To the Editor:
Allopurinol is the major drug used in the treatment of gout and hyperuricemia. Generally, the drug is well tolerated, although a minority of people, about 2%, develop a hypersensitivity reaction with rash or, less frequently, Stevens-Johnson syndrome (SJS)1. A multinational study reported that allopurinol is the most common drug associated with SJS and toxic epidermal necrolysis2. Genomic studies have shown that the HLA-B*5801 allele is a strong risk factor (OR 34–348) for developing allopurinol-induced SJS3,4,5,6,7,8. However, the clinical utility of HLA-B*5801 is unclear.
We conducted an observational study in the immediate family members of a male patient who experienced allopurinol-induced SJS in 1997 (index case). The patient, now 72 years old, was diagnosed with SJS by a dermatologist when he developed a generalized blistering rash, fever, and internal organ failure 1 week after the initiation of allopurinol 300 mg/day for gouty arthritis. He had a white blood cell count of 13 × 109/l and 19% eosinophilia. Mild hepatic dysfunction and raised inflammatory markers (C-reactive protein 86 mg/l) were observed. Skin biopsy excluded other blistering conditions and was consistent with SJS. Allopurinol was stopped and he was admitted to hospital but required transfer to the intensive care unit after 3 days. He was treated with topical glucocorticosteroids, oral prednisone 25 mg/day, and cyclosporine. He was deemed well and discharged after 2 weeks.
Subsequently, the patient was recruited into our study of patients with allopurinol hypersensitivity9. He resides in Australia with his son and daughter. His other immediate family members, including a sister and brother, live in Malaysia. His brother also has gouty arthritis but has taken allopurinol 300 mg/day for 10 years without any reactions. The other family members do not have gouty arthritis and do not take allopurinol. The patient and his family members identify themselves as Han Chinese; his parents and grandparents were all born in China. The demographics and clinical information for the family are summarized in Table 1. HLA-B locus genotyping was performed using 4-digit, high-resolution DNA sequencing (from saliva samples) based on previous methods3. Laboratory technicians were blinded to the clinical status of the patients. The patient and his sister were HLA-B*5801-positive. By contrast, the allopurinol-tolerant brother and the son and daughter of the patient were HLA-B*5801-negative.
The cost-effectiveness of HLA-B*5801 for primary screening of potential allopurinol-induced SJS before initiation of the drug remains uncertain. Across studies thus far, all Han Chinese patients with allopurinol-induced SJS were found to be HLA-B*5801-positive3,7,9 (Table 2). However, HLA-B*5801 has a 20% carriage rate in Han Chinese3, meaning 1 in 5 Han Chinese patients starting allopurinol would be denied it if primary screening were applied. Consequently, a large number of patients would be left with limited options for gout management. Febuxostat, a newer xanthine oxidase inhibitor, is expensive and its longterm safety profile is not yet established10. Uricosuric drugs including probenecid and benzbromarone are less effective in renal impairment and may cause urate precipitation in renal tubules.
The clear influence of positivity for HLA-B*5801 on the phenotypical response to allopurinol exposure in a family is illustrated by the contrast between these 2 Han Chinese brothers who have gout. Also, by inference, the risk of developing allopurinol-induced SJS appears to be negligible in the children of the patient, whereas the sister is at high risk if allopurinol is commenced. HLA-B*5801 should be used to screen Han Chinese patients for the risk of allopurinol-induced SJS prior to initiation of the drug, particularly if a family member has experienced this serious adverse reaction. Han Chinese patients with recurrent or tophaceous gout ought to have the option of HLA typing for the B*5801 locus when considering therapeutic decisions.
Acknowledgment
The authors acknowledge the family of the patient; for HLA typing, Associate Professor Elizabeth Phillips and Dr. David Nolan, Centre for Clinical Immunology and Biomedical Statistics and Department of Clinical Immunology and Immunogenetics, Royal Perth Hospital, Perth, Western Australia, Australia; for consultation and review of manuscript, Professor Garry Graham, Department of Clinical Pharmacology and Toxicology, St. Vincent’s Hospital Sydney and Faculty of Medicine, University of New South Wales.
Footnotes
-
Supported by Australian National Health and Medical Research Council Programme Grant 568612; Lexy Davies Bequest, St. Vincent’s Hospital Sydney.
REFERENCES
- 1.
- 2.
- 3.
- 4.
- 5.
- 6.
- 7.
- 8.
- 9.
- 10.