A Semi-Mechanistic pharmacokinetic enzyme model for the characterisation of rifampicin pharmacokinetics in South African pulmonary tuberculosis infected adults
W Smythe(1), H McIlleron(1), C Merle (2), J Horton(3), P Smith(1), and USH Simonsson(4)
(1) Division of Clinical Pharmacology, Department of Medicine, University of Cape Town, South Africa; (2) Department of Epidemiology and Population Health, Tropical Epidemiological Group, London School of Hygiene and Tropical Medicine, United Kingdom; (3) Tropical Projects, Hitchin, United Kingdom; (4) Department of Pharmaceutical Biosciences, Uppsala University, Sweden.
Objectives: Rifampicin, together with other first line regimen drugs, is used to treat drug sensitive Mycobacterium Tuberculosis. Rifampicin is known to have highly variable absorption (1, 2) and to induce its own metabolism (3). These characteristics result in low rifampicin concentrations in many patients, and may increase the likelihood of treatment failure and emergent drug resistance. The primary objective of this pharmacokinetic analysis was to determine the population pharmacokinetics of rifampicin at pre-induced and fully auto-induced states amongst African patients with pulmonary tuberculosis using mixed-effects modelling.
Methods: Adults (n=173) with pulmonary tuberculosis received once daily doses of either 450 mg (below 50 kg) or 600 mg (above 50 kg) of rifampicin together with isoniazid, pyrazinamide and ethambutol for 6 days of the week. Three blood samples per patient were taken after the first dose (pre-induction) and repeated after approximately 28 days (steady state) yielding a total of 998 samples for analysis of rifampicin concentrations in plasma. A semi-mechanistic pharmacokinetic model incorporating an enzyme turn over model to address rifampicin's auto-inductive properties, together with a multiple dosing transit absorption compartment model to describe the drug's highly variable absorption was developed using the first order conditional method in NONMEM.
Results: Rifampicin displayed potent auto-induction with an estimated EC50 of 0.133 mg/L which is less than most plasma concentrations following a standard rifampicin dose. The model estimated un-induced oral clearance at 5.97 L.h-1. The enzyme turn-over half-life was fixed to approximately 24 hours (kENZ fixed to 0.029 h-1) reaching steady state in approximately 1 week (5) since samples were collected only at pre- and post-induced occasions. Based on the VPC stratified by occasion, the model adequately predicted rifampicin pharmacokinetics both at the pre-induced and induced state.
Conclusions: The semi-mechanistic model describing the pharmacokinetics of rifampicin at pre-induced and induced states will be extended to investigate potential drug-drug interactions seen between RIF and the other drug components of the anti-tuberculosis regimens.
 Peloquin CA, Jaresko GS, Yong CL, et al. Population Pharmacokinetic Modeling of Isoniazid, Rifampin, and Pyrazinamide. Antimicrobial Agents Chemother. 1997; 41: 2670-9.
 Wilkins JJ, Savic RM, Karlsson MO, Langdon G, McIlleron H, Pillai G,. Smith PJ, and USH Simonsson. Antimicrobial Agents Chemother. 2008; 52: 2138-48.
 Loos U, Musch E, Jensen JC, Schwabe H K, and M Eichelbaum. Influence of the enzyme
induction by rifampicin on its presystemic metabolism. Pharmacol. Ther. 1987; 33: 201-4.
 Mitchison DA. Role of individual drugs in the chemotherapy of tuberculosis. Int J Tuberc Lung Dis. 2000; 4: 796-806.
 Fromm MF, Busse D, Kroemer HK, and M Eichelbaum. Hepatology. 1996; 24: 796-801.