A semi-mechanistic model of the population pharmacokinetics and bactericidal activity of high-dose isoniazid against multi-drug-resistant tuberculosis
Kamunkhwala Gausi (1); Maxwell Chirehwa (1); Elisa Ignatius (2); Richard Court (1); Xin Sun (3); Laura Moran (4); Richard Hafner (5); Lubbe Wiesner (1); Susan L Rosenkranz (3); Tawanda Gumbo (6); Susan Swindells (7); Andreas Diacon (8); Helen McIlleron (1); Kelly Dooley (2); Paolo Denti (1)
1. Division of Clinical Pharmacology, Department of Medicine, University of Cape Town, Cape Town, South Africa; 2.Johns Hopkins University School of Medicine, Baltimore, MD, USA; 3. Harvard T.H. Chan School of Public Health, Boston, MA, USA; 4. Social & Scientific Systems, a DLH Company, Silver Spring, MD, USA; 5. Division of AIDS, National Institute of Allergy and Infectious Diseases, Bethesda, MD, USA; 6. Center for Infectious Diseases Research and Experimental Therapeutics, Baylor Research Institute, Baylor University Medical Center, Dallas, Texas, USA; 7. Department of Internal Medicine, University of Nebraska Medical Center, Omaha, NE, USA; 8. Stellenbosch University and Task Applied Science, Cape Town, South Africa.
The emerging threat of drug-resistant Mycobacterium tuberculosis (M.tb) strain has made it essential to optimise and customise old and new drugs for individual patients or subgroups.1,2 Isoniazid is a good candidate for dose optimisation and customisation since higher doses have been reported to be effective against tuberculosis with low to moderate levels of isoniazid resistance.3–5
The present work aimed to i) characterise the pharmacokinetics of standard (5mg/kg) to high dose isoniazid (10-5mg/kg), ii) analyse the drug-drug interaction between isoniazid and concomitant drugs within the multidrug-resistant TB regimen. iii) describe the association between isoniazid pharmacokinetics (at high or standard dose) and its early bactericidal activity against M.tb (drug-sensitive and inhA-mutated).
Methods: Isoniazid PK was evaluated using pooled data from two studies: INHindsight6 and PODRtb7, with participants on standard to high isoniazid dose (5-5mg/kg). INHindsight was a 7-day early bactericidal activity (EBA) study with isoniazid monotherapy, while PODRtb was an observational study in which isoniazid was coadministered as part of MDR-TB treatment regimen. The MDR-TB treatment included terizidone/cycloserine, pyrazinamide, moxifloxacin, ethionamide, ethambutol, kanamycin, ethionamide or/and isoniazid.
PK samples were captured on day 6 (INHindsight) or ≥2 weeks (PODRtb study) after treatment initiation. Blood samples were drawn at pre-dose, 0.5, 1, 2, 4, 6, 8, 12, 24 hours and at pre-dose, 2, 4, 6, 8, and 10 hours in the INHindsight and PODRtb study, respectively. N-acetyltransferase 2 (NAT2) genotype information was captured, categorising patients into rapid, intermediate or slow acetylators.8,9 PODRtb participant NAT2 information was not available; therefore, a mixture model was used to assign phenotype.10
Isoniazid pharmacodynamics (PD), i.e. its bactericidal effect, was evaluated using only the INHindsight study, which performed intensive PD sampling. Each participant’s sputum was collected overnight over 16 hours at pre-entry (Day 1), Day 0 (pre-treatment), and daily while on isoniazid monotherapy for 7 days. The samples were cultured on solid media to quantify colony-forming units (CFU) and liquid media to measure time-to-positivity (TTP). Isoniazid minimum inhibitory concentration (MIC) information was also measured at baseline. CFU and TTP were jointly modelled to quantify the bacterial load. CFU “directly” characterised the bacterial load at the time of sputum collection, while TTP was obtained by observing the M.tb growth in Mycobacterial Growth Indicator Tubes (MGIT) using bacterial load (CFU) as the initial number of bacteria transferred into the tube, which then grows and is detected after crossing a threshold. To link the PK and PD, the effect of isoniazid exposure was tested on the bacterial kill rate using overall AUC (constant kill), or instantaneous isoniazid concentration, either in plasma or delayed using an effect compartment.11
Isoniazid pharmacokinetics/pharmacodynamics was modelled using NONMEM version 188.8.131.52 and ancillary software.13
Results: PK samples were available for 161 participants (58 from INHindsight and 103 from PODRtb study). Isoniazid PK was well described using a two-compartment disposition model with first-order absorption through transit compartments, and saturable elimination implemented using a well-stirred liver model14. The liver model assumptions were as follows: isoniazid plasma protein binding of 5%, a hepatic blood flow of 90L/h, and a liver volume of distribution of 1L for a typical individual weighing 56.1kg fat-free mass. Allometric scaling based on fat-free mass was applied on all clearance and volume of distribution parameters, including hepatic blood flow and volume. As expected, NAT2 genotype significantly affected intrinsic hepatic clearance of the drug, as the model predicted a 4-folds difference between slow and rapid acetylator clearances. Saturation of first-pass was parameterised using a Michaelis-Menten model with a Km value of 19.5mg/L. All participants treated with the MDR-TB regimen had a 65.6% lower AUC compared to participants on isoniazid monotherapy. Ethionamide coadministration was associated with a 29% increase in isoniazid AUC, irrespective of NAT2 genotype.
PD data (CFU and TTP) were available from 59 INHindsight participants, of which only 27 (46%) had corresponding MIC information. The decline in sputum CFU was interpreted using a one-compartment first-order kill model, while an exponential bacterial growth model was used to describe the growth in MGIT that produced the TTP data. Isoniazid concentration in the effect compartment modulated the first-order bacterial kill through a sigmoidal Emax relationship. The model predicted lower potency (i.e. higher EC50) but similar maximum-kill of isoniazid against inhA-mutated isolates compared to drug-sensitive. A 6-hours delay in the onset of growth in the MGIT was observed for sputum samples collected after isoniazid initiation (p≪0.001). The inhA-mutated isolates doubling growth time was identified to be 4 hours slower than the drug-sensitive isolates (p≪0.001). A weak positive relationship between MIC and EC50 was observed (p=0.035), but this was not included in the final model due to missingness. The effect compartment resulted in a “delay” with a half-life of approximately 17 hours before isoniazid concentration in plasma becomes effective against M.tb. Based on the PK/PD model simulations, to achieve a drop in bacterial load comparable to 5mg/kg against drug-sensitive TB, 10, 15, and 20mg/kg doses are necessary against inhA-mutated isolates in slow, intermediate, and rapid NAT2 acetylators, respectively.
Conclusions: We show nonlinear pharmacokinetics of isoniazid, particularly at doses >10mg/kg. Markedly lower isoniazid exposures in participants on combination MDR-TB treatment compared to monotherapy exposures were observed. There is evidence that the drug-drug interaction perpetrator might be terizidone/cycloserine, however further investigations are required, especially since the effect is major. In contrast, we observed a modest increase in isoniazid exposure when coadministered with ethionamide.
Additionally, the study results show that a higher isoniazid dose against inhA-mutated isolates has a high probability to achieve EBA similar to that of 5mg/kg against drug-sensitive isolates. Individualised dosing of isoniazid based on NAT2 acetylator status may help patients attain effective exposures against inhA-mutated isolates while mitigating toxicity risks associated with higher doses. Individualised dosing at the start of TB treatment is possible since simple molecular tests to detect and characterise the TB strain’s levels of isoniazid resistance are currently available, and recently, there has been progress in the development of a genetic test for isoniazid acetylator status.15,16
- Mouton JW, Ambrose PG, Canton R, et al. Conserving antibiotics for the future: New ways to use old and new drugs from a pharmacokinetic and pharmacodynamic perspective. Drug Resist Updat 2011; 14: 107–17. Available at: https://linkinghub.elsevier.com/retrieve/pii/S1368764611000215. Accessed February 6, 2020.
- Singh R, Dwivedi SP, Gaharwar US, Meena R, Rajamani P, Prasad T. Recent updates on drug resistance in Mycobacterium tuberculosis. J Appl Microbiol 2020; 128: 1547–67. Available at: https://sfamjournals.onlinelibrary.wiley.com/doi/full/10.1111/jam.14478. Accessed July 26, 2021.
- Rieder HL, Van Deun A. Rationale for high-dose isoniazid in the treatment of multidrug-resistant tuberculosis. Int J Tuberc Lung Dis 2017; 21: 123–4. Available at: http://www.ingentaconnect.com/content/10.5588/ijtld.16.0619. Accessed October 27, 2019.
- Lempens P, Meehan CJ, Vandelannoote K, et al. Isoniazid resistance levels of Mycobacterium tuberculosis can largely be predicted by high-confidence resistance-conferring mutations. Sci Rep 2018; 8: 3246. Available at: http://www.nature.com/articles/s41598-018-21378-x. Accessed October 27, 2019.
- Böttger EC. Chapter 14: Drug Resistance in Mycobacterium tuberculosis: Molecular Mechanisms and Laboratory Susceptibility Testing. In: Donald P., van Helden P., eds. Antituberculosis Chemotherapy. Vol 10. S. Karger AG, 2011; 128–44. Available at: https://www.karger.com/Article/FullText/324630. Accessed October 27, 2019.
- Dooley KE, Miyahara S, von Groote-Bidlingmaier F, et al. Early Bactericidal Activity of Different Isoniazid Doses for Drug-Resistant Tuberculosis (INHindsight): A Randomized, Open-Label Clinical Trial. Am J Respir Crit Care Med 2020; 201: 1416–24. Available at: https://www.atsjournals.org/doi/10.1164/rccm.201910-1960OC. Accessed May 11, 2020.
- Court R, Chirehwa MT, Wiesner L, et al. Effect of tablet crushing on drug exposure in the treatment of multidrug-resistant tuberculosis. Int J Tuberc Lung Dis 2019; 23: 1068–74. Available at: https://www.ingentaconnect.com/content/10.5588/ijtld.18.0775. Accessed July 5, 2020.
- Gausi K, Wiesner L, Norman J, et al. Pharmacokinetics and Drug‐Drug Interactions of Isoniazid and Efavirenz in Pregnant Women Living With HIV in High TB Incidence Settings: Importance of Genotyping. Clin Pharmacol Ther 2020; 109: 1034–44. Available at: https://onlinelibrary.wiley.com/doi/10.1002/cpt.2044. Accessed November 30, 2020.
- Sabbagh A, Darlu P, Vidaud M. Evaluating NAT2PRED for inferring the individual acetylation status from unphased genotype data. BMC Med Genet 2009; 10: 148. Available at: /pmc/articles/PMC2806877/. Accessed May 17, 2021.
- Keizer RJ, Zandvliet AS, Beijnen JH, Schellens JHM, Huitema ADR. Performance of Methods for Handling Missing Categorical Covariate Data in Population Pharmacokinetic Analyses. AAPS J 2012; 14: 601–11. Available at: http://www.ncbi.nlm.nih.gov/pubmed/22648902. Accessed February 28, 2019.
- Upton RN, Mould DRDR. Basic concepts in population modeling, simulation, and Model-Based drug development: Part 3—Introduction to pharmacodynamic modeling methods. CPT Pharmacometrics Syst Pharmacol 2014; 3: 1–16. Available at: http://doi.wiley.com/10.1038/psp.2013.71. Accessed February 7, 2020.
- Boeckmann AJ, Beal SL, Sheiner LB. NONMEM User’s Guide, Part V. Introductory Guide. NONMEM Proj Gr 2011: 48.
- Keizer RJ, Karlsson MO, Hooker A. Modeling and Simulation Workbench for NONMEM: Tutorial on Pirana, PsN, and Xpose. CPT Pharmacometrics Syst Pharmacol 2013; 2: e50. Available at: http://www.ncbi.nlm.nih.gov/pubmed/23836189. Accessed January 24, 2019.
- Chirehwa MT, Rustomjee R, Mthiyane T, et al. Model based evaluation of higher doses of rifampicin using a semi-mechanistic model incorporating auto-induction and saturation of hepatic extraction.
- Verma R, Patil S, Zhang N, et al. A Rapid Pharmacogenomic Assay to Detect NAT2 Polymorphisms and Guide Isoniazid Dosing for Tuberculosis Treatment. Am J Respir Crit Care Med 2021; 204: 1317–26. Available at: https://www.atsjournals.org/doi/10.1164/rccm.202103-0564OC.
- Schluger NW. Using Isoniazid More Safely and More Effectively: The Time Is Now. Am J Respir Crit Care Med 2021; 204: 1248–50. Available at: https://www.atsjournals.org/doi/10.1164/rccm.202108-1938ED.