Objective: Carriage of human leukocyte antigen (HLA)-B*57:01 allele increases the risk of abacavir hypersensitivity reaction. Therefore, since 2008 HIV treatment guidelines recommend HLA-B*57:01 screening before abacavir administration, greatly reducing hypersensitivity reaction rate. However, clinically suspected abacavir-related hypersensitivity reactions are described in allele non-carriers. Major aim of this study was to evaluate the relationship between HLA-B*57:01 pattern and abacavir-related hypersensitivity reaction, focusing on hypersensitivity reaction prevalence in allele non-carriers.
Methods: We included all outpatients aged >18 years old with HIV infection and known HLA-B*57:01 pattern, followed at our Department from January 2000 until December 2017. Patients were divided according to HLA-B*57:01 pattern and first antiretroviral treatment prescribed (containing or not abacavir) as follows: HLA-B*57:01 allele carriers treated with abacavir and HLA-B*57:01 allele non-carriers treated with abacavir. We considered all adverse events reported during first abacavir administration, differentiating between confirmed hypersensitivity reactions and non-hypersensitivity reactions, according to abacavir hypersensitivity reaction definition included in the abacavir EU Summary of Product Characteristics and the US Prescribing Information.
Results: A total of 3144 patients had a known HLA-B*57:01 pattern. About 5.4% of them showed allele polymorphism; Caucasian ethnicity was the most represented. In this cohort, 1801 patients were treated with a first abacavir-containing regimen (98.2% of them was represented by allele non-carriers). 191 out of 1801 patients discontinued abacavir because of toxicity/intolerance; among them 107 described adverse events fulfilled the criteria of confirmed abacavir hypersensitivity reaction (22/32 allele-positive patients and 85/1769 allele-negative patients). After having experienced a confirmed abacavir hypersensitivity reaction, abacavir was re-administered to eight HLA-B*57:01 negative patients. Seven of them re-experienced a syndrome consistent with hypersensitivity reaction, finally leading to drug discontinuation. Overall, no fatal reactions were described.
Conclusion: Not all abacavir-related side effects occur as a result of classic HLA-B*57:01-mediated hypersensitivity reaction, as they can develop irrespective of HLA-B*57:01 status. Clinical vigilance must be an essential part of the management of individuals starting abacavir, at any time during treatment. In a 'real-life' setting, clinical diagnosis of suspected abacavir hypersensitivity reaction in allele non-carriers remains crucial for further clinical decision making.