Laure Lalande (1), Laurent Bourguignon (1,2,3), Pascal Maire (1,2), Sylvain Goutelle (1,2,3)
(1) Université Lyon 1, CNRS, UMR5558, Laboratoire de Biométrie et Biologie Evolutive, Villeurbanne, France. (2) Service Pharmacie, Groupement Hospitalier de Gériatrie, Hospices Civils de Lyon, Lyon, France. (3) Université Lyon 1, ISPB – Faculté de Pharmacie, Lyon, France.
Objectives: To develop a mathematical model describing the time-course of tuberculosis infection and its early treatment by isoniazid (INH) in the human lung. Then, to simulate the antibacterial effect of various INH dosage regimens and explore the drug and host determinants of the killing effect of INH against Mycobacterium tuberculosis (MTB).
Methods: We combined a pharmacokinetic (PK) model describing the plasma and lung disposition of INH in human [1], a pharmacodynamic (PD) model describing the relationship between INH concentration and its antibacterial effect [2], and an immune response model [3] to build a full systems’ pharmacology model of TB treatment by INH. This model was used to simulate various INH dosing therapy. The antibacterial effect on extracellular MTB predicted over the first days of therapy was compared to published values of early bactericidal activity (EBA). Global sensitivity analysis was performed using GUI-HDMR software tool to identify the model parameters that mostly influenced the early effect [4].
Results: For the standard INH dose of 300 mg/day, the predicted EBAs for the first two days of therapy were 0.454 ± 0.272 and 0.594 ± 0.243 log10CFU/mL/day, for fast and slow acetylators respectively. These values were in accordance with values of EBA reported in patients as they ranged between 0.371 and 0.770 log10CFU/mL/day [5]. Importantly, the model reproduced features of the antibacterial effect of INH observed in humans: the early antibacterial effect of INH appeared to level off for doses above 300 mg, and the killing effect displayed the typical biphasic shape of INH effect. The global sensitivity analysis revealed that the early antibacterial effect of INH was mostly influenced by PK and PD parameters such as plasma elimination rate constant, rate transfer from plasma to the lung, and maximal killing effect. However, an influence of pathophysiological parameters, although limited, appeared during the second phase of the decline.
Conclusions: Our model qualitatively and quantitatively reproduced important characteristics of the antibacterial effect of INH observed in patients. Simulations indicated that INH early killing effect was largely influenced by PK variability. This model framework may be extended to other anti-TB drugs and may be useful for preclinical development of anti-TB drug regimens.
References:
[1] Lalande, L., Bourguignon, L., Bihari, S., Maire, P., Neely, M., Jelliffe, R., Goutelle, S., 2015. Population Modeling and Simulation Study of the Pharmacokinetics and Antituberculosis Pharmacodynamics of Isoniazid in Lungs. Antimicrob. Agents Chemother. 59, 5181–5189.
[2] Gumbo, T., Louie, A., Liu, W., Ambrose, P.G., Bhavnani, S.M., Brown, D., Drusano, G.L., 2007. Isoniazid’s bactericidal activity ceases because of the emergence of resistance, not depletion of Mycobacterium tuberculosis in the log phase of growth. J. Infect. Dis. 195, 194–201.
[3] Marino, S., Kirschner, D.E., 2004. The human immune response to Mycobacterium tuberculosis in lung and lymph node. J. Theor. Biol. 227, 463–486.
[4] Ziehn, T., Tomlin, A.S., 2009. GUI–HDMR – A software tool for global sensitivity analysis of complex models. Environ. Model. Softw. 24, 775–785.
[5] Donald, P.R., Diacon, A.H., 2008. The early bactericidal activity of anti-tuberculosis drugs: a literature review. Tuberc. Edinb. Scotl. 88 Suppl 1, S75–83.
Reference: PAGE 25 (2016) Abstr 5776 [www.page-meeting.org/?abstract=5776]
Poster: Drug/Disease modeling - Infection