THE SOUTHERN AFRICAN JOURNAL OF HIV MEDICINE december 2009

Abacavir (ABC), a nucleoside reverse transcriptase in-
hibitor (NRTI), combined with lamivudine (3TC), has a
better short- and long-term outcome than 3TC com-
bined with zidovudine (ZDV) as first-line HIV therapy.1,2
In addition, children failing ABC/3TC-based first-line
therapy do not select thymidine NRTI-related muta-
tions, allowing for better choice in second-line thera-
py.2 With current first-line options, both first-line (sta-
vudine (d4T)) and second-line therapy (ZDV) include a
thymidine-based NRTI, thus compromising second-line
regimens.3-5 In well-selected children, ABC is also an
important drug in second-line and salvage therapy.6

Of all the NRTIs, ABC is associated with the lowest rate
of mitochondrial dysfunction. Types of dysfunction in-
clude lactic acidosis, peripheral neuropathy and lipo-at-
rophy. Substitution of d4T for ABC improves mitochon-
drial indices and reduces adipocyte apoptosis.7 In adults,
switching from d4T to ABC was superior to switching
from d4T to ZDV.8 In older children, once-daily use of
ABC has also been shown to be effective, thereby fa-
cilitating adherence and improving patient satisfaction,
particularly when all drugs are given once daily.9,10

Despite these advantages, ABC is rarely used as part of
first-line therapy in South Africa owing to cost. Tenofo-
vir, commonly used in adults experiencing NRTI adverse
events, is not licensed for children. With large cohorts
of children now on antiretroviral therapy for long peri-
ods of time, increased use of ABC is likely as NRTI ad-
verse events become apparent. Currently, the National
Department of Health permits using ABC when there
have been adverse events related to other NRTIs.

Of concern is the severe and life-threatening hypersen-
sitivity reaction (HSR) that occasionally occurs, neces-
sitating permanent discontinuation of ABC.

ABC HSR has been reported in adults and children. The
prevalence in clinical trials varies.11 In a European trial
of first-line therapy, where 92 children were initiated
on ABC, 4 (4.3%) terminated ABC for adverse reactions,
1 case (1%) being considered an HSR. There is clear
heterogeneity in risk according to ethnic groups, with
Caucasians at higher risk and a 40% reduction in risk
for African Americans. In the ARROW study of >1 200
HIV-infected children in Uganda and Zimbabwe, HSR
was reported in 0.2% of the children.12

Other factors that may be protective are male sex and
more advanced disease. However, this assessment was
performed before identification of the genetic link to
HSR.13

HLA-B*5701 AND HSR

An association with ABC HSR was described with HLA-
B*5701, HLA-DR7 and HLA-DQ3. If all three markers are
present, the positive predictive value for HSR is 100%
with a negative predictive value of 97%. HLA-B*5701
alone is highly predictive.14

It is clear that the varied distribution of the HLA-B*5701
genotype is responsible for variability of the risk of ABC
HSR between races and studies.15 Studies from the USA
indicated that this mutation is more prevalent among
white and Hispanic persons than African Americans.16
In Korea the HLA-B*5701 genotype and ABC HSR are
rare.17

In the PREDICT-1 study, where patients with HLA-
B*5701 did not receive ABC, 3.4% of patients given
ABC were diagnosed with HSR but no cases could be
confirmed with patch testing (a research tool only).18
Prospective screening for HLA-B*5701 in patients and

ABACAVIR: ITS USE AND HYPERSENSITIVITY

C L i N i C A L

Helena Rabie1, FCP (Paed)
Kristin Lorenc Henning1, MB ChB, DCH

Pierre Schoeman2, MB ChB, MMed (Clin Path)
Nico de Villiers2, PhD (Hum Gen)

G H J (Oubaas) Pretorius2, PhD (Biochem)
Mark F Cotton1, PhD

1Department of Paediatrics and Child Health, Tygerberg Children’s Hospital and Stellenbosch University, Tygerberg, W Cape
2PathCare Laboratories, N1 City, Goodwood, W Cape

Abacavir, a nucleoside reverse transcriptase inhibitor, is useful in first- and second-line HIV therapy and as a
substitute for stavudine and zidovudine when toxicity is a problem. Although it is safe and well tolerated, a life-
threatening hypersensitivity reaction can occur. The risk for developing this reaction relates to the presence of
specific genotypes, especially HLA-B*5701.

EPiDEMiOLOGY AND ESTiMATiON OF RiSK FOR HSR

december 2009 THE SOUTHERN AFRICAN JOURNAL OF HIV MEDICINE

avoidance of ABC in positive patients is effective in re-
ducing HSR, and this is now the standard of care in the
First World. Over-diagnosis of HSR is well documented
in the absence of testing.19 A reduction in confirmed
cases occurs when routine testing is performed.18

Despite the availability of testing and the recommen-
dation to test, there is a debate as to the cost effec-
tiveness and cost benefit of testing in ethnic groups
where HLA-B*5701 is not prevalent.17

There are no data on the prevalence of HLA-B*5701 in
the various South African ethnic groups. Full genetic
screening for HLA-B*5701 is very costly. Cheaper meth-
ods involving PCR for small sequences of the gene are
currently under review. Although full testing is avail-
able in South Africa, patients in the public sector do
not have access. We recommend that testing be offered
to all patients where affordable, regardless of ethnic
group, until more information is available. However,
it is reasonable to use ABC without prior screening if
there is no alternative.

It is important to remember that HSR has been report-
ed in patients negative for HLA-B*5701.20 In patients
in whom HSR reaction was diagnosed and who subse-
quently tested negative for HLA-B*5701, ABC remains
contraindicated.

Diagnosis of ABC HSR is complicated by its subtle initial
features. Also, other drugs such as trimethoprim-sul-
famethoxazole, nevirapine and efavirenz are known to
cause hypersensitivity and should be recognised. Dis-
tinguishing the ABC HSR from other drug-related ad-
verse events, intercurrent infections and even immune
reconstitution inflammatory syndrome may be particu-
larly difficult when ABC is used as first-line therapy,
as all drugs are initiated simultaneously. In addition,
ABC initiation may lead to symptoms that are similar
but not related to HSR, including nausea and vomiting,
fever and rash. These reactions are usually mild.

Ninety-four per cent of patients who experience HSR do
so within 6 weeks after initiation of therapy. The me-
dian time to onset is 11 days, but symptoms can start
on the first day and have been reported up to 318 days
later. ABC HSR has occurred in patients who interrupted
therapy without having had hypersensitivity and subse-
quently restart, but this is believed to be rare.11,21 A single
case of ABC HSR after switching from twice daily to once
daily administration has also been reported.22 Vigilance
for the duration of ABC exposure is required.

The ABC HSR is a multi-organ process manifesting
signs or symptoms from at least two of the following
groups:

n Fever is the most common manifestation of ABC
HSR, occurring in 80% of cases. Chills have been
reported to accompany fever.

n Rash is experienced by 70% of cases, and pruritus
can also occur. In contrast to the rash caused by
non-NRTIs and sulphonamides, it is often mild and
may go unnoticed by patients. When rash occurs in
the absence of other features of HSR, ABC should
not be discontinued.

n Gastro-intestinal symptoms such as nausea, vom-
iting, diarrhoea and abdominal pain are all features
of HSR but may also occur in the absence of HSR,
particularly when ABC is used with ZDV. Therefore,
as with rash, patients with isolated gastro-intestinal
symptoms should not discontinue ABC but should
be followed up closely.

n Constitutional symptoms include fatigue, myalgias
and generalised malaise.

n Respiratory symptoms occur in 18% of cases and
include dyspnoea, cough and pharyngitis. Symptoms
may be difficult to distinguish from those caused by
influenza and other respiratory viruses. Respiratory
symptoms together with abdominal symptoms sug-
gest HSR rather than influenza or other respiratory
illness.23

Clusters and combinations of symptoms are important
in the diagnosis of ABC HSR. Table I illustrates the fre-
quency of some combinations.11,24

With ABC HSR, there is an accentuation of symptoms
in the hours immediately after the dose and worsen-
ing of symptoms with each subsequent dose. A number
of case reports illustrate the varied clinical presenta-
tion, with Kawasaki-like illness, prominent exanthema
and even disseminated intravascular coagulation being
seen.25-29

If ABC is not terminated, or if it is re-initiated after
temporary cessation, the HSR will progress to hypoten-
sion, renal dysfunction, bronchospasm and ultimately
death.11

Abnormal laboratory findings may include leucopenia,
anaemia and thrombocytopenia, as well as elevations

Systems and combinations %
3 or 4 organ systems 49
Fever and rash 20
Fever and GIT 8
Skin and GIT 3
Skin and constitutional 3
Other combinations 17

GIT = gastro-intestinal.

TABLE i. FREQUENCY OF SYMPTOM COMBiNATiONS
iN ABACAViR HYPERSENSiTiViTY (ADAPTED FROM

CLAY24)

CLiNiCAL FEATURES AND DiAGNOSiS OF ABC HSR

THE SOUTHERN AFRICAN JOURNAL OF HIV MEDICINE december 2009

in transaminases, urea, creatinine and lactate dehy-
drogenase (LDH). Eosinophilia is usually absent.11 Patch
testing is currently only a research tool.

Termination of therapy is followed by rapid improve-
ment in the symptoms.

Rechallenging with ABC leads to anaphylaxis and
should be avoided even in cases where there was diag-
nostic uncertainty.

In Table II we set out the features of the first 3 cases
of suspected HSR seen at the Tygerberg Children’s Hos-
pital Family Clinic for HIV. Of note is that HSR was
documented in children across the racial spectrum. In
all patients there was progression of symptoms over
time and in 1 case there was a clear increase in sever-
ity associated with dosing. All children had abdominal
symptoms and nonspecific rash. In these cases, child-
ren were stable on other ART drugs as they had all
switched to ABC because of d4T toxicity.

On commencement of ABC, patients should be coun-
selled in detail about the possible signs of HSR and be
advised to contact their care provider should any occur.
To avoid confusion, therapy should not be initiated in
patients with intercurrent symptoms.
It is advisable for patients to discuss symptoms early
with the clinician rather than terminating therapy
without consultation. Where termination without con-
sultation occurs, ABC cannot be reinitiated. Patients

should also be made aware of the special ‘patient alert
card’ that comes in the packaging. This card should be
presented to any health care provider who sees the
child, especially when care is not given by the usual
provider. Providers at emergency facilities may be less
familiar with this condition, and where possible con-
tact information for the usual care provider should be
supplied as well.

Deciding whether to terminate therapy in a patient
with suggestive symptoms can be difficult given the
very nonspecific nature of the presentation. A detailed
medical history should be obtained. The following
should be considered:
n When was ABC initiated? In the case of ABC HSR,

usually within the past 6 weeks.
n Are two or more systems involved?
n Do the symptoms increase with each dose?
n Are the symptoms exacerbated just after the dose?
n Do the symptoms fit into the well-recognised clus-

ters?
n What other medications/medication is the patient

taking, and what was the timing of their initiation
related to the ABC?

If patients present with mild symptoms and it is not
clear whether symptoms are due to HSR, the clinician
may consider allowing an additional dose. The patient
should be able to report back, or hospitalisation may
be required for observation. If symptoms worsen, ABC
should be terminated immediately and permanently. If
symptoms do not worsen, ABC can be carefully con-

Case 1 Case 2 Case 3

Race White Coloured Black
Age (years) 9 5 10
Gender Female Male Male
Time to onset of symptoms <1 day 9 days 2 months
Accentuation with dose Yes Uncertain Uncertain
Increasing severity Yes Yes Yes
Time after onset to
presentation to TCH (days) 1 5 3
Fever No Yes No
Rash Blotchy, erythematous on Extensive maculopapular Fine papular rash on
neck and hands on trunk, arms and legs the chest
Papules on the trunk and Exanthema in mouth
left arm Non-purulent conjunctivitis
Gastrointestinal Loss of appetite Nausea Abdominal pain and
Epigastric and right upper Loose stools tenderness
quadrant tenderness Vomiting
Loss of appetite
Constitutional Myalgias Lethargy Weight loss (1 kg)
Malaise
Respiratory No No Cough
Red throat
Number of systems affected 3 4 4
Time to resolution 48 hours 5 days 2 - 3 days
HLA-B*5701 Negative – tested after Positive – tested after Negative – tested after
the HSR the HSR the HSR

TABLE ii. CLiNiCAL FEATURES iN 3 CHiLDREN DiAGNOSED WiTH ABC HSR AT TYGERBERG CHiLDREN’S HOSPiTAL
AFTER A SiNGLE DRUG SUBSTiTUTiON OF STAVUDiNE FOR LiPO-ATROPHY

MANAGEMENT OF PATiENTS iNiTiATiNG ABC

december 2009 THE SOUTHERN AFRICAN JOURNAL OF HIV MEDICINE

tinued while other possible reasons for the patient’s
symptoms are investigated. When the diagnosis is
thought to be clear or there is sufficient concern, ABC
should be terminated immediately and permanently.

Hospitalisation and special investigations will depend
on the severity of symptoms. Corticosteroids do not
prevent or alter the natural history of ABC HSR.30 The
reaction usually improves within 48 hours.

Clinicians treating children need to be very aware of
the usefulness of ABC. Although there is no informa-
tion on the prevalence of either ABC hypersensitivity or
HLA-B*5701 in South African children, available data
suggest that black children are at lower risk than Cau-
casian children, with no data on children of mixed race.
Although screening for HLA-B*5701 is recommended
and will prevent cases, research is needed to assess its
cost effectiveness in the South African public health
setting.

REFERENCES
1. Comparison of dual nucleoside-analogue reverse-transcriptase inhibitor regimens

with and without nelfinavir in children with HIV-1 who have not previously been
treated: the PENTA 5 randomised trial. Lancet 2002; 359: 733-740.

2. Green H, Gibb DM, Walker AS, et al. Lamivudine/abacavir maintains virological
superiority over zidovudine/lamivudine and zidovudine/abacavir beyond 5 years
in children. AIDS 2007; 21: 947-955.

3. de Ronde A, van Dooren M, de Rooij E, van Gemen B, Lange J, Goudsmit J.
Infection by zidovudine-resistant HIV-1 compromises the virological response to
stavudine in a drug-naive patient. AIDS 2000; 14: 2632-2633.

4. Kuritzkes DR, Bassett RL, Hazelwood JD, et al. Rate of thymidine analogue
resistance mutation accumulation with zidovudine- or stavudine-based regimens.
J Acquir Immune Defic Syndr 2004; 36: 600-603.

5. Maxeiner HG, Keulen W, Schuurman R, et al. Selection of zidovudine resistance
mutations and escape of human immunodeficiency virus type 1 from antiretroviral
pressure in stavudine-treated pediatric patients. J Infect Dis 2002; 185: 1070-
1076.

6 . Saez-Llorens X, Nelson RPJ, Emmanuel P, et al. A randomized, double-blind
study of triple nucleoside therapy of abacavir, lamivudine, and zidovudine versus
lamivudine and zidovudine in previously treated human immunodeficiency virus
type 1-infected children. The CNAA3006 Study Team. Pediatrics 2001; 107: E4.

7. McComsey GA, Paulsen DM, Lonergan JT, et al. Improvements in lipoatrophy,
mitochondrial DNA levels and fat apoptosis after replacing stavudine with
abacavir or zidovudine. AIDS 2005; 19: 15-23.

8. Carr A, Workman C, Smith DE, et al. Abacavir substitution for nucleoside analogs
in patients with HIV lipoatrophy: a randomized trial. JAMA 2002; 288: 207-215.

9. Scherpbier HJ, Bekker V, Pajkrt D, Jurriaans S, Lange JM, Kuijpers TW. Once-daily
highly active antiretroviral therapy for HIV-infected children: safety and efficacy
of an efavirenz-containing regimen. Pediatrics 2007; 119: e705-715.

10. LePrevost M, Green H, Flynn J, et al. Adherence and acceptability of once daily
lamivudine and abacavir in human immunodeficiency virus type-1 infected
children. Pediatr Infect Dis J 2006; 25: 533-537.

11. Hewitt RG. Abacavir hypersensitivity reaction. Clin Infect Dis 2002; 34: 1137-
1142.

12. Nahirya-Ntege P, Naidoo B, Nathoo KJ, et al. Successful management of suspected
abacavir hypersensitivity reactions among African children in theARROW
(AntiRetroviral Research fOr Watoto) trial. Presented at the International AIDS
Society Conference, 19 - 22 July 2009, Cape Town (Poster TUPEB18).

13. Symonds W, Cutrell A, Edwards M, et al. Risk factor analysis of hypersensitivity
reactions to abacavir. Clin Ther 2002; 24: 565-573.

14. Mallal S, Nolan D, Witt C, et al. Association between presence of HLA-B*5701,
HLA-DR7, and HLA-DQ3 and hypersensitivity to HIV-1 reverse-transcriptase
inhibitor abacavir. Lancet 2002; 359: 727-732.

15. Hetherington S, Hughes AR, Mosteller M, et al. Genetic variations in HLA-B
region and hypersensitivity reactions to abacavir. Lancet 2002; 359: 1121-1122.

16. Hughes AR, Mosteller M, Bansal AT, et al. Association of genetic variations in
HLA-B region with hypersensitivity to abacavir in some, but not all, populations.
Pharmacogenomics 2004; 5: 203-211.

17. Park WB, Choe PG, Song KH, et al. Should HLA-B*5701 screening be performed in
every ethnic group before starting abacavir? Clin Infect Dis 2009; 48: 365-367.

18. Munoz de Benito RM, Arribas Lopez JR. [Prospective validation of a
pharmacogenetic test: the PREDICT-1 study.] Enferm Infecc Microbiol Clin 2008;
26: Suppl 6, 40-44.

19. Rauch A, Nolan D, Thurnheer C, et al. Refining abacavir hypersensitivity diagnoses
using a structured clinical assessment and genetic testing in the Swiss HIV
Cohort Study. Antivir Ther 2008; 13: 1019-1028.

20. Calza L, Rosseti N, Biagetti C, Pocaterra D, Colangeli V, Manfredi R. Abacavir-
induced reaction with fever and severe skin rash in a patient tested human
leukocyte antigen-B*5701 negative. Int J STD AIDS 2009; 20: 276-277.

21. Frissen PH, de Vries J, Weigel HM, Brinkman K. Severe anaphylactic shock after
rechallenge with abacavir without preceding hypersensitivity. AIDS 2001; 15:
289.

22. Gervasoni C, Vigano O, Grinelli E, Ortu M, Galli M, Rusconi S. Abacavir
hypersensitivity reaction after switching from the twice-daily to the once-daily
formulation. AIDS Patient Care STDs 2007; 21: 1-3.

23. Keiser P, Nassar N, Skiest D, et al. Comparison of symptoms of influenza A with
abacavir-associated hypersensitivity reaction. Int J STD AIDS 2003; 14: 478-481.

24. Clay PG. The abacavir hypersensitivity reaction: a review. Clin Ther 2002; 24:
1502-1514.

25. Abacavir warning: certain respiratory symptoms can indicate hypersensitivity
reaction. AIDS Treat News 2000; No. 337.

26. Aquilina C, Mularczyk M, Lucas F, Viraben R. Unusual clinical presentation of
hypersensitivity reaction to abacavir. AIDS 2003; 17: 2403-2404.

27. Dargere S, Verdon R, Bouhier K, Bazin C. Disseminated intravascular coagulation
as a manifestation of abacavir hypersensitivity reaction. AIDS 2002; 16: 1696-
1697.

28. Lanzafame M, Trevenzoli M, Lattuada E, Faggian F, Vento S, Concia E. Enanthema
as the first clinical manifestation of abacavir hypersensitivity reaction: a case
report. Infez Med 2003; 11: 40-41.

29. Toerner JG, Cvetkovich T. Kawasaki-like syndrome: abacavir hypersensitivity? Clin
Infect Dis 2002; 34: 131-133.

30. Wit FW, Wood R, Horban A, et al. Prednisolone does not prevent hypersensitivity
reactions in antiretroviral drug regimens containing abacavir with or without
nevirapine. AIDS 2001; 15: 2423-2429.

CONCLUSiON

Acknowledgement

We thank the staff from PathCare for their assist-
ance and Drs B Leibbrandt and Clair Edson for pro-
viding patient details. We also thank Dr Leon Levin
and Dr Tammy Meyers for their editorial contribu-
tion.