Chunli Chen (1), Sebastian G. Wicha (1), Gerjo J. de Knegt (2), Jurriaan E.M. de Steenwinkel (2), Ulrika SH Simonsson (1)
(1) Department of Pharmaceutical Biosciences, Uppsala University, Uppsala, Sweden, (2) Erasmus MC, Department of Medical Microbiology and Infectious Disease, University Medical Centre Rotterdam, Rotterdam, The Netherlands
Objectives: The aim of this work was to investigate pharmacodynamic (PD) interactions between the standard treatment drugs of sensitive Mycobacterium tuberculosis in an chronic mouse model using the Multistate Tuberculosis Pharmacometric (MTP) model [1] and the general pharmacodynamic interaction (GPDI) model based on the Bliss Independence criterion [2].
Methods: Pharmacokinetic (PK) models for rifampicin (RIF), isoniazid (INH), pyrazinamide (PZA) and ethambutol (EMB) were developed using sparse PK data from separate infected BALB/c mice, combined with rich PK data from healthy BALB/c mice [3]. Infected BALB/c mice randomized to monotherapy received either 4 weeks of RIF (5, 10 or 20 mg/kg) or INH (12.5, 25 or 50 mg/kg) or EMB (50, 100 or 200 mg/kg) or PZA (75, 150 or 300 mg/kg). The PD biomarker colony forming unit (CFU) was measured after 1, 2 and 4 weeks treatment using 9 mice at each occasion. For the different duo, triple or quatro combinations of the drugs, fixed doses of 10 mg/kg RIF, 25 mg/kg INH, 100 mg/kg EMB and 150 mg/kg PZA were used and CFU was measured after 1, 2, 4, 8, 12 and 24-weeks of treatment using 3 mice at each occasion. Natural growth data was collected at 1, 3, 7, 14 and 21 days after infection. All modeling was done using NONMEM® 7.3[4] together with Perl-speaks-NONMEM [5], Xpose [5] and Pirana [6].
Results: In monotherapy, RIF was found to kill all three sub-states i.e. fast-multiplying (F), slow-multiplying (S) and non-multiplying bacteria (N) as well as to inhibit the growth rate of the F sub-state. INH had no effect on N, but killed F and S bacteria. Monotherapy of EMB and PZA displayed no detectable killing effects, because of lack of longitudinal PD data. Yet, in the presence of PZA, INH killed N bacteria. Antagonism was quantified between RIF and INH against S and N bacteria. This interaction increased log10 CFU/ml by approximately 0.79 and 0.86 compared with expected additivity on day 28 after treatment. EMB, itself inactive, synergized killing of RIF against N bacteria, which decreased log10 CFU/mL by 2.84 compared to expected additivity.
Conclusions: The present study results suggest that the proposed MTP model together with the GPDI model can be applied to both mono and combination therapy CFU data originating from animal studies. This approach provides a quantitative evaluation framework of potential PD interactions among anti-tuberculosis drugs in TB drug development.
References:
[1] Clewe O, Aulin L, Hu Y, et al. (2015) A multistate tuberculosis pharmacometric model: a framework for studying anti-tubercular drug effects in vitro. J Antimicrob Chemother: 1460-2091
[2] Wicha SG, Chen C, Clewe O, Simonsson US. (2016) A general pharmacodymamic interaction model based on the Bliss Independence criterion. PAGE abstract, Lisbon
[3] de Steenwinkel JEM, Aarnoutse RE, de Knegt GJ, et al. (2013) Optimization of the rifampin dosage to improve the therapeutic efficacy in tuberculosis treatment using a murine model. Am J Respir Crit Care Med 187:1127–1134
[4]Beal S, Sheiner LB, Boeckmann A, Bauer RJ. (2009) NONMEM User’s Guides (1989-2009), Icon Development Solutions, Ellicott City, MD, USA
[5]Keizer RJ, Karlsson MO, Hooker A. (2013) Modeling and Simulation Workbench for NONMEM: Tutorial on Pirana, PsN, and Xpose. CPT Pharmacomet Syst Pharmacol 2:e50
[6]Keizer RJ, van Benten M, Beijnen JH, et al. (2011) Piraña and PCluster: a modeling environment and cluster infrastructure for NONMEM. Comput Methods Programs Biomed 101:72–79
Reference: PAGE 25 () Abstr 5986 [www.page-meeting.org/?abstract=5986]
Poster: Drug/Disease modeling - Infection