HLA-B*58:01 is an Incomplete Predictor of Allopurinol-Induced Severe Cutaneous Adverse Reactions

Importance Carriage of HLA-B*58:01 has been shown to have a strong association with the development of allopurinol-induced Stevens-Johnson syndrome and toxic epidermal necrolysis (SJS/TEN) and drug reaction and eosinophilia and systemic symptoms (DRESS) in many populations globally; however, there is a critical need to determine if this is generalizable to varying populations including those in the United States (US). Objective To perform HLA class I and II association studies in a cohort of US patients diagnosed with allopurinol-induced SJS/TEN or DRESS compared to allopurinol tolerant and population controls. Design, Setting, and Participants We enrolled consenting individuals who had specialist adjudicated allopurinol-induced SJS/TEN or DRESS (collectively allopurinol-SCAR). HLA carriage in these cases was compared to allopurinol tolerant and population controls identified through Vanderbilt University Medical Center (VUMC) BioVU, a biobank which includes 94,489 individuals with imputed human leukocyte antigen (HLA) class I and II typing from genotyping array data. Main Outcomes and Measures We performed HLA class I and II conditional logistic regression case-control analyses between allopurinol-SCAR cases and both population controls and allopurinol tolerant controls matched on age, sex, and self-identified race. We reported odds ratio (OR) and 95% confidence interval (CI) with Bonferroni corrected P ( Pc ) <.05. Results The conditional logistic regression analyses included allopurinol-SCAR cases (n=16) and 10:1 matched allopurinol tolerant controls (n=160). We found two HLA class I alleles independently associated with increased risk of allopurinol-induced SCAR: HLA-B*58:01 (OR 28 [95% CI, 8.6 – 100.6]) and HLA-A*34:02 (OR 20.6 [95% CI, 3.3 – 131.1]). We did not identify any HLA class II alleles meeting the Pc level of significance. Conclusions and Relevance We found HLA-B*58:01 to be strongly associated with allopurinol-induced SCAR, generalizing findings from previous studies. Additionally, we found HLA-A*34:02 to be a second independent genetic risk factor for allopurinol-SCAR. These findings underscore the need to conduct specific population-based studies that both reproduce known and uncover novel HLA associations in order to reduce harm through contributions to screening, risk stratification, and diagnosis. KEY POINTS Question Is the association between HLA-B*58:01 and allopurinol-SCAR generalizable to admixed populations in the US or are additional HLA associations involved? Findings In this HLA association study with 16 patients of primarily self-identified Black race of adjudicated allopurinol-induced SJS/TEN or DRESS, we demonstrate a strong association with the established risk allele, HLA-B*58:01, and, for the first time, identified HLA-A*34:02 as an additional independent risk factor. Meaning HLA-B*58:01 is absent in more than one-third of our US cohort of allopurinol-SCAR cases suggesting that more comprehensive screening and diagnostic approaches are necessary to prevent additional cases in genetically heterogenous populations.