III-47 Federico Romano

Extrapolation of the early bacterial activity of pretomanid to inform dose selection for the treatment of pulmonary tuberculosis

Federico Romano (1) Morris Muliaditan (2) Ruth Casanueva (4) Oscar Della Pasqua (1,3,5)

(1) Department of Clinical Pharmacology and Therapeutics, University College London (London), UK, (2) Leiden Experts on Advanced Pharmacokinetics and Pharmacodynamics (LAP&P), Leiden, The Netherlands (3) Clinical Pharmacology Modelling and Simulation, GlaxoSmithKline, Brentford, UK (4) Drug Discovery Biology, Therapeutic Efficacy Unit, TB group, Tres Cantos, Spain (5) Consiglio Nazionale delle Ricerche (CNR), Rome, Italy

INTRODUCTION

A limitation in antitubercular (anti-TB) drug development remains the empiricism based on which companion drugs and their dosing regimens are selected for clinical testing. Model-based approaches can offer a method of discriminating whether the dose and exposure to an anti-tubercular compound is appropriate to ensure optimal antibacterial activity (i.e, bactericidal and sterilizing effects), either as monotherapy or in combination with other drugs. Recently, the use of a bacterial growth dynamics model enabled the characterisation of the contribution of companion drugs to the overall antibacterial activity of rifampicin and bedaquiline in murine TB infection models [1]. This semi-mechanistic model was developed to include fast-(F) and slow-(S) growing bacterial subpopulations, which represent the log-growth and stable phase of M.tuberculosis infection [1, 2]. In addition, its parameterisation allows one to disentangle system-specific from drug-specific properties, enabling translation from preclinical species to humans. Pretomanid (P) has been selected as a companion drug in next-generation anti-TB regimens at a daily 200mg dose. However, based on early bactericidal activity (EBA) studies alone, it was unknown whether P has long-term sterilising activity. Hence, we illustrate the application of the growth dynamics model as a translational tool for the prediction of the EBA of P, including the effect of doses higher than 200 mg to achieve potentially higher bacterial clearance.

METHODS

We first characterised the pharmacokinetics (PK) of P using blood concentrations in non-infected mice (n=22) following administration of either oral or intravenous 2-75mg/kg doses. Secondly, using internal and published data from acute and chronic murine models of M. tuberculosis (H37Rv) infection [3-6], the antibacterial activity of P (n=50:1.4-600 mg/kg) was estimated and scaled to humans taking into account interspecies differences in PK, pharmacodynamics and disease [1, 7]. All animal studies were ethically reviewed and carried out in accordance with European Directive 2010/63/EEC and the GSK Policy on the Care, Welfare and Treatment of Animals. Subsequently, a population PK model [8] was used to describe the concentrations of P at steady state (Css,avg) in drug-sensitive TB patients (n=120) from two clinical studies (CL-007 and CL-010) [9, 10]. The predicted Cssavg in these patients was then used in conjunction with the scaled system-specific parameters to predict the EBA (n=118) over 14 days. Simulated results were compared to the observed data to assess the predictive performance. Clinical trial simulations were implemented based on the same study population enrolled in the EBA studies and included untested doses (100-1200mg/daily) [9, 10]. A target threshold of 90% of the estimated maximum killing rate (EC90) for both F and S bacterial subpopulations was used as rationale for human dose selection. All analysis were performed using NONMEM version 7.3 [11].

RESULTS

Potency estimates (EC50) for P were 0.491mg/L and 2.08mg/L against F- and S- subpopulations. Differences were also observed for the maximal effect (Emax-F= 0.0568h-1 and Emax-S= 0.0105 h-1). Model-based extrapolation of the antibacterial activity showed good agreement between the predicted and observed CFU/ml counts over time. Simulation scenarios predicted the median antimicrobial activity to range between 0.092-0.138 log10 CFU/ml/day for daily doses that ranged between 100-1200mg. This effect corresponded to average predicted Css, avg in TB patients in the range between 0.67 and 8 mg/L Based on our simulations, the recommended 200mg dose yields a Css, avg that is 85% of the Emax-F, but far below the EC90 for the Emax-S.

CONCLUSIONS

Whilst EBA has been considered a standard step in the clinical evaluation of anti-TB drugs, the implications of EBA results for dose selection for subsequent progression of candidate compounds in combination therapy have been overlooked. Compared to empirical approaches, the use of a bacterial growth dynamics model provides a robust basis for quantifying the antibacterial activity on different metabolic phenotypes of M.tuberculosis. Importantly, our results show that the 14-day EBA of a 200mg dose of P is comparable to that of rifampicin, linezolid and bedaquiline (0.113, 0.104 and 0.104 log10 CFU/ml/day). On the other hand, it can be anticipated that higher doses may contribute to a potential sterilizing activity following longer treatment duration.

References:

  1. Muliaditan, M. and O. Della Pasqua, Evaluation of pharmacokinetic-pharmacodynamic relationships and selection of drug combinations for tuberculosis. Br J Clin Pharmacol, 2021. 87(1): p. 140-151.
  2. Muliaditan, M. and O. Della Pasqua, Bacterial growth dynamics and pharmacokinetic-pharmacodynamic relationships of rifampicin and bedaquiline in BALB/c mice. Br J Pharmacol, 2022. 179(6): p. 1251-1263.
  3. Ahmad, Z., et al., PA-824 exhibits time-dependent activity in a murine model of tuberculosis. Antimicrob Agents Chemother, 2011. 55(1): p. 239-45.
  4. Tyagi, S., et al., Bactericidal activity of the nitroimidazopyran PA-824 in a murine model of tuberculosis. Antimicrob Agents Chemother, 2005. 49(6): p. 2289-93.
  5. Tasneen, R., et al., Sterilizing activity of novel TMC207- and PA-824-containing regimens in a murine model of tuberculosis. Antimicrob Agents Chemother, 2011. 55(12): p. 5485-92.
  6. Lenaerts, A.J., et al., Preclinical Testing of the Nitroimidazopyran PA-824 for Activity against Mycobacterium tuberculosis in a Series of In Vitro and In Vivo Models. Antimicrobial Agents and Chemotherapy, 2005. 49(6): p. 2294-2301.
  7. Muliaditan M, D.P., Model-based rationale for drug combinations in tuberculosis. 2017(PAGE 26 (2017) Abstr 7272 [www.page-meeting.org/?abstract=7272]).
  8. Salinger, D.H., et al., Population Pharmacokinetics of the Antituberculosis Agent Pretomanid. Antimicrobial agents and chemotherapy, 2019. 63(10): p. e00907-19.
  9. Diacon, A.H., et al., Phase II dose-ranging trial of the early bactericidal activity of PA-824. Antimicrob Agents Chemother, 2012. 56(6): p. 3027-31.
  10. Diacon, A.H., et al., Early bactericidal activity and pharmacokinetics of PA-824 in smear-positive tuberculosis patients. Antimicrob Agents Chemother, 2010. 54(8): p. 3402-7.
  11. ICON Development Solutions, E.C., MD. NONMEM. [v.] 2018.

Reference: PAGE 30 (2022) Abstr 10223 [www.page-meeting.org/?abstract=10223]

Poster: Drug/Disease Modelling - Infection