The estimated number of children younger than 15 years living with HIV worldwide increased to 2.5 million in 2009, and 90% of them lived in developing countries.1 Scaling up of antiretroviral therapy (ART) for children has lagged behind that for adults but is increasing with 29% of children younger than 15 years receiving the therapy they need in 2009.2 Morbidity and mortality of HIV-infected children has decreased significantly due to the provision of highly active antiretroviral therapy (HAART), and now, more children are surviving to adolescence and adulthood.3 However, treatment failure may appear in patients who have received long-term therapy, and in these patients, drug-resistant virus inevitably emerges. Children are particularly susceptible to treatment failure because of additional adherence challenges, their unique pharmacokinetics without adequate dosing guidelines, and a lack of age-appropriate formulations among other challenges.4-6 In high-income countries, it has been reported that children have more extensive drug resistance than adults7 and experience virologic failure at 12 months of treatment that is twice the rate seen among adults in resource-limited countries.8
The China national-free pediatric antiretroviral therapy program (NPATP) sponsored by National Center for AIDS/STD Control and Prevention (NCAIDS) was started in July 2005.9 The current first-line HAART regimens recommended for children with HIV in China are zidovudine (AZT)/stavudine (D4T) plus lamivudine (3TC) plus nevirapine (NVP)/efavirenz (EFV). The cumulative number of HIV-infected children in care was 5162 by the end of June 2009, and 1529 received ART.9 The short-term treatment efficacy and resistance patterns of Chinese children in the NPATP were reported in a previous study.10 Because resources were limited, regular viral load (VL) monitoring and genotypic resistance testing were not widely available. Most of the pediatric antiretroviral drugs for this program were donated. The limited variety and complex import process have restricted the timely availability of drugs. Second-line medicine was not provided until donated drugs were received in 2008.9
Limited availability of second-line ART regimens and accumulation of resistance mutations to the current nonnucleoside reverse transcriptase inhibitor (NNRTI)-based combination is a major challenge for many developing countries. The effectiveness of second-line ART regimens in these settings is not clear. In this article, we focus on those cases experiencing virologic failure, with an analysis of their resistance profiles before a second-line ART pilot program was started. Then, we observed 12-month treatment outcomes after switching to second-line regimens.
MATERIALS AND METHODS
Study Site and Participants
The NPATP provided patients with free antiretroviral medicine and free CD4 testing. Basing on the NPATP, NCAIDS established the first observational pediatric ART cohort in the rural areas of central China in July 2005. This site is located in Shangcai County, Henan province, which was affected by previous plasma donation practices. The HIV-positive children in this site account for nearly 20% of the treated children in all of China through March 2009. Because this site has the largest number of treated children and the longest duration of providing treatment, we aimed to monitor the treatment outcome in this open-label cohort by collecting more detailed treatment data and providing extra laboratory testing beyond what was available through the national program. Due to a lack of second-line medicine before 2008, patients who met virologic failure criteria continued to take their failing regimen. Beginning in December 2007, we screened patients with virologic failure, performed resistance testing, and provided second-line regimens. Participants who had received HAART for more than 12 months and had VL >1000 copies per milliliter on 2 consecutive measurements with or without immunological failure were enrolled. We aimed to assess the resistance associated with treatment failure to first-line combinations and the feasibility and efficacy of the standard second-line treatment regimen. The 2 nucleoside reverse transcriptase inhibitor (NRTIs) + NNRTI was the most common first-line regimen. Abacavir (ABC), the fixed dose coformulated combination of AZT/3TC/ABC, and lopinavir/ritonavir (LPV/r) were the medicines available for use in second line, which could be combined with any first-line drug as deemed appropriate. This study was approved by the institutional review board of NCAIDS. Participants and all the parents signed informed consent.
We collected ART information from local medical records, including demographic characteristics, medical history, hematology, and chemistry laboratory results at the same time of the resistance test. We performed CD4 and VL testing every 3 months after switching to the second-line regimen. The follow-up data about second-line treatment were collected by trained local doctors using specified case report forms.
Absolute CD4 cell counts were performed on whole blood by FACS Calibur (BD Biosciences, San Jose, CA) at the local treatment site within 24 hours of phlebotomy. Virology specimens were processed locally within 6 hours of collection. After plasma separation, all plasma samples were frozen at −80°C and transported to central laboratories using dry ice. VL and genotypic resistance assays were performed at Beijing Ditan Hospital. VL testing was performed using the Cobas Amplicor HIV-1 Monitor (Roche Diagnostics, Indianapolis, IN), with a lower detection limit for HIV-1 RNA of 50 copies per milliliter. All samples with HIV-1 RNA >1000 copies per milliliter were subsequently sequenced for drug resistance mutations. Genotypic resistance mutation analysis was performed using the ViroSeq HIV-1 Genotyping System (Abbott Molecular, Des Plains, IL). In this reaction, the entire protease gene and two-thirds of the reverse transcriptase gene are amplified. ViroSeq HIV-1 Genotyping System Software with a proprietary algorithm analyzes the mutations and generates the final drug resistance report. Thymidine analogue mutations (TAMs) were defined as those that contain M41L, D67N, K70R, L210W, T215Y/F, and K219Q/E.
Characteristics at baseline such as age, sex, route of transmission, CD4 T-cell count, VL, duration on first-line treatment, first-line regimen, second-line regimen, and mutation status were described with median [interquartile range (IQR)] or proportion as appropriate. To compare the CD4 and VL of baseline with 12 months later, the Wilcoxon signed-rank test was used. A P value less than 0.05 was considered to be statistically significant. Analysis was performed using SPSS version 15.0.
From July 2005 to March 2009, 256 HIV-infected children consecutively received ART at Shangcai County pediatric AIDS clinic, with 1 lost to follow-up, 5 stopped ART, and 5 deaths. As of March 2009, 124 (48%) children were identified as having failed their first-line ART with a VL >50 copies per milliliter. The remaining 121 (47%) children were VL <50 copies per milliliter. Between December 2007 and March 2009, among the 124 children with virologic failure, 48 children had 1 time VL <1000 copies per milliliter or between 50 and 1000 copies per milliliter, whereas the remaining 76 children with more than 2 consecutive VL >1000 copies per milliliter provided samples for resistance testing (Fig. 1). Among these 76 patients, the median time between the first laboratory evidence of virologic failure with VL >50 copies per milliliter and the performance of resistance testing was 23.9 (IQR: 16.7-24.0) months. All remaining patients who had virologic failure continued to receive VL monitoring with routine switching to second-line regimen after the screening period as indicated.
Baseline Clinical Characteristics at the Time of Resistance Testing
The median age of these 76 children was 13.9 (IQR: 11.1-16.0) years, and 51 (67.1%) were male. Sixty (78.9%) were infected through mother-to-child transmission and 15 (19.7%) by contaminated blood products. The median first-line HAART duration was 33.1 (IQR: 24.2-40.8) months. The regimens at enrollment were all 2NRTIs + 1NNRTI. Sixty-nine (90.8%) children received NVP-based regimen, whereas 7 (9.2%) received EFV-based regimen. The AZT to D4T ratio was nearly 1:1. Sixty-seven (88.2%) children had 3TC in their regimen, whereas 9 (11.8%) children had didanosine (ddI) in their regimen. Forty-nine of 76 children in this study had a history of receiving adult formulation medications before the start of the national pediatric treatment program. Because the pediatric program did not start until July 2005, patients or family members often shared their adult formulation drugs with their children and could not describe clearly the dose or duration. The median CD4 count was 143 (IQR 61, 255) cells per cubic millimeter, and the median VL was 4.6 log10 copies per milliliter before regimen switch. Among these children, 7 (9.2%) were double orphans and 19 (25%) were single orphans. Nineteen (25%) children had their ART administered by grandparents or relatives (Table 1).
Antiretroviral Resistance-Associated Mutations
All 76 patients had resistance mutations. All of them were subtype B, which is the predominant subtype associated with plasma donation in central China.11 The frequency of relevant mutations in the reverse transcriptase enzyme for NNRTI class was Y181C/I 37 (48.7%) and K103N/S 38 (50%) (Table 2). Mutations affecting NRTI class included M184V/I, observed in 59 (77.6%) children. Twenty-five (32.9%) children harbored at least 3 TAMs and 18 (23.7%) children had 1-2 TAMs. T215Y/F and M41L were more common than other mutations and accounted for 44 (57.9%) and 34 (44.7%), respectively. Additionally, 9 (11.8%) children harbored the Q151M mutation, 7 (9.2%) the K65R mutation, and 4 (5.3%) the L74V/I mutation (Table 2). Twenty-six percent of children had multiple resistance with M184V/I, ≥3 TAMs, and 1 major NNRTI resistant.
98.7% and 81.6% of genotyped children harbored high-level resistance mutations to NVP and EFV, respectively (Fig. 2). In NRTI class, high-level resistance to 3TC reached 82.9%, followed by 57.9% for D4T and 52.6% for AZT (Fig. 2). Even for second-line drugs, resistance genotyping analysis revealed intermediate- to high-level resistance to ABC (72.4%), ddI (69.7%), and tenofovir (56.6%) (Fig. 2). Meanwhile, the prevalence of resistance to multiple drugs was high. Fifty-three percent of the children had concurrent 3TC, AZT, or D4T and NNRTI high resistance, and another 17% children had concurrent 3TC, ABC, or tenofovir or ddI and NNRTI high resistance.
In the protease gene, there were polymorphisms and mutations such as V77I and A71TV but no major protease inhibitor (PI)-associated mutations.
Among 76 children, 65 switched to second-line regimen medications, 5 died before switching regimens was possible, and 6 remained on their failing first line because of uncorrectable adherence problems or parental refusal to change regimen. The combination of ABC, 3TC, and LPV/r is the standard available second-line regimen in China and was the most common regimen used, accounting for 43 (66.2%) regimens in this study population. Four (6.2%) children continued on their original nucleosides due to residual susceptibility to AZT and only had their NNRTI replaced with LPV/r. Eighteen (27.7%) used coformulated ABC, 3TC, and AZT with LPV/r, but after 2.63 (IQR: 2.56-4.04) months, 13 of these (72%) changed to ABC + 3TC + LPV/r due to moderate anemia, with median hemoglobin of 75 (IQR: 60, 82) g/L. The remaining 5 patients continued taking coformulated AZT/3TC/ABC with LPV/r during the period of follow-up observation.
Treatment Efficacy After Switching to Second-Line Regimen
Among the 65 cases switched to a second-line regimen, CD4 counts increased from a median of 143 (IQR: 61-255) cells per cubic millimeter at switch baseline to 399 (IQR: 236-643) cells per cubic millimeter after 12 months of receiving second-line therapy (Fig. 3). The CD4 absolute increase was 256 cells per cubic millimeter. This increase was statistically significant (P < 0.001). There were 54 (83.1%) children with undetectable VL (<50 copies/mL) after 12 months of second-line therapy (Fig. 3), this was statistically significant compared with switch baseline (P < 0.001). Among the remaining 11 cases with detectable VL, 4 (6.2%) were between 50 and 1000 copies per milliliter and 7 (10.8%) had VLs over 1000 copies per milliliter after 12 months. Logistic regression analysis failed to detect any statistically significant differences between responders and nonresponders with regard to second-line regimen, duration of time on first-line regimen, resistance mutations at failure, CD4 cell count, VL, or age.
HAART has turned HIV infection from a fatal disease into a chronic and manageable illness, and more HIV-infected children are surviving into adolescence.12 Presently, most HIV-infected children live in developing countries, which make access to lifelong ART a major challenge. In our study, we found that nearly 50% of patients had virologic failure to their first-line therapy. This result is substantially higher compared with other studies. A study from Tanzania demonstrated that 31.6% children with a median first-line ART duration of 2.4 years had virologic failure.13 The virologic failure rate was 19.3% among 5485 children with median follow-up of 16 months in South Africa.14 Predictors of virologic failure that have been identified elsewhere include younger age, the use of an NVP-containing regimen compared with other combinations, exposure to prevention of mother-to-child transmission regimens, low baseline CD4, and reported nonadherence.13,14 There may be several reasons why our cohort had a higher rate of virologic failure compared with other cohorts. Patients in our cohort all lived in a rural area with transportation challenges and limited access to sophisticated facilities, training, and support services. The patients with virologic failure who were enrolled had a mean age of 13.9 years. Having to manage a chronic medical illness during the difficult developmental stages of adolescence is an additional challenge to pediatric HIV care that can affect adherence and treatment durability. Twenty-five percent of children were in the care of grandparents or other relatives. Elderly caregivers in other settings have been reported to have difficulty in adhering to the strict requirements of ART because of deteriorating memory, impaired vision, or poor comprehension about modern biomedical therapeutics.15 Most of these patients had a history of taking adult formulations with potentially inaccurate dosing, which likely affected the subsequent virological outcome and patterns of resistance mutations. Higher rates of virologic failure and resistance have been previously reported among children in this area receiving adult formulation ART.10,16
When virologic failure occurs, genotypic resistance testing is recommended before starting a second-line therapy. Our observational study shows the profile of drug resistance in HIV-infected children who received long-term first-line regimen therapy despite virologic failure. Many HIV-1-infected children with virologic failure harbored strains with multiple resistance mutations. Because of a low genetic barrier to resistance to 3TC and NNRTIs, resistance to these drugs was commonly observed in our cohort, consistent with other reports.17-19 One study from Thailand showed a similar pattern of genotypic resistance mutations, with 85% of 120 virologically failing children harboring the M184V/I mutation and high-level resistance to 3TC after 2 years of first-line treatment.20 In that study, the proportion of patients with at least 1 NNRTI mutation, at least 4 TAMS, Q151M complex, and K65R were 98%, 23%, 12%, and 5%, respectively. Complex resistance mutations often accumulate gradually. In a cohort from Uganda, M184V- and NNRTI-associated mutations appeared within 6 months of virological failure but the emergence of TAMs occurred after 12 months.21 The higher rate of resistance may reflect the delay in regimen switching where the availability of VL and resistance testing is limited. Long-term treatment with a failing first-line regimen leads to the accumulation of more resistance mutations.7 It will be a significant challenge for future ARV choices in these young patients harboring highly resistant virus.
As with most pediatric programs in the developing world, second-line drugs were available only with difficulty despite a long apparent need. In our cohort, extensive cross-resistance was observed to nearly all reverse transcriptase inhibitors of the nucleoside and nonnucleoside classes. This poses significant challenges for the choice of subsequent regimens for these children. Considering the accumulation of multiple mutations and the evidence that AZT prevents future development of K65R,22,23 we chose to use coformulated ABC, 3TC, and AZT as a second-line drug combination for those who had severe resistance to NRTIs, although the durability of this combination was limited by anemia. First-line regimen failures in this cohort did not have major PI resistance mutations, which made a boosted PI the optimal choice for their second-line regimen after failure of NNRTIs.
In reality, the currently available regimens are often a result of compromise and practical considerations rather than optimal drug selection.24 But what was observed unexpectedly in our research was that despite intermediate or high-level resistance to commonly available second-line nucleoside agents, children failing first-line ART had good 1-year virologic and immunologic outcomes when switched to new regimens. CD4 counts increased 256 cells per cubic millimeter, and 83.1% had undetectable VL (<50 copies/mL). This boosted PI-based regimen demonstrated satisfactory effect even without an optimal nucleoside companion drug. Comparable studies of the efficacy of second-line ART are limited in children. A recently published study from Thailand showed that 75% of children on 1.5 years PI-based ART had VL <50 copies per milliliter among children who failed dual NRTIs.25 Another study of 50 children showed a 14% increase in CD4% with 74% achieving VL <50 copies per milliliter after 96 weeks of receiving double-boosted PI as second-line therapy.26 It has been shown that ritonavir-boosted PIs result in a higher rate of virologic suppression and have a high genetic barrier to viral resistance, even at lower adherence rates.27-29 From our study, the implication is that the functional monotherapy effect of a boosted PI was sufficient for some patients with multiple resistance mutations. The efficacy of LPV/r monotherapy as a simplification strategy was tested in a Thai cohort of children on a dual PI combination as second line who had achieved virological suppression.30 In that study, 72% of subjects maintained their virologic suppression. A recent study of adults treated with LPV/r as second-line monotherapy after failing first line found that 87% of patients had VL <400 copies per milliliter after 24 weeks in an as-treated analysis.31 Another study of LPV/r monotherapy in pregnant women for prevention of mother-to-child HIV transmission found that 88% of women had VL <200 copies per milliliter after 8 weeks in the intent-to-treat analysis.32 These studies all suggest the use of LPV/r or an alternative ritonavir-boosted PI monotherapy as a strategy for second-line therapy in children deserves further study.
Most of the children in this cohort were delayed in switching to a second-line regimen and remained on first-line therapy after virologic failure for additional 23.9 (IQR: 16.7, 24.0) months because of a lack of availability of second-line drugs at the early stage of our ART program. A study in the United States showed that children who experienced a delayed regimen switch were less likely to maintain immunological response.7 Early detection of virologic failure might allow for more timely switching and a more favorable outcome from second-line therapy.
There are some limits in our data. First, all these children live in rural areas where medical resources including skilled medical workers are limited. Many children with HIV have their disease managed by clinicians who lack pediatric training and are not accustomed to monitoring growth and adjusting treatments accordingly.
Second, although adherence rate is highly associated with the treatment success, we were unable to include this parameter in our analysis. It is difficult to have accurate assessments of adherence in rural areas because the local health care workers were unable to perform objective measures such as pill counts. Adherence by self-report commonly was 100% in all patients, but this was often at odds with other information. Further study is needed to determine ways we may intervene effectively with our patients to improve their adherence given their unique challenges.
Third, resistance testing can only be performed in the national laboratory, and even virologic testing cannot be carried out routinely at local sites, which creates barriers for timely laboratory monitoring. Access to such monitoring is improving, and it is hoped that earlier detection of virologic failure before broad cross-class resistance emerges can help to preserve treatment options for the construction of more effective salvage regimens.
Finally, this was an observational study, and all the failure cases came from 1 clinic from the free pediatric ART program, which may not reflect the characteristics of resistance and efficacy of second-line therapy in other settings in China. However, these patients come from the areas in China where the greatest number of HIV-infected children reside and as such are likely to represent the situation for most children with HIV in the country.
In conclusion, HIV-1-infected children in this study population who continued their first-line regimen despite virologic failure harbored multiple resistance mutations. VL monitoring needs to be strengthened to identify treatment failure earlier and preserve treatment options. Although the extent of resistance to the NRTI class limits subsequent treatment options, boosted PI-containing second-line regimens remain effective during this short-term observation. Further efforts on patient education and improved adherence are needed to enhance the antiretroviral therapies for children.
The authors would like to thank the health care staff in this pediatric antiretroviral treatment center.
3. Brady MT, Oleske JM, Williams PL, et al. Declines in mortality rates and changes in causes of death in HIV-1-infected children during the HAART era. J Acquir Immune Defic Syndr
4. Nakhmanina NY, van den Anker JN. Treating an HIV-infected paediatric patient: an easy task? Antivir Ther
5. Burger D, Ewings F, Kabamba D, et al. Limited sampling models to predict the pharmacokinetics of nevirapine, stavudine, and lamivudine in HIV-infected children treated with pediatric fixed-dose combination tablets. Ther Drug Monit
6. Puthanakit T, van der Lugt J, Bunupuradah T, et al. Pharmacokinetics and 48 week efficacy of low-dose lopinavir/ritonavir in HIV-infected children. J Antimicrob Chemother
7. Agwu A, Lindsey JC, Ferguson K, et al. Analyses of HIV-1 drug-resistance profiles among infected adolescents experiencing delayed antiretroviral treatment switch after initial nonsuppressive highly active antiretroviral therapy. AIDS Patient Care STDS
8. Kamya MR, Mayanja-Kizza H, Kambugu A, et al. Predictors of long-term viral failure among Ugandan children and adults treated with antiretroviral therapy. J Acquir Immune Defic Syndr
9. Zhao Y, Sun X, He Y, et al. Progress of the National Pediatric Free Antiretroviral Therapy program in China. AIDS Care
10. Zhang FJ, Haberer J, Wei H, et al. Drug resistance in the Chinese National Pediatric Highly Active Antiretroviral Therapy Cohort: implications for paediatric treatment in the developing world. Int J STD AIDS
11. Li JY, Li HP, Li L, et al. Prevalence and evolution of drug resistance HIV-1 variants in Henan, China. Cell Res
12. Violari A, Cotton MF, Gibb DM, et al. Early antiretroviral therapy and mortality among HIV-infected infants. N Engl J Med
13. Emmett SD, Coleen K, Cunningham, et al. Predicting virologic failure among HIV-1-infected children receiving antiretroviral therapy in Tanzania: a cross-sectional study. J Acquir Immune Defic Syndr
14. Davies MA, Moultrie H, Eley B, et al. Virologic failure and second-line antiretroviral therapy in children in South Africa—The IeDEA Southern Africa Collaboration. J Acquir Immune Defic Syndr
15. Skovdal M, Campbell C, Madanhire C, et al. Challenge faced by elderly guardians in sustaining the adherence to antiretroviral therapy in HIV infected children in Zimbabwe. AIDS Care
16. Zhang FJ, Haberer JE, Zhao Y, et al. Chinese pediatric highly active antiretroviral therapy observational cohort—a 1-year analysis of clinical, immunologic, and virologic outcomes. J Acquir Immune Defic Syndr
17. Jittamala P, Puthanakit T, Chaiinseeard S, et al. Predictors of virologic failure and genotypic resistance mutation patterns in Thai children receiving non-nucleoside reverse transcriptase inhibitor-based antiretroviral therapy. Pediatr Infect Dis
18. Janssens B, Raleigh B, Soeung S, et al. Effectiveness of highly active antiretroviral therapy in HIV-positive children: evaluation at 12 months in a routine program in Cambodia. Pediatrics
19. 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
20. Puthanakit T, Jourdain G, Hongsiriwon S, et al. HIV-1 drug resistance mutations in children after failure of first-line nonnucleoside reverse transcriptase inhibitor-based antiretroviral therapy. HIV Med
21. Ruel TD, Kamya MR, Pelin Li, et al. Early virologic failure and the development of antiretroviral drug resistance mutations in HIV-infected Ugandan children. J Acquir Immune Defic Syndr
22. Hurwitz SJ, Asif G, Kivel NM, et al. Development of an optimized dose for coformulation of zidovudine with drugs that select for the K65R mutation using a population pharmacokinetic and enzyme kinetic simulation model. Antimicrob Agents Chemother
23. Anderson PL, Rower JE. Zidovudine and lamivudine for HIV infection. Clin Med Rev Ther
24. Sohn AH, Nuttall JJC, Zhang FJ. Sequencing of antiretroviral therapy in children in low- and middle-income countries. Current Opinion in HIV and AIDS
25. Bunupuradah T, Suntarattiwong P, Li A, et al. Antiretroviral treatment outcome following genotyping in Thai children who failed dual nucleoside reverse transcriptase inhibitors. Int Soc Infect Dis
26. Bunupuradah T, van der Lugt J, Kosalaraksa P, et al. Safety and efficacy of a double-boosted protease inhibitor combination, saquinavir and lopinavir/ritonavir, in pretreated children at 96 weeks. Antivir Ther
27. Swindells S, DiRienza AG, Wilkin T,et al. Regimen simplification to atazanavir-ritonavir alone as maintenance antiretroviral therapy after sustained virologic suppression. JAMA
28. Llibre JM. First-line boosted protease inhibitor-based regimens in treatment-naive HIV-1-infected patients—making a good thing better. AIDS Rev
29. von Hentig N. Lopinavir/ritonavir: appraisal of its use in HIV therapy. Drugs Today (Barc)
30. Bunupuradah T, Kosalaraksa P, Puthanakit T, et al. Monoboosted lopinavir/ritonavir as simplified second-line maintenance therapy in virologically suppressed children. AIDS
31. Bartlett JA, Aga E, Ribaudo H, et al. A pilot study of LPV/r monotherapy following virologic failure of first line NNRTI-containing regimens in resource-limited settings: week 24 primary analysis of ACTG5230 [abstract 583]. Presented at: XVIII Conference on Retroviruses and Opportunistic Infections; March 2, 2011; Boston, MA.
32. Tubiana R, Mandelbrot L, Delmas S, et al. LPV/r monotherapy during pregnancy for PMTCT of HIV-1: the PRIMEVA/ANRS 135 randomized trial, pregnancy outcomes [abstract 125lb]. Presented at: XVIII Conference on Retroviruses and Opportunistic Infections; March 2, 2011; Boston, MA.