Lessons from practice
Volume 204 - Issue 4
HLA-B*5801: a genetic susceptibility to allopurinol-induced DRESS
Authors: Augustine N Mugwagwa, Roy Fischer and Izwan Zailan
Med J Aust 2016; 204 (4): 159-160. || doi: 10.5694/mja15.01113
Published online: 7 March 2016
Published online: 7 March 2016
Prescribers should be aware of the importance of HLA-B*5801 testing before initiating allopurinol in Han Chinese people
Clinical record
A 40-year-old Chinese man presented to our hospital complaining of a non-itchy rash 5 weeks after he had commenced taking allopurinol 100 mg daily, prescribed by his general practitioner for gout. The rash, involving his back and limbs, began to appear 2–3 weeks after starting allopurinol, with associated fevers, arthralgia and back pain. The patient denied having any rash or pain in his mouth. He had no history of recent overseas travel, unprotected sexual intercourse or intravenous drug use. He denied experiencing any vomiting, diarrhoea or abdominal pain. The patient’s background medical history included hypothyroidism. His baseline renal function was normal. His current medications were: allopurinol, 100 mg daily; thyroxine, 100 μg daily; and paracetamol, 1 g four times daily (as required) for back pain. He had no known allergies.
On examination, the patient was febrile, with a temperature of 40°C, but was haemodynamically stable. His skin showed a morbilliform, blanching rash involving the arms, axillae and trunk, anteriorly and posteriorly. Mucous membranes were intact. Important negative findings included the absence of lymphadenopathy, jaundice, hepatomegaly, murmurs, meningism and joint tenderness. Chest and abdominal examinations were unremarkable.
The initial differential diagnoses were viral illness or a delayed drug reaction to allopurinol. Allopurinol was promptly ceased and the patient was admitted under a general medicine team for further work-up and supportive treatment. His initial treatment involved empirical antibiotics and intravenous fluids. Two days into his admission, his clinical condition deteriorated and he was admitted to the intensive care unit (ICU). Further assessment showed multiorgan involvement: reactive lymphocytosis with thrombocytopenia; rhabdomyolysis with acute kidney injury and hyperkalaemia; and profoundly elevated transaminase levels (Box). The aetiology of acute kidney injury was presumed to be multifactorial.
On admission to the ICU, intravenous hydrocortisone 100 mg 6-hourly was initiated and an urgent skin punch biopsy sample was taken. The biopsy result was consistent with drug reaction with eosinophilia and systemic symptoms (DRESS), but was not diagnostic. Results of virological, blood culture and serological investigations for atypical organisms were negative. A transthoracic echocardiogram was negative for vegetations. Results of a comprehensive autoimmune screening test were negative, except for a weakly positive speckled antinuclear antibody result, with a low titre of 1:160. Complement levels were within reference intervals. An abdominal ultrasound showed a normal-sized liver with increased echogenicity, normal biliary tract and normal renal tracts.
As the diagnosis was narrowed down to likely allopurinol-induced DRESS, the patient was tested for the human leucocyte antigen B*5801 (HLA-B*5801) allele, which returned a positive result. Management of his condition required multidisciplinary teams, with input from infectious diseases, immunology, gastroenterology, acute surgical unit, histopathology and nephrology staff. After 8 days in the ICU, where emergency haemodialysis via a femoral vascular catheter had been initiated, the patient was transferred to the renal unit, where he underwent haemodialysis via a right internal jugular permacatheter for another 2 weeks. He was then discharged home, not requiring further dialysis, after spending a total of 25 days in hospital. His hospital stay was complicated by nosocomial infections, which were treated with appropriate antimicrobial therapy.
Advances in immunogenetics are enabling us to understand why certain people are more susceptible than others to some conditions.1 There is now increasing evidence to suggest that individuals with the HLA-B*5801 allele are more susceptible to allopurinol-induced DRESS.2-4 This allele is most common in people of Han Chinese origin.2-4 About 4% of the Australian population, or 866 000 people, identify as being of Chinese descent, and this number has been increasing since 2003.5
Allopurinol is a xanthine oxidase inhibitor. It is currently indicated as a first-line urate-lowering therapy in gout management. American College of Rheumatology (ACR) guidelines for managing gout suggest that pharmacological urate-lowering therapy should only be initiated in patients with an established diagnosis of gouty arthritis with: tophus or tophi identified by clinical examination or imaging study, two or more attacks per year, chronic kidney disease (Stage 2 or worse) or past urolithiasis.6
DRESS is believed to be a type IV hypersensitivity reaction4 and typically occurs 2–8 weeks after exposure to the offending drug. The name can be a misnomer, as about 50% of patients with DRESS, including our patient, demonstrate normal eosinophil counts. Clinically, DRESS results in multiorgan involvement, with varying presentations depending on the organ systems involved.
No cost-effectiveness studies determining the feasibility of HLA-B*5801 genotyping before prescription of allopurinol in Australia have yet been carried out. However, current literature from some Asian countries suggests that prior genotyping for HLA-B*5801 in Han Chinese individuals is cost-effective, considering the incidence of allopurinol-induced DRESS in this population.7 This is in keeping with the recommendations of the ACR guidelines.6 The Australian Medicines Handbook also suggests considering HLA-B*5801 testing in patients of Asian origin who are starting allopurinol; however, MIMS Online and the Therapeutic Guidelines do not suggest such testing.8,9 HLA-B*5801 genotyping is readily available in most laboratories in Australia and the average turnaround time for the test is 2–5 days. First-degree relatives of individuals who test positive for HLA-B*5801 are also advised to avoid taking allopurinol because of the genetic link.
For people who are positive for HLA-B*5801, the ACR guidelines recommend using either alternative urate-lowering agents, such as febuxostat or probenecid, or lower initiating doses of allopurinol, with vigilant monitoring for the development of adverse reactions 2–8 weeks after commencing therapy.6 There is a lack of consensus on a safe lower starting dose for allopurinol.6 Febuxostat and probenecid are both covered under the Pharmaceuticals Benefits Scheme in Australia, but are more expensive than allopurinol.10
Due to its propensity for multiorgan involvement, DRESS is a potentially fatal condition or may have long-term adverse effects on affected individuals. Subsequently, treatment of this condition represents a large financial burden on the health care system. The mainstay of management of DRESS involves prompt withdrawal of the offending drug, supportive treatment of the manifesting complications and directed treatment for more serious complications.4Lessons from practice
Allopurinol can cause drug reaction with eosinophilia and systemic symptoms (DRESS). This is more common in individuals who are positive for the human leucocyte antigen B*5801 (HLA-B*5801) allele.
DRESS can be fatal; therefore, if a patient develops signs or symptoms of DRESS (including a rash), the suspected offending drug should be ceased promptly.
As HLA-B*5801 testing is readily accessible across Australia and is potentially cost-effective, testing should be strongly considered before initiating allopurinol in Han Chinese individuals, in whom the allele is most common.
A normal blood eosinophil count does not exclude the diagnosis of DRESS.
Box –
Test |
Result by event and date |
Reference interval |
|||||||||||||
Admission, 8 Aug 2015 |
Day 1 in ICU, 10 Aug 2015 |
Day 1 in renal unit, 18 Aug 2015 |
Discharge, 2 Sep 2015 |
Day 6 at home, 8 Sep 2015 |
|||||||||||
Haemoglobin (g/L) |
131 |
135 |
111 |
72 |
81 |
135–175 |
|||||||||
Platelets (× 109) |
120 |
98 |
59 |
91 |
135 |
150–450 |
|||||||||
White cells (× 109/L) |
9.62 |
18.3 |
17.0 |
6.08 |
17.0 |
4.00–11.0 |
|||||||||
Neutrophils (× 109/L) |
6.16 |
13.71 |
11.69 |
4.18 |
11.03 |
1.80–7.50 |
|||||||||
Eosinophils (× 109/L) |
0 |
0 |
0.22 |
0 |
0 |
0.02–0.50 |
|||||||||
Lymphocytes (× 109/L) |
3.17 |
3.47 |
2.21 |
1.24 |
3.05 |
1.50–3.50 |
|||||||||
Potassium (mmol/L) |
4.5 |
6.7 |
5.3 |
4.0 |
4.6 |
3.5–4.9 |
|||||||||
Sodium (mmol/L) |
129 |
122 |
134 |
144 |
143 |
137–145 |
|||||||||
Urea (mmol/L) |
5.5 |
19.6 |
16.9 |
38.0 |
16.2 |
2.7–8.0 |
|||||||||
Creatinine (μmol/L) |
92 |
311 |
318 |
345 |
138 |
50–120 |
|||||||||
eGFR (mL/min/1.73 m2) |
89 |
21 |
20 |
18 |
55 |
90–120 |
|||||||||
Creatine kinase (U/L) |
not tested |
355 912 |
6831 |
268 |
not tested |
< 250 |
|||||||||
GGT (U/L) |
464 |
475 |
1169 |
699 |
1086 |
< 60 |
|||||||||
ALP (U/L) |
178 |
182 |
515 |
396 |
447 |
30–110 |
|||||||||
ALT (U/L) |
899 |
3118 |
633 |
277 |
213 |
< 55 |
|||||||||
AST (U/L) |
1257 |
4930 |
367 |
53 |
76 |
< 45 |
|||||||||
Bilirubin (μmol/L) |
14 |
45 |
172 |
45 |
40 |
2–24 |
|||||||||
Albumin (g/L) |
39 |
26 |
19 |
21 |
26 |
34–48 |
|||||||||
LDH (U/L) |
2081 |
17 795 |
569 |
460 |
643 |
110–230 |
|||||||||
INR |
1.3 |
1.4 |
1.0 |
1.1 |
0.9 |
0.9–1.2 |
|||||||||
ALP = alkaline phosphatase. ALT = alanine aminotransferase. AST = aspartate aminotransferase. eGFR = estimated glomerular filtration rate. GGT = γ-glutamyltransferase. ICU = intensive care unit. INR = international normalised ratio. LDH = lactate dehydrogenase. | |||||||||||||||
Competing interests
No relevant disclosures.
References
- Hoffman R, Benz EJ, Silberstein LE, et al. Haematology: basic principles and practice. 6th ed. Philadelphia: Elsevier Saunders, 2013.
- Lee MH, Stocker SL, Anderson J, et al. Initiating allopurinol therapy: do we need to know the patient’s HLA status? Intern Med J 2011; 42: 411-416.
- Hung SI, Chung WH, Liou LB, et al. HLA-B*5801 allele as a genetic marker for severe cutaneous adverse reactions caused by allopurinol. Proc Natl Acad Sci U S A 2005; 102: 4134-4139.
- Hershfield MS, Callaghan JT, Tassaneeyakul W, et al. Clinical Pharmacogenetics Implementation Consortium guidelines for human leukocyte antigen-B genotype and allopurinol dosing. Clin Pharmacol Ther 2013; 93: 153-158.
- Australian Bureau of Statistics. 2011 Census of Population and Housing: basic community profile: Australia. Canberra: ABS, 2013. http://www.censusdata.abs.gov.au/CensusOutput/copsub.NSF/All%20docs%20by%20catNo/2011∼Community%20Profile∼0/$File/BCP_0.zip?OpenElement (accessed Sep 2015).
- Khanna D, Fitzgerald J, Khanna P, et al. 2012 American College of Rheumatology Guidelines for Management of Gout Part I: systematic nonpharmacologic and pharmacologic therapeutic approaches to hyperuricemia. Arthritis Care Res (Hoboken) 2012; 64: 1431-1446.
- Saokaew S, Tassaneeyakul W, Maenthaisong R, et al. Cost-effectiveness analysis of HLA-B*5801 testing in preventing allopurinol-induced SJS/TEN in Thai population. PLoS One 2014; 9: e94294.
- Australian Medicines Handbook. Adelaide: Australian Medicines Handbook Pty Ltd, 2015.
- Rheumatology Expert Group. Therapeutic guidelines: rheumatology. Version 2. Melbourne: Therapeutic Guidelines Limited, 2010.
- Australian Government Department of Health. Pharmaceutical Benefits Scheme [website]. http://www.pbs.gov.au (accessed Sep 2015).