Should abacavir be a first-line option for adults with HIV in sub-Saharan Africa?
Background: Despite a poor
toxicity profile, zidovudine supersedes abacavir as a first-line agent in many international
treatment guidelines due to concerns about HLA-B*57:01-related
abacavir-hypersensitivity. An important
question is whether abacavir should become a first-line option in resource
limited settings where HLA-B*57:01 is sufficiently rare. In one of the largest HLA-typing efforts in
sub-Saharan Africa, we assessed HLA alleles among 581 HIV-infected patients in
Kampala (n=81) and Mbarara (n=500) to guide regional clinical decision-making.
Methods: Specimens were drawn from the UARTO cohort, including 81 pilot subjects
enrolled from Kampala during 2002-2004 and 500 subjects enrolled from Mbarara
during 2005-2010, just prior to antiretroviral therapy initiation. HLA-typing was performed using Roche/454/Fluidigm
HLA Typing Kits following the Roche protocols.
HLA alleles and genotypes were called using the Conexio ATF 454 HLA
typing software.
Results: Samples from 52 (64%) individuals in Kampala,
and 461 (92%) individuals in Mbarara yielded successful HLA-genotyping results.
HLA-B*57:01 was not observed in the 52 Kampala subjects. In the main Mbarara
cohort, one subject was heterozygous for HLA-B*57:01 among 461 subjects (0.2%
prevalence). This subject did not
receive abacavir
during study observation from 2006 - 2011. Overall, during the entire follow up duration
from 2005-2015 in Mbarara, only two other patients ever-used abacavir (0.4% usage rate); neither had HLA-B*57:01. The low prevalence of HLA-B*5701 is
consistent with other smaller studies conducted in sub-Saharan Africa.
Conclusions: Given
the low prevalence of HLA*B57:01 carriage in our cohort and others in
sub-Saharan Africa (Table 1), the cautious use abacavir-based therapy may be
more advisable than zidovudine-based therapy when tenofovir is not a viable
option. The benefits of abacavir over zidovudine-based therapy should be
considered in regional guidelines given abacavir has superior efficacy,
once-daily dosing, and has been studied as a component of regimens containing integrase
strand transfer inhibitors.
| Country | Region/ Ethnicity | Sample size | Allele frequency | Received abacavir? | Abacavir hypersensitivity reaction rate | References |
| Kenya | Nandi | 240 | 0.0083 | No | NA | Cao, 2004 |
| Kenya | Luo | 265 | 0.0076 | No | NA | Cao, 2004 |
| Mali | Dogon | 138 | NA | No | NA | Cao, 2004 |
| Uganda | Kampala | 161 | 0.0311 | No | NA | Cao, 2004 |
| Zambia | Lusaka | 44 | 0.0114 | No | NA | Cao, 2004 |
| Guiné-Bissau | NA | 65 | 0 | No | NA | Spínola, 2005 |
| Uganda | Kampala, Entebbe | 300 | 0 | Yes | 6/300 (2%) | Munderi, 2011 |
| Uganda | Kampala | 52 | 0 | No | NA | This study |
| Uganda | Mbarara | 461 | 0.001 | No | NA | This study |
G.Q. Lee1,2,3, S. McCluskey3,4, Y. Boum5, P.W. Hunt6, J.N. Martin6, D.R. Bangsberg3,7, G. Xiaojiang8, P.R. Harrigan1, J.E. Haberer7, M.J. Siedner3,5,7
1BC Centre for Excellence in HIV/AIDS, Vancouver, Canada, 2Ragon Institute of MGH, MIT and Harvard, Cambridge, United States, 3Harvard Medical School, Boston, United States, 4Massachusetts General Hospital, Infectious Disease Division, Boston, United States, 5Mbarara University of Science and Technology, Mbarara, Uganda, 6University of California, San Francisco, United States, 7Massachusetts General Hospital, Boston, United States, 8National Cancer Institute, Center for Cancer Research, Frederick, United States