[Skip to Content]
Access to paid content on this site is currently suspended due to excessive activity being detected from your IP address 54.167.142.229. Please contact the publisher to request reinstatement.
Sign In
Individual Sign In
Create an Account
Institutional Sign In
OpenAthens Shibboleth
[Skip to Content Landing]
Download PDF
Figure.
Flow of HIV-Infected Children Randomized to Continue a Ritonavir-Boosted Lopinavir–Based Regimen or Switch to an Efavirenz-Based Regimen
Flow of HIV-Infected Children Randomized to Continue a Ritonavir-Boosted Lopinavir–Based Regimen or Switch to an Efavirenz-Based Regimen

HIV indicates human immunodeficiency virus. “Transferred” indicates children whose care was transferred to other health care facilities because of relocation.

aBecause of an administrative error, this patient was not switched to efavirenz and continued taking ritonavir-boosted lopinavir all the way through the study but was analyzed per intention to treat in the efavirenz group.

Table 1.  
Baseline Characteristics of HIV-Infected Children Randomized to Continue a Ritonavir-Boosted Lopinavir–Based Regimen or Switch to an Efavirenz-Based Regimena
Baseline Characteristics of HIV-Infected Children Randomized to Continue a Ritonavir-Boosted Lopinavir–Based Regimen or Switch to an Efavirenz-Based Regimena
Table 2.  
Cumulative Probabilities of Primary End Points and Tests of Noninferiority Among HIV-Infected Children Randomized to Continue a Ritonavir-Boosted Lopinavir–Based Regimen or Switch to an Efavirenz-Based Regimen
Cumulative Probabilities of Primary End Points and Tests of Noninferiority Among HIV-Infected Children Randomized to Continue a Ritonavir-Boosted Lopinavir–Based Regimen or Switch to an Efavirenz-Based Regimen
Table 3.  
Characteristics of 7 Children With Viral Failurea
Characteristics of 7 Children With Viral Failurea
Table 4.  
Lipid Levels at Baseline and 40 Weeks After Randomization Among HIV-Infected Children Randomized to Continue a Ritonavir-Boosted Lopinavir–Based Regimen or Switch to an Efavirenz-Based Regimena
Lipid Levels at Baseline and 40 Weeks After Randomization Among HIV-Infected Children Randomized to Continue a Ritonavir-Boosted Lopinavir–Based Regimen or Switch to an Efavirenz-Based Regimena
1.
Bolton-Moore  C, Mubiana-Mbewe  M, Cantrell  RA,  et al.  Clinical outcomes and CD4 cell response in children receiving antiretroviral therapy at primary health care facilities in Zambia. JAMA. 2007;298(16):1888-1899.
PubMedArticle
2.
Abrams  EJ, Simonds  RJ, Modi  S,  et al.  PEPFAR scale-up of pediatric HIV services. J Acquir Immune Defic Syndr. 2012;60(suppl 3):S105-S112.
PubMedArticle
3.
Consolidated Guidelines on the Use of Antiretroviral Drugs for Treating and Preventing HIV Infection. Geneva, Switzerland: World Health Organization; 2013.
4.
Palumbo  P, Lindsey  JC, Hughes  MD,  et al.  Antiretroviral treatment for children with peripartum nevirapine exposure. N Engl J Med. 2010;363(16):1510-1520.
PubMedArticle
5.
Violari  A, Lindsey  JC, Hughes  MD,  et al.  Nevirapine vs ritonavir-boosted lopinavir for HIV-infected children. N Engl J Med. 2012;366(25):2380-2389.
PubMedArticle
6.
Sohn  AH, Nuttall  JJ, Zhang  F.  Sequencing of antiretroviral therapy in children in low- and middle-income countries. Curr Opin HIV AIDS. 2010;5(1):54-60.
PubMedArticle
7.
Nachega  JB, Parienti  JJ, Uthman  OA,  et al.  Lower pill burden and once-daily antiretroviral treatment regimens for HIV infection. Clin Infect Dis. 2014;58(9):1297-1307.
PubMedArticle
8.
World Health Organization. Antiretroviral drugs for treating pregnant women and preventing HIV infection in infants. http://whqlibdoc.who.int/publications/2010/9789241599818_eng.pdf?ua=1. Accessed December 14, 2014.
9.
Kuhn  L, Hunt  G, Technau  KG,  et al.  Drug resistance among newly diagnosed HIV-infected children in the era of more efficacious antiretroviral prophylaxis. AIDS. 2014;28(11):1673-1678.
PubMedArticle
10.
Coovadia  A, Abrams  EJ, Stehlau  R,  et al.  Reuse of nevirapine in exposed HIV-infected children after protease inhibitor-based viral suppression. JAMA. 2010;304(10):1082-1090.
PubMedArticle
11.
Kuhn  L, Coovadia  A, Strehlau  R,  et al.  Switching children previously exposed to nevirapine to nevirapine-based treatment after initial suppression with a protease-inhibitor-based regimen [published online March 16, 2012]. Lancet Infect Dis. 2012;12(7):521-530. doi:10.1016/S473-3099(12)70051-8.
PubMedArticle
12.
Moorthy  A, Kuhn  L, Coovadia  A,  et al.  Induction therapy with protease-inhibitors modifies the effect of nevirapine resistance on virologic response to nevirapine-based HAART in children. Clin Infect Dis. 2011;52(4):514-521.
PubMedArticle
13.
Goodman  R.  The Strengths and Difficulties Questionnaire. J Child Psychol Psychiatry. 1997;38(5):581-586.
PubMedArticle
14.
Pillay  V, Ledwaba  J, Hunt  G,  et al.  Antiretroviral drug resistance surveillance among drug-naive HIV-1-infected individuals in Gauteng Province, South Africa in 2002 and 2004. Antivir Ther. 2008;13(suppl 2):101-107.
PubMed
15.
Strehlau  R, Martens  L, Coovadia  A,  et al.  Absence seizures associated with efavirenz initiation. Pediatr Infect Dis J. 2011;30(11):1001-1003.
PubMedArticle
16.
Hirt  D, Urien  S, Olivier  M,  et al.  Is the recommended dose of efavirenz optimal in young West African HIV-infected children? Antimicrob Agents Chemother. 2009;53(10):4407-4413.
PubMedArticle
17.
Musiime  V, Fillekes  Q, Kekitiinwa  A,  et al.  The pharmacokinetics and acceptability of lopinavir/ritonavir minitab sprinkles, tablets, and syrups in African HIV-infected children. J Acquir Immune Defic Syndr. 2014;66(2):148-154.
PubMed
18.
Musiime  V, Kendall  L, Bakeera-Kitaka  S,  et al.  Pharmacokinetics and acceptability of once- vs twice-daily lamivudine and abacavir in HIV type-1-infected Ugandan children in the ARROW Trial. Antivir Ther. 2010;15(8):1115-1124.
PubMedArticle
19.
Ramjan  R, Calmy  A, Vitoria  M,  et al.  Systematic review and meta-analysis: patient and programme impact of fixed-dose combination antiretroviral therapy. Trop Med Int Health. 2014;19(5):501-513.
PubMedArticle
20.
Global Tuberculosis Report. Geneva, Switzerland: World Health Organization; 2014.
21.
Martinson  NA, Moultrie  H, van Niekerk  R,  et al.  HAART and risk of tuberculosis in HIV-infected South African children. Int J Tuberc Lung Dis. 2009;13(7):862-867.
PubMed
22.
Guidelines for the Management of HIV in Children. 2nd ed. Pretoria: National Department of Health, South Africa; 2010.
23.
Frohoff  C, Moodley  M, Fairlie  L,  et al.  Antiretroviral therapy outcomes in HIV-infected children after adjusting protease inhibitor dosing during tuberculosis treatment. PLoS One. 2011;6(2):e17273.
PubMedArticle
24.
Reitz  C, Coovadia  A, Ko  S,  et al.  Initial response to protease-inhibitor-based antiretroviral therapy among children less than 2 years of age in South Africa. J Infect Dis. 2010;201(8):1121-1131.
PubMedArticle
25.
Zanoni  BC, Phungula  T, Zanoni  HM,  et al.  Impact of tuberculosis cotreatment on viral suppression rates among HIV-positive children initiating HAART. AIDS. 2011;25(1):49-55.
PubMedArticle
26.
Soeters  HM, Napravnik  S, Patel  MR,  et al.  The effect of tuberculosis treatment on virologic and CD4+ cell count response to combination antiretroviral therapy. AIDS. 2014;28(2):245-255.
PubMedArticle
27.
Grennan  JT, Loutfy  MR, Su  D,  et al.  Magnitude of virologic blips is associated with a higher risk for virologic rebound in HIV-infected individuals. J Infect Dis. 2012;205(8):1230-1238.
PubMedArticle
28.
Dahri  K, Ensom  MH.  Efavirenz and nevirapine in HIV-1 infection. Clin Pharmacokinet. 2007;46(2):109-132.
PubMedArticle
29.
Decloedt  EH, Maartens  G.  Neuronal toxicity of efavirenz. Expert Opin Drug Saf. 2013;12(6):841-846.
PubMedArticle
30.
Paredes  R, Marconi  VC, Lockman  S,  et al.  Impact of antiretroviral drugs in pregnant women and their children in Africa. J Infect Dis. 2013;207(suppl 2):S93-S100.
PubMedArticle
31.
Ton  Q, Frenkel  L.  HIV drug resistance in mothers and infants following use of antiretrovirals to prevent mother-to-child transmission. Curr HIV Res. 2013;11(2):126-136.
PubMedArticle
32.
Roberts  T, Bygrave  H, Fajardo  E, Ford  N.  Challenges and opportunities for the implementation of virological testing in resource-limited settings. J Int AIDS Soc. 2012;15(2):17324.
PubMedArticle
33.
Basson  AE, Rhee  SY, Parry  CM,  et al.  Impact of drug resistance-associated amino acid changes in HIV-1 subtype C on susceptibility to newer NNRTIs. Antimicrob Agents Chemother. 2015;59(2):960-971.
PubMedArticle
34.
Mukonzo  JK.  The challenge of paediatric efavirenz dosing. AIDS. 2014;28(13):1855-1857.
PubMedArticle
Original Investigation
November 3, 2015

Efavirenz-Based Antiretroviral Therapy Among Nevirapine-Exposed HIV-Infected Children in South AfricaA Randomized Clinical Trial

Author Affiliations
  • 1Empilweni Services and Research Unit, Rahima Moosa Mother and Child Hospital, Faculty of Health Sciences, University of the Witwatersrand, Johannesburg, South Africa
  • 2ICAP, Mailman School of Public Health, Columbia University, New York, New York
  • 3Department of Epidemiology, Mailman School of Public Health, Columbia University, New York, New York
  • 4Gertrude H. Sergievsky Center, College of Physicians and Surgeons, Columbia University, New York, New York
  • 5Center for HIV and STIs, National Institute for Communicable Diseases, Johannesburg, South Africa
  • 6Department of Biostatistics, Mailman School of Public Health, Columbia University, New York, New York
JAMA. 2015;314(17):1808-1817. doi:10.1001/jama.2015.13631
Abstract

Importance  Advantages of using efavirenz as part of treatment for children infected with human immunodeficiency virus (HIV) include once-daily dosing, simplification of co-treatment for tuberculosis, preservation of ritonavir-boosted lopinavir for second-line treatment, and harmonization of adult and pediatric treatment regimens. However, there have been concerns about possible reduced viral efficacy of efavirenz in children exposed to nevirapine for prevention of mother-to-child transmission.

Objective  To evaluate whether nevirapine-exposed children achieving initial viral suppression with ritonavir-boosted lopinavir–based therapy can transition to efavirenz-based therapy without risk of viral failure.

Design, Setting, and Participants  Randomized, open-label noninferiority trial conducted at Rahima Moosa Mother and Child Hospital, Johannesburg, South Africa, from June 2010 to December 2013, enrolling 300 HIV-infected children exposed to nevirapine for prevention of mother-to-child transmission who were aged 3 years or older and had plasma HIV RNA of less than 50 copies/mL during ritonavir-boosted lopinavir–based therapy; 298 were randomized and 292 (98%) were followed up to 48 weeks after randomization.

Interventions  Participants were randomly assigned to switch to efavirenz-based therapy (n = 150) or continue ritonavir-boosted lopinavir–based therapy (n = 148).

Main Outcomes and Measures  Risk difference between groups in (1) viral rebound (ie, ≥1 HIV RNA measurement of >50 copies/mL) and (2) viral failure (ie, confirmed HIV RNA >1000 copies/mL) with a noninferiority bound of −0.10. Immunologic and clinical responses were secondary end points.

Results  The Kaplan-Meier probability of viral rebound by 48 weeks was 0.176 (n = 26) in the efavirenz group and 0.284 (n = 42) in the ritonavir-boosted lopinavir group. Probabilities of viral failure were 0.027 (n = 4) in the efavirenz group and 0.020 (n = 3) in the ritonavir-boosted lopinavir group. The risk difference for viral rebound was 0.107 (1-sided 95% CI, 0.028 to ∞) and for viral failure was −0.007 (1-sided 95% CI, −0.036 to ∞). We rejected the null hypothesis that efavirenz is inferior to ritonavir-boosted lopinavir (P < .001) for both end points. By 48 weeks, CD4 cell percentage was 2.88% (95% CI, 1.26%-4.49%) higher in the efavirenz group than in the ritonavir-boosted lopinavir group.

Conclusions and Relevance  Among HIV-infected children exposed to nevirapine for prevention of mother-to-child transmission and with initial viral suppression with ritonavir-boosted lopinavir–based therapy, switching to efavirenz-based therapy compared with continuing ritonavir-boosted lopinavir–based therapy did not result in significantly higher rates of viral rebound or viral failure. This therapeutic approach may offer advantages in children such as these.

Trial Registration  clinicaltrials.gov Identifier: NCT01146873

Introduction

Implementation of pediatric antiretroviral treatment (ART) programs in sub-Saharan Africa has resulted in significant reductions in morbidity and mortality among children infected with human immunodeficiency virus (HIV), changing a rapidly fatal disease into a chronic condition.1 The success of ART programs in low-resource settings has been attributed to a public health approach whereby standardized population guidelines facilitate individual patient management.2 For infants and young children, ritonavir-boosted lopinavir–based therapy is recommended as first-line ART.3 Initially, ritonavir-boosted lopinavir was recommended only for infants exposed to nevirapine for prevention of mother-to-child transmission (PMTCT), but it later was shown to also have better virologic efficacy in unexposed infants and young children.4,5 In adults and older children, efavirenz is recommended as part of first-line ART.3

For HIV-infected children older than 3 years, efavirenz has advantages for long-term maintenance therapy. Recommending efavirenz for older children would harmonize their regimen with adult guidelines and reduce the cost of national programs. Efavirenz may avoid some of the metabolic toxicities associated with ritonavir-boosted lopinavir and simplifies co-treatment for tuberculosis.6 Ritonavir-boosted lopinavir has an unpleasant taste, presenting major adherence challenges for parents administering this drug in syrup form to their children who are still too young to swallow tablets.6 Efavirenz has the advantage of once-daily dosing, which has been shown to improve adherence and virologic outcome.7

Nonnucleoside reverse transcriptase inhibitors continue to be recommended for PMTCT. This includes efavirenz or nevirapine as part of maternal therapy and infant nevirapine prophylaxis, which is recommended regardless of maternal regimen.3,8 With improved PMTCT coverage, the majority of the (albeit shrinking number of) children who acquire HIV infection have resistance to nonnucleoside reverse transcriptase inhibitors prior to starting therapy.9 We previously evaluated whether children who initially started ritonavir-boosted lopinavir–based therapy could safely transition to nevirapine-based therapy soon after achieving viral load suppression. Our results supported the clinical utility of this strategy with some caveats. Resistance selected during PMTCT led to a higher rate of virologic failure in the group transitioning to nevirapine.1012 In the new trial presented herein, we evaluated whether a switch to efavirenz can overcome this limitation. Specifically, we tested among children perinatally exposed to nevirapine as part of PMTCT whether those whose viral load was initially suppressed by ritonavir-boosted lopinavir–based therapy could transition to efavirenz-based therapy without increased risk of viral failure.

Methods
Study Design

We conducted a randomized, open-label noninferiority trial between June 2010 and December 2013 at Rahima Moosa Mother and Child Hospital in Johannesburg, South Africa (trial protocol available in Supplement 1). Children were randomized to switch to efavirenz-based therapy or to continue ritonavir-boosted lopinavir–based therapy and were followed up for 48 weeks after randomization. The noninferiority design was chosen because efavirenz was not expected to have better virologic outcomes than the standard regimen. The study was approved by the institutional review boards of Columbia University and the University of the Witwatersrand. Children’s mothers or legal guardians provided written informed consent.

Children were eligible for enrollment if they had nevirapine exposure as part of PMTCT, were currently receiving ritonavir-boosted lopinavir–based therapy started prior to age 36 months and continued for at least 1 year, and had an HIV RNA measurement of less than 50 copies/mL. All children in the control group of our prior trial10,11 who were still in follow-up were screened for eligibility. In addition, clinicians responsible for the care of HIV-infected children at other clinics in the area were approached about referring children meeting our eligibility criteria. Random assignments were generated by the study statistician using a permuted-block design with block sizes between 8 and 12 and were concealed in opaque envelopes opened on site at the time of randomization. Children were followed up at 4, 8, 16, 24, 32, 40, and 48 weeks after randomization.

Drug Regimens

Efavirenz was prescribed once daily in the evening at 200 mg for weights of 10 kg to 13.9 kg (22-30 lb) and 300 mg for weights of 14 kg to 24.9 kg (31-55 lb). Efavirenz was available in 50-mg and 200-mg capsules. If children were unable to swallow capsules, caregivers were shown how to open the capsules and dissolve the contents in water. Ritonavir-boosted lopinavir syrup was given twice per day at 230 mg/m2 per dose. Children able to swallow tablets were given 1 tablet twice per day (200 mg lopinavir/50 mg ritonavir) if body surface area was less than 0.9 m2 or 2 tablets twice per day if body surface area was 0.9 m2 or higher. Both groups received adherence counseling at the time of randomization and at each study visit.

At the time the study was undertaken, local guidelines advised against the use of stavudine for new patients but provided no guidance for those already receiving it. At enrollment, children who were receiving stavudine were screened for eligibility for a substudy of preemptive switching to abacavir compared with continuing stavudine. Children not eligible for the substudy continued to receive the other 2 antiretroviral drugs that they were already receiving. Stavudine was given at 1 mg/kg twice daily and abacavir at 8 mg/kg twice daily. Some children were receiving zidovudine (180 mg/m2 twice daily). Lamivudine (4 mg/kg twice daily) was used as the third drug for all children. All medications were dose-adjusted at every visit based on growth.

Study End Points

Quantity of HIV RNA in plasma was measured at 4, 8, 16, 24, and 48 weeks. Based on the results of our prior trial,10,11 we identified 2 primary virologic end points: (1) viral rebound, defined as 1 or more HIV-1 RNA measurements of greater than 50 copies/mL, and (2) viral failure, defined as confirmed (ie, ≥2) HIV-1 RNA measurements of greater than 1000 copies/mL by 48 weeks after randomization. All children with greater than 50 copies/mL of HIV-1 RNA at a scheduled study visit were recalled for a repeat test.

CD4 cell counts and CD4 cell percentages were measured at baseline and 24 and 48 weeks. Complete blood cell count was performed at baseline and 24 weeks and alanine aminotransferase was measured at baseline and 32 weeks. A fasting lipid panel was performed at baseline and 40 weeks. Weight and height, concomitant medications, and other clinical conditions were recorded at each visit. The Strengths and Difficulties Questionnaire, a validated standardized screening questionnaire of emotional/behavioral problems,13 was administered at baseline and 40 weeks. Caregivers were asked to return all unused medications at each study visit. These were reconciled by the pharmacist with the expected use of each drug as a measure of adherence.

Plasma HIV RNA measurements (AmpliPrep/COBAS TaqMan HIV-1 Test, version 2.0, Roche), CD4 cell determinations, blood counts, and liver function tests were conducted by Clinical Laboratory Services in Johannesburg and reported directly to the site for use in clinical management. The quantification range of the HIV RNA assay was 20 to 10 000 000 copies/mL. All samples with more than 1000 copies/mL were tested at the National Institute for Communicable Diseases for drug resistance using population sequencing as previously described.14

Statistical Analysis

The sample size of 300 was selected to detect a noninferiority bound of −0.15 around the risk difference for viral rebound (end point 1) and a bound of −0.11 for viral failure (end point 2). Our prior trial10,11 observed the risk of viral rebound to be 0.55, which means that based on our noninferiority bound, we were prepared to tolerate a risk of 0.70 or less in the intervention group. At a data and safety monitoring board (DSMB) review of a planned interim analysis, it was noted that virologic end points were less common than anticipated and that for end point 2, the stopping criteria had been attained; ie, noninferiority at the prespecified bound had been confirmed. The DSMB requested recalculation of expected noninferiority bounds based on the interim results. These were recalculated to be −0.10 for both end points. The DSMB advised completion of the study with this narrower noninferiority bound, which would allow for stronger conclusions to be drawn from the noninferiority analysis.

Intention-to-treat analyses were conducted using all available follow-up data. The cumulative probabilities of virologic end points were calculated using Kaplan-Meier methods. Follow-up time for children not followed up to 48 weeks was censored at their last follow-up visit. To address the noninferiority design, the risk difference for each end point was calculated as the difference in Kaplan-Meier probabilities. The standard errors of the difference were calculated with the Δ method. P values for the noninferiority analysis were calculated from a 1-sided t test and tested the null hypothesis that Δ (probability in the control group − probability in the efavirenz group) was −0.10 or less. The threshold to define significance was P = .05. Differences in outcomes between the randomized groups by variation in the other antiretroviral drugs contained in the regimens were investigated in stratified analyses. Other outcomes were compared across groups using t tests for continuous variables and χ2 or Fisher exact tests for categorical variables. For comparison of adherence outcomes between the 2 groups, generalized estimating equation models were used. All P values other than for the noninferiority analysis were 2-sided and P<.05 was considered statistically significant. Weight- and height-for-age z scores were calculated using World Health Organization software. Analyses were performed using SAS, version 9.1.3 (SAS Institute Inc).

Results
Study Population

A total of 300 children were enrolled in the trial, 223 of 236 children referred from pediatric HIV clinics in the area and 77 of 85 children from the control group of our prior trial.10,11 Two children discontinued the study prior to randomization, resulting in 150 children randomly assigned to switch to efavirenz-based therapy and 148 to continue ritonavir-boosted lopinavir–based therapy (Figure).

At randomization, children had started ART at an average of 9.3 months of age (range, 3 weeks to 32 months), had been receiving ART for an average of 3.5 years, and were an average of 4.3 years of age. Most (73.5%) had been exposed to both maternal and infant nevirapine for PMTCT, the remainder to either maternal or infant nevirapine. Fifty-three percent of the study population was female. Other characteristics are shown in Table 1. No children died in the 48 weeks after randomization, and retention in the study was excellent, with 292 (98%) of 298 followed up through 48 weeks. All 6 children not retained through the end of the study were in the efavirenz group. All of these dropouts were due to relocations out of the area, with no apparent link to randomized group.

Primary End Points

The Kaplan-Meier probability of viral rebound of greater than 50 copies/mL by 48 weeks was 0.176 (n = 26) in the efavirenz group and 0.284 (n = 42) in the ritonavir-boosted lopinavir group. Probabilities of viral failure (ie, confirmed HIV RNA >1000 copies/mL) were 0.027 (n = 4) in the efavirenz group and 0.020 (n = 3) in the ritonavir-boosted lopinavir group (Table 2). For viral rebound of greater than 50 copies/mL, the risk difference was 0.107 (1-sided 95% CI, 0.028 to ∞), and for viral failure, −0.007 (1-sided 95% CI, −0.036 to ∞). The lower bounds of both of these 1-sided 95% CIs exceeded the −0.10 prespecified noninferiority bound. We rejected the null hypothesis that efavirenz is inferior to ritonavir-boosted lopinavir (P < .001) for both end points, accepting the alternative hypothesis that efavirenz, relative to a standard approach using ritonavir-boosted lopinavir, is noninferior for both primary end points.

Details on the 7 children with viral failure, including 4 in the efavirenz group and 3 in the ritonavir-boosted lopinavir group, are shown in Table 3. Two of the 4 children in the efavirenz group resumed a ritonavir-boosted lopinavir–based regimen and later experienced viral resuppression. Both children had nonnucleoside reverse transcriptase inhibitor (K103N) and nucleoside reverse transcriptase inhibitor (M184V/I) resistance. Viral elevation in 1 child occurred following treatment interruption due to an elevated alanine aminotransferase level. After resolution of the hepatitis and resumption of the efavirenz-based regimen, the child experienced resuppression. This child had no resistance detected. The fourth child in the efavirenz group had viral failure in association with severe household disruption. Both K103N and M184V resistance mutations were detected. The child was transitioned to lamivudine monotherapy as a holding therapy until adherence could be attained. The child did not reach a point at which a suppressive regimen could be reintroduced and did not complete follow-up. All children in the ritonavir-boosted lopinavir group with viral failure had resuppression without regimen change despite 2 of 3 children having M184V mutations and the other having an E138A mutation (Table 3).

There were no differences in viral outcomes in those receiving stavudine vs abacavir. Effects of efavirenz vs ritonavir-boosted lopinavir were also unchanged when stratified by stavudine vs abacavir.

Secondary Outcomes

The lipid profile of children in the efavirenz group was better than that of the ritonavir-boosted lopinavir group 40 weeks after randomization. Children in the efavirenz group had lower total cholesterol, low-density lipoprotein cholesterol, and triglyceride levels than those in the ritonavir-boosted lopinavir group (Table 4).

Both groups remained within the normal range for CD4 cell percentages and CD4 cell counts. At baseline and at 24 weeks, there were no differences in CD4 cell percentages or CD4 cell counts between the groups. By 48 weeks, CD4 cell percentage was 2.88% (95% CI, 1.26%-4.49%) higher in the efavirenz group than in the ritonavir-boosted lopinavir group. There were no significant differences in weight- or height-for-age z scores between groups at visits after randomization. Neither anemia nor neutropenia were more common in the efavirenz group. Alanine aminotransferase elevations were more common in the efavirenz group but were primarily of grade 1 or 2. Skin manifestations did not differ across the groups. There were no deaths in either group, and hospital admissions were rare. Two children in the efavirenz group had tuberculosis medication initiated after randomization (eTable in Supplement 2).

Four weeks after randomization, 26% of children in the efavirenz group reported trouble sleeping or having nightmares compared with no children in the ritonavir-boosted lopinavir group (P < .001). However, this difference between groups was no longer present after 8 weeks through the end of the study. There were no significant differences between the groups in behavioral problems on the Strengths and Difficulties Questionnaire, although a high proportion of children in both groups had an abnormal Strengths and Difficulties Questionnaire total difficulties score (30.8% vs 34.0%; P = .30). Nausea was reported slightly more frequently in the efavirenz group (28.0% vs 17.6%; P = .03). Other symptoms were similar across the groups (eTable in Supplement 2).

Two children experienced seizure disorders suspected to be related to efavirenz. One child was diagnosed as having absence seizures believed to be related to delayed efavirenz clearance due to cytochrome P450 (CYP) 2B6 516 TT homozygosity, which resolved after discontinuing the drug (previously reported15). The other child was diagnosed as having generalized tonic-clonic seizures attributed to abnormally high efavirenz blood levels of 20 mg/L (reference range, 1-4 mg/L)16 resulting from 2 genetic mutations—CYP2B6 516G>T and CYP2B6 785A>G. The seizures resolved after efavirenz was stopped and ritonavir-boosted lopinavir was restarted.

Across all follow-up visits, 73.2% of children in the ritonavir-boosted lopinavir group returned the medication containers for adherence calculations and 86.1% of children in the efavirenz group did so. Nonadherence, defined as returning more than 10% of the expected volume of either ritonavir-boosted lopinavir or efavirenz, was similar in the 2 groups at 12.9% and 15.8%, respectively (P = .23). Nonadherence to the other 2 drugs in the regimen was also similar in the 2 groups (eTable in Supplement 2).

Discussion

Switching to efavirenz-based therapy compared with continuing ritonavir-boosted lopinavir–based therapy did not result in significantly higher rates of viral rebound (ie, HIV RNA >50 copies/mL) or viral failure (ie, confirmed HIV RNA >1000 copies/mL) in this cohort of nevirapine-exposed children achieving initial viral suppression with ritonavir-boosted lopinavir.

There are several potential advantages of switching to efavirenz, including preserving ritonavir-boosted lopinavir for second-line treatment, harmonizing pediatric and adult treatment guidelines, and reducing the cost of national programs. Ritonavir-boosted lopinavir syrup is a medication of poor palatability that has to be dosed twice per day and requires cold storage to maintain long-term stability.6 New formulations will address some but not all of these limitations.17 Efavirenz offers a more adherence-friendly formulation, including being dosed once daily. Because abacavir and lamivudine can also be used once daily in children who already have viral suppression,18 a once-daily pediatric fixed-dose combination could be formulated. In adults, simplifications in formulations have had significant adherence benefits.19

Tuberculosis is a common coinfection among children with HIV in South Africa.20,21 Rifampicin is a potent inducer of the CYP enzyme class. The use of ritonavir-boosted lopinavir as part of pediatric ART is problematic in that lopinavir is metabolized by these enzymes, which may in turn lead to diminished virologic efficacy. World Health Organization guidelines for co-treatment of tuberculosis and HIV in children aged 3 years or older recommend addressing this issue by substituting efavirenz or a third nucleoside reverse transcriptase inhibitor for ritonavir-boosted lopinavir. Alternatively, the guidelines also suggest the option of using ritonavir-boosted lopinavir combined with additional ritonavir to maintain adequate lopinavir concentrations; the South African pediatric treatment guidelines recommend this latter approach.3,22 The poor palatability of ritonavir-boosted lopinavir is made considerably worse by the introduction of additional ritonavir syrup, thus complicating adherence and possibly affecting virologic suppression. Reduced viral suppression in young children receiving tuberculosis co-treatment with a ritonavir-boosted lopinavir–based regimen has been previously reported.2326 Although efavirenz is also metabolized by the same hepatic enzyme system, there does not appear to be a requirement to increase its dose, nor is there a need for any additional drug to boost its levels, thereby making this regimen option the more suitable alternative.

Long-term use of ritonavir-boosted lopinavir also raises concerns about metabolic toxicities, and our trial demonstrated a more favorable lipid profile with efavirenz, with lower levels of total cholesterol, low-density lipoprotein cholesterol, and triglycerides. Our data suggest that further benefits of efavirenz include less frequent low-level viremia and more robust CD4 cell response. We previously noted both of these benefits with nevirapine relative to ritonavir-boosted lopinavir–based therapy.10,11 Although this study was designed as a noninferiority study, we observed higher rates of low-level viral rebound in the ritonavir-boosted lopinavir group. One study has suggested that low-level viremia may portend a higher risk of future virologic failure.27 The dose of ritonavir-boosted lopinavir used in this study was on the lower end of what is now recommended, which may have played a role in low-level viremia. No statistically significant differences in adherence were observed between the groups.

Neuropsychiatric adverse effects of efavirenz are well known. Efavirenz adverse events are generally associated with high drug concentrations, and specific genetic mutations predispose to inadequate drug clearance.28 In this study, 2 children developed seizure disorders while receiving efavirenz; both instances were likely due to these issues and resolved with change in treatment. Almost a quarter of children reported having had nightmares or trouble sleeping 4 weeks after switching to efavirenz, but these adverse events subsequently resolved. Efavirenz is associated with psychiatric adverse events in adults, but its manifestations in children are less well described.29 Reassuringly, on a standardized instrument assessing emotional/behavioral problems, there was no additional evidence of increased risk. In summary, although for the majority of children the drug was safe and well tolerated, clinical vigilance is necessary.

We believe it is unlikely that shifts away from short-duration nevirapine for PMTCT to wider use of efavirenz-based ART for mothers combined with longer durations of nevirapine prophylaxis for infants in the current era will affect the generalizability of our results. This is because a single base-pair mutation is sufficient to confer resistance to efavirenz. There is now an extensive body of literature that almost all infected infants exposed to even 1 dose of nevirapine will have at least 1 of these mutations.30,31 Hence, there is a ceiling effect that longer durations of nevirapine cannot plausibly exceed. One would need to speculate that resistance selected in the current era is qualitatively different either in persistence or in mutation mix, but this has not been observed in recent surveillance data.9 Nevertheless, ongoing monitoring of the resistance profile of children acquiring infection in the current era is warranted. Our strategy requires viral load monitoring to identify children who may benefit from switching. Postswitch virologic monitoring is also advisable. We are conducting a follow-up study of the children in the trial reported herein. Our strategy fits well with efforts to expand access to viral load testing within treatment programs in low-resource settings.32

Three of 4 children in the efavirenz group with viral failure had K103N mutations. They also had M184V/I mutations, and 2 of 3 children in the ritonavir-boosted lopinavir group with viral failure had M184V mutations. All children in the ritonavir-boosted lopinavir group experienced resuppression with the same regimen (including lamivudine) with adherence counseling. Both children in the efavirenz group who switched back to ritonavir-boosted lopinavir had resuppression. It is interesting that the K103N mutation predominated in viral failure in these children because all studies of PMTCT-exposed infants prior to ART initiation have observed a predominance of the Y181C mutation. This contrasts to patterns observed among their mothers and other adults.30,31 The Y181C mutation confers high-level resistance to nevirapine but only intermediate resistance to efavirenz.33 This may explain why we observed excellent efficacy of efavirenz in this trial but more viral failure in a prior trial attempting the reintroduction of nevirapine following induction of viral suppression with ritonavir-boosted lopinavir,10,11 both trials among nevirapine-exposed children.

Several limitations should be considered in interpreting this study. All children in this trial were older than 3 years, and although the eligibility criteria required only a year of treatment, the mean duration of treatment was 3.5 years. Thus, we are not able to determine whether such a long initial period of treatment with ritonavir-boosted lopinavir is necessary. A further limitation is that these results cannot be generalized to children younger than 3 years. We designed our trial before efavirenz dosing for children younger than 3 years was available, and dosing below this age remains controversial given the volatility in drug metabolism related to enzyme system maturation.34 Third, the treatment strategy was designed for a population achieving suppression with a ritonavir-boosted lopinavir–based regimen, and we cannot address the generalizability of this approach to children in whom ritonavir-boosted lopinavir therapy fails.

There is little guidance available as to what clinicians ought to do when confronted with a child older than 3 years who has begun treatment with ritonavir-boosted lopinavir. As a result, it has been left to individual interpretation, and there are anecdotal reports of clinicians switching to efavirenz in the absence of data to support such a practice. This study provides evidence to support the safety and efficacy of switching to efavirenz, the recommended drug for children older than 3 years, among children with viral suppression.

Conclusions

Among HIV-infected children exposed to nevirapine for PMTCT and achieving initial viral suppression with ritonavir-boosted lopinavir–based therapy, switching to efavirenz-based therapy compared with continuing ritonavir-boosted lopinavir–based therapy did not result in significantly higher rates of viral rebound or viral failure.

Back to top
Article Information

Corresponding Author: Louise Kuhn, PhD, Gertrude H. Sergievsky Center, College of Physicians and Surgeons, Columbia University, 630 W 168th St, New York, NY 10032 (lk24@columbia.edu).

Author Contributions: Drs Coovadia and Kuhn had full access to all of the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis.

Study concept and design: Coovadia, Abrams, Kuhn.

Acquisition, analysis, or interpretation of data: All authors.

Drafting of the manuscript: Coovadia, Kuhn.

Critical revision of the manuscript for important intellectual content: All authors.

Statistical analysis: Shiau, Tsai, Kuhn.

Obtained funding: Coovadia, Abrams, Kuhn.

Administrative, technical, or material support: Coovadia, Abrams, Hunt, Kuhn.

Study supervision: Coovadia, Abrams, Kuhn.

Conflict of Interest Disclosures: All authors have completed and submitted the ICMJE Form for Disclosure of Potential Conflicts of Interest and none were reported.

Funding/Support: The study was supported by grant HD061255 from the Eunice Kennedy Shriver National Institute of Child Health and Human Development.

Role of the Funder/Sponsor: The funders had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; or decision to submit the manuscript for publication.

Previous Presentation: An abstract of this report was presented at the Conference on Retroviruses and Opportunistic Infections; Boston, Massachusetts; March 5, 2014.

Additional Contributions: We thank Lynne Mofenson, MD (formerly of the Eunice Kennedy Shriver National Institute for Child Health and Human Development), for assistance with the study, as well as the members of the data and safety monitoring board: Mark Cotton, MD (Stellenbosch University), Brian Eley, MD (University of Cape Town), Mary-Glenn Fowler, MD (Johns Hopkins University), Carl Lombard, PhD (South African Medical Research Council), Paul E. Palumbo, MD (Geisel School of Medicine at Dartmouth), and Avy Violari, MD (Perinatal HIV Research Unit). The named individuals were not compensated for their participation. We also thank all members of the study team in South Africa and New York. We appreciate the dedication and commitment to the study of the participants and their caregivers.

References
1.
Bolton-Moore  C, Mubiana-Mbewe  M, Cantrell  RA,  et al.  Clinical outcomes and CD4 cell response in children receiving antiretroviral therapy at primary health care facilities in Zambia. JAMA. 2007;298(16):1888-1899.
PubMedArticle
2.
Abrams  EJ, Simonds  RJ, Modi  S,  et al.  PEPFAR scale-up of pediatric HIV services. J Acquir Immune Defic Syndr. 2012;60(suppl 3):S105-S112.
PubMedArticle
3.
Consolidated Guidelines on the Use of Antiretroviral Drugs for Treating and Preventing HIV Infection. Geneva, Switzerland: World Health Organization; 2013.
4.
Palumbo  P, Lindsey  JC, Hughes  MD,  et al.  Antiretroviral treatment for children with peripartum nevirapine exposure. N Engl J Med. 2010;363(16):1510-1520.
PubMedArticle
5.
Violari  A, Lindsey  JC, Hughes  MD,  et al.  Nevirapine vs ritonavir-boosted lopinavir for HIV-infected children. N Engl J Med. 2012;366(25):2380-2389.
PubMedArticle
6.
Sohn  AH, Nuttall  JJ, Zhang  F.  Sequencing of antiretroviral therapy in children in low- and middle-income countries. Curr Opin HIV AIDS. 2010;5(1):54-60.
PubMedArticle
7.
Nachega  JB, Parienti  JJ, Uthman  OA,  et al.  Lower pill burden and once-daily antiretroviral treatment regimens for HIV infection. Clin Infect Dis. 2014;58(9):1297-1307.
PubMedArticle
8.
World Health Organization. Antiretroviral drugs for treating pregnant women and preventing HIV infection in infants. http://whqlibdoc.who.int/publications/2010/9789241599818_eng.pdf?ua=1. Accessed December 14, 2014.
9.
Kuhn  L, Hunt  G, Technau  KG,  et al.  Drug resistance among newly diagnosed HIV-infected children in the era of more efficacious antiretroviral prophylaxis. AIDS. 2014;28(11):1673-1678.
PubMedArticle
10.
Coovadia  A, Abrams  EJ, Stehlau  R,  et al.  Reuse of nevirapine in exposed HIV-infected children after protease inhibitor-based viral suppression. JAMA. 2010;304(10):1082-1090.
PubMedArticle
11.
Kuhn  L, Coovadia  A, Strehlau  R,  et al.  Switching children previously exposed to nevirapine to nevirapine-based treatment after initial suppression with a protease-inhibitor-based regimen [published online March 16, 2012]. Lancet Infect Dis. 2012;12(7):521-530. doi:10.1016/S473-3099(12)70051-8.
PubMedArticle
12.
Moorthy  A, Kuhn  L, Coovadia  A,  et al.  Induction therapy with protease-inhibitors modifies the effect of nevirapine resistance on virologic response to nevirapine-based HAART in children. Clin Infect Dis. 2011;52(4):514-521.
PubMedArticle
13.
Goodman  R.  The Strengths and Difficulties Questionnaire. J Child Psychol Psychiatry. 1997;38(5):581-586.
PubMedArticle
14.
Pillay  V, Ledwaba  J, Hunt  G,  et al.  Antiretroviral drug resistance surveillance among drug-naive HIV-1-infected individuals in Gauteng Province, South Africa in 2002 and 2004. Antivir Ther. 2008;13(suppl 2):101-107.
PubMed
15.
Strehlau  R, Martens  L, Coovadia  A,  et al.  Absence seizures associated with efavirenz initiation. Pediatr Infect Dis J. 2011;30(11):1001-1003.
PubMedArticle
16.
Hirt  D, Urien  S, Olivier  M,  et al.  Is the recommended dose of efavirenz optimal in young West African HIV-infected children? Antimicrob Agents Chemother. 2009;53(10):4407-4413.
PubMedArticle
17.
Musiime  V, Fillekes  Q, Kekitiinwa  A,  et al.  The pharmacokinetics and acceptability of lopinavir/ritonavir minitab sprinkles, tablets, and syrups in African HIV-infected children. J Acquir Immune Defic Syndr. 2014;66(2):148-154.
PubMed
18.
Musiime  V, Kendall  L, Bakeera-Kitaka  S,  et al.  Pharmacokinetics and acceptability of once- vs twice-daily lamivudine and abacavir in HIV type-1-infected Ugandan children in the ARROW Trial. Antivir Ther. 2010;15(8):1115-1124.
PubMedArticle
19.
Ramjan  R, Calmy  A, Vitoria  M,  et al.  Systematic review and meta-analysis: patient and programme impact of fixed-dose combination antiretroviral therapy. Trop Med Int Health. 2014;19(5):501-513.
PubMedArticle
20.
Global Tuberculosis Report. Geneva, Switzerland: World Health Organization; 2014.
21.
Martinson  NA, Moultrie  H, van Niekerk  R,  et al.  HAART and risk of tuberculosis in HIV-infected South African children. Int J Tuberc Lung Dis. 2009;13(7):862-867.
PubMed
22.
Guidelines for the Management of HIV in Children. 2nd ed. Pretoria: National Department of Health, South Africa; 2010.
23.
Frohoff  C, Moodley  M, Fairlie  L,  et al.  Antiretroviral therapy outcomes in HIV-infected children after adjusting protease inhibitor dosing during tuberculosis treatment. PLoS One. 2011;6(2):e17273.
PubMedArticle
24.
Reitz  C, Coovadia  A, Ko  S,  et al.  Initial response to protease-inhibitor-based antiretroviral therapy among children less than 2 years of age in South Africa. J Infect Dis. 2010;201(8):1121-1131.
PubMedArticle
25.
Zanoni  BC, Phungula  T, Zanoni  HM,  et al.  Impact of tuberculosis cotreatment on viral suppression rates among HIV-positive children initiating HAART. AIDS. 2011;25(1):49-55.
PubMedArticle
26.
Soeters  HM, Napravnik  S, Patel  MR,  et al.  The effect of tuberculosis treatment on virologic and CD4+ cell count response to combination antiretroviral therapy. AIDS. 2014;28(2):245-255.
PubMedArticle
27.
Grennan  JT, Loutfy  MR, Su  D,  et al.  Magnitude of virologic blips is associated with a higher risk for virologic rebound in HIV-infected individuals. J Infect Dis. 2012;205(8):1230-1238.
PubMedArticle
28.
Dahri  K, Ensom  MH.  Efavirenz and nevirapine in HIV-1 infection. Clin Pharmacokinet. 2007;46(2):109-132.
PubMedArticle
29.
Decloedt  EH, Maartens  G.  Neuronal toxicity of efavirenz. Expert Opin Drug Saf. 2013;12(6):841-846.
PubMedArticle
30.
Paredes  R, Marconi  VC, Lockman  S,  et al.  Impact of antiretroviral drugs in pregnant women and their children in Africa. J Infect Dis. 2013;207(suppl 2):S93-S100.
PubMedArticle
31.
Ton  Q, Frenkel  L.  HIV drug resistance in mothers and infants following use of antiretrovirals to prevent mother-to-child transmission. Curr HIV Res. 2013;11(2):126-136.
PubMedArticle
32.
Roberts  T, Bygrave  H, Fajardo  E, Ford  N.  Challenges and opportunities for the implementation of virological testing in resource-limited settings. J Int AIDS Soc. 2012;15(2):17324.
PubMedArticle
33.
Basson  AE, Rhee  SY, Parry  CM,  et al.  Impact of drug resistance-associated amino acid changes in HIV-1 subtype C on susceptibility to newer NNRTIs. Antimicrob Agents Chemother. 2015;59(2):960-971.
PubMedArticle
34.
Mukonzo  JK.  The challenge of paediatric efavirenz dosing. AIDS. 2014;28(13):1855-1857.
PubMedArticle
×