HLA-Associated Severe Adverse Reactions

Certain severe, immune-mediated adverse drug reactions are strongly linked to specific human leukocyte antigen (HLA) alleles. These alleles, which encode molecules that present peptides and small molecules to T cells, predispose carriers to reactions such as drug hypersensitivity syndromes and severe cutaneous reactions, making them some of the most clinically actionable examples in pharmacogenomics.

Definition

HLA-associated severe adverse reactions are drug-induced immune reactions whose risk is strongly determined by carriage of particular HLA class I or class II alleles that present the drug or its metabolite to T cells.

Scope

This topic covers the principal HLA-drug associations and the immunological model that explains them: how an HLA allele can confer high relative risk for a specific reaction to a specific drug. It addresses the landmark associations for abacavir, carbamazepine, and allopurinol, the variation of risk alleles across populations, and the evidence supporting predictive screening. It is reference-educational and does not provide testing or prescribing instructions.

Core questions

• Which HLA alleles are robustly associated with which drug reactions?
• How does an HLA molecule mediate recognition of a small-molecule drug?
• Why do the same alleles confer different population-level risk across ancestries?
• What evidence supports screening for a given HLA-drug pair?

Key concepts

• HLA class I and class II restriction
• HLA-B*57:01 and abacavir hypersensitivity
• HLA-B*15:02 and carbamazepine-induced Stevens-Johnson syndrome
• HLA-B*58:01 and allopurinol severe cutaneous reactions
• Negative and positive predictive value of screening
• Population-specific allele frequencies

Key theories

Altered-repertoire / pharmacological-interaction models of HLA-restricted drug presentation

Proposed mechanisms explain how an HLA allele triggers a drug-specific T-cell response: a drug or metabolite may bind within the HLA peptide-binding groove and alter the displayed self-peptide repertoire, or interact non-covalently with HLA and T-cell receptor; the abacavir-HLA-B*57:01 association is a paradigm for allele-specific binding.

Mechanisms

HLA molecules present peptides to T cells as part of normal immune surveillance. In HLA-associated drug reactions, a particular allele can accommodate a drug or its reactive metabolite, either by binding within the peptide-binding groove and shifting the repertoire of self-peptides displayed, or by interacting directly with the HLA molecule and T-cell receptor. This generates a drug-specific cytotoxic T-cell response that drives tissue injury, most visibly in skin and mucosa. The exquisite specificity of these interactions explains why a single allele can confer high relative risk for a reaction to one drug while being irrelevant to others.

Clinical relevance

HLA-drug associations are among the few pharmacogenomic findings with high enough effect size to inform pre-prescription testing strategies, and they illustrate how genotype relates to serious drug toxicity. This entry explains the associations and their evidence base for educational appraisal; decisions about testing or prescribing belong to clinicians following current local guidance and are outside its scope.

Epidemiology

The strength and clinical value of an HLA-drug association depend on allele frequency in the population. HLA-B*15:02 is common in several Southeast and East Asian populations and rare in Europeans, so carbamazepine-related Stevens-Johnson syndrome risk and the value of screening differ accordingly. HLA-B*57:01 carriage underlies abacavir hypersensitivity across populations, and HLA-B*58:01 is associated with allopurinol-induced severe cutaneous reactions, with higher frequency in some Asian groups.

Evidence & guidelines

The evidence spans case-control discovery studies and at least one randomized screening trial: PREDICT-1 demonstrated that prospective HLA-B*57:01 screening eliminated immunologically confirmed abacavir hypersensitivity. The Clinical Pharmacogenetics Implementation Consortium has issued guidelines for HLA genotype and carbamazepine/oxcarbazepine, translating these associations into structured recommendations that lie outside the individualized advice provided here.

History

The field began with the 2002 case-control discovery that HLA-B*57:01 was associated with abacavir hypersensitivity, followed almost immediately by the identification of HLA-B*15:02 as a marker for carbamazepine-induced Stevens-Johnson syndrome and HLA-B*58:01 for allopurinol reactions in 2004-2005. The 2008 PREDICT-1 trial then provided randomized evidence that screening could prevent a severe reaction, establishing HLA testing as a model for actionable pharmacogenomics.

Debates

How should screening thresholds account for population allele frequency?

Because risk-allele frequencies differ greatly across ancestries, the number needed to test and the cost-effectiveness of screening vary by population, and recommendations must balance high negative predictive value against modest positive predictive value.

Key figures

• Simon Mallal
• Elizabeth Phillips
• Wen-Hung Chung
• Shuen-Iu Hung

Seminal works

• mallal-2002
• mallal-2008
• chung-2004
• hung-2005

Frequently asked questions

Why does an HLA allele predispose to a reaction to only one drug?

Because the HLA molecule must physically accommodate that particular drug or its metabolite to present it to T cells; the structural fit is drug-specific, so a given allele typically confers risk for one or a few drugs rather than all.

Does carrying a risk allele mean a reaction is certain?

No. Most associations have high negative predictive value, so absence of the allele largely excludes risk, but only a minority of carriers exposed to the drug actually develop the reaction, so the positive predictive value is much lower.