Approximately 2.3 million children are living with HIV globally.1 Without antiretroviral therapy (ART), 75% of children born with the virus die before the age of 5 years.2 Although 20% of HIV-positive adults in developing countries now have access to ART, <8% of children who need treatment today are getting it.3
Several initial treatment programs in the developing world have shown considerable success in reducing clinical symptoms and increasing CD4 cell counts in children. As in the developed world, however, detectable viral loads are common, with reports of 45% at 9 months, 41% at 12 months, 24% at 72 weeks, and 50% at 2 years.4-7
The Chinese National Center for AIDS has identified approximately 1800 children with HIV infection, although vertical transmission rates suggest that as many as 9000 babies may have been infected in 2005.8 In that same year, the Ministry of Health began a national pediatric HIV/AIDS treatment program. Previously, appropriate pediatric ART formulations and guidelines were unavailable. This prospective analysis describes the clinical, immunologic, and virologic effects of highly active antiretroviral therapy (HAART) in the first 83 children treated in the national program.
Study Site and Participants
All participants were treated at the Chinese Center for Disease Control (CCDC) clinic in rural Shangcai County, Henan Province. All children were enrolled in July and August 2005, were <16 years old, and had HIV-1 infection per enzyme-linked immunosorbent assay (ELISA) and confirmatory Western blot analysis. ART-naive patients were selected if they had World Health Organization (WHO) clinical stage III or IV disease or severe immunologic suppression (age-adjusted CD4 count <200 cells/mm3). All ART-experienced patients were enrolled in the pediatrics program and this study, and they were switched from adult to pediatric formulations. No patients in the cohort had contraindications to HAART (ie, active additional infections, severe anemia, elevated liver enzymes). Caregivers for all children gave informed written consent.
The default HAART regimen was zidovudine (AZT) plus lamivudine (3TC) plus nevirapine (NVP). Stavudine (d4T) was substituted for AZT in the setting of anemia or other side effects, and efavirenz (EFV) was substituted for NVP in the setting of concurrent tuberculosis treatment, elevated liver enzymes, rash, or other side effects. Generic AZT, 3TC, and NVP (Cipla, Mumbai, India) and branded d4T (Bristol-Myers Squibb, New York, NY) and EFV (Merck, Whitehouse Station, NJ) were dosed as follows: 180 mg/m2 of AZT administered twice daily; 1 mg/kg of d4T administered twice daily; 4 mg/kg of 3TC administered twice daily; 120 mg/m2 of NVP administered once daily for 2 weeks and increased to 7 mg/kg twice daily for children younger than 8 years of age and 4 mg/kg twice daily for children 8 years of age and older if no complications arose; and 200 mg of EFV daily for children weighing 10 to 14.9 kg, 250 mg for children weighing 15.0 to 19.9 kg, 300 mg for children weighing 20 to 24.9 kg, 350 mg for children weighing 25 to 32.4 kg, 400 mg for children weighing 32.5 to 40 kg, and 600 mg for children weighing ≥40 kg. Prior treatment regimens for the ART-experienced children are shown in Table 1; exact dosing is not known and was likely variable. These children were continued on their prior regimens using these noted doses, with the exception of substituting 3TC for didanosine (ddI) as a result of drug availability. There were no disruptions in ART supply.
All children received several weeks of adherence counseling through a summer camp, sponsored by a nongovernmental organization (NGO) before enrollment. They were seen clinically at enrollment, every 2 weeks for 2 months, every month for the next 4 months, every 3 months thereafter, and as needed. Patients came to the clinic monthly to pick up medications. The children received further adherence support through the NGO several times a week for the first 6 months of the program. The CCDC clinic provided subsequent support.
Blood samples were collected at 0, 1, 3, 6, 9, and 12 months after the initiation of HAART. CD4 cell counts were assessed locally using a BD FACSCOUNT machine (BD Biosciences, San Jose, CA). Equipment for testing CD4 percentage (CD4%) was not available. Specimens for viral load were processed locally within 6 hours of collection. The plasma fraction was separated and transported at −20°C to Beijing Ditan Hospital (a national HIV/AIDS training center) within 24 hours for analysis using the COBAS Amplicor HIV-1 Monitor Test (Roche Diagnostics, Indianapolis, IN). The 6-month viral load samples had to be repeated at 7 months because of machine malfunction. The limit of detection was 400 copies/mL.
All analyses reflect an intention-to-treat model and were performed using SAS 9.0 (SAS Institute, Cary, NC). Continuous variables were assessed with the Student t test or Mann-Whitney test as appropriate, and categoric variables were assessed with the Fisher exact test. General estimating equations were performed to compare groups of categoric data. The Wilcoxon signed rank test was used to compare median values of continuous data between groups. Statistical significance was set at P ≤ 0.05 for 2-sided tests.
The baseline characteristics of the 83 children in the cohort are shown in Table 1. One child was lost to follow-up. Duration of infection in nonperinatally infected children is not known. The median duration for treatment in the ART-experienced group was 59 (interquartile range [IQR]: 44.1 to 80.3) weeks.
After 1 year of HAART, median weight increased by a weight for age (and gender) z-score (WAZ) of 0.3 (IQR: −0.2 to 0.6; P < 0.0001) for previously ART-naive children and did not change for ART-experienced children (IQR: −0.2 to 0.1; P = 0.53). One child was hospitalized for transient jaundice and cough of unclear cause. Two children died, 1 of whom died within 3 days of starting HAART. All other children tolerated HAART.
The absolute median CD4 count was 201 (IQR: 125 to 406) cells/mm3 at 3 months, 485 (IQR: 168 to 651) cells/mm3 at 6 months, 365 (IQR: 199 to 578) cells/mm3 at 9 months, and 340 (IQR: 315 to 538) cells/mm3 at 12 months in the previously ART-naive children. For ART-experienced patients, the absolute median CD4 count was 241 (IQR: 116 to 335) cells/mm3 at 3 months, 328 (IQR: 199 to 490) cells/mm3 at 6 months, 302 (IQR: 153 to 493) cells/mm3 at 9 months, and 318 (IQR: 132 to 423) cells/mm3 at 12 months.
The median change in CD4 count was calculated by comparing each time point with baseline for each patient (Fig. 1). When comparing the change in median CD4 cell count between time points within the previously ART-naive group, differences were found to be significant (P < 0.0001 to 0.006), except among the 6-, 9-, and 12-month time points (P = 0.27 to 0.43). In the ART-experienced group, significance was seen only from 0 to 6 months (P = 0.007), 0 and 3 to 9 months (P = 0.02 and P = 0.04, respectively), and 6 to 12 months (P = 0.04).
The absolute median viral load was 2.65 (IQR: <2.60 to 3.41) log10 copies/mL at 3 months, 2.99 (IQR: <2.60 to 3.57) log10 copies/mL at 7 months, 3.35 (IQR: 2.78 to 4.30) log10 copies/mL at 9 months, and <2.60 (IQR: <2.60 to 4.11) log10 copies/mL at 12 months for the previously ART-naive patients. For the ART-experienced patients, the absolute median viral load was 3.78 (IQR: 2.61 to 4.82) log10 copies/mL at 3 months, 3.99 (IQR: 3.06 to 4.79) log10 copies/mL at 7 months, 4.44 (IQR: 3.84 to 5.12) log10 copies/mL at 9 months, and 4.58 (IQR: 3.88 to 5.37) log10 copies/mL at 12 months.
The median change in viral load was also calculated by comparing each time point with baseline for each patient (see Fig. 1). In previously ART-naive patients, comparisons of the change in median viral load between time points revealed significance among all time points (P < 0.0001 to 0.006) except 7 to 12 months (P = 0.74) and 3 to 12 months (P = 0.17). In ART-experienced patients, significance was seen for 0 to 3 months (P = 0.001), 3 to 9 months and 12 months (both P < 0.0001), and 7 to 9 months and 12 months (both P = 0.003).
The proportion of children with <400 copies/mL is shown in Figure 2. Many detectable viral loads were <10,000 copies/mL in the previously ART-naive and ART-experienced children: 42 (80%) and 16 (50%) patients, respectively, at 3 months, 39 (77%) and 16 (50%) at 7 months, 32 (63%) and 10 (31%) at 9 months, and 32 (63%) and 9 (28%) at 12 months. The immunologic response was analyzed by virologic response and found to be statistically significant in ART-naive patients, but not previously ART-experienced patients (Table 2).
This report presents the first data on HAART outcomes for HIV-infected children in China and indicates many successes in the national program. After 1 year of HAART, all but 2 children survived and none stopped HAART because of side effects or toxicity. The increase in median CD4 count of nearly 300 cells/mm3 in the first 6 months and the initial decrease in median viral load of 2.58 log10 copies/mL for ART-naive patients are robust and indicate the ability of the program to deliver high-quality care.
The increases in median viral load at 7 and 9 months and lack of increase in median CD4 count at 9 and 12 months, however, indicate challenges within the program. Some decrease in the initial CD4 count rise would be consistent with reports of bimodal expansion of CD4 cells;9 however, a plateau, along with increasing viral replication, suggests problems in treatment. Further, the similar trends in both groups suggest that the causal factors occurred at the program level rather than at the individual, level. Potential treatment problems include waning adherence, inaccurate dosing, and/or drug interactions. Adherence is a particularly complex issue and can be affected by factors such as the child's age, disease severity, availability and degree of social support, and caregiver illness.10 The extent of concurrent ART and traditional medicine use is unknown, and studies of the prevalence and pharmacodynamics are needed.
The drop in median viral load at 12 months, however, suggests improvements in the treatment program. The cohort clinicians were aware of the trends in CD4 cell count and viral load data and strengthened the program midyear. Dosing accuracy was reviewed, and more emphasis was placed on adherence support. Alternatively, the variation in laboratory trends could reflect laboratory inaccuracy. Both laboratories, however, meet national quality standards, and the CD4 cell counts rose significantly in the ART-naive patients with fully suppressed viral loads, as would be expected.
Significant improvements in outcomes were not seen in the ART-experienced children. Rural China, where this cohort was located, is poor, and few medical resources are available. Local doctors treated these children to the best of their ability; however, before initiation of the national program, none were educated about pediatric-specific needs, including interpretation of CD4 cell counts by age, adjustment of medicines by weight, and adherence challenges. All these children were treated with adult formulations, typically requiring extensive pill splitting, and they probably received inaccurate doses. Moreover, caregivers for the children are poorly educated and were not well supported in ensuring their children's adherence to ART. Most of the ART-experienced children, therefore, likely had extensive ART resistance at entry into this study.
Many children in this cohort need second-line HAART, which is currently unavailable in China. Further, resistance to NVP, EFV, and 3TC is known to develop quickly,11,12 and the transient loss of viral suppression in numerous children may reflect dual therapy. Sustainability of their virologic suppression may thus be limited. Indeed, the high levels of virologic failure seen in this and other studies4,6 using WHO-recommended first-line HAART13 (NVP or EFV plus 2 nucleoside agents) suggest that alternative regimens, including protease inhibitors (PIs), may be useful in some settings. Whether for first- or second-line therapy, however, PIs present many difficulties, such as cost, heat instability, side effects, and drug-drug interactions.
Treatment of pediatric HIV/AIDS in developing areas is challenging, but the experience in China and elsewhere4-7 indicates that many children respond quite well. Accumulating data of high virologic failure rates, however, suggest that aggressive efforts to optimize access, adherence, and social support are critical to defer the need for second-line HAART.
The authors thank the clinical and CCDC staff in Shangcai County, Henan and Beijing Ditan Hospital laboratories; the Chi Heng Foundation; the Clinton Foundation HIV/AIDS Initiative; and Drs. Stephen Spector, Joseph H. Harwell, and Torsten Neilands.
1. The Joint United Nations Programme on AIDS, World Health Organization. AIDS Epidemic Update: December 2006
. Geneva, Switzerland: WHO; 2006.
2. Taha T, Graham S, Kumwenda N, et al. Morbidity among human immunodeficiency virus-1-infected and -uninfected African children. Pediatrics
4. Song R, Jelagat J, Dzombo D, et al. Efficacy of highly active antiretroviral therapy in HIV-1-infected children in Kenya. Pediatrics
5. Fassinou P, Elenga N, Rouet F, et al. Highly active antiretroviral therapies among HIV-1-infected children in Abidjan, Cote d'Ivoire. AIDS
6. Puthanakit T, Oberdorfer A, 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
7. Doerholt K, Duong T, Tookey P, et al, and the Collaborative HIV Paediatric Study (CHIPS). Outcomes for human immunodeficiency virus-1-infected infants in the United Kingdom and Republic of Ireland in the era of effective antiretroviral therapy. Pediatr Infect Dis J
8. Ministry of Health, People's Republic of China; Joint United Nations Programme on HIV/AIDS, World Health Organization. 2005 Update on the HIV/AIDS Epidemic and Response in China
. Beijing, China: Chinese Center for Disease Control and Prevention, 2006.
9. Essajee S, Kim M, Gonzalez C, et al. Immunologic and virologic responses to HAART in severely immunocompromised HIV-1 infected children. AIDS
10. Steele RG, Grauer D. Adherence to antiretroviral therapy for pediatric HIV infection: review of the literature and recommendations for research. Clin Child Fam Psychol Rev
11. Machado ES, Lambert JS, Afonso AO, et al. Overview of genotypic and clinical profiles of human immunodeficiency virus type 1-infected children in Rio De Janeiro, Brazil. An Acad Bras Cienc
12. Eshleman SH, Krogstad P, Jackson JB, et al. Analysis of human immunodeficiency virus type 1 drug resistance in children receiving nucleoside analogue reverse-transcriptase inhibitors plus nevirapine, nelfinavir, or ritonavir (Pediatric AIDS Clinical Trials Group 377). J Infect Dis
13. World Health Organization. Antiretroviral Therapy of HIV Infection in Infants and Children. Towards Universal Access
. Geneva, Switzerland: WHO; 2006.
Keywords:© 2007 Lippincott Williams & Wilkins, Inc.
developing world; highly active antiretroviral therapy; outcomes; pediatric