B Ngara (1, 4), S P Zvada (2), T D Chawana (1), B Stray-Pedersen (3, 4), C Nhachi (1), S Rusakaniko (1)
(1) University of Zimbabwe, Zimbabwe, (2) Stellenbosch University, South Africa, (3) Letten Foundation Research Centre, Zimbabwe, (4) University of Oslo, Norway
Introduction: An estimated 36.9 million people were living with Human Immunodeficiency Virus (HIV) worldwide in 2017. Of these approximately 3.0 million were children and adolescents under 20 years of age [1]. Zimbabwe has a prevalence of 13.3%, with 1.3 million people living with HIV including, 77,000 children and adolescents [2]. Adolescence experience higher levels of non-adherence to treatment of HIV [3]. Measuring drug concentration in hair promises to be reliable method for assessing exposure to antiretroviral drugs due to accumulation from plasma [4]. Modelling and simulation approach are necessary to explore the usefulness of quantifying drug concentrations in hair for the benefit of measuring long term adherence. Drug plasma measurements cannot reliably be used for adherence monitoring especially in settings where patients took the drugs only towards clinic visits.
Objectives:
- To develop a pharmacokinetic model based on drug concentrations quantified in the hair
- To identify population characteristics associated with variability in ritonavir-boosted concentrations in hair
Materials and methods: Data used in model development and validation was obtained from a study conducted in Zimbabwean adolescents on HIV treatment for at least one month[3]. Participants were randomised to the intervention or control study arms. Hair samples and other data variables were collected at enrolment and three months follow-up. The structural model was characterised using a two-compartmental model structure, which included an output compartment to predict measurements observed from the hair compartment. A generalised nonlinear model was fit using ADVAN13 in NONMEM 7.3 [5]. Previously published models describing population pharmacokinetics of as atazanavir or ritonavir in plasma were utilised, and parameter estimates were fixed to literature values [4], [5]. Then the fraction of the drug that accumulated in hair was estimated while the hair volume of distribution was fixed to unit for both drugs. Stepwise covariate modelling strategy was used for covariate selection in PsN [8]. Model assessment was done using goodness of fit plots in Xpose4 [8].
Results: Our findings showed that there is 16% and 18% of the respective atazanavir and ritonavir concentrations in hair relative to steady-state trough plasma concentrations. At follow-up event, we estimated an increase of 30% and 42% in concentrations of the respective atazanavir and ritonavir concentrations that accumulated in hair compared with accumulation at enrolment. A unit increase in self-reported adherence measured was associated with a 2% increase for both atazanavir and ritonavir concentrations in hair. Thinner participants had 54% higher hair concentrations while overweight had 21% lower compared to normal weight participants. Adolescents receiving care from fellow siblings had atazanavir concentrations of at least 54% less compared to receiving care from mature guardians. Participants in the control arm and those in earlier stages of disease progression had volume of distribution for atazanavir concentrations 53% higher and for ritonavir concentrations 37% less compared to their counterparts.
Conclusion: The work demonstrated methods for hair quantitative pharmacology, to compliment efforts working towards establishing point of care methods based on quantifying drug in hair. Hair collection is easy, analysis is cheap and samples can be transported without biohazardous precautions and cold chain. Drug concentrations in hair provide information on drug exposure and adherence, predicts virologic treatment outcome, and segmental analysis tells us drug exposure at different time points. Most important determinants of increased concentrations in hair were monitoring at follow up event, body weight and care. Measuring ART levels in hair promises to be more accurate and feasibly accomplished. It is crucial to perform follow-up work which involves establishing the relationship between hair pharmacokinetic parameters and a measure of treatment response such as viral loads. Additionally, comparing the predictive accuracy for exposure-response models when exposure is based on either plasma or hair drug concentrations is also to check.
References:
[1] “Global and Regional Trends,” UNICEF DATA. .
[2] “ZWE_2018_countryreport.pdf.” .
[3] T. D. Chawana*, D. Katzenstein, K. Nathoo, B. Ngara, and C. F. B. Nhachi, “Evaluating an enhanced adherence intervention among HIV positive adolescents failing atazanavir/ritonavir-based second line antiretroviral treatment at a public health clinic,” J. AIDS HIV Res., vol. 9, no. 1, pp. 17–30, Jan. 2017.
[4] T. D. Chawana 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. 1999, May 2017.
[5] Alison J. Boeckmann, Lewis B. Sheiner, and Stuart L. Beal, “NONMEM Users Guide – Part V.” .
[6] F. Foissac et al., “Population pharmacokinetics of atazanavir/ritonavir in HIV-1-infected children and adolescents,” Br. J. Clin. Pharmacol., vol. 72, no. 6, pp. 940–947, Dec. 2011.
[7] C. Zhang, P. Denti, E. H. Decloedt, Y. Ren, M. O. Karlsson, and H. McIlleron, “Model-based evaluation of the pharmacokinetic differences between adults and children for lopinavir and ritonavir in combination with rifampicin,” Br. J. Clin. Pharmacol., vol. 76, no. 5, pp. 741–751, Nov. 2013.
[8] R. J. Keizer, M. O. Karlsson, and A. Hooker, “Modeling and Simulation Workbench for NONMEM: Tutorial on Pirana, PsN, and Xpose,” CPT Pharmacomet. Syst. Pharmacol., vol. 2, no. 6, p. e50, Jun. 2013.
Reference: PAGE 28 (2019) Abstr 9048 [www.page-meeting.org/?abstract=9048]
Poster: Methodology - New Modelling Approaches