Population PK of Isoniazid in South African Adults.
Paolo Denti(1), Roxana Rustomjee(2), Thuli Mthiyane(2), Philip Onyebujoh(3), Peter Smith(1), HelenMcIlleron(1)
(1) Division of Clinical Pharmacology, Department of Medicine, University of Cape Town, South Africa, (2) TB Research Unit: Clinical and Biomedical, South African Medical Research Council, Durban, South Africa, (3) Special Programme for Research and Training in Tropical Diseases, World Health Organization, Geneva, Switzerland
Objectives: Isoniazid, together with rifampicin, forms the backbone of 1st line antitubercular treatment. The rate of isoniazid metabolism is highly dependent on genetic polymorphisms of N-acetyltranferase 2 [1]. The objective of this study was to develop a population PK model of isoniazid.
Methods: A cohort of 62 South African HIV infected subjects with pulmonary TB was treated with a fixed dose combination of rifampicin, isoniazid, pyrazinamide, and ethambutol, according to the WHO weight-based dosing recommendations [2]. Intensive PK sampling was performed on 4 occasions, on the 1st, 8th, 15th and 29th day of antituberculosis treatment. A population PK model was built in NONMEM VII. Several absorption and disposition models were tested, and variability between subjects (BSV) and occasions (BOV) in the PK parameters was quantified. Samples below the limit of quantification (LOQ) were handled with the M6 method from [3], and a dedicated additive error structure was used. Since no genotyping or metabolite information was available, a mixture model was used to explore the presence of subpopulations with respect to acetylator phenotype.
Results: Isoniazid PK was described by a two compartment disposition model with absorption through a series of transit compartments [4] and first-order elimination. A large variability was observed in absorption and bioavailability and mainly at BOV, rather than BSV level. The clearance, instead, was found to be strongly subject-specific and did not vary much between occasions. The large BSV in CL (initially more than 50%) was explained with the introduction of a mixture model, which estimated a CL of 26 L/h for about 45% of subjects, and a value more than doubled for the rest. After the implementation of this mixture, BSV in CL dropped to 20%. The subpopulation with fast clearance was also found to have lower bioavailability (about 70%).
Conclusions: Although no direct acetylator status information was available for the subjects in this study, and thus no verification was possible, the improvement in fit obtained with the introduction of the mixture modeling indicated the presence of at least 2 subpopulations, likely corresponding to slow and fast acetylators. In fast metabolizers, a reduction in bioavailability was also detected, possibly due to a larger extent of first-pass metabolism.
References:
[1] D.P. Parkin, S. Vandenplas, F.J. Botha, M.L. Vandenplas, H.I. Seifart, P.D. van Helden, B.J. van der Walt, P.R. Donald, and P.P. van Jaarsveld, “Trimodality of isoniazid elimination: phenotype and genotype in patients with tuberculosis.,” American journal of respiratory and critical care medicine, vol. 155, May. 1997, pp. 1717-22.
[2] World Health Organization, Treatment of tuberculosis: guidelines for national programmes, 3rd edition. WHO/CDS/TB/2003.313, 2004.
[3] S.L. Beal, “Ways to fit a PK model with some data below the quantification limit.,” Journal of pharmacokinetics and pharmacodynamics, vol. 28, Oct. 2001, pp. 481-504.
[4] R.M. Savic, D.M. Jonker, T. Kerbusch, and M.O. Karlsson, “Implementation of a transit compartment model for describing drug absorption in pharmacokinetic studies.,” Journal of pharmacokinetics and pharmacodynamics, vol. 34, Oct. 2007, pp. 711-26.
