Access to third-line antiretroviral therapy (ART) for HIV-infected individuals in low- and middle-income countries (LMICs) is limited, making prevention of second-line ART failure an urgent global HIV priority.1 HIV-infected children and adolescents on ART have suboptimal virologic outcomes compared with adults, which may be due to both adherence challenges and the flux in pharmacokinetics unique to these age groups.2 For children and adolescents on second-line ART, antiretroviral (ARV) drug levels—integrating both adherence and pharmacokinetics—could help explain virologic outcomes and guide subsequent treatment decisions. Earlier detection of those at risk for treatment failure through quantitative measurement of ARV concentrations could be important for targeting adherence interventions to at-risk individuals to prolong the duration of second-line ART, especially among pediatric populations in LMIC settings where alternative regimens and resistance testing may not readily be available.
Hair concentrations of ARVs are a noninvasive measure that reflect cumulative systemic drug exposure and predict viral suppression in adults better than self-reported adherence,3,4 single plasma ARV levels,5 and host or viral characteristics, such as immune status and pretreatment viral loads.4,6,7 Hair sampling may be a particularly desirable approach for monitoring adherence/exposure in pediatric populations, because hair collection is painless and avoids phlebotomy.8 We previously demonstrated an association of ARV hair concentrations with virologic responses among HIV-infected children on second-line ART in Southeast Asia.5 The previous analysis was cross sectional and did not allow for the predictive utility of hair concentrations on viremia to be examined. In this longitudinal study, we determine for the first time whether ARV hair concentrations predict viremia in a cohort of HIV-infected Asian children and adolescents on lopinavir (LPV)-based second-line ART.
Study Design and Population
We analyzed data from the Prospective Monitoring of Second-line Antiretroviral Therapy Failure and Resistance in Children (TASER-P) study, a longitudinal cohort study that monitored second-line ART failure and resistance from 2011 to 2014 in HIV-infected children from Vietnam, Thailand, and Indonesia. Eligibility and recruitment at the 8 sites of TASER-P have been previously described.9 Briefly, children were enrolled if they were HIV infected, younger than 18 years, experienced first-line treatment failure, and had already switched or were currently switching to second-line ART [defined as an ARV class switch from a nonnucleoside reverse transcriptase inhibitor to a protease inhibitor (PI)]. Children on nonstandard treatment regimens [ie, once-daily boosted LPV/ritonavir (LPV/r) or a boosted PI without any nucleoside transcriptase inhibitors] were ineligible. Clinical and laboratory assessments for treatment response and safety were performed every 6 months. Self-reported adherence assessments (reported by parent/guardian when appropriate) were collected using the WHO 30-day visual analog scale.10 Small hair samples for LPV concentrations and plasma for HIV-1 RNA levels were collected every 6 months by study personnel.
Hair Collection and Analyses of ARV Hair Concentrations
Small samples of hair were collected according to previously described methods.11 Approximately, 20 strands of hair (1–3 mg) are required to measure the concentration of PIs.12 Hair samples were cut from underneath the top layer of hair from the occiput,7 placed in aluminum foil, stored at room temperature, and shipped to the University of California, San Francisco (UCSF) for analysis.
The Hair Analytical Laboratory (HAL) at UCSF has developed, validated, and published the methods for analyzing LPV in hair using liquid chromatography-tandem mass spectrometry (LC-MS/MS).13 Briefly, the LPV in each finely cut hair specimen is extracted with methanol (MeOH): trifluoroacetic acid (vol:vol) overnight at 37°C. The resulting solution is evaporated, sodium phosphate buffer added, followed by extraction with methyl t-butyl ether:ethyl acetate (1:1) (vol:vol). The organic layer is isolated, evaporated, and reconstituted for LC-MS/MS analysis. The extracted samples are separated by reverse phase chromatography and detected by tandem mass spectrometry using electrospray positive ionization with multiple reaction monitoring mode. The assay range is 0.05–20.0 ng/mg for LPV and the UCSF HAL assay has been peer reviewed and approved by the Division of AIDS (DAIDS) Clinical Pharmacology and Quality Assurance (CPQA) program.14
Participants were included in the present analysis if they had 1 or more hair specimens collected at least 3 months after switching to second-line therapy and were on a LPV-based regimen (91% of participants were on LPV/r-based regimens in the TASER-P cohort).9 At any visit where the HIV-RNA level was >1000 copies/mL, standard genotyping of the viral reverse transcriptase and protease gene was performed. If ARV resistance to any class was detected at a visit, we censored the data at that point and only used previous observations from that person. Because virologic failure from viral resistance can be independent of adherence, we wanted to focus in this analysis on viremia likely to be related to adherence. In 2 cases, virologic failure due to resistance occurred at the participant's first visit in the hair substudy, so data from these 2 participants did not contribute to the time-to-event analyses.
Demographic and clinical characteristics were summarized with medians for continuous measures and proportions for categorical variables. We used separate time-to-first-event discrete-time Cox regression models to identify predictors of experiencing viremia defined as either an HIV-1 RNA level >400 copies/mL or an HIV-1 RNA level >1000 copies/mL. Characteristics assessed as potential predictors of viremia were those evaluated in the parent study9 and included age ≤11 years, sex, country, self-reported adherence over the last 30 days, LPV hair concentrations, CD4 percentage and HIV-1 RNA level on switching to second-line ART; hair levels and self-reported adherence were included in models as time-varying covariates. Statistical significance was based on a 2-sided P value <0.05. Analyses were performed using SAS version 9.4 (SAS Institute Inc., Cary, NC).
All participating study sites and coordinating centers obtained local institutional review board approvals for study participation. Informed consent was provided by primary caregivers; assent was provided by children older than 7 years who were aware of their own HIV status when required by the local review board.
Characteristics of Study Participants
Overall, 277 children were enrolled in the TASER-P study between February 2011 and December 20129; of whom, 244 met the inclusion criteria for the present analysis. At enrollment, 55% of participants were Vietnamese, 34% were Thai, and 11% were Indonesian (Table 1). There were slightly more boys (55%) than girls, and the median age was 10 years [interquartile range (IQR) 7–13]; 40% of participants were older than 11 years. Overall, 88% of children or guardians reported more than 95% adherence to the ART regimen over the past 30 days. At switch to second-line ART, the median CD4 count was 300 (IQR 146–547) cells/mm3 and the median HIV-RNA level was 5.0 (IQR 4.3–5.6) log10/mL.
Predictors of Time-to-First Viremia
Median time of study follow-up was 48 weeks and a median of 3 (range 1–5) hair samples were collected from each participant. Overall, 83 cases of viremia with HIV-1 RNA >400 copies/mL and 56 episodes with HIV-1 RNA >1000 copies/mL occurred during the follow-up period without accompanying ARV resistance.
In unadjusted models, higher hair concentrations of LPV were strongly associated with a decreased likelihood of experiencing viremia with HIV-1 RNA >400 copies/mL [Table 2, odds ratio (OR) = 0.56 per doubling in hair concentration, 95% confidence interval (CI): 0.47 to 0.67, P < 0.001]; the association was similar with the outcome of HIV-1 RNA >1000 copies/mL (OR = 0.54, 95% CI: 0.45 to 0.65, P < 0.001). Self-reported adherence ≥95% of the time was also associated with a decreased likelihood of having viremia in unadjusted analyses (OR = 0.24 ≥95% vs <95% adherence 95% CI: 0.13 to 0.44, P < 0.001 for HIV-1 RNA >400 copies/mL; OR = 0.26, 95% CI: 0.13 to 0.51, P < 0.001 for HIV-1 RNA >1000 copies/mL). Compared with participants with CD4 percentage >15% at switch to second-line ART, those with CD4 percentage ≤15% had a 2-fold higher odds of having viremia of HIV-1 RNA >1000 copies/mL (OR = 1.97, 95% CI: 0.99 to 3.90, P = 0.053).
After adjustment for age, sex, country, self-reported adherence, CD4 percentage, and HIV-RNA copies/mL, LPV hair concentrations remained the strongest independent predictor of viremia, both for HIV-1 RNA >400 copies/mL [Table 2, adjusted odds ratio = 0.41 per doubling in hair concentration, 95% CI: 0.29 to 0.58, P < 0.001] and HIV-1 RNA >1000 copies/mL (adjusted odds ratio = 0.54, 95% CI: 0.45 to 0.65, P < 0.001). Self-reported adherence was no longer associated with viremia in the multivariate models (Table 2).
Our study is the first to prospectively examine the association of LPV hair concentrations with viremia in a cohort of HIV-infected children and adolescents on second-line ART. We show that LPV hair concentrations are the strongest predictor of viremia among children and adolescents on second-line LPV-based ART, even when controlling for sex, age, self-reported adherence, and other characteristics. The association between hair levels and virologic outcomes was very similar for viremia defined as either HIV-1 RNA >400 copies/mL or HIV-1 RNA >1000 copies/mL. Our results add to the previously limited data on using noninvasive objective measures of ART adherence and drug exposure in pediatric populations. These findings—and a recently-published study showing a similar finding for atazanavir hair levels among adolescents on atazanavir-based second-line ART15—suggest that hair concentrations of ARVs can inform clinical decision making around the management of viremia among HIV-infected children and adolescents by identifying those at risk for treatment failure because of chronically low drug levels.
There are several constraints currently impeding the availability and use of third-line ART for pediatric populations in LMICs, including limited laboratory capacity to diagnose treatment failure, lagging development of ARV formulations tailored for children, limited access to genotypic resistance testing, and cost.1 Given these barriers to delivery of third-line ART, understanding who may be at risk for virologic failure among pediatric populations on second-line ART is critical. Our results suggest that hair concentrations could guide clinical decision making in at-risk children, and future studies of real-time point-of-care hair assays are needed.16 Low hair concentrations in the face of virologic failure could trigger adherence interventions to prevent resistance and prolong the duration of second-line ART. Adequate drug concentrations in the face of persistent virologic failure could inform cost-effective decision-making around when to perform genotypic resistance testing. In addition, identifying patients with low hair concentrations who indeed have adequate adherence could trigger more personalized and potentially more aggressive dosing strategies.
Our study has some limitations. Hair testing for ARV levels has not been implemented at a regulated laboratory within Asia to date, requiring shipment of hair samples to CPQA-validated laboratories outside the region. Our study used standardized and validated high-performance assays, but required the use of an expensive LC-tandem MS machine. Low-cost hair assays, currently in development, are needed before widespread applicability of this noninvasive approach to assessing ARV adherence and exposure in LMIC clinical settings.16 In addition, plasma ARV levels were not assessed as a part of the parent study and therefore, we were unable to compare the predictive ability of hair levels with plasma concentrations on viremia.
In conclusion, this is the first study to demonstrate a longitudinal and strong association of LPV hair concentrations with viremia among HIV-infected children and adolescents on second-line ART. Future studies that examine the clinical application and cost-effectiveness of adherence monitoring through ARV hair concentrations are needed, especially now that viral load testing in LMICs is being expanded, while genotypic resistance testing remains limited. Early detection of those at risk for treatment failure with noninvasive approaches such as hair sampling could be clinically important for pediatric populations in LMIC settings.
The authors would like to thank the participants and their families, as well as the TASER-P investigators and site staff for taking part in this study. The authors would also like to thank Dr. Leslie Z. Benet, General Director of the HAL.
TASER-Pediatrics Study Group
N. Kurniati and P. Wicaksana, Cipto Mangunkusumo General Hospital, Jakarta, Indonesia; V. Sirisanthana, L. Aurpibul, and T. Sudjaritruk, Chiang Mai University, Chiang Mai, Thailand; P. Kosalaraksa, P. Lumbiganon, and C. Sopharak, Khon Kaen University, Khon Kaen, Thailand; W. Prasitsuebsai, S. Kerr, T. Bunupuradah, S. Teeraananchai, T. Jupimai, J. Intasan, N. Thammajaruk, and C. Ruengpanyathip, The HIV Netherlands, Australia, Thailand Research Collaboration (HIV-NAT), Bangkok, Thailand; K. Chokephaibulkit, S. Sricharoenchai, W. Phongsamart, and N. Kongstan, Siriraj Hospital, Mahidol University; KH Truong and TPK Le, Children's Hospital 1, Ho Chi Minh City, Vietnam; VC Do, TM Ha, and VT An, Children's Hospital 2, Ho Chi Minh City, Vietnam; LV Nguyen, KDT Khu, and TTT Giang, National Hospital of Pediatrics, Hanoi, Vietnam; A.H. Sohn, N. Durier, T. Singtoroj, and J.L. Ross, TREAT Asia, amfAR—The Foundation for AIDS Research, Bangkok, Thailand.
1. World Health Organization. Access to Antiretroviral Drugs in Low- and Middle-income Countries: Technical Report July 2014. Geneva, Switzerland: World Health Organization Press; 2014.
2. Agwu AL, Fairlie L. Antiretroviral treatment, management challenges and outcomes in perinatally HIV-infected adolescents. J Int AIDS Soc. 2013;16:18579.
3. Gandhi M, Ameli N, Bacchetti P, et al. Protease inhibitor levels in hair strongly predict virologic response to treatment. AIDS. 2009;23:471–478.
4. Koss CA, Natureeba P, Mwesigwa J, et al. Hair concentrations
of antiretrovirals predict viral suppression in HIV-infected pregnant and breastfeeding Ugandan women. AIDS. 2015;29:825–830.
5. Prasitsuebsai W, Kerr SJ, Truong KH, et al. Using lopinavir
concentrations in hair samples to assess treatment outcomes on second-line regimens among Asian children. AIDS Res Hum Retroviruses. 2015;31:1009–1014.
6. van Zyl GU, van Mens TE, McIlleron H, et al. Low lopinavir
plasma or hair concentrations
explain second-line protease inhibitor failures in a resource-limited setting. J Acquir Immune Defic Syndr. 2011;56:333–339.
7. Hickey MD, Salmen CR, Tessler RA, et al. Antiretroviral concentrations in small hair samples as a feasible marker of adherence
in rural Kenya. J Acquir Immune Defic Syndr. 2014;66:311–315.
8. ter Heine R, Beijnen JH, Huitema AD. Bioanalytical issues in patient-friendly sampling methods for therapeutic drug monitoring: focus on antiretroviral drugs. Bioanalysis. 2009;1:1329–1338.
9. Prasitsuebsai W, Teeraananchai S, Singtoroj T, et al. Treatment outcomes and resistance patterns of children and adolescents on second-line antiretroviral therapy
. J Acquir Immune Defic Syndr. 2016;72:380–386.
10. Giordano TP, Guzman D, Clark R, et al. Measuring adherence
to antiretroviral therapy
in a diverse population using a visual analogue scale. HIV Clin Trials. 2004;5:74–79.
11. Gandhi M, Ameli N, Bacchetti P, et al. Atazanavir concentration in hair is the strongest predictor of outcomes on antiretroviral therapy
. Clin Infect Dis. 2011;52:1267–1275.
12. Kosuge K, Uematsu T, Araki SI, et al. Comparative dispositions of ofloxacin in human head, axillary, and pubic hairs. Antimicrob Agents Chemother. 1998;42:1298–1302.
13. Huang Y, Gandhi M, Greenblatt RM, et al. Sensitive analysis of anti-HIV drugs, efavirenz, lopinavir
and ritonavir, in human hair by liquid chromatography coupled with tandem mass spectrometry. Rapid Commun Mass Spectrom. 2008;22:3401–3409.
14. DiFrancesco R, Rosenkranz SL, Taylor CR, et al. Clinical pharmacology quality assurance program: models for longitudinal analysis of antiretroviral proficiency testing for international laboratories. Ther Drug Monit. 2013;35:631–642.
15. Chawana TD, Gandhi M, Nathoo K, et al. Defining a cut-off for atazanavir in hair samples associated with virological failure among adolescents failing second-line antiretroviral treatment. J Acquir Immune Defic Syndr. 2017;75:55–59.
16. Gandhi M, Yang Q, Bacchetti P, et al. Short communication: a low-cost method for analyzing nevirapine levels in hair as a marker of adherence
in resource-limited settings. AIDS Res Hum Retroviruses. 2014;30:25–28.