The Universal Coverage Health Program (UC), managed by the National Health Security Office (NHSO) since the end of 2007, has been providing HIV treatment and care for more than 70% of the Thai HIV-infected population. The UC provides health care benefit to all Thai people who are not covered by social security and the civil service welfare schemes. In 2014, approximately 75.29% of Thai people have been covered by UC, 15.24% covered by social security, and 7.89% covered by the civil service welfare scheme. The NHSO administered all these schemes, and the NHSO database was used to evaluate UC schemes.1 In 2014, 4502 HIV children were receiving antiretroviral therapy (ART) through UC.2 The program has expanded coverage and availability of ART drugs rapidly. Treatment regimens follow the national guidelines, which are developed by country experts, and are periodically revised, mostly in line with World Health Organization guidelines.
The best outcomes of ART require early treatment initiation and good adherence.3,4 A previous study in a multiregional cohort in Asia reported that the probability of achieving immunological recovery (IMR) increased to 77% by 4 years after ART initiation, and was associated with younger age and higher CD4% at ART initiation.5 A previous study in Thailand reported a higher chance of IMR in children who initiated ART at a higher CD4%, and with either nonnucleoside reverse-transcriptase inhibitor (NNRTI)-based or protease inhibitor (PI)-based highly active antiretroviral therapy (HAART).6,7 Despite IMR, virological failure (VF) still occurred especially during the first 2 years of ART because of suboptimal adherence.8–11 Consequently, there have been an increasing cumulative number of children and adolescents failing treatment and requiring second-line regimens. The cost, pill load, and long-term adverse effects of the second-line regimen have become challenging issues for long-term management of HIV-infected Thai children.
To date, most Thai HIV-infected children were receiving ART through UC. All clinical data of the children receiving ART through UC have been registered in the NHSO database, which continues to record treatment characteristics of each visit, and includes periodic linkage with death registry to ensure vital status is complete. In this study, we assessed the treatment outcomes in first-line ART, and identify factors associated with switching to second-line regimens, in HIV-infected children treated through the Thai national program.
The UC was expanded from National AIDS Program and administered by the NHSO to provide free treatment for HIV-infected patients in Thailand. Patients who registered into UC scheme were recorded at registration and any follow-up visit through the NHSO database. The UC provides free ART with CD4 testing twice a year and HIV-RNA testing once a year after ART initiation. This study included children who initiated ART between 2008 and 2014 aged <15 years, and had received ART for at least 6 months and also had at least 1 CD4 and Viral load (VL) results recorded in the NHSO database. Children were prescribed a first ART regimen using at least 3 drugs including an NNRTI or PI, plus 2-3 nucleoside reverse-transcriptase inhibitors (NRTIs), in line with the national guidelines. According to national guidelines during the study period, NNRTI [nevirapine (NVP) in younger than 3 years and efavirenz (EFV) in older than 3 years] with combination of 2 NRTIs were the recommended first-line regimen for children. For infants younger than 12 months exposed to NVP as a part of prevention of mother-to-child transmission (PMTCT), the recommended first regimen was lopinavir/ritonavir plus 2 NRTI.12 We excluded children who were treated before enrollment in UC due to the lack of information at ART initiation. Baseline was defined as the time of ART initiation. Baseline CD4 level results included those performed within 3 months before ART initiation and baseline viral load results within 1 year of ART initiation. Children were considered lost to follow-up if they had no clinic contact for more than 6 months from their last visit. This study was approved by the Institutional Review Board of the Institute for Development of Human Research Protection, Ministry of Public Health, Thailand. The NHSO database was deidentified by NHSO personnel before analysis.
Definition of Treatment Outcomes and Switching ART Regimens
All children were assessed for IMR after ART initiation. IMR was defined as CD4% ≥25% at any age, or CD4 ≥500 cells/mm3 at age ≥5 years. VF was defined as HIV-RNA ≥1000 copies/mL and immunological failure (IMF) was defined as CD4 levels <200 cells/mm3 or <10% at age <5 years, and CD4 levels <100 cells/mm3 at age ≥5 years.13 Children who had VF or IMF were considered as treatment failure and censored at the time of switching to second-line regimen. We defined switching to second-line regimens as a class switch from NNRTI to PI or PI to NNRTI after at least 6 months of treatment following national guidelines. We did not consider those who switched among single drug (NRTI) because of toxicities or drug out of stock as second-line switches. The reasons of changing regimens were ascertained as treatment failures (VF or IMF) based on laboratory or clinical results where possible. Children who had not switched ART were censored at the last visit date.
Baseline characteristics, including baseline demographics, age, year, regions, antiretroviral regimens, and clinical stage, and CD4 levels for all children and children who switched to second-line regimen, were summarized using descriptive statistics. The Kaplan–Meier method was used to estimate IMR and VF. Cox models were used to assess predictors of IMR, IMF, and VF. Covariates included baseline age, sex, year of ART initiation or calendar year by time update, baseline CD4 percentage, country regions, and first ART regimen. The competing risk methods of Fine and Gray14 was used to calculate the subdistribution hazard ratios (SHRs) to assess associations between baseline characteristics and switching ART regimen, with death considered as a competing risk. Cumulative incidence rates of switching ART regimen and death were generated using competing risk estimators. Covariates assessed for association with switch to second-line ART included baseline age, sex, World Health Organization stage at baseline, baseline CD4%, first ART regimen, year of starting ART, and regions where children started treatment.
Variables with P <0.10 were considered for inclusion in multivariate models. Statistical significance was identified using a 2-sided P value less than 0.05. Statistical analysis was performed with SAS version 9.4 (SAS Institute Inc, Cary, NC) and with Stata version 14 (StataCorp, College Station, TX).
A total of 4810 children were eligible. We excluded 315 children who had been on ART <6 months and 375 children who had no CD4 and viral load testing after treatment. Four thousand one hundred twenty children were included in the analysis (Supplemental Digital Content, Figure S1, http://links.lww.com/QAI/A987).
At ART initiation (Table 1), the median age was 9.3 years [interquartile range (IQR) 5.9–12.0] and 55% (n = 2283) were girls. The majority were in the Centers for Disease Control and Prevention classification (CDC) stage N or A (n = 3156, 77%) and were prescribed NNRTI-based ART (25% EFV and 66% NVP) that largely contained 3 TC/d4T (49%) as the first regimen. Most children came from Northeastern Thailand, and started ART in 2008. The number of children initiating ART has decreased continuously over the time. Of the 2794 children who had a baseline CD4% measured, the median CD4% was 9.7 (IQR 3.2%–17.0%). One year after ART initiation, 3107 (75%) had viral load measured with a median log10 VL of 1.6 (IQR 1.6–2.6) copies/mL. The median duration of follow-up was 3.7 (2.1–5.5) years after ART initiation with 17,950 person-years of follow-up. One hundred eighty-two children died and 270 children were lost to follow-up. Eighty-four (46%) deaths occurred before switching regimen (Supplemental Digital Content, Table S1, http://links.lww.com/QAI/A987).
Treatment Outcomes and Switching Regimens
The overall estimated probability of IMR was 40% [95% confidence interval (CI): 38% to 44%] at 1 year after ART and 69% (95% CI: 67% to 70%) at 2 years, and increased to 76% (95% CI: 75% to 78%) by 3 years after ART initiation (Fig. 1).Table 2 shows that factors associated with IMR. The multivariate model revealed that girls [adjusted hazard ratio (aHR) 1.13, 95% CI: 1.05 to 1.21] had higher change of IMR than boys and also children who initiated ART at younger age were more likely to achieve IMR than those who were aged ≥12 years. ART initiation in more recent years was associated with improved IMR, whereas first regimens, baseline CDC stage, and region had no significant association with IMR (Table 2).
After at least 6 months ART, 449 children had IMF, with a median duration from baseline to IMF of 48 months (26–66).Of the 1412 (34%) children who had VF, the median duration from baseline to VF was 35 months (16–58). The estimated probabilities of VF at 1, 2, and 3 years were 11%, 25%, and 32%, respectively. In multivariate analysis, the same predictors were associated with both VF and IMF. Children who started ART at age ≥12 years had the highest chance of treatment failure. The children who were treated before the year 2011 also were at higher risk of VF and IMF. Moreover, CD4% of <16% and unknown CD4% were associated with an increased risk of treatment failure. Children living in Bangkok had a lower risk of treatment failure (Supplemental Digital Content, Table S2, http://links.lww.com/QAI/A987).
Of the 1054 (26%) children who switched to a second-line regimen, the median age was 10 years. The cumulative incidence rates of changing regimen were 4% at 1 year, and increased to 13% by 2 years, and 20% by 3 years after ART initiation (Fig. 2). Of the 182 deaths, there was no trend of change in mortality over the 3 years of ART. The ascertained reasons of switching ART regimen were VF (n = 706; 67%), IMF (n = 16; 2%) and 181 (17%) children had both VF and IMF, whereas 151 (14%) children had no information on the reason for switching. The median duration from VF, and IMF, to switching was 6 and 8.4 months, respectively. In multivariate analysis, children who initiated ART at age <5 years [adjusted SHR (aSHR) 1.29, 95% CI: 1.07 to 1.55] and age ≥12 years (aSHR 1.33, 95% CI: 1.12 to 1.57) had a higher chance of switching to second-line regimen than those who were aged 5 to < 9 years. The children who started with NNRTI-based regimens as the first-line regimen (aSHR 1.61, 95% CI: 1.23 to 2.12) were more likely to switch regimen than those who were prescribed with PI-based regimen. The children who had baseline CD4% < 10% had a higher chance of switching regimen than those having baseline CD4% ≥16%. There was an increasing trend of changing regimen in more recent years of ART initiation (Table 3).
This is the first report of treatment outcomes in children receiving care through the UC in Thailand. It reflects real-life practice and outcomes in a resource-limited setting with established national guidelines and a systematic program. Our findings showed good immunologic outcomes, with high rate of IMR of 76% by 3 years of treatment. We found that 13% children switched to second-line regimen by 2 years, and 20% by 3 years of first-line ART. According to the national guidelines, the patients should be switched to second line as soon as possible after treatment failure to prevent extensive development of resistance mutations.12 We found that most children switched to a second-line regimen within 6 months of VF; those initiating ART during early adolescence (ie, aged 12–15 years) had a high risk of switching treatment regimens.
We found that children who started ART at a younger age had a higher probability of achieving IMR, similar to the reports from high-income countries.15,16 Moreover, other factors associated with immune recovery including female sex, higher baseline CD4%, and more recent years of ART initiation were similar to other previous reports from the region and underscored the need for early initiation of ART.5–8 We also found that a low IMF rate of 11% indicated early detection and switching of ART in those who had treatment failure in this study. It is known that CD4 monitoring is insensitive to detecting treatment failure; however, CD4 measurement is still used for immunological treatment failure where viral load testing is limited.11,17,18 Our findings demonstrate that annual VL and semiannual CD4 testing are important for patient monitoring and seem to be working well in the Thai UC program for guiding appropriate regimen switching.
We estimated the probability of VF to be approximately 10% per year. However, the 1-year probability of VF (the first detectable VL ≥1000 copies/mL) in our data was less than reported by the IeDEA Southern Africa Collaboration study, but with a similar rate by 3 years after ART initiation.9 Most children who switched to a second-line regimen did so within 2 years of the first-line treatment in both young children and adolescents. This finding was similar to previous studies in Western countries10 and Southern Africa.9 We also found that a lower baseline CD4%, more recent year of ART initiation, and initiating with NNRTI-based regimens were associated with an increased risk of regimen switching, similar to data from UK and Ireland cohorts.19 In accordance with the national guidelines, NNRTI-based ART was preferred as the first-line regimen and PI-based ART was provided as the second-line regimen for children who had first-line failure. Twenty-six percentage of children developed VF, IMF, or clinical failure from our findings. They switched to second-line within 6 months after VF. VL monitoring was able to detect treatment failure early, and UC has been providing free VL testing once a year after ART initiation in Thailand.12 Our findings suggest that initiating first-line ART with a PI-based regimen should be considered for children who are adolescents and those who had low baseline CD4 count.
The strength of our study is that it reflects the entire ART program in Thailand since 2008 and shows treatment outcomes on a national scale. More than 98% of HIV-infected children were receiving ART through UC and 2% covered by other schemes from this database. However, challenges do remain for the management of long-term adverse effects of second-line regimens, and the inevitable subsequent need for third-line regimens into the future as these children grow up to adolescents and adults. There are a number of limitations to our analysis. First, there are no information of exposure and PMTCT history collected in the UC program which may be predictors of treatment failure and regimen switching. However, we believe that 97% of HIV-infected children by the year 2011 were perinatally infected6 and approximately 8% of children were exposed to NVP from PMTCT, because these children were started with PI based as their first regimen. Second, information on adherence, physical status, weight, and height were mostly missing in the database, including these covariates in analyses. Third, there are no reasons recorded in the database to identify the cause of switching regimen. We defined the reason for switching based on CD4 and VL measurements at visits before a second regimen, but we could not identify IMF or VF as the reason for switching in 14% of children who did switch. Last, in terms of social factors, there were no information on parent status, caregiver status, education status, disclosure, stigma, traveling, and distance between home and the health care centers. Despite these limitations, we believe that the CD4, VL, and ART data are highly reliable, as these data are directly linked between the laboratory and antiretroviral supply.
In conclusion, children receiving ART through UC in Thailand had good treatment outcomes, with over 76% of reaching CD4 recovery, and most children who switched regimen did so appropriately within 6 months after VF. Earlier treatment was associated with IMR, and IMR outcomes improved over time. A switch to a second-line regimen was found at approximately 10% per year. Our findings demonstrate the effectiveness of the Thai national UC, and support early first-line treatment in all children, and also underscored treatment support for adolescents.
The data set was provided by the Thai NHSO and the Ministry of Health in Thailand.
1. UN. Universal Health Care Scheme (UHCS). 2014. Available at: https://webapps01.un.org/nvp/indpolicy.action?id=2849
. Accessed February 1, 2017.
2. National AIDS committeeTHAILAND AIDS RESPONSE PROGRESS REPORT 2015. 2015. Available at: http://www.unaids.org/sites/default/files/country/documents/THA_narrative_report_2015.pdf
. Accessed July 8, 2016.
3. Lumbiganon P, Kariminia A, Aurpibul L, et al. Survival of HIV-infected children: a cohort study from the Asia-Pacific region. J Acquir Immune Defic Syndr. 2011;56:365–371.
4. 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. 2010;53:86–94.
5. Hansudewechakul R, Sirisanthana V, Kurniati N, et al. Antiretroviral therapy outcomes of HIV-infected children in the TREAT Asia pediatric HIV observational database. J Acquir Immune Defic Syndr. 2010;55:503–509.
6. Phongsamart W, Hansudewechakul R, Bunupuradah T, et al. Long-term outcomes of HIV-infected children in Thailand: the Thailand pediatric HIV observational database. Int J Infect Dis. 2014;22:19–24.
7. Puthanakit T, Kerr S, Ananworanich J, et al. Pattern and predictors of immunologic recovery in human immunodeficiency virus-infected children receiving non-nucleoside reverse transcriptase inhibitor-based highly active antiretroviral therapy. Pediatr Infect Dis J. 2009;28:488–492.
8. Bunupuradah T, Sricharoenchai S, Hansudewechakul R, et al. Risk of first-line antiretroviral therapy failure in HIV-infected Thai children and adolescents. Pediatr Infect Dis J. 2015;34:e58–62.
9. 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. 2011;56:270–278.
10. Desmonde S, Eboua FT, Malateste K, et al. Determinants of durability of first-line antiretroviral therapy regimen and time from first-line failure to second-line antiretroviral therapy initiation. AIDS. 2015;29:1527–1536.
11. Costenaro P, Penazzato M, Lundin R, et al. Predictors of treatment failure in HIV-positive children receiving combination antiretroviral therapy: cohort data from Mozambique and Uganda. J Pediatr Infect Dis Soc. 2015;4:39–48.
12. Puthanakit TTA, Ananworanich J, Wongsawat J, et al. Thai national guidelines for the use of antiretroviral therapy in pediatric HIV infection in 2010. Asian Biomed. 2010;4:505–513.
13. WHO. WHO definitions of clinical, immunological and virological failure for the decision to switch ART regimens. 2010. Available at: http://www.who.int/hiv/pub/guidelines/arv2013/art/WHO_CG_table_7.15.pdf
. Accessed July 8, 2016.
14. Fine JP, Gray RJ. A proportional hazards model for the sub distribution of a competing risk. J Am Stat Assoc. 1999;94:496–509.
15. Yin DE, Warshaw MG, Miller WC, et al. Using CD4 percentage and age to optimize pediatric antiretroviral therapy initiation. Pediatrics. 2014;134:e1104–e1116.
16. Dollfus C, Le Chenadec J, Faye A, et al. Long-term outcomes in adolescents perinatally infected with HIV-1 and followed up since birth in the French perinatal cohort (EPF/ANRS CO10). Clin Infect Dis. 2010;51:214–224.
17. Bacha T, Tilahun B, Worku A. Predictors of treatment failure and time to detection and switching in HIV-infected Ethiopian children receiving first line anti-retroviral therapy. BMC Infect Dis. 2012;12:197.
18. Yirdaw KD, Hattingh S. Prevalence and predictors of immunological failure among HIV patients on HAART in southern Ethiopia. PLoS One. 2015;10:e0125826.
19. Childs T, Shingadia D, Goodall R, et al. Outcomes after viral load rebound on first-line antiretroviral treatment in children with HIV in the UK and Ireland:an observational cohort study. Lancet HIV. 2015;2:e151–158.