Noha Abdelgawad (1), Jose Calderin Miranda (1), Sean Wasserman (2, 3), Faye Lanni (4), Rosleine Antilus-Sainte (4), Firat Kaya (4), Matthew Zimmerman (4), Martin Gengenbacher (4), Veronique Dartois (4), Paolo Denti (1)
(1) Division of Clinical Pharmacology, Department of Medicine, University of Cape Town, South Africa, (2) Centre for Infectious Diseases Research in Africa, Institute of Infectious Disease and Molecular Medicine, University of Cape Town, Observatory 7925, South Africa (3) Institute for Infection and Immunity, St George’s University of London, United Kingdom (4) Center for Discovery and Innovation, Hackensack Meridian Health, Nutley, New Jersey, USA
Objectives:
Tuberculosis meningitis (TBM) is the most devastating manifestation of M. tuberculosis infection causing high rates of mortality and disability [1]. Recommended TBM treatment is based on that of pulmonary TB, without considerations for protective barriers that limit drug entry into the CNS. Drug levels are also affected by disease-related physiological changes in the barriers. To design better regimens, it is essential to improve understanding of antitubercular pharmacokinetics (PK) at the site of disease. Drug concentrations are difficult to monitor in the CNS in patients, requiring preclinical tools, such as rabbit infection models, to study TBM. This study aims to investigate the PK of rifampicin (RIF), isoniazid (INH), pyrazinamide (PZA), and linezolid (LZD) in CNS tissues using a preclinical model of TBM in rabbits [2].
Methods:
A rabbit model of TBM that recapitulates neurological and immunopathological features of human disease was used for this study [2]. Three separate experiments were carried out: RIF only, INH and PZA co-administered, and LZD only. All drugs were orally administered daily for 3 days at 90 mg/kg or 120 mg/kg for RIF, 30 mg/kg for INH, 175 mg/kg for PZA, and 90 mg/kg for LZD. Doses were estimated to achieve human-equivalent exposures based on PK studies in uninfected rabbits. In all experiments, blood samples were collected pre-dose and at 0.5, 1, 2, 3, 5, 7, and 24 h post-drug on day 1 and again on day 3 until necropsy, which was scheduled at 2, 3, 6, 10, or 24 h post-dose. Terminal samples were collected for plasma, CSF, meninges, lumbar and cervical spinal cord, brain, and lung tissues. Drug concentrations were quantified in all samples using liquid chromatography coupled with mass spectrometry assay. The lower limits of quantification in plasma were 0.001 mg/L for RIF and LZD, 0.002 mg/L for INH, and 0.01 mg/L for PZA, while in tissues they were 0.01 mg/L for RIF, INH, and LZD and 0.05 mg/L for PZA.
A plasma model was developed separately for each drug, with CL and V allometrically scaled by weight. Then, individual plasma pharmacokinetic parameters were fixed, and the CSF and tissue concentrations were modelled using an “effect” compartment approach with each tissue represented as a compartment connected to the central plasma compartment. For each “effect” compartment we estimated a pseudo-partition coefficient (PPC) i.e., CSF or tissue to plasma drug ratio, and an equilibration half-live (T1/2eq).
Results:
For RIF, 16 TBM-infected rabbits were available, providing 230 plasma, 155 CSF and tissue observations; for INH and PZA, 4 rabbits were included, providing 50 plasma, and 40 CSF and tissue observations; for LZD, 13 rabbits, 117 plasma and 54 CSF and tissue concentrations. No observations were below the limit of quantification (BLQ) for any of the drugs, except 3 PZA plasma observations. A one-compartment disposition model with transit absorption and first-order elimination was the best fit for the plasma pharmacokinetics of all drugs, except for INH, which displayed two-compartment disposition. All drugs had good penetration in lung tissues, with PPCs >0.65 and T1/2eq of <1 h. Conversely, drug penetration into the CNS compartments ranged more widely, both between drugs and between compartments. The values of PPC and T1/2eq were most favorable (the highest ratio and fastest equilibrium) for INH and PZA (PPC range between tissues was 0.8–1 and 0.5–1, respectively and T1/2eq range was 2–4 min and 4–12 min, respectively) and were less favorable for LZD (PPC of 0.1–0.4 and T1/2eq of 0.5–1 h) and RIF (PPC of 0.07–0.3 and T1/2eq of 0.6–3 h).
Conclusion:
A rabbit model for TBM infection [2] was successfully used to develop PK models of antitubercular drug penetration into CSF, various CNS tissues, and lungs. The estimated PPC and T1/2eq for CSF align with previously reported values in humans [3–5] which provides reassurance that these preclinical tissue pharmacokinetic models could be used to predict drug levels in CNS tissues in humans. While the small numbers of rabbits/observations for some of the drugs is a limiting factor, this study provides invaluable observations from tissues that cannot be obtained in humans which are critical to investigating and optimizing TBM treatment regimens. Further research is needed to establish exposure targets that ensure optimal drug levels are attained in the CNS.
References:
- Davis AG, Rohlwink UK, Proust A, Figaji AA, Wilkinson RJ. The pathogenesis of tuberculous meningitis. J Leukoc Biol. 2019;105:267–80.
- Lanni F, Antilus Sainte R, Hansen, M, Parigi P, Kaya F, LoMauro K, et al. A preclinical model of TB meningitis to determine drug penetration and activity at the sites of disease. Silverman JA, editor. Antimicrob Agents Chemother [Internet]. 2023;67. Available from: https://journals.asm.org/doi/10.1128/aac.00671-23
- Abdelgawad N, Tshavhungwe M, Rohlwink U, McIlleron H, Abdelwahab MT, Wiesner L, et al. Population Pharmacokinetic Analysis of Rifampicin in Plasma, Cerebrospinal Fluid, and Brain Extracellular Fluid in South African Children with Tuberculous Meningitis. Antimicrob Agents Chemother [Internet]. 2023 [cited 2024 Mar 6];67. Available from: https://journals.asm.org/journal/aac
- Ding J, Thuy Thuong Thuong N, Pham T Van, Heemskerk D, Pouplin T, Tran CTH, et al. Pharmacokinetics and Pharmacodynamics of Intensive Antituberculosis Treatment of Tuberculous Meningitis. Clin Pharmacol Ther [Internet]. 2020 [cited 2022 Jul 26];107:1023–33. Available from: https://pubmed.ncbi.nlm.nih.gov/31956998/
- Abdelgawad N, Wasserman S, Abdelwahab MT, Davis A, Stek C, Wiesner L, et al. Linezolid Population Pharmacokinetic Model in Plasma and Cerebrospinal Fluid Among Patients With Tuberculosis Meningitis. J Infect Dis [Internet]. 2023 [cited 2024 Mar 6]; Available from: https://dx.doi.org/10.1093/infdis/jiad413
Reference: PAGE 32 (2024) Abstr 10886 [www.page-meeting.org/?abstract=10886]
Poster: Drug/Disease Modelling - Infection