLA-A molecules sit on the surface of most nucleated cells and present short peptide fragments to CD8 T cells. This system helps the immune system distinguish normal cells from cells affected by infection, cancer, or other changes.
Each person normally inherits one HLA-A allele from each biological parent. A targeted HLA-A31:01 test asks whether at least one of those two copies is HLA-A31:01. It is not a test for a damaged gene, immune deficiency, or carbamazepine concentration. The allele is a normal form of HLA-A found in healthy people.
Carbamazepine or a related molecule can interact with immune recognition in a way that activates drug-reactive T cells in susceptible people. The exact mechanism is complex and may differ among reaction phenotypes. HLA-A*31:01 changes the probability that carbamazepine-derived signals will be presented or recognized in a harmful way, but other genes, viral reactivation, dose, concurrent medicines, and chance also contribute.
The association is broader than one specific rash. Studies have linked HLA-A*31:01 to:
- Maculopapular exanthema, a widespread red or raised drug eruption
- DRESS, which may include fever, rash, facial swelling, eosinophilia, swollen lymph nodes, and liver or other organ injury
- Stevens–Johnson syndrome and toxic epidermal necrolysis, uncommon blistering disorders involving skin and mucous membranes
- Carbamazepine hypersensitivity syndrome, an older term that often overlaps with DRESS
The strength of association is not identical for every phenotype or population. HLA-A*31:01 has shown clinically important associations in people of European, Japanese, Korean, and some other ancestries. Allele frequency also varies among populations. An ancestry label, however, is an imperfect substitute for a person’s genotype, particularly in families with mixed or uncertain ancestry.
A general HLA typing test may identify HLA-A31:01 if it reaches adequate resolution, but a broad result such as “HLA-A31” is not always sufficient. Medication guidance is tied to the exact allele with the asterisk and four-digit field: HLA-A31:01.
Who May Benefit From Testing
Testing is mainly considered before carbamazepine is prescribed to someone who has never taken it. The prescriber weighs the expected value of the result against treatment urgency, local guidance, available alternatives, and the person’s ancestry and clinical history.
Carbamazepine may be considered for focal seizures, trigeminal neuralgia, glossopharyngeal neuralgia, or bipolar disorder. The value of testing can differ among these situations because alternative drugs have different effectiveness, adverse effects, interactions, and pregnancy considerations.
Testing may be especially useful when:
- Carbamazepine is a realistic treatment option and therapy can wait for the result
- The person has ancestry in a population where HLA-A*31:01 is present at a meaningful frequency
- The person’s ancestry is mixed, unknown, or not well represented by broad population labels
- A health system uses preemptive pharmacogenetic testing before selected antiseizure medicines
- A previous report shows HLA-A31 at low resolution and exact confirmation is needed
- A family member had a serious carbamazepine reaction, although the patient still needs their own clinical evaluation
A family history can raise concern but cannot determine the result. Close relatives may share HLA alleles, yet siblings do not inherit identical HLA combinations unless they happen to receive the same parental haplotypes. A relative’s negative result also does not prove that the patient is negative.
Testing after a suspected reaction can sometimes support the investigation, but it does not prove which drug caused the event. Drug causality depends on timing, clinical features, laboratory findings, competing infections, all medicines taken, and sometimes specialist testing. A positive HLA-A*31:01 result is supporting evidence rather than a stand-alone diagnosis of DRESS or SJS/TEN.
Testing is less likely to change care after a person has taken carbamazepine continuously for a long period without hypersensitivity. Most immune-mediated cutaneous reactions occur early, commonly within the first several weeks or months. Late reactions can occur, and other toxicities remain possible, so long-term tolerance is not an absolute guarantee. A person already taking carbamazepine should not stop or alter it solely to obtain genetic testing without speaking with the prescriber.
The test is not a broad allergy panel. It does not predict ordinary dizziness, double vision, low sodium, blood-count changes, liver enzyme changes, or every possible carbamazepine adverse effect. It also does not establish which antiseizure medicine will control seizures best.
How Testing Is Performed
The test usually uses a blood sample or cheek swab. No fasting is needed, and carbamazepine does not have to be present in the body. Because the result comes from inherited DNA, the test can be done at any age and before any drug exposure.
Laboratories may use allele-specific polymerase chain reaction, sequence-specific primers, probe-based methods, Sanger sequencing, or next-generation sequencing. A targeted assay reports whether HLA-A*31:01 was detected. A broader pharmacogenetic or HLA panel may report both HLA-A alleles and other relevant markers.
A typical report may say:
- HLA-A*31:01 detected
- HLA-A*31:01 not detected
- Positive for one copy of HLA-A*31:01
- HLA-A genotype: HLA-A02:01/HLA-A31:01
For prescribing, one copy is generally enough to classify the person as a carrier. There is limited reason to make different routine recommendations for one versus two copies because the allele is uncommon and guidance usually treats any positive result as increased risk.
The report should identify the exact allele and describe the assay. A tag single-nucleotide variant can sometimes serve as a surrogate for HLA-A*31:01, but its accuracy may vary across ancestry groups if the tag is not perfectly linked to the allele. Direct HLA typing or a well-validated targeted assay provides stronger confirmation for a clinical decision.
Potential reasons for repeat or confirmatory testing include:
- The result came from research or direct-to-consumer data rather than a clinical laboratory
- The report lists only HLA-A31 without allele-level resolution
- The assay’s population validation is unclear
- The specimen may have been mislabeled or contaminated
- The person has received an allogeneic stem-cell transplant
After donor stem-cell transplantation, blood DNA may largely represent the donor. A cheek swab can also contain donor-derived white cells. If the recipient’s original inherited drug-risk genotype is needed, the laboratory may request a pretransplant sample, hair follicles, cultured skin cells, or another source selected to reduce donor-cell interference.
Turnaround time ranges from a few days to several weeks depending on the laboratory and whether the test is local or sent out. Treatment urgency may influence whether the prescriber waits, chooses another drug, or starts a different short-term plan.
What a Positive Result Means
A positive result means HLA-A*31:01 was detected and the person has a higher-than-baseline risk of carbamazepine hypersensitivity. It does not mean the person is allergic now, and it cannot predict the exact reaction, severity, or timing.
The absolute risk remains much lower than 100%. Many carriers could take carbamazepine without developing a severe reaction, but exposing every carrier to find out would create avoidable risk when effective alternatives are available. This is why pharmacogenetic guidance often favors another medicine for a carbamazepine-naive carrier when clinically appropriate.
The result is most relevant to carbamazepine. It should not automatically be applied to every medication with a similar chemical structure. Oxcarbazepine, eslicarbazepine, lamotrigine, phenytoin, and phenobarbital have their own evidence, HLA associations, and reaction profiles. Cross-reactivity can occur among aromatic antiseizure medicines, especially after a true severe hypersensitivity reaction, but a positive HLA-A*31:01 result alone does not provide a complete cross-reactivity map.
A positive result should be recorded in several places:
- The prescribing clinician’s note
- The medication allergy or pharmacogenomic alert section of the health record
- The patient’s personal medication list
- Pharmacy records when possible
- A medical alert card if recommended locally
The record should state “HLA-A*31:01 positive—increased carbamazepine hypersensitivity risk,” rather than simply “carbamazepine allergy” if the person has never been exposed. This distinction preserves accurate history while still preventing inadvertent prescribing.
If carbamazepine is considered essential despite a positive result, the decision requires specialist discussion. It is not a situation for a casual monitored trial at home. The clinician must consider the indication, alternative treatments, severity of the untreated condition, guideline recommendations, and the limits of rescue once a severe immune reaction begins. Routine desensitization is not an appropriate way to overcome genetic risk for SJS/TEN or DRESS.
A positive HLA result does not explain unrelated rashes. Eczema, contact dermatitis, viral exanthems, and reactions to other drugs still require their own assessment. Likewise, HLA-A*31:01 is not a diagnosis of an autoimmune disease or a measure of immune strength.
What a Negative Result Means
A negative result means the laboratory did not detect HLA-A*31:01. This removes one recognized genetic risk factor and may support standard carbamazepine use when the medicine is otherwise appropriate. It does not reduce risk to zero.
A negative person can still develop:
- A mild drug rash
- DRESS
- SJS or TEN
- Liver injury
- Blood-cell abnormalities
- Hyponatremia
- Neurologic dose-related effects
- Reactions associated with another HLA allele
The negative result must be interpreted with HLA-B15:02 status when that allele is relevant. HLA-B15:02 has a particularly strong association with carbamazepine-induced SJS/TEN in several Asian populations. Someone can be HLA-A31:01 negative and HLA-B15:02 positive, or the reverse.
A negative test also cannot guarantee that a broad panel detected every rare risk allele. Most clinical tests target alleles with established evidence and actionable guidance. New associations continue to be studied, but they are not all ready for routine prescribing.
If carbamazepine is started after a negative result, ordinary safety measures still apply. The prescriber should review interactions, baseline blood count and liver function when indicated, sodium risk, pregnancy considerations, and gradual dose titration. The patient should receive clear instructions about early symptoms of hypersensitivity.
Do not repeat the test after a negative result merely because the person ages or changes medication. Germline HLA genotype does not change. Repeating may be reasonable only when the original report lacked adequate resolution, used an unvalidated method, or may reflect donor DNA after stem-cell transplantation.
A negative result should not be used to restart carbamazepine after a convincing previous severe reaction. Clinical history takes priority. Re-exposure after DRESS, SJS, or TEN can be dangerous even when the tested allele is absent.
HLA-A*31:01 Versus HLA-B*15:02
These two tests are related but not interchangeable. Both concern carbamazepine hypersensitivity, yet their strongest associations, population distributions, and clinical emphasis differ.
| Feature | HLA-A*31:01 | HLA-B*15:02 |
|---|---|---|
| Strongest clinical signal | Broader hypersensitivity spectrum, especially maculopapular eruption and DRESS, with some SJS/TEN association | Very strong association with carbamazepine-related SJS/TEN in several Asian ancestry groups |
| Population pattern | Found across multiple populations, including European and East Asian groups | More frequent in parts of Southeast, East, and South Asia; uncommon in many European populations |
| Positive result | Usually supports choosing an alternative to carbamazepine in a treatment-naive patient | Usually supports avoiding carbamazepine because of SJS/TEN risk |
| Negative result | Does not address HLA-B*15:02 risk | Does not address HLA-A*31:01 risk |
An ancestry-based rule can miss people because populations are diverse and family histories are incomplete. Some guidelines recommend broader testing than others. Local prescribing labels, professional guidance, and test availability should therefore be considered together.
A broad pharmacogenetic test may include both markers. The clinician should verify that the panel directly identifies the required alleles rather than inferring them from a limited set of nearby variants.
Other alleles, including HLA-B15:11 and HLA-B15:21, have been associated with carbamazepine cutaneous reactions in some populations. Their routine use varies because evidence and guideline coverage are less uniform. A laboratory panel may report them, but the prescriber should not assume every listed association carries the same level of actionability.
Medication Decisions After the Result
The result informs prescribing; it does not choose the medication by itself. The best alternative depends on the condition being treated, prior response, age, pregnancy potential, kidney and liver function, other medicines, and cost or access.
For epilepsy, alternatives may include medicines such as levetiracetam, lamotrigine, lacosamide, valproate, or others, but seizure type and patient characteristics determine suitability. Lamotrigine itself can cause serious rash and requires slow titration. Valproate has major pregnancy-related risks. One drug cannot be substituted safely from a genetic report alone.
For trigeminal neuralgia, oxcarbazepine is sometimes considered, but it is chemically related and has its own HLA-associated risks. Other options or procedural treatments may be appropriate in selected patients. For bipolar disorder, lithium, valproate, lamotrigine, or atypical antipsychotics may be considered depending on the phase of illness and individual risks.
A clinician reviewing a positive result should:
- Confirm that the patient has not already had a carbamazepine reaction.
- Verify the exact allele and laboratory method.
- Review HLA-B*15:02 or other testing required by ancestry and guidance.
- Reassess why carbamazepine was selected.
- Compare effective non-carbamazepine options.
- Document the result so another prescriber does not unknowingly repeat the decision.
A patient with a negative result still needs a start-up plan. Carbamazepine is commonly titrated rather than immediately given at a full maintenance dose. The prescriber may obtain baseline laboratory tests and repeat them based on symptoms, dose, comorbidities, and local practice.
Genetic testing is preventive, not a substitute for follow-up. A person should know whom to call, which symptoms require same-day advice, and which symptoms require emergency care. Photographs can help document an early rash, but taking a picture should never delay evaluation of a rapidly evolving reaction.
Symptoms, Limitations, and Next Steps
Serious carbamazepine hypersensitivity usually begins after exposure, often during the first weeks. Early symptoms may resemble a viral illness. Any new rash deserves prompt contact with the prescriber, particularly when accompanied by fever or systemic symptoms.
Seek urgent medical assessment for:
- Blisters, skin peeling, painful skin, or purple target-like spots
- Sores or pain in the mouth, eyes, nose, or genital area
- Facial swelling or marked swelling around the eyes
- Fever with a spreading rash
- Shortness of breath, throat tightness, or faintness
- Yellow skin or eyes, dark urine, severe abdominal pain, or confusion
- Enlarged lymph nodes, profound weakness, or signs of organ involvement
SJS/TEN and DRESS can progress after the drug is stopped and may require hospital care. At the same time, suddenly stopping an antiseizure medicine can provoke seizures. The safest response is immediate professional assessment rather than an unsupervised stop-and-restart decision. Emergency symptoms take priority over routine medication instructions.
The test has several limits. It cannot determine the probability for one individual with perfect precision. Risk estimates vary with phenotype, ancestry, study design, and background incidence. It does not detect every genetic contributor, and a positive result cannot identify which alternative medicine will be tolerated.
Before acting on a result, ask:
- Does the report say HLA-A*31:01 exactly?
- Was the test performed in a clinical laboratory?
- Is HLA-B*15:02 testing also indicated?
- Has the patient ever taken carbamazepine, and for how long?
- Was there a previous rash or systemic reaction?
- Which alternative has comparable benefit for the condition?
- Is the result documented in the permanent record?
Keep the original report. HLA allele names can be lost when copied into a portal or allergy list, and “HLA-A31 positive” may be too vague for future prescribing. The full result should travel with the patient across neurology, psychiatry, pain care, primary care, and pharmacy settings.
HLA-A*31:01 testing works best as a one-time, pre-prescription safety measure. A positive result can prevent avoidable exposure, while a negative result supports—but never replaces—careful prescribing and early recognition of hypersensitivity.
Previous clinical hypersensitivity outweighs a reassuring genotype. Someone who developed a convincing carbamazepine-related DRESS, SJS, or TEN should not be re-exposed merely because HLA-A31:01 was not detected. The reaction may have involved HLA-B15:02, another allele, or a mechanism not captured by current testing. Conversely, a positive result in a person who once tolerated a brief course does not prove lifelong safety; duration, dose, and the accuracy of the exposure history still matter.
Panel reports may include many gene-drug pairs. Only the markers relevant to the contemplated medicine should drive the immediate choice, and each should be interpreted under a current guideline. A “normal pharmacogenetic panel” summary can hide the fact that HLA-A*31:01 was not assayed or was imputed rather than directly typed.
References
- HLA genotype testing for carbamazepine, oxcarbazepine and eslicarbazepine: A guideline developed by the UK Centre of Excellence in Regulatory Science and Innovation in Pharmacogenomics (CERSI-PGx) 2026 (Guideline)
- Pharmacogenetic biomarkers associated with risk of developing severe drug eruptions and clinical implementation of HLA genetic testing 2025 (Review)
- Dutch Pharmacogenetics Working Group (DPWG) guideline for the gene-drug interaction of CYP2C9, HLA-A and HLA-B with anti-epileptic drugs 2024 (Guideline)
- Association of HLA-B13:01 with DRESS and HLA-A31:01 with eosinophilia in SJS/TEN induced by carbamazepine 2024
- Associations of HLA genetic variants with carbamazepine-induced cutaneous adverse drug reactions: An updated meta-analysis 2022 (Meta-analysis)
- Association of HLA-A*31:01 Screening With the Incidence of Carbamazepine-Induced Cutaneous Adverse Reactions in a Japanese Population 2018
Disclaimer
This article provides general educational information and does not replace individualized prescribing advice. Do not start, stop, restart, or substitute carbamazepine based only on an HLA result; a qualified clinician should review the exact allele, prior reactions, ancestry, indication, and alternatives. A rapidly spreading rash, blistering, mucosal sores, facial swelling, fever, or breathing difficulty after carbamazepine requires urgent medical assessment.
