Answers are grounded in this paper and cite their sources.
To the Editor: According to Uniting the World Against AIDS (UNAIDS), more than 90% of the 2.3 million children worldwide infected by HIV at the end of 2006 currently live in sub-Saharan Africa.1 Most of antiretroviral (ARV) regimens in Africa are represented by the generic formulation combined with lamivudine (3TC), stavudine (d4T), and nevirapine (NVP), for economic and adherence reasons.2 Recent studies showed the feasibility and the efficiency of initiating ARV-based regimen in African HIV-infected children.3-4 However, few data are available among drug resistance profiles in African children infected with HIV non-B subtypes.5-7 In this study, we report on adherence, immunovirological monitoring, and ARV drug resistance profiles in 52 HIV-infected African children attending the Complexe Pédiatrique of Bangui, Central African Republic, and receiving ARV therapy for 6 months, according to the 2006 World Health Organization recommendations.2 The resistance analysis was carried out on the plasma sample collected after 6 months of treatment. HIV protease (PR) and reverse transcriptase gene sequencing was performed by the consensus technique of the Agence Nationale de Recherches sur le SIDA et les hépatites virales, Paris (ANRS).8 Resistance mutations were interpreted according to the 2008 ANRS algorithm (www.hivfrenchresistance.org). HIV-1 subtype was evaluated by comparing the polymerase sequence with consensus sequences using the Stanford database (http://www.hiv.lanl.gov). Adherence was assessed using an empirical questionnaire addressed to the parent or the child, according to the child's age, including the following: (1) the variable α: number of pill(s) forgotten during the period of the last week; (2) the variable β: number of pills taken inappropriately during the period of the last week; and (3) the variable γ: number of days without drug intake during the period of the last week. Quantitative estimation of adherence, Ad, was calculated as follows: AD(%) = (1 − [α/14) + (β/7) + (γ/28)]/3) × 100. The variables α, β, and γ were rounded up to the nearest integer. Finally, the adherence was estimated as very good if Ad ≥ 90%, good if 80% < Ad < 90%, middle if 60% < Ad ≤ 80%, and bad if Ad ≤ 60%. Fifty-four percent (28/52) of the children were females, the median age was 7.9 years (range, 1-16 years) (Table 1). Of the 52 children, 12 (23%) were at stage 2, 38 (73%) at stage 3, and 2 (4%) at stage 4, according to the World Health Organization clinical staging for pediatric AIDS.2 The major ARV therapy used was the association of d4T, 3TC, and NVP. Twenty-four (46%) children received this association as a generic tablet containing the combined formulation and 10 (19%) of 52 children in individual formulations. Values of the children adherence for ARV therapy were estimated as very good in 14 (27%) of 51 children, good in 3 (6%) of 51, middle in 6 (12%) of 51, and bad in 29 (57%) of 51 (Table 1). HIV subtype was determined in 26 children for whom HIV sequences were available. HIV CRF11_cpx was the most represented found in 10 children (38%). HIV viruses were found to be CRF01_AE in 4 (15%) children, subtype A1 in 3 (12%) children, CRF02_AG in 3 (12%) children, CRF06_cpx in 1 (4%) child, a recombinant subtype A1/CRF11_cpx in 1 (4%) child, and subtype A1 in 1 (4%) child. In 4 (15%) of 26 sequences, a defined HIV subtype could not be assessed.TABLE 1: Adherence and Biological Monitoring (CD4 T-Cell Count, HIV-1 RNA Viral Load, and Resistance Genotypes) in 52 HIV-Infected Children Treated in Bangui Over a 6-Month PeriodTABLE 1: (Continued) Adherence and Biological Monitoring (CD4 T-Cell Count, HIV-1 RNA Viral Load, and Resistance Genotypes) in 52 HIV-Infected Children Treated in Bangui Over a 6-Month PeriodAt baseline, the median CD4 T-cell percentage was 6% (range, 1%-18%) and median CD4 T-cell count was 128 cells per cubic millimeter (range, 8-950 cells/mm3) (Table 1). At 6 months of ARV therapy, the median CD4 T-cell percentage was 15% (range, 5%-35%) and median CD4 T-cell count was 420 cells per cubic millimeter (range, 197-1384 cells/mm3) (P < 0.0001) (Table 1). HIV viral load level at 6 months of treatment was below the detection limit of 1.70 log10 copies per milliliter in 13 (25%) children, all showing a very good adherence (≥90%). The median viral load among the remaining 39 children was 3.29 log10 copies per milliliter (range, 1.72-4.91 copies/mL). HIV viral load at 6 months of ARV treatment was found inversely associated with the level of adherence (r = 0.6970; P < 0.0001). Resistance genotypic analysis was assessed in 26 children, for whom PR and reverse transcriptase sequences were available (Table 1). Wild-type viruses were detected in 6 children (23%). Twenty (77%) children exhibited viruses with mutations conferring resistance to at least 1 ARV drug. The most detected resistance mutation was the M184V, found in 18 (69%) of the 26 sequences. Resistant viruses harboring thymidine analogue mutations were detected in 4 (15%) of 26 children. Mutations associated with resistance to nonnucleoside reverse transcriptase inhibitor (NNRTI) were detected in 14 (54%) of 26 sequences. Overall, 28 NNRTI resistance mutations were found in 14 specimens (median, 2; range, 1-4), among which the Y181C mutation was the most frequent (9/14, 64%). Noteworthy, galenic formulation of the mainly prescribed association d4T/3TC/NVP did not influence either the adherence or the frequency of resistance mutations. All the samples exhibited at least 3 mutations associated with resistance to protease inhibitors (median, 6; range, 3-8). Three of the 10 children receiving indinavir showed protease inhibitor-resistant viruses harboring major resistance mutations: M46I in patient 27, V82F in patient 39, and L90M in patient 65. Of interest, 4 PR sequences isolated from patients 30, 83, 141, and 178 showed a “possible resistance” to saquinavir (SQV), according to the ANRS algorithm, since the I15V and I62V mutations were present in all cases added to mutations at codon 10 or codon 20. Overall, among those 26 children, 5 (19%) exhibited plasma viruses resistant to 1 drug of their treatment, 14 (54%) harbored viruses resistant to 2 drugs of their treatment, and 1 (4%) harbored virus resistant to all 3 drugs of its ARV therapy. This study reported on the immunovirological status at 6 months of ARV-based treatment in 52 African HIV-infected children. To our knowledge, this series constitutes a unique evaluation of ARV drug resistance in children harboring such a variety of HIV non-B subtypes (A1, CRF01_AE, CRF02_AG, CRF06_cpx, and CRF11_cpx). In previous reports, the children were infected with a predominant subtype as follows: 95% of CRF02_AG in Chaix et al,5 exclusively subtype C in Averbuch et al,7 and 58% of subtype A1 in Lwembe et al.6 Despite a significant increase in CD4 T-cell count from baseline, HIV viral load remained detectable in 75% of ARV-treated children and a middle or bad adherence was observed in 69% of the children. Furthermore, viruses resistant to at least 1 ARV drug were found in 77% of the children with virological failure. High prevalence of resistance was observed with rates of 69% and 54% for the M184V and the NNRTI-resistance mutations, respectively. In addition, our study underlined a high prevalence of natural polymorphisms in PR gene, at already known codons (10, 20, and 36)9 but also at other resistance-associated positions. In particular, positions 69 and 89 are both found to harbor mutations in all 26 specimens, except one. Of interest, 4 viruses displayed a possible resistance to SQV, suggesting a potential role of the natural polymorphisms in viral susceptibility to SQV. The use of other genotypic interpretation systems, REGA (version 7.1.1) and Stanford (version 4.3.0),10,11 for these cases, both concluded to viruses susceptible to SQV. All these children were infected with HIV CRF11_cpx; to what extent the natural polymorphisms will possibly impact virological response to SQV is currently unknown. A recent report, based on the correlation between baseline PR sequence and virological response to tipranavir, showed a different set of mutations retained in the whole studied population than in the patients infected with HIV subtype B.12 So increasing data among non-B subtypes will actually improve the reliability of the interpretation algorithms, currently mostly based on data from HIV subtype B sequences. Finally, these findings highlight the interest of improving adherence and carry out HIV plasma viral load determination for ARV therapies monitoring of pediatric AIDS in Africa because of their high sensitivity to early diagnose treatment failure development. Jean-Chrysostome Gody, MD*† Charlotte Charpentier, PharmD, PhD‡§ Olivia Mbitikon, MD† Ali Si-Mohamed, PharmD, PhD‡§ Jérome LeGoff, PharmD, PhD‡§ Gérard Grésenguet, MPH, MD, PhD* Laurent Bélec, MPH, MD, PhD‡§ *Faculté des Sciences de la Santé Bangui, République Centrafricaine †Complexe Pédiatrique Bangui, République Centrafricaine ‡Laboratoire de Virologie, Hôpital Européen Georges Pompidou Paris, France §Université Paris V, et Unité Internationale INSERM U743 (Immunologie Humaine) Equipe Immunité et Biothérapie Muqueuse Paris, France Supported by the Association pour la Recherche en Infectiologie, Paris, France. Presented at the XVII International AIDS Conference, August 3-8, 2008, Mexico City, Mexico.
No takes yet. Share an insight, caveat, or question.
Gody et al. (2008) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: