Secondary Logo

Journal Logo

CLINICAL SCIENCE

Comparative response of African HIV-1-infected individuals to highly active antiretroviral therapy

Frater, Alexander J.a; Dunn, David T.b; Beardall, Alison J.a; Ariyoshi, Koyac; Clarke, John R.a; McClure, Myra O.a; Weber, Jonathan N.a

Author Information

Abstract

Introduction

At the end of 2000, an estimated 36.1 million individuals were living with HIV-1 infection, of whom 25.3 million were living in sub-Saharan Africa. The adult prevalence rate of HIV-1 infection in sub-Saharan Africa is estimated to be 8.8%, with an incidence of 7.2 new infections per minute [1].

In the industrialized world, the introduction of highly active anti-retroviral therapy (HAART) for the treatment of HIV-1 infection has had a dramatic effect on AIDS-related morbidity and mortality rates [2]. However, in sub-Saharan Africa severe financial limitations on health budgets and individual incomes mean that HAART is currently not available. Although there are high-level attempts to address this fundamental issue of inequity, the likely efficacy of HAART in African HIV-1-positive patients remains unproved. Here we show the responses of African HIV-1-positive individuals to HAART, ex Africa, in an inner London clinic.

Fourteen antiretroviral drugs are licensed in the UK, in three classes: nucleoside reverse transcriptase inhibitors (NRTI), non-nucleoside reverse transcriptase inhibitors (NNRTI) and protease inhibitors (PI) [3]. The efficacy of three-drug HAART regimens on mortality and morbidity has been demonstrated in randomized controlled trials [4,5]. Clinical efficacy is associated with a reduction in viral load to below the level of detection of reverse transcriptase–polymerase chain reaction RNA or branched DNA assays (generally, less than 50 RNA copies/ml of plasma) and the maintenance of undetectable viraemia for prolonged periods. Maintenance of an undetectable viral load is associated with a continuous increase in CD4 cell counts, representing immune reconstitution, as evidenced by a lack of clinical progression [6,7].

The great majority of HIV-1 infections in Europe and north America are with subtype B, whereas in sub-Saharan Africa, subtypes A, C and D predominate and subtype B is very rare. Inter-subtype nucleotide sequence divergence may exceed 20, 15 and 25% for gag, pol and env, respectively [8]. These highly heterogeneous sequences may affect pathogenesis [9], infectivity and response to therapy [10,11]. In-vitro biochemical data from protease enzymes from Ugandan isolates (subtypes A and C) indicate a decreased inhibition in the presence of protease inhibitors compared with subtype B enzymes [10].

Few data exist on the response of non-B HIV-1 subtypes to antiretroviral drugs. The NNRTI class has been described as being less effective in subtype O isolates [12] and HIV-2 [13]. Within the M group, pre-therapy resistance has been described for subtype F to 8-chloro-tetrahydro-imidazo (4,5,1-jk) (1,4)-benzodiazepin-2 (1H) thione derivatives [14] and for subtype D to all classes [15]. More recently, it has been reported that secondary protease mutations, which are very common in non-B HIV-1 subtypes (in particular at codons 10 and 36), may be associated with a poorer virological outcome on HAART [11]. Conversely, some studies have also suggested that there are no differences in outcome to therapy between subtypes. No differences were found in outcomes to tenofovir, adefovir and zidovudine for subtypes A–G [16], and between subtypes F and A [17], or between B and C [18] for NRTI and NNRTI. Two reports [19,20] have been published on African patients treated with antiretroviral agents, both concerning monotherapy in pregnancy, and one report [21] has shown no impact of the subtype on mother-to-child transmission in Uganda when single-dose nevirapine was used as prophylaxis.

Despite the absence of antiretroviral drugs in Africa, it was possible to study the impact of HAART on emigrant African HIV-1-positive patients treated overseas. At St Mary's Hospital, a computerized database detailed 4121 HIV-1-positive patients treated since 1982, of whom 371 patients were infected in Africa, the majority in Uganda. In this report, we were able to demonstrate the clinical outcome of African patients infected with non-B HIV-1 subtypes to HAART, and to compare this response with that of contemporary European patients treated with the same HAART regimens.

Patients and methods

Patient information stored on the St Mary's Hospital database is derived from questionnaires completed during the first clinic attendance, from laboratory viral load and CD4 cell count data and from clinical data collected after each clinic visit. Information recorded includes demographic details (age, sex, risk factor, ethnicity, country of origin, date of HIV-1 infection and date of AIDS diagnosis), virological data (viral loads with dates and assay used), immunological data and therapy-related data (antiretroviral drugs prescribed with start dates and stop dates). The database is updated monthly.

The database was used to identify patients putatively infected with non-B HIV-1 subtypes and to characterize their virological and immunological responses to HAART, compared with patients from the United Kingdom and western Europe. Patients who described themselves as ‘white British/European’ constituted the ‘European group’ and those who were ‘black African/Asian’ and born in Africa were the ‘African group'. Asians born and infected in Africa were included in the African group as it was expected that exposure would also have been to African non-B HIV-1 subtypes. Individuals who described themselves as homosexuals and intravenous drug users (IVDU) were excluded from the African group (except for one African patient with an IVDU risk factor whose non-B subtype had been confirmed previously). In our laboratory, sequence analyses from African homosexuals and IVDU had shown the infecting viruses to be predominantly subtype B, indicating that infection may have occurred within Europe or north America (data not shown).

From the two ethnic groups only patients who were initially drug-naive and started a three-drug HAART regime containing two NRTI, and either one NNRTI or one PI, were selected. Only patients with at least one viral load measurement pre- and post-therapy were analysed. All patients had post-therapy CD4 cell count data, although 12 (five African and seven European) had no pre-therapy value. In these cases the mean pre-therapy CD4 cell count for the corresponding ethnic group was used. Such an imputation might lead to an underestimation of the standard errors of the changes in CD4 cell counts, although as the proportions in each group were very small the effect on results was negligible. Data were collected until the end of November 2000.

In order to confirm that the questionnaires correctly identified patients infected with B and non-B HIV-1 subtypes, a random proportion of the African (59/97; 60%) and European (80/265; 30%) groups were sequenced in pol, using either the ABI vII ViroSeq HIV sequencing kit and an ABI 310 genetic analyser (Applied Biosystems, Foster City, CA, USA) or the Visible Genetics TruGene genotyping system (Visible Genetics, Toronto, Canada). Phylogenetic analysis used Jukes and Cantor neighbour-joining methodology to compare patient sequences with reference sequences from the Los Alamos database (http://www.hiv-web.lanl.gov) or by direct submission to the National Centre for Biotechnology Information (NCBI) website (http://www.ncbi.nlm.nih.gov/retroviruses/subtype/subtype.html).

Statistical analysis

Kaplan–Meier analysis was used to estimate the time to viral load undetectability, defined as less than 500 RNA copies/ml of plasma. This level corresponded to the highest ‘lower limit of undetectability’ of the assays used over the duration of the study. It should be noted that the more frequently viral load measurements are performed the earlier one is likely to observe an undetectable value. Therefore, the estimates from this analysis reflect clinical policy as well as intrinsic virological response to therapy.

Longer-term response to therapy was assessed by a change in CD4 cell count and viral load (measured on a log10 scale) from baseline. For each individual the value closest to the timepoints analysed (3, 6, 9, 12, 15, 18 and 21 months) was selected, provided the test was carried out within 1.5 months of the target assessment time. Analysis of covariance, assuming normal errors, was used to assess the independent effects of ethnic group, the use of PI or NNRTI in the initial regime, baseline viral load and baseline CD4 cell count. Analysis of change in viral load was complicated by frequent left censoring (below limit of assay detectability) of post-therapy values. This was addressed by using ‘interval regression’ as implemented in the INTREG command in stata (Stata Corporation, College Station, TX, USA), an approach that is superior to equating values to the lower threshold [22]. Absolute changes in viral load and CD4 cell count (see Fig. 2 and Fig. 3) referred arbitrarily to a patient on a PI regime with a baseline viral load of 10 000 copies/ml of plasma and a baseline CD4 count of 200 cells/μl. This choice had no bearing on the observed difference between the two ethnic groups. An overall test for an effect of ethnic group (Wald statistic with robust standard errors) was performed by including all data points, including a factor for time, and taking account of the fact that individuals could appear at multiple timepoints [23].

Fig. 2.
Fig. 2.:
Change in log10 viral load from baseline over time in African and European patients receiving highly active antiretroviral therapy. Solid line, African cohort; hashed line, European cohort. Figures below the X-axis represent numbers of European (Euro) and African (Afr) patients included at each timepoint. Error bars represent the standard deviation of the mean at each data point. The graph refers arbitrarily to a patient on a protease inhibitor regime with a baseline viral load (VL) of 10 000 copies/ml of plasma and a baseline CD4 cell count of 200 cells/μl.
Fig. 3.
Fig. 3.:
Change in CD4 cell count from baseline over time in African and European patients on highly active antiretroviral therapy. Solid line, African cohort; hashed line, European cohort. Figures below the X-axis represent numbers of European (Euro) and African (Afr) patients included at each timepoint. Error bars represent the standard deviation of the mean at each data point. The graph refers arbitrarily to a patient on a protease inhibitor regime with a baseline viral load of 10 000 copies/ml of plasma and a baseline CD4 cell count of 200 cells/μl.

Results

A total of 360 patients fulfilled the criteria for inclusion in the analysis; 265 constituted the ‘European’ group, and 97 formed the ‘African’ group. Table 1 shows the demographic details of the European and African cohorts. Imbalance is present in the African group towards heterosexual patients (91 versus 10% of the European patients, imposed by selection criteria) and women (61 versus 7% of the European patients). The baseline median viral load was similar (4.95 versus 4.74 log10 copies RNA/ml plasma for European and African patients, respectively), although the baseline median CD4 cell count in the African group was substantially lower (140 versus 240 cells/μl). The median duration of follow-up from the start of antiretroviral therapy to the last viral load measurement was 22.3 and 15.1 months for the European and African patients, respectively.

Table 1
Table 1:
Cohort demographics detailing the European and African groups.

Overall, 2725 viral load measurements were used in the analysis, with a median sampling interval of 12.3 weeks. A potential difficulty with comparing HIV-positive cohorts is that viral load values depend on the assay in use. The Roche Amplicor v.1.5 assay (Roche, Alameda, CA, USA) used a lower limit of detection (LLD) of 400 copies/ml of plasma, the Chiron Quantiplex bDNA 2.0 assay (Chiron, Emeryville, CA, USA) used an LLD of 500 copies/ml and the Chiron 3.0 assay used an LLD of 50 copies/ml. There is, therefore, a 1 log10 variation in range, impacting on the analyses of change in viral load from baseline and on the time taken to achieve undetectability. The majority (80%) of viral loads detailed on the database were measured using the Chiron 3.0 bDNA assay; 14% used the Chiron 2.0 bDNA assay and 6% the Roche Amplicor v1.5 assay. No significant differences were found between assays used for the European and African cohorts.

The drug regimes prescribed were similar for the two groups, except specific PI (Table 1). PI-based HAART was used in 56% of patients, with the remainder receiving NNRTI. Indinavir (42% of European patients receiving a PI) and nelfinavir (45% of African patients receiving a PI) were the most frequently prescribed PI. Nevirapine was used in 92% of both European and African patients on NNRTI-based HAART. Zidovudine and lamivudine (either individually or as combivir) were the most frequently prescribed NRTI (63 and 57% of European and African patients, respectively). Stavudine in combination with either lamivudine or didanosine were the next most frequent NRTI combinations used (Table 1). The questionnaire responses correctly distinguished between patients infected with B and non-B HIV-1 subtypes (Table 2).

Table 2
Table 2:
Numbers (%) of HIV-1 subtypes according to the African and European cohorts.

Time to viral load undetectability

The initial virological response to HAART was assessed by analysing the time taken for the European and African cohorts to achieve undetectable viral loads (Fig. 1). No significant differences existed between the response of the two cohorts (P = 0.9, log rank test). Kaplan–Meier estimates for the proportions achieving, but not necessarily maintaining, undetectability by 3, 6, 9 and 12 months were 56, 81, 89 and 91% for the European group and 60, 81, 86 and 91% for the African group (Fig. 1). The absence of the effect of ethnic group persisted after controlling for baseline viral load, baseline CD4 cell count and drug class in a multivariate Cox regression analysis (P = 0.5).

Fig. 1.
Fig. 1.:
Kaplan–Meier estimation of the time taken for African and European patients treated with highly active antiretroviral therapy to reach undetectable viral load. HAART, Highly active antiretroviral therapy; VL, viral load. Black line, African cohort; hashed line, European cohort. Figures below the X-axis represent numbers of European (Eur) and African (Afr) patients included at each timepoint.

Viral load change from baseline on highly active antiretroviral therapy

Fig. 2 shows the estimated mean change in viral load from baseline. In the European cohort the initial reduction appeared to be maintained until at least 21 months on average. In contrast, for the African group there was evidence of an increase in viral load after 9 months, resulting in a widening viral load gap between the two cohorts. Overall, the effect of ethnic group was statistically highly significant (P < 0.001), mainly caused by differences in the second year of therapy. The baseline CD4 cell count was weakly related to the average viral load reduction (0.15 log10 greater decrease in viral load for every 100 cells by which the baseline CD4 count was higher;P = 0.05). Interestingly, there was also some evidence of a drug class effect (0.60 greater mean reduction for the NNRTI compared with the PI regimes, P = 0.02); however, no distinct patterns were observed (data not shown) when ethnic groups were compared separately in the context of PI and NNRTI regimes (P = 0.3, test for interaction).

CD4 cell count change from baseline on highly active antiretroviral therapy

Fig. 3 shows the estimated mean change in the CD4 cell counts from pre-therapy baseline. The African patients had substantially lower median baseline CD4 cell counts than the European patients (140 versus 240 CD4 cells/μl, respectively). Nevertheless, there were no statistically significant differences between the two groups in changes in CD4 cell counts (P = 0.11). In the first 3 months of therapy the mean increase in CD4 cell count for the European and African patients was 36.7 and 28.7 cells/μl per month, and thereafter 4.6 and 5.0 cells/μl per month, respectively. Adjustment for drug class and baseline CD4 cell count made no difference to the results (data not shown).

Discussion

This is the first study to compare the clinical response to HAART of HIV-1-infected African and European patients. No significant difference was attributable to non-B HIV-1 subtype (predominantly A, C and D) in the initial virological response. In addition, not only were the results comparable between the viral subtypes, but they indicated that HAART is likely to be equally efficacious in Africa as in the industrialized world. This is fortuitous, because antiretroviral drugs were not designed to inhibit non-B subtype viruses. These data on the initial response to therapy agreed with the only other similar study, which found similar virological outcomes in two matched cohorts of 50 B and non-B-infected individuals [24]. The data on the time to viral load undetectability, the change in viral load from baseline in the first 12 months and the change in the CD4 cell count all suggest equivalence between patients infected with European and African HIV-1 subtypes. These findings concur with the widely held, but non-evidence-based, assumption that HAART is efficacious in non-B subtype infections.

Of concern is the relative increase in viral load in the African cohort after 9 months of antiretroviral therapy. Possible explanations for this observation could be virological, pharmacological or social. First, we have previously described that neither viral subtype nor individual baseline polymorphisms in pol in subtypes A, C and D impacted on the development of clinical failure in African patients receiving HAART. Only 50% of African patients with virological failure developed mutations in pol associated with resistance to antiretroviral drugs. Although multiple sequence changes were seen at other codons in all subtypes, the majority were either present in pre-therapy samples (i.e. polymorphisms) or only occurred sporadically [25]. However, in a subsequent study [11] certain baseline polymorphisms in protease, especially at codons 10 and 36, were associated with a poorer virological outcome.

Increasing data have been published on the development of resistance in non-B HIV-1 subtypes [17,20, 26,27]. However, compared with subtype B, no novel pathways to the development of resistance in African patients have been identified, and this is therefore unlikely to be the explanation for the disparity in the longer-term viral load changes. Although it has been claimed that in subtype C-infected patients on nelfinavir the mutation at L90M occurs more frequently than D30N, and that in patients failing NNRTI regimes V179I has been associated predominantly with subtype A and V106M with subtype C [28], the significance of these findings remains to be confirmed.

The African cohort presented with a substantially lower pre-therapy CD4 cell count, which corresponds with the findings of a previous epidemiological study of African patients in London [29]. The baseline CD4 cell count has been reported to correlate with virological response on HAART [30], and might therefore be considered to be an explanation for the discrepancy in viral load outcomes. In this analysis, the impact of baseline CD4 cell count on the change in viral load was weak (0.15 log10 greater decrease in viral load for every 100 cells by which the baseline CD4 cell count was higher), and not great enough to produce the reported ethnic group differences of at least 1 log10 observed by 12 months. Interestingly, the virological differences observed on therapy in the ethnic groups did not translate into poorer CD4 cell count responses in the African cohort (Fig. 3). This finding was possibly temporal, and had the study been continued for a longer duration poorer CD4 cell count responses may have become evident. Alternatively, despite the higher viral loads, the persisting benefit conferred by HAART may have been adequate to permit immunological reconstitution.

A further explanation for the discordant viral load results after 9 months was poorer adherence in the African cohort, especially in view of the cultural and language barriers that exist for emigrant populations being treated at a western clinic. Antiretroviral drugs, particularly those available in the late 1990s, were associated with a high pill burden, frequent side-effects and inconvenient dosage timing. Poor adherence was, and still is, one of the major causes of viral load rebound [31]. Unfortunately, no adherence records are maintained on the database, so this hypothesis cannot be formally tested. In addition, imbalances existed in the proportions of women, IVDU and homosexual patients in the two cohorts, and although these differences were unlikely to influence the initial response to therapy, it is possible that they acted as confounding factors in the longer-term data. If adherence was poorer within the African cohort, western clinics would have to re-assess patient education and support for ethnic minorities. The minor differences in drug regimes used may also have impacted on the responses to HAART. In particular, of the PI prescribed, indinavir was used in 42% of the European cohort but in only 20% of the African patients, and nelfinavir was used in 35% of the European and 45% of the African patients. This study was not designed to identify the different efficacies of one drug regime over another, although as there are no reports on in-vivo outcome differences between ethnic groups, these results warrant further investigation in randomized trials.

Conclusion

Overall, these data are extremely encouraging for those advocating the implementation of antiretroviral therapy for patients infected with HIV-1 in Africa. The initial virological and immunological responses of the cohorts to HAART were similar, the former suggesting equivalent drug susceptibility in all HIV-1 subtypes. Further studies to measure drug levels and assess adherence are necessary to explain the poorer long-term viral load data, as it is possible that the explanation for this is not virological. On the basis of these findings, there is no justification for withholding HAART from Africa on virological grounds.

Acknowledgements

The authors would like to thank Ms Trinh Duong and Ms Ruth Goodall for assistance with statistical analysis and Dr Abdel Babiker for helpful comments on the manuscript. They also thank Dr Duncan Churchill and Ms Sarah Galpin for developing the database at St Mary's Hospital, and Ms Sarah Galpin for technical assistance.

Contributors: A.J. Frater and D.T. Dunn performed the statistical analysis and wrote the paper. J.N. Weber and K. Ariyoshi conceived the study and, with M.O. McClure, supervised its progress. A.J. Beardall prepared the database for analysis. J.R. Clarke subtyped the European cohort.

References

1. Joint United Nations Programme on HIV/AIDS (UNAIDS). AIDS epidemic update. December 2000.
2. Palella FJJ, Delaney KM, Moorman AC. et al. Declining morbidity and mortality among patients with advanced human immunodeficiency virus infection.HIV Outpatient Study Investigators. N Engl J Med 1998, 338: 853–860.
3. BHIVA Writing Committee on behalf of the BHIVA Executive Committee. British HIV Association (BHIVA) guidelines for the treatment of HIV-infected adults with antiretroviral therapy. HIV Med 2000, 1: 76–101.
4. Hammer SM, Squires KE, Hughes MD. et al. A controlled trial of two nucleoside analogues plus indinavir in persons with human immunodeficiency virus infection and CD4 cell counts of 200 per cubic millimeter or less.AIDS Clinical Trials Group 320 Study Team. N Engl J Med 1997, 337: 725–733.
5. Cameron DW, Heath-Chiozzi M, Danner S. et al. Randomised placebo-controlled trial of ritonavir in advanced HIV-1 disease.The Advanced HIV Disease Ritonavir Study Group. Lancet 1998, 351: 543–549.
6. Mocroft A, Devereux H, Kinloch-de-Loes S. et al. Immunological, virological and clinical response to highly active antiretroviral therapy treatment regimens in a complete clinic population.Royal Free Centre for HIV Medicine. AIDS 2000, 14: 1545–1552.
7. Bart PA, Rizzardi GP, Tambussi G. et al. Immunological and virological responses in HIV-1-infected adults at early stage of established infection treated with highly active antiretroviral therapy. AIDS 2000, 14: 1887–1897.
8. Robertson DL, Anderson JP, Bradac JA. et al. HIV-1 nomenclature proposal. Science 2000, 288: 55–56.
9. Kaleebu P, Ross A, Morgan D. et al. Relationship between HIV-1 Env subtypes A and D and disease progression in a rural Ugandan cohort. AIDS 2001, 15: 293–299.
10. Velazquez-Campoy A, Todd MJ, Vega S, Freire E. Catalytic efficiency and vitality of HIV-1 proteases from African viral subtypes. Proc Natl Acad Sci U S A 2001, 98: 6062–6067.
11. Perno CF, Cozzi-Lepri A, Balotta C. et al. Secondary mutations in the protease region of human immunodeficiency virus and virologic failure in drug-naive patients treated with protease inhibitor-based therapy. J Infect Dis 2001, 184: 983–991.
12. Descamps D, Collin G, Letourneur F. et al. Susceptibility of human immunodeficiency virus type 1 group O isolates to antiretroviral agents: in vitro phenotypic and genotypic analyses. J Virol 1997, 71: 8893–8898.
13. Pauwels R, Andries K, Desmyter J. et al. Potent and selective inhibition of HIV-1 replication in vitro by a novel series of TIBO derivatives. Nature 1990, 343: 470–474.
14. Apetrei C, Descamps D, Collin G. et al. Human immunodeficiency virus type 1 subtype F reverse transcriptase sequence and drug susceptibility. J Virol 1998, 72: 3534–3538.
15. Palmer S, Alaeus A, Albert J, Cox S. Drug susceptibility of subtypes A, B, C, D, and E human immunodeficiency virus type 1 primary isolates. AIDS Res Hum Retroviruses 1998, 14: 157–162.
16. Palmer S, Margot N, Gilbert H, Shaw N, Buckheit RJ, Miller M. Tenofovir, adefovir, and zidovudine susceptibilities of primary human immunodeficiency virus type 1 isolates with non-B subtypes or nucleoside resistance. AIDS Res Hum Retroviruses 2001, 17: 1167–1173.
17. Caride E, Brindeiro R, Hertogs K. et al. Drug-resistant reverse transcriptase genotyping and phenotyping of B and non-B subtypes (F and A) of human immunodeficiency virus type I found in Brazilian patients failing HAART. Virology 2000, 275: 107–115.
18. Shafer RW, Eisen JA, Merigan TC, Katzenstein DA. Sequence and drug susceptibility of subtype C reverse transcriptase from human immunodeficiency virus type 1 seroconverters in Zimbabwe. J Virol 1997, 71: 5441–5448.
19. Guay LA, Musoke P, Fleming T. et al. Intrapartum and neonatal single-dose nevirapine compared with zidovudine for prevention of mother-to-child transmission of HIV-1 in Kampala, Uganda: HIVNET 012 Randomised Trial. Lancet 1999, 354: 795–802.
20. Jackson JB, Becker-Pergola G, Guay LA. et al. Identification of the K103N resistance mutation in Ugandan women receiving nevirapine to prevent HIV-1 vertical transmission. AIDS 2000, 14: F111–F115.
21. Eshleman SH, Becker-Pergola G, Deseyve M. et al. Impact of human immunodeficiency virus type 1 (HIV-1) subtype on women receiving single-dose nevirapine prophylaxis to prevent HIV-1 vertical transmission (HIV Network for Prevention Trials 012 Study). J Infect Dis 2001, 184: 914–917.
22. Hughes MD. Analysis and design issues for studies using censored biomarker measurements with an example of viral load measurements in HIV clinical trials. Stat Med 2000, 19: 3171–3191.
23. Royall RM. Robust confidence intervals using maximum likelihood estimators. Int Stat Rev 1986, 54: 221–226.
24. Loveday C, Burke A, Johnson M, et al.Equivalence in virological responses to HAART in patients with HIV-1 subtype non-B infections: a case–controlled study (n = 100). In:8th Conference on Retroviruses and Opportunistic Infections. Chicago, 4–8 February 2001 [Abstract 526].
25. Frater AJ, Beardall A, Ariyoshi K. et al. Impact of baseline polymorphisms in RT and protease on outcome of highly active antiretroviral therapy in HIV-1-infected African patients. AIDS 2001, 15: 1493–1502.
26. Weidle PJ, Kityo CM, Mugyenyi P. et al. Resistance to antiretroviral therapy among patients in Uganda. J Acquired Immune Defic Syndr 2001, 26: 495–500.
27. Adje C, Cheingsong R, Roels TH. et al. High prevalence of genotypic and phenotypic HIV-1 drug-resistant strains among patients receiving antiretroviral therapy in Abidjan, Cote d'Ivoire. J Acquired Immune Defic Syndr 2001, 26: 501–506.
28. Mital D, Pillay D. The impact of HIV-1 subtype on drug-resistant HIV-1 strains. J HIV Ther 2001, 6: 56–60.
29. Del Amo J, Petruckevitch A, Phillips A. et al. Disease progression and survival in HIV-1-infected Africans in London. AIDS 1998, 12: 1203–1209.
30. Miller V, Staszewski S, Sabin C. et al. CD4 lymphocyte count as a predictor of the duration of highly active antiretroviral therapy-induced suppression of human immunodeficiency virus load. J Infect Dis 1999, 180: 530–533.
31. Cinti SK. Adherence to antiretrovirals in HIV disease. AIDS Reader 2000, 10: 709–717.
Keywords:

Africa; antiretroviral therapy; HIV drug resistance; HIV subtypes; viral load

© 2002 Lippincott Williams & Wilkins, Inc.