Maxwell T. Chirehwa (1), Court Richard (1), De Kock Marianna (2), Wiesner Lubbe (1), de Vries Nihal (3), Harding Joseph (4), Gumbo Tawanda (5), Denti Paolo (1), Warren Rob (2), Maartens Gary (1), McIlleron Helen (1)
(1) Division of Clinical Pharmacology, Department of Medicine, University of Cape Town, South Africa; (2)DST/NRF Centre of Excellence in Biomedical Tuberculosis Research, SAMRC Centre for Tuberculosis Research, Stellenbosch University,South Africa (3) Brooklyn Chest Hospital, Cape Town, South Africa; (4) DP Marais Hospital, Cape Town, South Africa; (5)Center for Infectious Diseases Research and Experimental Therapeutics, Baylor Research Institute, Baylor University Medical Center, Dallas, Texas
Objectives: Cycloserine or its structural analogue terizidone which consists of two molecules of cycloserine (1), is a key drug recommended by WHO in long treatment regimens for multidrug-resistant tuberculosis (MDR-TB) (2). Considering the neurotoxicity associated with the use of cycloserine, and the relative weakness of many of the second-line anti-TB drugs, data defining target cycloserine exposures is needed. We describe (a) the population pharmacokinetics of cycloserine dosed as terizidone and (b) the distribution of the percentage of time the concentration is above the MIC (%Time>MIC) within a 24-hour dosing interval.
Methods: We recruited adult patients on treatment for MDR-TB at two hospitals in South Africa. During the study period, the standard MDR-TB treatment regimen consisted of moxifloxacin, kanamycin, pyrazinamide, cycloserine dosed as terizidone (dosed by weight band), ethambutol, and ethionamide or high-dose isoniazid. Blood samples were collected pre-dose and at 2, 4, 6, 8, and 10 hours post-dose, and in 9 of 133 participants additional samples were collected at 12, 24, and 26 hours post-dose. We quantified plasma concentrations of cycloserine using a validated LC/MS/MS assay and interpreted using nonlinear mixed-effects modelling in NONMEM version 7.4.2. Assuming two molecules of cycloserine for each molecule of terizidone, the dose of terizidone was converted to an equivalent dose of cycloserine, adjusting for molecular weight. We evaluated one- and two-compartment disposition models with first-order absorption (with or without a delay) and elimination. We included allometric scaling in the base model, and considered the effect of other physiologically plausible covariates including creatinine clearance (CrCL, calculated using the Cockroft-Gault formula). We determined baseline sputum MICs using Sensititre MYCOTB MIC plates and calculated the proportion of patients with %Time>MIC of at least 30% (3), assuming a lung cavity-to-serum penetration ratio of 0.09 (4).
Results: 927 plasma samples were available from 166 pharmacokinetic profiles. The median weight and fat-free mass were 47 kg (range: 30 – 85) and 40.7 kg (24.1 – 58.9), respectively. A one-compartment disposition model with first-order absorption (after a delay described by a chain of transit compartments) and elimination best described the pharmacokinetics of cycloserine. The parameter estimates of the final model were: CL/F (0.832 L/h), V/F (23.3 L), Ka (0.836 h-1), MTT (0.565 h). The model could detect two clearance pathways, non-renal and renal, with the latter being modulated by CrCL and accounting for approximately 50% of elimination. The predicted CrCL was adjusted to a median weight of 47 kg and its effect was included in the model relative to the median CrCL of 100 mL/min (5). Allometric scaling was included on both pathways and on the volume of distribution using FFM (52 points drop in OFV). Between-subject variability was supported on CL/F, and between-occasion variability was included on Ka, MTT, and F. MIC values (median [range]: 16 [2, 32] mg/L) were available for 103 patients. While the median plasma trough concentration (18 mg/L, range: 3 – 41 mg/L) was predicted to be above the MIC of 82% of the patients, the %Time>MIC in the lungs was >30% in 2/103 patients (MIC was 2 and 4 mg/L, respectively) and zero in the rest of the cohort.
Conclusions: To our knowledge, this is the largest population pharmacokinetic study describing cycloserine dosed as terizidone. Cycloserine pharmacokinetics was best described by a one-compartment disposition model with first-order absorption and elimination. The estimated proportion of cycloserine eliminated via the renal pathway (50%) is lower than the range previously reported of 60-70% (6).
References:
[1] Farrington M, Sharma SK. Chemotherapy of bacterial infections. In: Bennett PN, Brown MJ, Sharma P, eds. Clinical Pharmacology. 11th ed. Toronto: Churchill Livingstone, 2012; 191–212.
[2] WHO. 2018. Rapid Communication: Key changes to treatment of multidrug- and rifampicin-resistant tuberculosis (MDR/RR-TB). WHO/CDS/TB/2018.18.
[3] Deshpande D, Alffenaar J-WC, Köser CU, et al. D-cycloserine pharmacokinetics/pharmacodynamics, susceptibility, and dosing implications in multidrug-resistant tuberculosis: a faustian deal. Clin Infect Dis 2018; 67(suppl_3): S308–16.
[4] Dheda K, Lenders L, Magombedze G, et al. Drug Penetration Gradients Associated with Acquired Drug Resistance in Tuberculosis Patients. Am J Respir Crit Care Med 2018: 198(9): 1208-19.
[5] Holford N, Heo Y-A, Anderson B. A Pharmacokinetic Standard for Babies and Adults. J. Pharm. Sci. 2013;102(9):2941–2952.
[6] Macleods. Cycloserine [package insert]. Mumbai; 2007.
Reference: PAGE 28 (2019) Abstr 9034 [www.page-meeting.org/?abstract=9034]
Poster: Drug/Disease Modelling - Infection