Abacavir hypersensitivity screening not a substitute for 'clinical vigilance' say investigators at UK's largest HIV clinic

HLA-B*5701 testing to predict which patients have a risk of developing an abacavir hypersensitivity reaction is not a substitute for clinical vigilance, a prospective UK study published in the November 30^th edition of AIDS suggests.

Investigators from the Chelsea and Westminster Hospital in London, the largest HIV treatment centre in Europe, found that two treatment-experienced patients who tested negative for HLA-B*5701, the gene associated with a risk of allergy to abacavir, and who subsequently started treatment with the drug, developed a suspected hypersensitivity reaction, with one patient having a skin patch test which strongly suggested an allergic reaction to abacavir.

In addition, two treatment-naïve patients with positive HLA-B*5701 results started therapy with abacavir before the results of their screen were known and went on to develop a hypersensitivity reaction.

The main reason why patients stop treatment with abacavir (Ziagen, also in the combination pills Kivexa and Trizivir) is because of a suspected hypersensitivity reaction. Recommencing therapy with abacavir after its use has been discontinued because of a hypersensitivity reaction can be fatal.

Historically, approximately 5% - 8% of patients who started abacavir treatment developed a suspected hypersensitivity reaction and discontinued therapy with the drug. Researchers had noted that patients of northern European ethnicity had the greatest risk of experiencing such a reaction and it was then shown that the presence of the HL- B*5701 gene was a strong predictor of hypersensitivity to the drug.

Prospective cohort studies in Australia and the UK have shown a statistically significant decrease in the incidence of abacavir hypersensitivity since the introduction of HLA-B*5701 screening of patients before they initiate treatment with the drug.

Investigators at the Chelsea and Westminster Hospital wished to determine the impact of HLA-B*5701 screening on the incidence of abacavir hypersensitivity in their patients.

Routine HLA-B*5701 screening was introduced in the clinic in 2005 and all treatment-naïve, and treatment-experienced patients, who are about to start or change anti-HIV therapy have the test.

The rate of abacavir hypersensitivity since the introduction of screening was compared to that observed in the twelve month period before such screening was introduced.

A total of 739 HLA B*5701 screens were performed between August 2005 and July 2006. Of these tests, 735 were successful and yielded a result. A total of 54 tests (7%) were positive, including eleven of the 111 tests performed on women (9.9%) and 43 of the 624 tests in men (6.9%). This meant that there was no statistically significant difference in the incidence of a positive test result in men and women (p = 0.35).

Although an abacavir hypersensitivity reaction has historically been associated with patients of white ethnicity, the Chelsea and Westminster investigators found that 7.6% of white patients had a positive HLA-B*5701 test compared to 9% of patients of black ethnicity, and again this difference was not statistically significant.

Of the 681 patients with a negative HLA-B*5701 result, 285 were treatment-naïve. A total of 122 of these patients started antiretroviral therapy, 47 with an abacavir-containing regimen. None developed a hypersensitivity reaction to the drug.

A total of 396 treatment-experienced patients had a negative HLA-B*5701 screen prior to changing treatment and 151 of these switched to treatment with abacavir. The investigators established that eight (5.3%) patients subsequently stopped abacavir treatment, four (2.6%) because of a suspected abacavir hypersensitivity reaction.

Skin-patch tests were performed on two of these patients and one had a strongly positive result along with symptoms suggestive of a hypersensitivity reaction.

The investigators noted that two treatment-naïve patients positive for HLA-B*5701 started treatment with abacavir. Therapy was initiated before the results of their HLA screen were known. Both patients had a hypersensitivity reaction.

In the twelve months before HLA-B*5701 screening became routine at the clinic, ten of the 144 (7.5%) patients starting treatment with abacavir experienced a suspected hypersensitivity reaction. In the twelve month period after screening was introduced, six of 207 (3%) patients who initiated therapy with the drug had a suspected allergic reaction to it. The statistical difference in the incidence of suspected abacavir hypersensitivity between these two period was not statistically significant (p = 0.10).

But when the investigators excluded the two patients who had a positive HLA-B*5701 test result, and who started abacavir therapy, and confined analysis to patients with a negative test result, then the difference between the two time periods was statistically significant, 7.5% vs. 2% (p = 0.03).

The investigators conclude, “our results highlight the need to review results carefully before commencing therapy, and that it is essential to maintain clinical vigilance even in the presence of a negative HLA-B*5701 result.”