HIV Articles  
Resistance to Antiretroviral Drugs: A Threat to the Prevention and Treatment of Pediatric HIV Infection EDITORIAL COMMENTARY
  Journal of Infectious Diseases May 15, 2007
Paul A. Krogstad
UCLA AIDS Institute, Departments of Pediatric Infectious Diseases and Medical and Molecular Pharmacology, David Geffen School of Medicine, University of California, Los Angeles
(See the article by Persaud et al., on pages 1402-10.)
Mother-to-child transmission (MTCT) of HIV was recognized early in the worldwide AIDS pandemic, and, by the late 1980s, abnormal psychomotor development, candida esophagitis, and recurrent invasive bacterial infections were recognized as signatures of HIV infection in infants. Pneumocystis pneumonia occurred in many infants within the first 3-6 months of life, often preceding death by as little as 1 month [1, 2]. Although symptoms of HIV infection generally appeared much later, children who developed symptoms in the first year of life generally died by 2-3 years of age, and AIDS became the seventh leading cause of mortality in young children in the United States by 1996 [3]. Nucleoside analogue reverse-transcriptase inhibitor (NRTI) antiretroviral therapy delayed the progression of disease, and zidovudine sometimes caused dramatic improvements in cognitive function, but these salutary effects were usually modest and transient in nature [4-6].
Fortunately, striking progress in both the prevention and treatment of HIV infection of children was made in early to mid-1990s. Administration of zidovudine during pregnancy, labor, and delivery, accompanied by 6 weeks of postnatal therapy for the infant, was shown to reduce MTCT of HIV by 70%. A series of studies in the 1990s identified virological, immunological, and obstetrical variables linked with perinatal infection of infants. Armed with this information, modifications of obstetrical care and increased use of antiretroviral therapy (ART) during pregnancy reduced the rate of MTCT of HIV in the United States from 25% to its current rate of <2% [7]. At about the same time, the prognosis for HIV-infected children also began to change. With the availability of highly active ART (HAART) regimens for children built around HIV protease inhibitors (PIs) and, to a lesser extent, nonnucleoside reverse-transcriptase inhibitors (NNRTIs), the annual mortality rate for infected children in the United States and Europe began to dip sharply by 1997 [8, 9]. In the mid-1990s, children who reached 9 years of age were considered long-term survivors [10]. Survival to adulthood is now the norm for perinatally infected children with access to specialized medical care and HAART.
Despite these successes, it has been unclear how to best proceed with treatment for perinatally infected infants, particularly those who are asymptomatic at the time of diagnosis. Current US and World Health Organization (WHO) treatment guidelines reflect the general agreement among pediatric HIV experts that ART should be initiated in infants and children who have either HIV-related clinical symptoms or laboratory evidence of immune compromise [11, 12]. However, the risk of life-threatening complications of HIV infection is greatest in the youngest of infants [1, 2, 13, 14], and no laboratory parameter accurately identifies those at highest risk for precipitous immunological deterioration, developmental delay, and premature death. Consequently, US treatment guidelines suggest that antiretroviral therapy be considered for all infants, regardless of CD4 lymphocyte parameters and HIV viral load [11]. Several small studies have demonstrated that durable suppression of HIV viremia can be achieved in the youngest and most immunologically immature infants. For example, in the Pediatric AIDS Clinical Trials Group (PACTG) 356 study, 52 infants were treated with 3- and 4-drug ART regimens. Those who began treatment at a median age of 8 weeks were twice as likely to maintain suppression of plasma HIV RNA levels of <400 copies/mL for 200 weeks than infants who initiated therapy at a median age of 7.6 months (60% vs. 30%; P = .03) [15].
However, the rates of suppression of HIV RNA to below the limits of detection were lower in PACTG 356 and other studies than those typically reported for adults and older children. Many factors are probably involved, including the immaturity of innate and adaptive immunity during early childhood, the practical difficulties of administering oral medications to infants, and the pharmacological challenges posed by the dynamic changes in body composition and drug metabolism that occur in infancy [16]. Early ART for infants has also been complicated by the increase in HIV drug resistance among adults, including pregnant women, and the lack of pediatric formulations and pharmacokinetic data for many antiretroviral drugs.
In this issue of the Journal, Persaud et al. present results from a virologic substudy of blood specimens taken from infants <6 months of age enrolled in PACTG 1030, a dose-finding, open-label study of the liquid coformulation of lopinavir and ritonavir [17]. Infants were enrolled at a median age of 10 weeks, after HIV infection was documented by virologic methods. Although this article focuses on studies of HIV drug resistance, lopinavir/ritonavir appeared to be effective: after 24 weeks of therapy, HIV load was below detection limits in 70% of infants enrolled in the study. Using reverse-transcriptase polymerase chain reaction to amplify HIV pol gene sequences from pretreatment plasma specimens, Persaud et al. found evidence of resistance to reverse transcriptase inhibitors in 5 (24%) of the 21 patients. In 4 of these cases (enrolled between 2002 and 2005), resistance to NNRTI was identified. Similar data were reported from PACTG 356: 9% of infected infants had HIV with genotypic evidence of drug resistance. It is particularly noteworthy that in both of these studies, maternal history did not always provide a clear explanation for the drug-resistance patterns encountered before the treatment of the infected infants. MTCT of resistant virus is well known, but most studies to date have emphasized the impact of ART use by the mother. Particular attention has understandably been given to the selection and persistence of resistance that follows perinatal use of nevirapine to prevent MTCT [18]. In the current report, nevirapine was not used during pregnancy or for chemoprophylaxis for 2 infants found to be infected with virus containing NNRTI resistance mutations (G190A and V106I/Y188H mutations in HIV reverse transcriptase). These and other data indicate that MTCT of resistant virus is becoming common [19], if not fully understood.
Going a step beyond other reports, Persaud et al. provide data about the stability of drug-resistant HIV in the infected infants. Using exquisitely sensitive culture methods optimized for use with the small amounts of blood available from newborns, they isolated >50 replication-competent HIV variants from the infected infants with genotypic evidence of drug resistance at baseline. Strikingly, they found that all infants with drug-resistant HIV before HAART retained a cellular reservoir that contained the drug-resistant virus for periods of up to 21 months after initiation of therapy. Viral suppression in these infants had been achieved using a PI-based regimen, and it is conceivable that treatment failure would occur if an NNRTI-based regimen (e.g., efavirenz) were substituted later on. These findings reinforce the current recommendation in US treatment guidelines that drug-resistance testing be performed before the initiation of therapy [11] and heighten interest in ongoing studies of the impact of nevirapine use for perinatal chemoprophylaxis on the subsequent use of this agent for therapy of infected infants and children.
Pediatric HIV infection is now preventable, treatable, and potentially survivable. The report by Persaud et al. helps highlight current impediments to the elimination of pediatric AIDS. First, the difficulties and uncertainties involved in treating the steadily increasing number of HIV-infected children underscore the importance of preventing MTCT. Second, their data reinforce the need for early and thorough evaluation of new antiretroviral agents for pediatric use, especially new classes of drugs active against highly resistant HIV variants. (It is worth noting that lopinavir/ritonavir is not yet approved for use in infants under the age of 6 months, although it was approved in 2000 for use in older children and adults.) Additional clinical trials are needed to identify optimal antiretroviral regimens for treatment-naive infants and children infected with drug-resistant virus. Finally, the data from this study also reinforce the importance of monitoring rates of HIV drug resistance in pregnant women and their transmission to infants, particularly in view of the emergence and persistence of high-level resistance to NNRTIs after perinatal exposure to nevirapine [18] and of WHO recommendations for the use of NNRTIs as first-line therapy [12].
Unfortunately, drug-resistance testing is currently expensive and inaccessible in many countries, at least as currently performed with sequence-based genotyping methods. Potential alternatives have been identified to detect point mutations that most commonly confer resistance to NRTI, NNRTI, and PI drugs [20-22], but these would not detect less-common resistance-associated HIV genotypes. The need for more affordable and accessible technology is not limited to methodology for the detection of HIV drug resistance. There is also a need for newer, simpler approaches for the detection of HIV infection in infants, because transplacental transfer of maternal IgG precludes the use of antibody-detection methods. Early detection of HIV infection and the presence of drug resistance could make all the difference for many infants [23].
We must keep in mind that pediatric AIDS, now rare in the United States, remains a major cause of death among children elsewhere in the world, and that antiretroviral treatment options remain limited for infants, who are at the greatest risk of rapid progression of disease.
  icon paper stack View Older Articles   Back to Top