Abacavir is a nucleoside reverse transcriptase inhibitor indicated for the treatment of HIV infection, in combination with other medications, as part of highly active antiretroviral therapy. Abacavir competitively inhibits viral reverse transcriptase, suppressing HIV’s ability to convert its RNA genome into DNA before insertion into a host cell’s genome. It is commercially available as a single agent (Ziagen) or coformulated as a fixed-dose combination with other nucleoside reverse transcriptase inhibitors, lamivudine (Epzicom/Kivexa) and lamivudine/zidovudine (Trizivir). As compared with a tenofovir-based highly active antiretroviral therapy regimen, an abacavir-based one showed a significantly shorter time to virologic failure and also a shorter time to first adverse event in patients with baseline viral loads >100,000 copies/ml18
but showed no differences in virologic failure rates in patients with lower baseline viral loads. Abacavir received significant attention after the report of an association of the drug with an increased risk of myocardial infarction19
as compared with other nucleoside reverse transcriptase inhibitors; however, subsequent analyses,20
including a meta-analysis conducted by the US Food and Drug Administration (FDA), have failed to show any such association.
Although abacavir is generally well tolerated, ~5–8% of patients experience HSR during the first 6 weeks of treatment if genetic prescreening is not performed. Symptoms of HSR increase in severity over time if the drug is continued despite the progressive symptoms. Symptoms of an HSR include at least two of the following: fever, rash, gastrointestinal symptoms (e.g., nausea, vomiting, abdominal pain), fatigue, cough, and dyspnea. Suspicion of an HSR warrants immediate discontinuation of abacavir. If the symptoms of clinically diagnosed HSR resolve after discontinuation of abacavir, drug rechallenge is contraindicated because immediate and life-threatening reactions, including anaphylaxis and even fatalities, can occur.21
In addition, an allergy to abacavir should be noted in the patient’s medical record. Previous data have shown that peripheral blood mononuclear cells from hypersensitive patients have a detectable immune response when cultured with abacavir in vitro
including increased expression of interferon-γ, tumor necrosis factor-α, and other inflammatory cytokines, showing a clear role of the immune system in mediating abacavir HSR.
Linking genetic variability to variability in drug-related phenotypes
There is substantial evidence linking the presence of the HLA-B57:01
genotype with phenotypic variability (see Supplementary Table S3
online). The application of a grading system to the evidence linking genotypic variability to phenotypic variability indicates a high quality of evidence in the majority of cases (see Supplementary Table S3
). The evidence described below and in Supplementary Table S3
provides the basis for the recommendations in and .
Treatment algorithm for clinical use of abacavir based on HLA-B*57:01 genotype. HLA-B, human leukocyte antigen B; HSR, abacavir hypersensitivity reaction.
Recommended therapeutic use of abacavir in relation to HLA-B genotype
In 2002, two independent research groups reported the initial association between HLA-B57:01
and abacavir HSR24
using cohort and case–control designs. The association was replicated in a UK population in 2004.26
However, the results were not broadly generalizable because the populations studied were predominantly white males. Nevertheless, given the strength of the observed association, some centers began implementing prospective screening of HLA-B57:01
in all HIV-positive patients to exclude HLA-B57:01
positivity before starting abacavir. This approach led to significant reductions in the incidence of HSR.27
These studies, along with the retrospective SHAPE study,30
found that HLA-B57:01
was also predictive of HSR in females and in African Americans.
Moreover, the results of PREDICT-1, the first double-blind, prospective, randomized trial of a genetic test to reduce adverse drug events, showed that genetic prescreening for HLA-B57:01
resulted in no immunologically confirmed HSR events among HLA-B57:01
-negative patients in the genetic testing arm,31
vs. a 2.7% incidence of immunologically confirmed HSR among 842 unscreened patients in the standard-of-care control arm. The results of PREDICT-1 and the existing body of evidence prompted the FDA to implement a black box warning in 2008 about the high risk of HLA-B57:01
-associated HSR. The FDA recommended that all patients be screened before being treated with abacavir (including those who had previously tolerated the drug and were being restarted on the therapy) and that abacavir not be initiated in carriers of HLA-B57:01
. Abacavir is one of a limited number of drugs for which the FDA has recommended genetic testing prior to use, and it remains one of the best examples to date of pharmacogenetics being integrated into routine medical practice.
We agree with others32
screening should be performed in all abacavir-naive individuals before initiation of abacavir-containing therapy (see ); this is consistent with the recommendations of the FDA, the US Department of Health and Human Services, and the European Medicines Agency. In abacavir-naive individuals who are HLA-B57:01
-positive, abacavir is not recommended and should be considered only under exceptional circumstances when the potential benefit, based on resistance patterns and treatment history, outweighs the risk. HLA-B57:01
genotyping is widely available in the developed world and is considered the standard of care prior to initiating abacavir. Where HLA-B57:01
genotyping is not clinically available (such as in resource-limited settings), some have advocated initiating abacavir, provided there is appropriate clinical monitoring and patient counseling about the signs and symptoms of HSR, although this remains at the clinician’s discretion.
There is some debate among clinicians regarding whether HLA-B57:01
testing is necessary in patients who had previously tolerated abacavir chronically, discontinued its use for reasons other than HSR, and are now planning to resume abacavir. The presence of HLA-B57:01
has a positive predictive value of ~50% for immunologically confirmed hypersensitivity,31
indicating that some HLA-B57:01
-positive individuals can be, and have been, safely treated with abacavir. However, we were unable to find any data to show that HLA-B57:01
-positive individuals with previous, safe exposure to abacavir had zero risk of HSR upon re-exposure. Although there are isolated case reports of previously asymptomatic patients developing a hypersensitivity-like reaction after restarting abacavir,37
there were confounding circumstances. Many of the patients had complicating concomitant illnesses that could have masked an HSR during initial abacavir therapy, and none were immunologically confirmed, making the case reports difficult to interpret. Furthermore, most of these case reports precede the availability of HLA-B57:01
genetic testing, making it impossible to determine from the published data whether there could be a risk of HSR upon re-exposure to abacavir in previously asymptomatic HLA-B57:01
In addition, there may also exist a small group of patients who have been on chronic abacavir therapy since before the introduction of HLA-B57:01
genotyping. Given that virtually all abacavir HSR events occur within the first several weeks of therapy, and that ~50% of HLA-B57:01
carriers can safely take abacavir, we were unable to find any evidence to suggest that HLA-B57:01
-positive individuals on current, long-term, uninterrupted abacavir therapy are at risk of developing HSR. Existing clinical guidelines32
have a blanket recommendation that all HLA-B57:01
-positive individuals should avoid abacavir, regardless of patient history. Although HLA-B57:01
genotyping has proven utility in significantly reducing the incidence of both clinically diagnosed and immunologically confirmed hypersensitivity7
in patients being newly considered for abacavir therapy, the connection between HLA-B57:01
genotype and risk of HSR in patients with previous asymptomatic abacavir use is less clear.
Recommendations for incidental findings
Although other variants in HLA-B are associated with autoimmune diseases and drug response phenotypes, they have not been associated with abacavir HSR.
Abacavir skin patch testing may be performed after a case of clinically diagnosed HSR to determine whether it can be immunologically confirmed. At this time, skin patch testing is an investigational procedure, and the results should be interpreted only by an experienced immunologist. More details on skin patch testing can be found in the Supplementary Materials and Methods
Potential benefits and risks for the patient
A clear benefit of HLA-B57:01
testing is that it leads to a reduction in the incidence of abacavir HSR by identifying patients at significant risk so that alternative antiretroviral therapy can be prescribed for them. Importantly, a number of effective and safe antiretrovirals are available that can be substituted for abacavir in patients carrying this risk-related allele. HLA-B57:01
’s high negative predictive value (>99%)31
shows that it is extremely effective in identifying those at risk of immunologically confirmed hypersensitivity to abacavir. A potential problem would be an error in genotyping or in reporting a genotype. This could result in high-risk patients mistakenly being given abacavir and potentially having an HSR. However, given that patients testing negative for HLA-B57:01
also have a 3% risk of developing a clinically diagnosed HSR, standard practice would include patient counseling and careful monitoring for signs and symptoms of an HSR. Given the lifelong nature of genotype results, an error in genotyping may also have a broader adverse impact on a patient’s health care if other associations with HLA-B57:01
are found in the future.
Caveats: appropriate use and/or potential misuse of genetic tests
The positive predictive value of HLA-B57:01
genotyping is ~50%, which means that a significant number of patients will be denied abacavir on the basis of their genotyping results even though they would have been able to take abacavir without experiencing an HSR. There is currently no way to know a priori
carriers are and which are not likely to experience HSRs, although new genetic risk factors may be found in the future. Given the potential seriousness of HSRs, the moderate positive predictive value is greatly outweighed by the very high negative predictive value of HLA-B57:01
is not predictive of any other adverse reactions a patient may experience while on abacavir treatment, nor does it predict whether abacavir will be effective in treating a patient’s HIV. In addition, genotyping is not a replacement for appropriate patient education and clinical monitoring for the signs and symptoms of hypersensitivity. The development of signs and symptoms of an HSR warrants that serious consideration be given to discontinuing abacavir, regardless of the HLA-B
CPIC guidelines reflect expert consensus based on clinical evidence and peer-reviewed literature available at the time they are written and are intended only to assist clinicians in decision making and to identify questions for further research. New evidence may have emerged since the time a guideline was submitted for publication. Guidelines are limited in scope and are not applicable to interventions or diseases not specifically identified. Guidelines do not account for all variations among individual patients and cannot be considered inclusive of all proper methods of care or exclusive of other treatments. It remains the responsibility of the health-care provider to determine the best course of treatment for the patient. Adherence to any guideline is voluntary, with the ultimate determination regarding its application to be made solely by the clinician and the patient. CPIC assumes no responsibility for any injury to persons or damage to property related to any use of CPIC’s guidelines, or for any errors or omissions.