PURPOSE This study was conducted to evaluate the usefulness of human leukocyte antigen (HLA) typing in preventing allopurinol-induced severe cutaneous adverse reactions (SCARs) through the application of an allopurinol tolerance induction protocol or prescription of other alternative medications in high-risk patients.

METHODS HLA typing was performed in patients with chronic renal insufficiency who needed allopurinol. HLA-B*58:01-negative patients were prescribed the usual dose of allopurinol. For HLA-B*58:01-positive patients, administration of either allopurinol based on a 28-day tolerance induction protocol or alternative medications was initiated. Hypersensitivity reactions were surveyed for 90 days and compared with the result of a previous retrospective cohort study.

RESULTS Among a total of 401 study subjects, no SCARs were noted in HLA-B*58:01-positive patients with application of the tolerance induction protocol (n = 30) or alternative medications (n = 16), nor were any SCARs observed in HLA-B*58:01-negative patients who started allopurinol at the usual dose (n = 355). Compared with the previous retrospective cohort study, a significant reduction in SCARs was observed in HLA-B*58:01-positive patients (0 vs. 18%; P = 0.002).

CONCLUSION This study shows the usefulness of HLA-B*58:01 screening in identifying patients at high risk for the development of allopurinol-induced SCARs and suggests that application of a tolerance induction protocol or alternative medications could be an effective strategy to prevent allopurinol-induced SCARs in HLA-B*58:01-positive patients.