Albin Leding 1, Diana Angelica Aguilar-Ayala 2, Maxime R. Eveque-Mourroux 3, Natalya Serbina 3, Hans Lennernäs 1, Ainhoa LucÃa 5, Santiago Ramón-GarcÃa 6, Ulrika S.H. Simonsson 1
1 Uppsala University (Uppsala, Sweden), 2 Department of Microbiology, Faculty of Medicine, University of Zaragoza (Zaragoza, Spain), 3 Univ. Lille, Inserm, Institut Pasteur Lille, U1177 - Drugs and Molecules for living Systems (Lille, France), 4 TB Alliance (New York, USA), 5 Department of Microbiology, Faculty of Medicine, University of Zaragoza, Spanish Network for Research on Respiratory Diseases (CIBERES), Carlos III Health Institute (Zaragoza, Madrid, Spain), 6 Department of Microbiology, Faculty of Medicine, University of Zaragoza, Spanish Network for Research on Respiratory Diseases (CIBERES), Carlos III Health Institute, Research & Development Agency of Aragón Foundation (Fundación ARAID) (Zaragoza, Madrid, Zaragoza, Spain)
Introduction: Active metabolites of drugs under clinical development require evaluation of their quantitative and qualitative contributions to the overall pharmacological effect and safety profile. Moreover, given that parent compounds and their active metabolites often share similar mechanisms of action, potential pharmacodynamic (PD) interactions in humans should be predicted. TBAJ-587 is a diarylquinoline in clinical development. TBAJ-587 has been shown to have active metabolites, of which the metabolite M3 has been shown to have similar activity in vitro as TBAJ-587[1] and circulates in plasma to a lesser extent[2]. This work aimed to predict the impact of TBAJ-587’s metabolite M3 on relative human efficacy in phase 2a clinical trials across different TBAJ-587 oral doses, using a translational quantitative system pharmacology (QSP) framework.
Method: The translational QSP framework incorporated nonlinear mixed-effects models describing TBAJ-587’s and M3’s in vitro exposure response relationship with the multistate tuberculosis pharmacodynamic model[3], in vitro pharmacodynamic interactions between TBAJ-587 and M3 with the general pharmacodynamic interaction model[4], in vivo plasma to lung distribution from mice, population pharmacokinetic in healthy volunteers[2] and translational factors[5]. Translation factors included species difference in unbound fraction, state of infection at simulation start, bacterial dynamic differences between systems and bacterial susceptibility difference. The separate models were built in NONMEM and simulations were performed with the translational QSP framework built in mrgsolve[6]. The scenarios simulated were performed with the full translation QSP framework, the translation QSP framework without accounting for the PD interaction between TBAJ-587 and M3, the translation QSP framework with TBAJ-587 efficacy alone or the translation QSP framework with M3 efficacy alone. The simulations were performed to mimic a phase 2a antitubercular clinical trial lasting for 14 days on a typical individual level with once daily oral dosing of TBAJ-587 with 25, 50, 100 or 200 mg at the fed state.
Result: Inhibition of growth and killing of all described bacterial populations were identified for TBAJ-587 and M3. The unbound lung and plasma M3/TBAJ-587 ratio was predicted to be 1.60 and 1.16, respectively, when 100 mg once daily dosing was simulated for 14 days. The day 14 predicted percentage change from baseline in colony-forming units (CFU) in human sputum was between the range −17% and −50% for 25 mg and 200 mg once daily, when accounting for both TBAJ-587 and M3 efficacy, as well as the PD interaction between TBAJ-587 and M3. Furthermore, a predicted reduced range was seen in the change from baseline in the scenario without the PD interaction, resulting in the change from baseline range to be −19% to −36% over the dose levels. Simulated scenario with accounting for only TBAJ-587 resulted in a change from baseline of between −7% and −32% for 25 mg and 200 mg once daily. The predicted relative contribution at day 14 of M3 efficacy alone was less than TBAJ-587 efficacy alone, with a predicted median (min – max) M3/TBAJ-587 change from baseline ratio of 0.45 (0.15 – 0.68) over the dose levels. The negligible contribution of M3 to the total efficacy was seen after 100 mg and 200 mg doses, with a change from baseline ratio of 0.22 and 0.15, respectively. Pharmacodynamic interactions identified between TBAJ-587 and M3 were concentration-dependent. The PD interactions between TBAJ-587 and M3 showed on a CFU sputum level, were dependent on time as well as dose and at later timepoints showed more than expected additivity, except for 25 and 50 mg, where expected additivity was predicted at later timepoints.
Conclusion: In conclusion, accounting for both the active metabolite, M3, and the interaction between TBAJ-587 and M3 resulted in larger predicted change from baseline in phase 2a.
References:
1. Aguilar-Ayala DA, Rabodoarivelo MS, Eveque-Mourroux MR, Leding AAM, Sonnekalb L, Marco AP, et al. In vitro pharmacokinetics and pharmacodynamics of the diarylquinoline TBAJ-587 and its metabolites against Mycobacterium tuberculosis. bioRxiv; 2025 [cited 2026 Feb 2]. Available from: https://doi.org/10.1101/2025.11.05.686716
2. Leding AAM, Bruinenberg P, Conradie A, Nedelman J, Lombardi A, Hickman D, et al. Population pharmacokinetics of TBAJ-587 and its main metabolites-Evaluation of different loading dose strategies and early dose selection. Br J Clin Pharmacol. 2025 Nov 19;
3. Clewe O, Aulin L, Hu Y, Coates ARM, Simonsson USH. A multistate tuberculosis pharmacometric model: a framework for studying anti-tubercular drug effects in vitro. J Antimicrob Chemother. 2016 Apr;71(4):964–74.
4. Wicha SG, Chen C, Clewe O, Simonsson USH. A general pharmacodynamic interaction model identifies perpetrators and victims in drug interactions. Nat Commun. 2017 Dec;8(1):2129.
5. Wicha SG, Clewe O, Svensson RJ, Gillespie SH, Hu Y, Coates ARM, et al. Forecasting clinical dose-response from preclinical studies in tuberculosis research: translational predictions with rifampicin. Clin Pharmacol Ther. 2018 Dec;104(6):1208–18.
6. Elmokadem A, Riggs MM, Baron KT. Quantitative systems pharmacology and physiologically-based pharmacokinetic modeling with mrgsolve: a hands-on tutorial. CPT Pharmacometrics Syst Pharmacol. 2019 Dec;8(12):883–93.
Reference: PAGE 34 (2026) Abstr 12017 [www.page-meeting.org/?abstract=12017]
Poster: Drug/Disease Modelling - Infection