Case 1: Discussion

Question | Discussion | References | CME Credit

Abacavir hypersensitivity reaction is a multi-organ systemic illness that occurs in approximately 5-8% of HIV-infected patients who initiate therapy with abacavir [1]. The abacavir hypersensitivity reaction resembles a delayed hypersensitivity reaction and it can cause life-threatening complications if abacavir is continued despite progressive symptoms [1]. In addition, among persons who have experienced the abacavir hypersensitivity reaction, subsequent rechallenge with abacavir following discontinuation can cause an immediate and potentially fatal reaction. Abacavir is available as a single drug (Ziagen), as part of the Epzicom combination pill (abacavir plus lamivudine), and as part of the Trizivir combination pill (zidovudine plus lamivudine plus abacavir).

Incidence and Timing of Hypersensitivity Reaction

The incidence of abacavir hypersensitivity reported in clinical trials involving predominantly Caucasians is approximately 6-8% (range 2-9%) [1,2]. In contrast, the risk among African Americans is significantly lower and is approximately 2-3%. More than 90% of abacavir hypersensitivity reactions occur during the first 6 weeks of therapy, with a median time to onset of symptoms of 8-11 days (range 1-318 days) [3,4]. Failure to recognize the reaction has been associated with fatalities when abacavir was continued despite progressive symptoms. In multivariate analysis, the risk of developing hypersensitivity was almost 40% lower for individuals of African descent and for patients with prior antiretroviral experience [5]. There has been no clear correlation of the risk of abacavir hypersensitivity with the CD4 cell count or HIV-1 RNA level.

HLA-B*5701 and Hypersensitivity Reaction

In 2002, investigators first described an association of abacavir hypersensitivity with the genetic markers HLA-B*5701, HLA-DR7, and HLA-DQ3, with a sensitivity ranging from 33-78% in cohorts composed predominantly of Caucasian men; the presence of all three of the markers HLA-B*5701, HLA-DR7, and HLA-DQ3 had a positive predictive value for abacavir hypersensitivity of 100% and a negative predictive value of 97% (Figure 1) [6]. Subsequent studies have shown HLA-B*5701 alone serves as an excellent predictor of abacavir hypersensitivity syndrome [7-10]. In one study, 260 abacavir-naive patients were prospectively sorted into low-risk and high-risk hypersensitivity groups based on HLA-B*5701 status and no cases of abacavir hypersensitivity were identified in the 148 HLAB*5701-negative patients treated with abacavir [8]. In the PREDICT-1 trial, a multinational study that involved 1956 patients from 19 countries, patients were randomized to either prospective HLA-B*5701 screening (patients with positive HLA-B*5701 did not receive abacavir) or to the standard of care (no screening and all patients received abacavir) [9]. The overall prevalence of HLA-B*5701 in this study was 5.6%, but most of the patients were caucasian [9]. Among the 802 individuals who screened negative for HLA-B*5701, 3.4% had a clinical diagnosis of abacavir hypersensitivity reaction, but 0% had immunologically confirmed abacavir hypersensitivity using patch testing [9]. The SHAPE study confirmed the lower risk of abacavir hypersensitivity syndrome in blacks and validated the usefullness of HLA-B*5701 screening in both white and black populations [10].

Clinical Manifestations

The abacavir hypersensitivity reaction as a multi-organ process manifested by a sign or symptom from at least 2 of the following groups: (1) fever, (2) rash, (3) gastrointestinal, (4) constitutional, or (5) respiratory [1]. In contrast to the more prominent rash that can develop with sulfonamides or the non-nucleoside reverse transcriptase inhibitors (NNRTIs), the rash associated with abacavir is often mild and is usually not the predominant symptom. With abacavir hypersensitivity, constitutional symptoms (fatigue, myalgias, and generalized malaise) are often present, and less frequently respiratory symptoms, such as dyspnea, cough, and pharyngitis are prominent. The constellation of symptoms associated with abacavir hypersensitivity are often non-specific, frequently mimicking those of viral illnesses such as influenza, and thus potentially confound the recognition of abacavir hypersensitivity [11]. In order to identify features that help delineate these overlapping entities, investigators compared 15 patients with abacavir hypersensitivity with 30 patients who had culture proven influenza A and no abacavir exposure; the presence of gastrointestinal symptoms was strongly associated with abacavir hypersensitivity, whereas cough without gastrointestinal symptoms was much more likely to be caused by influenza (Figure 2) [11]. In general, the probability of abacavir hypersensitivity increases when a greater number of manifestations are present (Figure 3). Other important diagnostic clues, as seen in the patient described in this case, are the accentuation of symptoms within hours of taking each dose of abacavir and the progressive worsening of symptoms with each subsequent dose. Abnormal laboratory findings are nonspecific, but may include leukopenia, anemia, thrombocytopenia, as well as elevations in transaminases, BUN, creatinine, and LDH. Eosinophilia is usually not present [3,4]. In most patients, stopping abacavir will promptly result in resolution of the abacavir hypersensitivity reaction.

Rechallenge with Abacavir

In patients who have experienced abacavir hypersensitivity, rechallenge with abacavir has been associated with severe, life-threatening reactions manifested by hypotension, renal insufficiency, and bronchoconstriction [12]. During the early drug development of abacavir, patients with abacavir hypersensitivity were rechallenged, and immediate life-threatening reactions occurred in approximately 20% of these individuals [3]. Now, medical providers know never to rechallenge an individual who develops an abacavir hypersensitivity reaction. Although a life-threatening reaction has also been reported after restarting abacavir in the absence of preceding hypersensitivity, this appears to be an extremely rare event [13]. Moreover, subsequent studies have found no increase in hypersensitivity risk after abacavir treatment interruptions in the absence of a preceding hypersensitivity reaction [14,15].

HLA-B*5701 Screening Recommendations

The most important aspect of clinical care related to the abacavir hypersensitivity reaction is the use of effective prevention measures. Accordingly, in December 2007, the Department of Health and Human Services (DHHS) Panel on Antiretroviral Guidelines for Adults and Adolescents recommended screening for HLA-B*5701 before starting any patient on an abacavir-containing regimen [10]. This panel has also recommended that any patient with a positive test for HLA-B*5701 should not receive abacavir. Further, a positive test for HLA-B*5701 should prompt documentation of an abacavir allergy in the patient's chart. Although the HLA-B*5701 is an excellent screening tool for abacavir hypersensitivity, a negative test does not absolutely rule out the possibility a patient could develop an abacavir hypersensitivity reaction. In the setting where HLA-B*5701 screening is not readily available, the panel suggests that it is reasonable to initiate therapy with abacavir if the patient will receive appropriate clinical counseling and monitoring for any signs of abacavir hypersensitivity reaction.

Patch Test Screening

The epicutaneous abacavir patch test has been studied as a means to immunologically confirm an abacavir hypersensitivity reaction [9,10]. The test purportedly addresses the problem of false-positive clinical diagnoses of abacavir hypersensitivityâ€”a problem considered significant because of the overly sensitive criteria used to diagnose abacavir hypersensitivity reaction [10]. The patch test is performed by applying a dilute, non-irritating concentration of abacavir (in a petroleum base) to the skin surface, typically the mid-back region [16]. The skin is then evaluated for redness and swelling 48-72 hours later. One group of investigators examined the durability of abacavir patch testing in seven patients who had a remote history (mean 7 months) of a positive patch test and all seven patients had robust responses to the patch test [17]. Given that a positive patch test should only occur in persons who have previously taken abacavir and developed a hypersensitivity reaction, the patch test is not an appropriate screening test for abacavir hypersensitivity in persons who have never taken abacavir. In addition, at this point, the DHHS Panel on Antiretroviral Guidelines for Adults and Adolescents does not recommend using the abacavir patch test as a clinical tool to immunologically confirm an abacavir hypersensitivity reaction [10].

Management of A Potential Hypersensitivity Reaction

Clear communication with patients and vigilance in recognizing the constellation of symptoms associated with abacavir hypersensitivity reactions are key to effective management. Counseling regarding the hypersensitivity reaction prior to starting abacavir is critical, and patients should know when and where to call if they have any symptoms that suggest an abacavir hypersensitivity reaction. The most problematic decision for clinicians occurs when a patient develops mild symptoms that may suggest abacavir hypersensitivity. When it is not clear whether the patient is experiencing an abacavir hypersensitivity reaction, some experts have advised that it is reasonable to have the patient take another dose of the drug and report back within several hours to determine whether the symptoms have progressed. If the symptoms worsen, abacavir should be discontinued. If the symptoms do not worsen, it is reasonable to continue abacavir with extremely close follow-up. The clinician should permanently discontinue abacavir if they cannot rule out the hypersensitivity reaction.

In the case presented, the patient simultaneously started trimethoprim-sulfamethoxazole (Bactrim, Septra) and an antiretroviral regimen that included abacavir. The patient should have been tested for HLA-B*5701 before starting abacavir. In addition, from a practical perspective, it would have been preferable to have started the trimethoprim-sulfamethoxazole at least several weeks prior to starting abacavir. Otherwise, toxicity from a new medication other than abacavir could confound the ability to recognize an abacavir hypersensitivity reaction or alternatively could lead to the mistaken diagnosis of abacavir hypersensitivity reaction when the reaction was caused by another medication. The potential for confusion can be especially problematic when abacavir and an NNRTI are started simultaneously. In such cases, staggered introduction of these medications may be unwise for resistance reasons.

Although much emphasis is placed on the importance of recognizing abacavir hypersensitivity reactions, it is equally important not to overdiagnose them. Once abacavir has been discontinued because of presumed hypersensitivity, the patient has lost abacavir as a potential antiretroviral agent. When abacavir is discontinued prematurely or without adequate assessment of symptoms, therapeutic options may be lost. It is important that clinicians carefully assess patients with potential hypersensitivity reactions before stopping this drug.

If the hypersensitivity reaction occurs, abacavir should be discontinued and supportive care provided. After withdrawal of abacavir, symptoms usually resolve within a few days. In some patients symptoms continue to accelerate for several days after discontinuing abacavir. Corticosteroids have no proven benefit in the management or prevention of the hypersensitivity reaction [18]. Because of the potential for life-threatening rechallenge reactions, individuals who experience hypersensitivity to abacavir must be counseled never to take this drug again. In addition, both medical providers and patients must be aware that the patient cannot use abacavir in the form of Ziagen, Trizivir, or Epzicom.

[Back to Case 1 Question | See Case 1 References]

CME Credit | Back to Top

Figure 1
Figure 2
Figure 3