Philippine Eloy(1), Céline Verstuyft(2,3), Nilesh Bhatt(4), Maryline Bonnet(5), Anne-Marie Taburet(3,6), Julie Bertrand(7) and the ANRS12214 study team
(1) Department of Epidemiology, Biostatistic and Clinical Research, Bichat Hospital, AP-HP, F-75018 Paris, France; (2) Service de génétique moléculaire et pharmacogénétique, hôpital Bicêtre, AP-HP, France; (3) INSERM UMR 1184, Center for Immunology of Viral Infections and Autoimmune Diseases; (4) INS, Ministry of Health, Av. Eduardo Mondlane, number 1008, Maputo, Mozambique; (5) Epicentre, Paris, France; (6) Pharmacie Clinique, Hôpital Bicêtre, , AP-HP, France ; (7) IAME, UMR 1137, INSERM, Université Paris Diderot, Sorbonne Paris Cité Paris, France
Objectives: Nevirapine (NVP) is a non-nucleoside reverse-transcriptase inhibitor of human immunodeficiency virus (HIV) type 1. NVP can be coadministered safely with standard anti-tuberculosis (anti-TB) treatment in coinfected patients [1]. Standard anti-TB treatment includes rifampicin (RMP), a potent cytochrome P450 (CYP) inducer, and isoniazid for 6 months with ethambutol plus pyrazinamide for the first 2 months.
NVP is metabolized mainly through CYP3A and CYP2B6 both demonstrated to be inducible and encoded in highly polymorphic genes [2]. Indeed, the CYP2B6 G516T polymorphism has been associated with increased NVP exposure in a Cambodian population [3]. Here, we use a modelling approach to characterize the pharmacokinetics (PK) of NVP when co-administered with anti-TB drugs in TB-HIV coinfected patients in Mozambique and explore transport and metabolism genes SNP in this population.
Methods: HIV-TB coinfected patients were recruited from the CARINEMO-ANRS12146 study [4] who had consented for genetic testing to be enrolled in the CARINEMO-ANRS12214 sub-study. Anti-TB treatment (RMP, isoniazid, pyrazinamide, ethambutol) was initiated and patients were randomized 4 to 6 weeks later to receive ART based on efavirenz or NVP. Here, we focus on participants who received NVP plus two nucleoside reverse-transcriptase inhibitors.
NVP pre-dose concentration was assayed at months 1, 3, 6, (on anti-TB treatment) and 9 and 12 (off anti-TB treatment) by HPLC with a limit of quantification of 25 ng/mL [5]. Extensive PK sampling was performed in 20 participants at months 1 and 7 (sampling times: pre-dose, H0.5, H1, H1.5, H2, H4, H6, H8, H10, H12).
Plasma concentrations of NVP were analyzed using nonlinear mixed-effects software Monolix v2016R1. The structural, between-subject variability (BSV) and residual variability models were developed using the data from the month 7 extensive PK sampling, while patients were off anti-TB treatment. The between occasion variability (BOV) model was developed on months 1 and 7 extensive PK data and effect of anti-TB co-administration was investigated on these data. When including all patients at all occasions we added a second residual error variability model, specific to pre-dose concentrations, allowing more variability as dose intake was not supervised.
Genotyping for CYP2B6, NAT2, CYP2A6, CYP3A4, CYP3A5 and ABCB1 was performed.
Results: Among 252 patients who enrolled in NVP arm, 251 had at least 1 NVP concentration available. The number of patients with NVP pre-dose concentration available was 103, 217, 237, 229, 226 at months 1, 3, 6, 9 and 12 respectively. 20 patients enrolled in the extensive pharmacokinetic sub-study: all patients had extensive sampling while on anti-TB and among them, 16 after discontinuation of anti-TB. A 1-compatment (V/F= 69.7 L) model with one-order absorption (ka= 1.32 /h) and linear elimination (CL/F= 1.92 L/h, BSV=24%, BOV=28%) best fit the data. We used a fraction of dose absorbed parameter p fixed to 1 which BSV= 38% captured the correlation between V and CL. Two proportional error models (sPK= 2% and sres= 9%) were used for concentrations from the extensive PK sub-study and the pre-dose samples respectively. Model selection was performed based on Bayesian information criterion and goodness-of-fit plots. Anti-TB treatment was shown to increase NVP apparent clearance in patients taking both NVP and anti-TB treatment by 46% (p<10-10).
Among the 251 patients with at least 1 NVP concentration available, 146 also had at least 1 genotype available. Loss-of-function allele frequencies in our population were: ABCB1 rs1045642 : 11% ; CYP2B6 rs3745274 (G516T) : 43% ; CYP2B6 rs7251950 : 7% ; CYP2A6 rs8192726 : 94% ; CYP3A4 rs35599367 : none ; CYP3A5 rs776746 : 18% and 10% of participants were considered as slow NAT2 acetylators.
Conclusions: To our knowledge, this is the first study modelling NVP PK on and off anti-TB treatment and showing a significant 46% increase in NVP apparent clearance when co-administered with anti-TB drugs including RMP. Description of drug-metabolizing enzymes and transporters genes SNPs frequency in a Mozambique population is the first step before analyzing which of these genetic polymorphisms will be significant covariates of the model and how they will impact the NVP-TB drugs drug-drug interaction.
References:
[1] World Health Organization. Consolidated guidelines on the use of antiretroviral drugs for treating and preventing HIV infection: recommendations for a public health approach. [Internet]. 2016 [cited 2017 Jan 6]. Available from: http://www.ncbi.nlm.nih.gov/books/NBK374294/
[2] Erickson DA, Mather G, Trager WF, Levy RH, Keirns JJ. Characterization of the in vitro biotransformation of the HIV-1 reverse transcriptase inhibitor nevirapine by human hepatic cytochromes P-450. Drug Metab Dispos Biol Fate Chem. 1999;27:1488–95.
[3] Bertrand J, Chou M, Richardson DM, Verstuyft C, Leger PD, Mentré F, et al. Multiple genetic variants predict steady-state nevirapine clearance in HIV-infected Cambodians: Pharmacogenet Genomics. 2012;22:868–76.
[4] Bonnet M, Bhatt N, Baudin E, Silva C, Michon C, Taburet A-M, et al. Nevirapine versus efavirenz for patients co-infected with HIV and tuberculosis: a randomised non-inferiority trial. Lancet Infect Dis. 2013;13:303–12.
[5] Bhatt NB, Baudin E, Meggi B, da Silva C, Barrail-Tran A, Furlan V, et al. Nevirapine or efavirenz for tuberculosis and HIV coinfected patients: exposure and virological failure relationship. J Antimicrob Chemother. 2015;70:225–32.
Reference: PAGE 27 (2018) Abstr 8663 [www.page-meeting.org/?abstract=8663]
Poster: Drug/Disease Modelling - Infection