HLA-B*58:01 was absent in more than one-third of patients with allopurinol-induced severe cutaneous adverse reactions, according to a recent brief report.
HLA-B*58:01 and HLA-A*34:02 were identified as independent genetic risk factors for allopurinol-induced severe cutaneous adverse reactions (SCARs) in a heterogeneous US population. HLA-B*58:01 carried the highest risk, with an odds ratio of 28 and a 95% confidence interval of 8.6 to 100.6, while HLA-A*34:02 demonstrated an odds ratio of 20.6 and a 95% confidence interval of 3.3 to 131.1. A gene-dose effect was observed for HLA-B*58:01, with homozygous carriage yielding an odds ratio of 55.3 and a 95% confidence interval of 2.5 to 1208.6.
Researchers analyzed whether established human leukocyte antigen (HLA) risk markers apply to genetically diverse populations in the US. This genetic association brief report included 16 patients with specialist-adjudicated Stevens-Johnson syndrome, toxic epidermal necrolysis, or drug reaction with eosinophilia and systemic symptoms (DRESS). They were matched with 160 allopurinol-tolerant control participants by age, sex, and race. High-resolution human leukocyte antigen typing was conducted for all patients and compared with allele frequencies in both matched controls and a biobank cohort of 94,489 participants. Conditional logistic regression assessed class I and class II allele associations, and Bonferroni-corrected significance thresholds guided interpretation.
No class II alleles reached significance. HLA-B*35:01 and HLA-B*58:02 showed unadjusted associations that did not remain after statistical correction. HLA-A*34:02 was more common among patients (3 of 16) than among tolerant matched controls (7 of 160) and the biobank population (1,246 of 94,489). Its identification as a second independent risk factor was reported in the brief report.
HLA-A*34:02 was also associated with increased risk for allopurinol-induced SCARs and has been linked to allopurinol-related drug-induced liver injury that may precede DRESS with elevated liver enzymes. In a prior SCAR cohort, 5 of 9 HLA-B*58:01–negative patients carried an HLA-A34 allele, which included HLA-A*34:02.
The report, published in JAMA Dermatology, was limited by its small cohort size, although the diversity of the participants allowed meaningful assessment of genetic associations despite the rarity of severe cutaneous adverse reactions. The findings also indicated that relying solely on HLA-B*58:01 screening would not identify a substantial proportion of affected patients in the US population. HLA-B*58:01 was not present in over one-third of the overall SCAR cohort and was also lacking in 45% of self-identified Black patients, a group commonly treated with allopurinol.
“It is important to note, however, that an ideal screening test has 100% negative predictive value, and our US brief report identifies that carriage of HLA-B*58:01 alone is not capable of identifying all individuals at risk of allopurinol-induced SCARs,” noted lead author Chelsea N. Campbell, MS, of the Department of Pathology, Microbiology, and Immunology, and the Center for Drug Safety and Immunology, Department of Medicine at Vanderbilt University Medical Center in Nashville, Tennessee, with colleagues.
Elizabeth J. Phillips, MD, reported receiving personal fees from several pharmaceutical and medical organizations unrelated to this brief report, and no additional disclosures were noted.
Source: JAMA Dermatology