Pharmacogenetics predict abacavir hypersensitivity

In a demonstration of the potential clinical applications of recent progress in pharmacogenetics, two research teams have separately identified a genetic marker that may predict which HIV-infected patients are at risk of a hypersensitivity reaction to abacavir [Ziagen].

Patients with the HLA-B*5701 allele were more than 100 times more likely to experience a hypersensitivity reaction when treated with abacavir, a potent HIV-1 nucleoside reverse transcriptase inhibitor, according to lead investigator Dr. Simon Mallal, of the Royal Perth Hospital in Australia, who presented the first report at the 9th Conference on Retroviruses and Opportunistic Infections. Dr. Mallal’s study is also published in the March 2nd issue of The Lancet.

The Australian researchers evaluated data for 581 participants in the Western Australian HIV Cohort Study. Of these patients, 200 were treated with abacavir.

In general, “about 5% of patients treated with abacavir develop a severe, potentially life-threatening hypersensitivity reaction,” Dr. Mallal explained to conference participants. The prevalence ranges from 0% to 14%, and symptoms usually develop within the first six weeks of treatment. His team hypothesized that there may be a genetic basis for this hypersensitivity.

Eighteen of the 200 Australian patients developed abacavir hypersensitivity. The results of MHC region typing indicated that HLA-B*5701 was present in 14 of these 18 patients (78%). However, four of the 167 patients (2.3%) considered to be abacavir-tolerant also had this allele.

Furthermore, the HLA-DR7 and HLA-DQ3 combination was identified in 13 (72%) hypersensitive and in five (3%) abacavir-tolerant patients. Overall, 13 (72%) hypersensitive patients had the HLA-B*5701, HLA-DR7, HLA-DQ3 combination.

“All patients with the full haplotype had abacavir hypersensitivity,” Dr. Mallal told conference participants. He estimates that avoiding abacavir therapy in patients with this haplotype could significantly reduce the development of this potentially life-threatening drug reaction from 9% to 2.5% in his patient population.

Based on these findings, “we have now changed our clinical practice,” he continued. At his clinic, they now plan to withhold abacavir from patients with the HLA-B*5701, -DR7, -DQ3 haplotype. However, he cautioned that the current findings are not yet generally applicable. The patients he studied were all Caucasians, and the findings may not apply to patients of other races or to patients having a different genetic make-up.

Dr. Mallal is also concerned that the findings, if misinterpreted, “may do more harm than good.” Specifically, he cautioned that this is not yet a diagnostic test to confirm hypersensitivity, as testing for the haplotype was only 72% sensitive.

Also, “four out of the 18 hypersensitive patients developed classical abacavir hypersensitivity, but did not have the haplotypes.” Therefore, anyone who develops signs of abacavir sensitivity should stop the drug immediately.

“A negative test for the haplotype must not be used as evidence to rechallenge a patient” who shows signs of abacavir hypersensitivity, he added.

The second presentation, by Dr. Seth Hetherington, of GlaxoSmithKline of Research Triangle Park, North Carolina, involved a retrospective case-control study. This proof-of-principle study evaluated HIV-infected patients for polymorphisms associated with abacavir hypersensitivity. Data were available for 200 patients from whom allele frequencies for 114 candidate gene markers were analysed.

The cohort consisted primarily of men (92%) and caucasians (74%). Dr. Hetherington’s group found that 85 subjects developed abacavir sensitivity, while 115 (controls) did not.

Several polymorphisms within the HLA region were identified that were associated with susceptibility to hypersensitivity. Of the markers evaluated thus far, they found that the HLA-B57 gene allele was most highly associated with the reaction. In the 197 subjects with available data, HLA-B57 was present in 39 of 84 patients who developed hypersensitivity reactions (46%) compared with 4 of 113 controls (3.5%).

“These findings demonstrate the potential of pharmacogenetics in selecting drug treatment and improving treatment response,” Dr. Hetherington told conference participants.

He too cautioned that these findings may not necessarily apply to other patient populations. For example, there were a few cases of abacavir hypersensitivity in black patients who did not have the marker. Also, other genetic factors cannot be ruled out as causing or contributing to hypersensitivity.

Abacavir sensitivity usually starts with severe GI symptoms, followed by fever, rash, shortness of breath and myalgia, Dr. Mallal said. About 50% of hypersensitive patients begin to get symptoms by day 11.

Discontinuation of the drug reverses symptoms. However, without discontinuation, symptoms increase over time and the condition can be fatal. Rechallenge with abacavir can be especially dangerous, Dr. Mallal pointed out, because second reactions are usually more severe and more rapid.


S Hetherington and others (GlaxoSmithKline, Research Triangle Park, NC). HLA-B57 and TNF-alpha Variants Associated with Hypersensitivity Reactions to Abacavir among HIV-1 Positive Subjects. Abstract 92 (oral).

S Mallal and others (Center for Clinical Immunology and Biomedical Statistics, Royal Perth Hospital; Royal Perth Hospital; and University of Western Australia). The Presense of HLA-B 5701, -DRB1 and -DQ3 is Highly Predictive of Hypersensitivity to the HIV Reverse Transcriptase Inhibitor Abacavir. Abstract 91 (oral).

S Mallal and others. The Presense of HLA-B 5701, -DRB1 and -DQ3 is Highly Predictive of Hypersensitivity to the HIV reverse Transcriptase Inhibitor Abacavir. The Lancet. 359: 727-732. March 2, 2002.

Links to other websites are current at date of posting but not maintained.