585www.expert-reviews.com ISSN 1478-7210© 2008 Expert Reviews Ltd10.1586/14787210.6.5.585

Key Paper Evaluation

This article reviews the study of Prendergast et al., summarizes the paper’s key findings and discusses implications for pediatric HIV prac-tice in resource-limited settings [1]. For some time, children have not prominently featured in initiatives aimed at managing the global HIV epidemic. This culminated in the launch of the ‘Unite for Children, Unite against AIDS’ cam-paign by the United Nations Children’s Fund (UNICEF) in October 2005. This campaign focused international attention on the vulner-ability and unmet care needs of HIV-exposed and -infected children and called for the imple-mentation of comprehensive preventative and therapeutic interventions [101]. Underpinning this campaign was the increasing global awareness that the Millennium Development Goals rele-vant to the pediatric HIV epidemic, particularly Millennium Development Goal 4 addressing mortality of under 5-year olds and Millennium Development Goal 6 focusing on the HIV pan-demic in children and adults, would probably not be achieved by 2015, as has been previously envisaged [102]. One of the stated objectives of the UNICEF campaign was the provision of either highly active anti retroviral therapy (HAART) or cotrimoxazole prophylaxis, or both, to 80%

of children in need by 2010 [101]. At the end of 2005, an estimated 75,000 children were receiv-ing HAART. This figure increased to 198,000 by the end of 2007, a 2.6-fold improvement. Antiretroviral coverage in HIV-infected adults and children in all middle- and low-income countries was estimated to be 31% at the end of 2007 [103]. At the current rate of enrolling chil-dren on HAART, the UNICEF target will be extremely difficult to accomplish.

Treatment guidelines produced in 2006 by the WHO have played a pivotal role in shaping pedi-atric antiretroviral treatment programs in many resource-limited countries [104]. Criteria for start-ing children on HAART were based on a WHO clinical-staging system for children and on results of a meta-analysis of longitudinal immunological data and associated short-term risk of disease pro-gression in nearly 4000 HIV-infected children, situated in Europe and North America [2]. In infants, that is, children under 12 months of age, with advanced clinical disease and/or advanced immuno suppression, HAART was recommended [104]. Although the 2006 WHO guidelines were founded on the best-available evidence, several concerns were expressed, including the absence of randomized control trials to guide when to

Brian S EleyPaediatric Infectious Diseases Unit, Red Cross Children’s Hospital, School of Child and Adolescent Health, University of Cape Town, Rondebosch, 7701, Cape Town, South Africa Tel.: +27 216 585 111 Fax: +27 216 891 287 brian.eley@uct.ac.za

Evaluation of: Prendergast A, Mphatswe W, Tudor-Williams G et al. Early virological suppression with three-class antiretroviral therapy in HIV-infected African infants. AIDS 22, 1333–1343 (2008).

There is limited information regarding the outcome of infants treated with highly active antiretroviral therapy (HAART) in middle- and low-income countries. The mean/median age of children enrolled on HAART in these countries is generally high; however, it is acknowledged that untreated HIV-infected children under 2 years of age, including infected infants, are an extremely vulnerable group. This article assessed the findings of a recently published paper describing the outcome of 63 infants randomized to commence immediate or deferred HAART in a resource-poor setting and documented favorable short-term outcome in both groups, with correspondingly high adherence rates. The recent change in global treatment guidelines, recommending that all HIV-infected infants should be commenced on HAART soon after diagnosis, irrespective of their clinical status and/or immunological severity, is discussed in relation to the key findings of the article.

Keywords: highly active anti retroviral therapy • infancy • resource-limited setting

Antiretroviral therapy during infancy: essential intervention for resource-limited settingsExpert Rev. Anti Infect. Ther. 6(5), 585–589 (2008)

Expert Rev. Anti Infect. Ther. 6(5), (2008)586

Key Paper Evaluation Eley

commence HAART in children. Following their publications, pediatric treatment programs mushroomed in resource-limited countries. Responses, as measured by changes in HIV-related mor-bidity and mortality and CD4 measurements, have generally been very favorable. Striking features of recent publications describing pediatric HAART programs in middle- and low-income countries include the high age of children at the time of starting HAART and the small number of children aged under 2 years commenced on HAART [3–10]. By contrast, natural history studies have shown that young, untreated children are an extremely vulnerable group. In resource-limited settings, more than 50% of HIV-infected children die before their second birthday [11]. The study by Prendergast et al. is important because it is one of the first to describe early outcomes of HAART in a group of infants in a resource-limited setting.

MethodsA pilot, randomized controlled trial was conducted at St Mary’s Hospital, Mariannhill and Prince Mshiyeni Hospital, Umlazi, with both institutions located in KwaZulu-Natal, South Africa. The study was designed to evaluate three strategies for treating HIV-infected infants with antiretroviral therapy. Approval for the study was obtained from the Biomedical Research Ethics Committee of the University of KwaZulu-Natal and the Institutional Review Board of Massachusetts General Hospital, Boston, MA, USA.

The participants were enrolled between July 2003 and September 2005. Pregnant women at the two hospitals were tested for HIV after 28 weeks of gestation. HIV-infected women were invited to participate in the study. A single dose of nevirap-ine (200 mg) was given to infected women to take at the start of labor. After delivery, infants were provided with a single dose of nevirapine (2 mg/kg bodyweight) within 72 h of birth. Venous blood was drawn from the infants on day 1 and 28 to test for intrauterine and intrapartum HIV infection, respectively. PCR technology was used to diagnose HIV infection. Diagnosis was confirmed by repeating the diagnostic test within 8 days of com-pleting the first test. Infants were included in the study if they had intra-uterine or intrapartum HIV infection and providing that written consent had been obtained. Infants were excluded if they were preterm (<37 weeks gestation), underweight at birth (<2 kg bodyweight), had evidence of another congenital infection or had severe congenital abnormalities.

Infants were randomized to receive:

Deferred HAART, commenced once South African national •criteria were met, that is, WHO stage 3 or advanced stage 2 disease, or CD4 levels below 20% if under 18 months of age or CD4 levels below 15% if above 18 months of age

Immediate HAART for 1 year, before stopping therapy•

Immediate HAART, with up to three structured treatment •interruptions to 18 months of age, before stopping therapy

Children in the immediate study arms would be restarted on HAART after treatment interruption, once the South African national criteria were fulfilled. All infants received cotrimoxazole prophylaxis between 6 weeks and 1 year of age.

Infants were reviewed at monthly intervals. These visits included clinical assessments, CD4 counts and viral loads. Mothers received infant-feeding counseling. Nursing staff pro-vided ongoing feeding support and education throughout the study. Mothers with CD4 counts of less than 200 cells/mm3 were provided with HAART.

The first-line HAART regimen used in the study was zido-vudine, lamivudine, nevirapine and nelfinavir. Nevirapine was discontinued when the viral load reached less than 50 copies/ml. Infants who developed virological failure of the first-line regi-men were genotyped and commenced on a second-line regimen, guided by the results of resistance testing. First-line regimens were modified if infants developed adverse events. Adherence to HAART was assessed by verbal report and measurement of returned medication at each monthly visit.

The authors used conventional statistical methodology to analyze the results of the study, including the use of the student’s t-test and the Mann–Whitney U test to compare continuous variables, Fisher’s exact test to evaluate categorical variables and Kaplan–Meier plots and the log rank test to compare survival data.

The primary study end point was the proportion of infants progressing to AIDS by 3 years. However, in the present paper, the authors used the study results to describe the outcome of HAART administered to deferred and immediate groups after 1 year of therapy.

ResultsOf 740 infants born to 719 HIV-infected mothers, 75 were diag-nosed with HIV infection by 28 days of life. A total of 12 infants were ineligible for the study. The remaining 63 were random-ized to immediate (n = 43) or deferred (n = 20) HAART. Of 63 randomized children, 36 in the immediate arms and 13 in the deferred arm commenced HAART during infancy and com-pleted 1 year or more of follow-up. The present paper primarily describes the short-term outcome of these 49 children. A total of 13 out of 60 mothers (22%) of randomized infants were started on HAART during the study.

All infants had been exposed to single-dose nevirapine, admin-istered perinatally. Pretreatment viral resistance testing was per-formed on 51 of the 63 randomized infants. All were infected with subtype C HIV-1 virus and 20 out of the 51 infants (39%) had non-nucleoside reverse transcriptase-resistant mutants in their viral populations. The frequency of resistant mutants was similar in intra-uterine-infected and intrapartum-infected infants. During the first year on HAART, the overall verbal adherence was 99.3% and measured adherence was 95.4%.

The 13 infants in the deferred group, who started HAART dur-ing the first year of life, reached the start criteria at a median age of 41 days (range: 1–106 days) but started HAART at a median age of 142 days (range: 81–227 days). Delay was caused by treat-ment with TB medication, social problems or care-giver prepara-tion. After 1 year of HAART, 100% had a viral load of less than 400 copies/ml and 92% had a viral load of less than 50 copies/ml. In the immediate group (n = 36), HAART was started on day 28 of life (range: 8–164 days) in the intrapartum-infected infants and

www.expert-reviews.com 587

Key Paper EvaluationAntiretroviral therapy during infancy

on day 55 of life (range: 36–90 days) in the peripartum-infected infants. After 1 year on HAART, viral load was less than 400 cop-ies/ml in 100% of individuals/infants and less than 50 copies/ml in 94% of individuals/infants.

A total of 46 children (94%) started the preferred four-drug regimen. Nevirapine was omitted in three infants due to safety concerns or adverse events. Ten children (20%) were commenced on second-line therapy within the first year because of cotreatment with TB medication and/or virological failure. Of five children with virological failure, one exhibited non-nucleoside reverse transcriptase resistance before starting HAART. In summary, 39 out of the 49 infants (80%) achieved virological suppression on first-line therapy and the remaining ten infants required second-line therapy to suppress viral replication by 1 year. There was no significant difference in median time required to suppress viral load to less than 50 copies/ml (i.e., 121 days in the immediate group and 115 days in the deferred group; p = 0.4).

There was no significant difference in mortality during infancy between the immediate and deferred groups. However, infants in the immediate group experienced fewer outpatient illnesses (seven vs 12 episodes; p = 0.003) and were less likely to be hospitalized (16 vs 42%; p = 0.09).

Discussion & significanceThis prospective study demonstrated that it is possible to treat infants successfully with HAART in resource-limited settings. A total of 100% of infants in both the immediate and deferred groups achieved a viral load of less than 400 copies/ml after 1 year on HAART. Suppression of virological replication is a surrogate marker of good quality care. Virological responses were consistent with excellent verbal and measured adherence rates and are in agreement with previously published studies that show satisfac-tory levels of adherence in resource-limited settings, including Africa [12].

Differences between the immediate and deferred groups were documented. In particular, those infants assigned to the immediate-treatment group experienced significantly fewer out-patient illnesses and a lower hospitalization rate. Previous studies have also reported lower morbidity when HAART was started early during infancy. In one observational study, the outcome of 40 infants who received HAART before the age of 6 months was compared with 43 infants started on therapy after 6 months of age. In those infants who received early HAART, opportunistic infections and encephalopathy did not develop during the first 24 months of life. By contrast, of those started on therapy after 6 months, seven experienced opportunistic infections (p = 0.01) and three developed encephalopathy (p = 0.08) [13]. An interesting observation recorded in the pres-ent study was the nonsignificant difference in mortality between the immediate and deferred groups. This finding is inconsis-tent with a recent, prospective, randomized control clinical trail, also from South Africa, the Children with HIV Early Antiretroviral Therapy (CHER) study, which compared the outcome of 252 infants who received immediate HAART with 125 infants in whom HAART was deferred until the national

criteria for starting children on HAART were fulfilled. In the CHER study, immediate therapy was associated with a 75% reduced risk of dying during the first 32 weeks of follow-up. This was a highly statistically significant finding, which effec-tively caused the closure of the deferred arm of that study [14]. As suggested by Prendergast et al., the small sample size of the present study may have masked early mortality differences.

The majority of participants in the present study were started on a four-drug/three-class first-line regimen. While the authors draw attention to a previous study that suggested that four-drug regimens may be superior to three-drug regimens in infants, the study in question was a nonrandomized Phase I–II trial and, therefore, was not designed to address the comparative efficacy of four-drug regimens [15].

The authors express concern that five children were switched to second-line therapy because of virological failure, despite high levels of adherence. They speculate that high baseline viral loads, frequently recorded in HIV-infected infants, may be associated with slow reduction of viral load after commencing HAART and, therefore, may increase the risk of emergence of viral resistance.

Prendergast et al. discuss the death of six of the 63 infants randomized during the first year of life (i.e., five in the immedi-ate group and one in the deferred group; p = 0.65). Three of the infants were on HAART at the time of their demise. These six infants died despite having had access to regular medical care, including HAART. While disturbing, the authors correctly con-clude that these deaths were not unanticipated, given the high level of mortality of infants and children under the age of 5 years.

Expert commentaryThe publication of the present study coincides with the release of revised global treatment guidelines for HIV-infected children by the WHO [105]. These new guidelines are mainly a response to the find-ings of the Prendergast and CHER studies [1,14]. Before these two studies were conducted, there were no randomized controlled clini-cal trials to guide the development of criteria for starting children on HAART. The data safety board monitoring the CHER study recommended that the deferred arm be discontinued, owing to a 32-week interim analysis that provided compelling evidence of the benefit of immediate HAART during infancy [14]. It is unlikely that a similar trial will be conducted in the future. Consequently, the WHO now recommends that all HIV-infected infants aged under 1 year should commence HAART as soon as possible, irrespective of their clinical condition or immunological severity [105].

The recommendation to treat all HIV-infected infants soon after diagnosis should substantially reduce HIV-related mortality during childhood. A major limiting factor remains the challenges associated with diagnosing HIV infection during infancy. Unless HIV infection is diagnosed, it cannot be treated. Virological tests are required to diagnose HIV infection in children aged under 18 months; however, the laboratory capacity to perform such virological tests is not well developed in many resource-limited settings. In addition, a shortage of technically trained staff and weak laboratory quality-control systems compound this problem, particularly in rural areas [16].

Expert Rev. Anti Infect. Ther. 6(5), (2008)588

Key Paper Evaluation Eley

Several management questions should be addressed to improve the treatment of infants with HAART, including issues surrounding:

The relative efficacy of four-drug first-line regimens for •infants. This should be evaluated in adequately powered effi-cacy trials. Trials in treatment-naive adults have not supported the routine use of four-drug regimens. In one study, adults were randomized to receive two nucleoside reverse tran-scriptase inhibitors (didanosine and stavudine) plus a non-nucleoside reverse transcriptase inhibitor (efavirenz) or a protease inhibitor (nelfinavir), or a four-drug/three-class regimen of didanosine, stavudine, efavirenz and nelfinavir. The efavirenz regimen was most effective in terms of viro-logical suppression and the durability of first-line therapy [17]. In another study, the efficacy of zidovudine, lamivudine plus efavirenz was compared with a four-drug/two-class regimen, comprising zidovudine, lamivudine, abacavir plus efavirenz. After 3 years, there was no difference in the proportion of patients reaching virological failure [18]. Efficacy studies are required before four-drug regimens can be considered for routine use in infants;

Optimal regimens for infants who were exposed to perinatal •antiretroviral agents, particularly nevirapine. In their latest guidelines, the WHO recommended that infants who were exposed to perinatal nevirapine or a maternal HAART regimen containing a non-nucleoside reverse transcriptase inhibitor should receive a protease inhibitor-containing regimen. How-ever, ideal regimens in this setting have been incompletely researched [105];

Whether therapy should be disrupted after starting HAART •in infancy. Whether structured drug interruptions should be employed to extend the effect of first-line therapy in this context is being addressed in the Prendergast and CHER studies;

The pharmacokinetics of antiretroviral agents. For some drugs •pharmacokinetics have not been adequately studied in young infants, such as that of lopinavir–ritonavir coformulation, and is needed to establish optimal dosing recommendations.

Notwithstanding these information gaps, global implementa-tion of HAART for all HIV-infected infants is proceeding. This intervention should lead to reduced pediatric HIV-associated morbidity and mortality, and increased survival.

Five-year viewDuring the next few years, the medium- and long-term outcome of infants commenced on HAART will become evident. Improved approaches to treatment will depend on identifying optimal first-line regimens for infants, correct dosing recommendations and whether structured treatment interruptions are able to extend the effectiveness of first-line therapy. The role of newer classes of anti-retroviral agents during infancy and childhood should be clearer, including recently licensed fusion inhibitors, chemokine receptor inhibitors and integrase inhibitors. These developments will proceed in parallel with ongoing initiatives to escalate preventative measures to disrupt vertical transmission and so reduce the global pediatric HIV burden.

Financial & competing interests disclosureThe author was a member of the WHO Technical Reference Group that formulated the recently updated global HIV/ART guidelines. The author has no other relevant financial involvement with any organization or entity with a financial interest in or financial conflict with subject matter or materi-als discussed in this manuscript. This includes employment, consultancies, honoraria, stock ownership, expert testimony, grants or patents received or pending, or royalties.

No writing assistance was utilized in the production of this manuscript.

Key issues

Perinatally HIV-infected children are at a high risk of dying in the • first few years of life.

Infants in resource-limited settings can be treated successfully • with highly active antiretroviral therapy (HAART). In the study reviewed, infants who received immediate therapy were less likely to experience minor outpatient illnesses or hospitalization.

Although global treatment guidelines have recently changed • and now recommend that all HIV-infected infants be commenced on HAART after diagnosis, further clinical trials are needed to optimize the administration of HAART during infancy and childhood.

ReferencesPrendergast A, Mphatswe W, 1

Tudor-Williams G et al. Early virological suppression with three-class antiretroviral therapy in HIV-infected African infants. AIDS 22, 1333–1343 (2008).

HIV Paediatric Prognostic Markers 2

Collaborative Study Group. Short-term risk of disease progression in HIV-1-infected children receiving no antiretroviral therapy or zidovudine monotherapy: a meta-analysis. Lancet 362, 1605–1611 (2003).

Janssens B, Raleigh B, Soeung S3 et al. Effectiveness of highly active antiretroviral therapy in HIV-positive children: evaluation at 12 months in a routine program in Cambodia. Pediatrics 120, E1134–E1140 (2007).

George E, Noël F, Bois G4 et al. Antiretroviral therapy for HIV-1-infected children in Haiti. J. Infect. Dis. 195, 1411–1418 (2007).

Wamalwa, DC, Farquhar C, Obimbo EM5 et al. Early response to highly active antiretroviral therapy in HIV-1-infected Kenyan children. J. Acquir. Immune Defic. Syndr. 45, 311–317 (2007).

Reddi A, Leeper SC, Grobler AC6 et al. Preliminary outcomes of a paediatric highly active antiretroviral therapy cohort from KwaZulu-Natal, South Africa. BMC Pediatr. 7, 13 (2007).

Kamya MR, Mayanja-Kizza H, 7

Kambugu A et al. Predictors of long-term viral failure among Ugandan children and adults treated with antiretroviral therapy. J. Acquir. Immune Defic. Syndr. 46, 187–193 (2007).

Eley B, Davies M, Apolles P8 et al. Antiretroviral treatment for children. S. Afr. Med. J. 96, 988–993 (2006).

www.expert-reviews.com 589

Key Paper EvaluationAntiretroviral therapy during infancy

Rouet, F, Fassinou P, Inwoley A9 et al. Long-term survival and immuno-virological response of African HIV-1-infected children to highly active antiretroviral therapy regimens. AIDS 20, 2315–2319 (2006).

Puthanakit T, Oberdorfer A, 10

Akarathum N et al. Efficacy of highly active antiretroviral therapy in HIV-infected children participating in Thailand’s national access to antiretroviral program. Clin. Infect. Dis. 41, 100–107 (2005).

Newell M, Coovadia H, Cortina-Borja M, 11

Rollins N, Gaillard P, Dabis F; for the Ghent International AIDS Society (AIS) working group on HIV infection in women and children. Mortality of infected and uninfected infants born to HIV-infected mothers in Africa: a pooled analysis. Lancet 364, 1236–1243 (2007).

Mills EJ, Nachega JB, Buchan I12 et al. Adherence to antiretroviral therapy in sub-Saharan Africa and North America: a meta-analysis. J. Am. Med. Assoc. 296, 679–690 (2006).

Faye A, Le Chenadec J, Dollfus C13 et al. Early versus deferred antiretroviral multidrug therapy in infants infected with HIV type 1. Clin. Infect. Dis. 39, 1692–1698 (2004).

Violari A, Cotton M, Gibb D14 et al. Antiretroviral therapy initiated before 12 weeks of age reduces early mortality

in young HIV-infected infants: evidence from the Children with HIV Early Antiretroviral Therapy (CHER) Study. Presented at: 4th IAS Conference Sydney, Australia, 22–25 July 2007.

Luzuriaga K, McManus M, Mofenson L15 et al. A trial of three antiretroviral regimens in HIV-1 infected children. N. Engl. J. Med. 350, 2471–2480 (2004).

Cherutich P, Inwani I, Nduati R, 16

Mbori-Ngacha D. Optimizing paediatric HIV care in Kenya: challenges in early infant diagnosis. Bull. World Health Organ. 86, 155–160 (2008).

INITIO Trial International Co-ordinating 17

Committee. Virological and immunological outcomes at 3 years after starting antiretroviral therapy with regimens containing non-nucleoside reverse transcriptase inhibitor, protease inhibitor, or both in INITIO: open-label randomized trial. Lancet 368, 287–298 (2006).

Gulick RM, Ribaudo HJ, Shikuma CM18 et al. Three- vs four-drug antiretroviral regimens for the initial treatment of HIV-1 infection: a randomized controlled trial. J. Am. Med. Assoc. 296, 769–781 (2006).

Websites

UNAIDS and UNICEF. The global 101

campaign on children and AIDS: Unite for Children. Unite against AIDS, October 2005 www.unicef.org/uniteforchildren

United Nations. Official list of MDG 102

indicators, January 2008 http://mdgs.un.org/unsd/mdg/resources/attach/indicators/officiallist2008.pdf

WHO, UNAIDS and UNICEF. 103

Towards universal access. Scaling up priority HIV/AIDS interventions in the health sector. Progress report 2008 http://data.unaids.org/pub/report/2008/200806_towardsuniversalaccessreport2008_en.pdf

World Health Organization. Antiretroviral 104

therapy of HIV infection in infants and children in resource-limited settings: towards universal access. Recommendations for a public health approach, 2006 www.who.int

World Health Organization. Report of the 105

WHO technical reference group. Paediatric HIV/ART care guideline group meeting, Geneva, Switzerland, April 2008 www.who.int/hiv/pub/paediatric/WHO_paediatric_ART_guideline_rev_mreport_2008.pdf

AffiliationBrian S Eley •Paediatric Infectious Diseases Unit, Red Cross Children’s Hospital, School of Child and Adolescent Health, University of Cape Town, Rondebosch, 7701, Cape Town, South Africa Tel.: +27 216 585 111 Fax: +27 216 891 287 brian.eley@uct.ac.za