Thanakorn Vongjarudech (1), Anne-Gaëlle Dosne (2), Bart Remmerie (2), Mats O Karlsson (1), Elin M Svensson (1,3)
(1) Department of Pharmacy, Uppsala University, Sweden (2) Janssen R&D, Beerse, Belgium (3) Department of Pharmacy, Radboud university medical center, The Netherlands
Introduction: M2, the main metabolite of bedaquiline (BDQ), is associated with QTc prolongation. Assessing QT liabilities in persons with active tuberculosis (TB) is challenging as a result of the bias introduced by the Frederica correction which performs poorly with tachycardia[1, 2]. This can lead to missed detection of QT prolongation (early and late overestimation of change from baseline [ΔQT]), resulting in unnecessary interruption of treatment [3]. Earlier model-based analysis addressed this non-drug-related secular trend of ΔQTcF increase over time by estimating a time-dependent QT effect [2, 4, 5]. It, however, requires skilled personnel and is unsuitable for clinical practice. Our previous work found TB-related tachycardia normalized with treatment, leading to inaccurately corrected QT intervals. We proposed a time-varying corrected QT (QTcTBT) to adjust for heart rate changes [6]. This project aimed to evaluate the time effect of treatment on QTc prolongation and characterize the M2 concentration effect on QT using QTcTBT.
Method: Data from 429 patients with multidrug-resistant tuberculosis (MDR-TB) were collected from two Phase IIb trials. In C208, a 2-stage, randomized, double-blind, placebo-controlled study, newly diagnosed patients received BDQ or a placebo for 8 weeks in Stage 1 and 24 weeks in Stage 2 [7, 8]. In C209, a single-arm, open-label trial, both newly diagnosed and treatment-experienced patients received BDQ for 24 weeks [9]. All patients received a background regimen. This study included baseline and on-treatment ECG measurements, and M2 pharmacokinetic (PK) metrics, which were derived using a published pharmacokinetic model along with the observed PK data [10]. The QTcTBT model, adapted from Tanneau’s model, consists of a time on treatment effect (asymptotic change in QT), 24- and 12-hour circadian rhythm cycles, the effect of M2, and patient covariates. The evaluation of the effect of time on treatment involved comparing 3 models: i) without the time effect, ii) with a time effect where the typical QTmax (maximal changes in QT over time) was set at 0 with inter-individual variability (IIV) of QTmax (zero typical change but allowing individuals to change in both directions), and iii) with a time effect estimating QTmax and its IIV. To assess the effect of M2 on QTc, linear, Emax, and power drug effects were examined. Parameter estimates and likelihood testing were compared between models. The parameter uncertainty and 95% confidence interval (CI) were obtained using sampling importance resampling (SIR) [11]. Ten-fold cross-validation was performed with 10 repetitions to evaluate the impact of time on treatment and M2 effects, as well as predictive performance. This analysis was conducted using NONMEM 7.5 with assistance from PsN 5.3.0[12], and R 4.2.2.
Results: Cross-validation demonstrated that models incorporating time effects, assuming either no change in typical QT over time (normalized OFV 0, 95% CI -6 to 6), or estimating QT changes (OFV -7, 95% CI -12 to 1), performed comparably while the model without time effects (OFV +283, 95%CI 280 to 286) performed worst among three models. The model assuming no change in typical QT with IIV on QTmax over time was chosen, as the estimation of QT max from the model with time effect estimating QT change was considered clinically minor at 24 weeks (-1.83 ms) , and IIV on QTmax could describe the random changes in QT. The Emax model was the best to describe the M2 effect, and the estimation of the M2 effect indicated a consistent drug effect across models, accounting for either no change in typical QT or a change in QT (Emax = 23.916 ms, 95% CI 17.614-32.340, EC50 787.104 ng/mL, 95% CI 472.865-1266.740, vs. Emax = 22.915 ms, 95% CI 17.255-27.010, EC50 = 679.376 ng/mL, 95% CI 425.390-922.962). Furthermore, it was estimated that the therapeutically relevant M2 concentration of 300 ng/mL prolongs the QTcTBT interval by around 6-7 ms (6.598 ms 90%CI 5.576-7.731).
Conclusion: In this study, QTcTBT was used instead of QTcF to adequately address the non-drug related secular trend in QT that is driven by TB disease improvement on treatment, and simplify quantification of BDQ’s QT prolonging effect. The QT prolongation of BDQ due to an M2 effect at the therapeutically relevant concentration was robust across investigated QTc models and less than 10 ms.
[1] Olliaro PL, Merle C, Mthiyane T, et al. Effects on the QT Interval of a Gatifloxacin-Containing Regimen versus Standard Treatment of Pulmonary Tuberculosis. Antimicrob Agents Chemother 2017; 61: e01834-16.
[2] Li H, Salinger DH, Everitt D, et al. Long-Term Effects on QT Prolongation of Pretomanid Alone and in Combinations in Patients with Tuberculosis. Antimicrobial Agents and Chemotherapy; 63. Epub ahead of print October 2019. DOI: 10.1128/AAC.00445-19.
[3] van Beek SW, Tanneau L, Meintjes G, et al. Model-Predicted Impact of ECG Monitoring Strategies During Bedaquiline Treatment. Open Forum Infect Dis 2022; 9: ofac372.
[4] Tanneau L, Svensson EM, Rossenu S, et al. Exposure–safety analysis of QTc interval and transaminase levels following bedaquiline administration in patients with drug‐resistant tuberculosis. CPT Pharmacom & Syst Pharma 2021; 10: 1538–1549.
[5] Tanneau L, Karlsson MO, Rosenkranz SL, et al. Assessing Prolongation of the Corrected QT Interval with Bedaquiline and Delamanid Coadministration to Predict the Cardiac Safety of Simplified Dosing Regimens. Clin Pharmacol Ther 2022; 112: 873–881.
[6] Vongjarudech T, Dosne A-G, Remmerie B, et al. Optimal longitudinal QT interval correction method considering changes in heart rate variability in patients treated for drug-resistant tuberculosis. PAGE, https://www.page-meeting.org/default.asp?abstract=10448 (accessed 14 March 2024).
[7] Diacon AH, Donald PR, Pym A, et al. Randomized pilot trial of eight weeks of bedaquiline (TMC207) treatment for multidrug-resistant tuberculosis: long-term outcome, tolerability, and effect on emergence of drug resistance. Antimicrob Agents Chemother 2012; 56: 3271–3276.
[8] Diacon AH, Pym A, Grobusch MP, et al. Multidrug-Resistant Tuberculosis and Culture Conversion with Bedaquiline. N Engl J Med 2014; 371: 723–732.
[9] Pym AS, Diacon AH, Tang S-J, et al. Bedaquiline in the treatment of multidrug- and extensively drug-resistant tuberculosis. Eur Respir J 2016; 47: 564–574.
[10] Svensson EM, Dosne A, Karlsson MO. Population Pharmacokinetics of Bedaquiline and Metabolite M2 in Patients With Drug-Resistant Tuberculosis: The Effect of Time-Varying Weight and Albumin. CPT: pharmacometrics & systems pharmacology; 5. Epub ahead of print December 2016. DOI: 10.1002/psp4.12147.
[11] Dosne A-G, Bergstrand M, Harling K, et al. Improving the estimation of parameter uncertainty distributions in nonlinear mixed effects models using sampling importance resampling. J Pharmacokinet Pharmacodyn 2016; 43: 583–596.
[12] Lindbom L, Pihlgren P, Jonsson EN. PsN-Toolkit–a collection of computer intensive statistical methods for non-linear mixed effect modeling using NONMEM. Comput Methods Programs Biomed 2005; 79: 241–257.
Reference: PAGE 32 (2024) Abstr 11276 [www.page-meeting.org/?abstract=11276]
Poster: Drug/Disease Modelling - Infection