Utility of patch testing in patients with hypersensitivity syndromes associated with abacavir.

Abstract

A diagnostic test would be useful to help eliminate the associated morbidity and mortality after full-dose rechallenge with abacavir in patients with suspected hypersensitivity. We describe seven patients with syndromes compatible with abacavir hypersensitivity who had positive patch tests. Immunohistochemistry on skin biopsies from the acute prospectively identified patients with rash (n = 3) matched those from positive-patch patients (n = 7), suggesting an identical pathophysiological process. No patients developed systemic symptoms or signs during patch testing. Abacavir is a nucleoside analogue used in combination with other antiretroviral drugs for the treatment of HIV. Approximately 2–5% of patients starting abacavir will experience a hypersensitivity syndrome, most commonly characterized by combinations of fever, rash, malaise, nausea, vomiting and diarrhoea. Severe reactions such as shock and even death have been described with full dose rechallenge after the occurrence of the hypersensitivity syndrome [1]. Patch testing is an ‘in-vivo’ test that involves the application of dilute, non-irritating concentrations of the substance or drug in a vehicle such as petrolatum to the surface of the skin. There is significant experience with patch testing for the diagnosis of T cell-mediated processes such as allergic contact dermatitis. Patch testing is simple and inexpensive to perform. Currently there is limited information as to the use of patch testing for other potentially cell-mediated processes such as drug reactions [2]. The rationale for use in these reactions is supported by the potential metabolism of the parent drug to reactive metabolite in skin and the presence of resident CD8 cells from skin biopsies of patients with acute drug-induced hypersensitivity syndromes [3]. We describe seven patients who developed positive patch tests to abacavir within 4 months of a presumed abacavir hypersensitivity reaction (Table 1, Fig. 1). One patient (case 6) was identified 33 months after the initial reaction. Four patients were identified prospectively at the time of the reaction, and three were identified retrospectively after the initial reaction. All patients met a minimum case definition of probable abacavir hypersensitivity syndrome, defined as either having at least two intensifying symptoms of rash, fever, gastrointestinal complaints, headache and resolution within 24 h of stopping the drug without an alternative explanation or a positive rechallenge to abacavir. Rash occurred in three out of four prospectively identified patients as part of the acute reaction, with immunohistochemistry documented (Table 1). Patients and HIV-negative controls had one patch panel of 1% abacavir and 10% abacavir in a petrolatum base applied to the mid-back. All patches were taken off at 48 h. Reading took place at 48 and 96 h in cases 1 to 3. Cases 4 to 7 and four additional HIV-infected controls tolerating or known to have tolerated abacavir also had 0.1, 5, 15 and 25% abacavir added to the patch panel, with additional readings at 1, 24, 48 and 96 h. Two HIV-negative controls with no previous abacavir exposure and five HIV-positive controls with previous abacavir exposure matched for age, sex, race, CD4 cell count and viral load had negative patch tests at 48 and 96 h.Fig. 1.: Patch testing at 24 h illustrating dose reponse (case 5).Table 1: Clinical summary of patch test-positive patients.Patch testing was positive in seven patients with probable abacavir hypersensitivity and negative in seven controls. This may be an extremely useful diagnostic modality for identifying patients with true abacavir reactions. HIV patients are often started on multiple concurrent medications, making the implication of a single drug in a reaction difficult. The clinical diagnosis of abacavir hypersensitivity syndrome was further strengthened in our six out of seven patch test-positive patients by subsequent rechallenge and tolerance of of of other antiretroviral drugs that they were taking at the time of the initial reaction. The pathological concordance between the two biopsies and patch testing further strengthens this, and sheds light on the pathophysiology of the acute skin reaction and positive patch tests. The immunohistochemistry on the acute skin rash biopsies and patch biopsies are in fact identical to previously presented data on skin biopsies from patients with acute abacavir hypersensitivity syndrome [4]. The absence of B cell markers and the presence of HLA-DR in all of the biopsies suggests a primarily cell-mediated, T helper cell type 1 response. These preliminary results suggest that an inexpensive and safe procedure such as patch testing may be a useful adjunct to the clinical diagnosis of abacavir hypersensitivity syndrome. Cases 4 to 7 illustrates that patch testing can become positive within 24 h of applying the patch. It should be emphasized that this test is currently a research tool, and a negative patch test at this time cannot be interpreted as grounds for full-dose unsupervised rechallenge with abacavir. Case 6 illustrates that patch tests can be positive remote from the initial exposure. However, the exact time course of when the test becomes positive in relation to the hypersensitivity syndrome and the expected duration of positivity in these patients is currently unknown and is the subject of further investigation. Acknowledgements The authors would like to acknowledge GlaxoSmith Kline US, Research Triangle, NC, USA, and the Canadian Infectious Disease Society for their support of this study; Dr Rodney Miller, Propath Laboratories, Dallas TX, USA, who processed the skin biopsies; and Drs A. Rachlis, S. Walmsley and C. Kovacs for their contribution of patients.