Yu Fu1, Hadi Taghvafard1, Medhat M. Said1, Piet H. van der Graaf 1,2, J. G. Coen van Hasselt 1, Nelleke Snelder3
1. Leiden Academic Centre for Drug Research, Leiden University, the Netherlands 2. Certara QSP, Canterbury, United Kingdom 3. LAP&P Consultants BV, Leiden, the Netherlands
Objectives: Cardiovascular safety issues are a main causes associated with attrition during drug development. Most attention has been attributed to effects associated with QT-prolongation, even though non-QT drug effects on hemodynamic endpoint such as blood pressure or cardiac contractility also represent serious cardiovascular safety issues in drug development. Previously a hemodynamic systems model (Snelder model) was developed in rat to characterize mode-of-action and concentration-effect relationships on specific hemodynamic variables.1,2 Cardiac output (CO) was challenging to measure but essential for identification of the system parameters of the Snelder model. Measures for contractility such as left ventricular dP/dtmax, are of interest to replace the need for CO data, as it can be non-invasively measured during the experiments. dP/dtmax is determined by myocardial contractility and the loading conditions on the ventricle, which are key variables in the left ventricle pressure-volume (PV) loop, this makes dP/dtmax a relevant variable to replace CO measurements. Here, we developed a novel systems model which integrates dP/dtmax with CO, and other hemodynamic biomarkers, using multiple dog telemetry studies for the selective β1-blocker atenolol, as proof-of-concept. We characterized the model through identifiability analyses with respect to drug- and system- specific parameters, performed external validation studies, and investigated if dP/dtmax can replace CO.
Methods: Previously collected experimental data was used to developed the systems model, and included measurements for heart rate (HR), CO, mean arterial pressure (MAP) and dP/dtmax (CTRM) from three preclinical in vivo telemetry studies in conscious Beagle dog, of which two studies were used for model development, and one study was used for external validation. Three cosine functions were used to model the circadian rhythm. Pharmacokinetic parameters for atenolol in dog were based on literature data. The hemodynamic data were analysed using a non-linear mixed-effects modelling approach implemented in NONMEM (version 7.4.3, Icon Development Solutions, Ellicott, MD, USA) with Perl speak to NONMEM toolkit (version 4.8.1, Uppsala University, Sweden). To investigate the importance of CO data in the estimation, the final model was fitted to a dataset without CO readouts. We performed structural identifiability analysis of the model using the MATLAB toolbox GenSSI 2.0, which is a MATLAB toolbox3 together with MATLAB (version R2020a). We performed stochastic simulations and re-estimation (SSE) to identify if the model can be used to identify drug mode of action (MoA).
Results: The developed hemodynamic systems model adequately described the effect of atenolol on HR, CO, dP/dtmax and MAP dynamics. The model adequately captured the data in the external validation dataset. Atenolol was found to inhibit both HR and CTR, which is in line with its mechanism of action. The concentration-effect relationships were best described by Emax models. The EC50 for HR and CTR was fixed to the reported KD of 58.3 ng/ml for binding of atenolol to the β1 receptor.4 Emax for HR and CTR were estimated to be 0.415 (RSE 11.6%) and 0.422 (RSE 9.56%), respectively. Omission of CO data lead to parameter estimates within the 95% confidence interval of the parameters estimates of the model that included CO data. The identifiability analysis showed that the proposed model is structurally identifiable for both set of observable outputs ((HR, CTRM and MAP) and (HR, CTRM, CO and MAP)). In the SSE analysis, the models with correct MoA showed highest drop in OFV with all the three simulation scenarios. We show that model parameters are both structurally and practically identifiable.
Conclusions: We successfully developed a novel cardiovascular systems model to characterize hemodynamic drug effects integrating contractility data and CO data in a mechanism-based fashion based on PV loop theory. It is anticipated that this model can be applied to describe dynamic changes of contractility (dP/dtmax) and other hemodynamic biomarkers (HR, CO and MAP) for novel drugs in dog, and predict the drug effect on HR and contractility to assess the cardiovascular safety. CO data can be omitted for further development of the cardiovascular translational modelling platform. Future studies may focus on applying the developed model to additional compounds to confirm consistency of system-specific parameters.
References:
[1] Snelder, N., Ploeger, B.A., Luttringer, O., Rigel, D.F., Fu, F., Beil, M., et al. (2014). Drug effects on the CVS in conscious rats: Separating cardiac output into heart rate and stroke volume using PKPD modelling. Br. J. Pharmacol. 171: 5076–5092.
[2] Snelder, N., Ploeger, B.A., Luttringer, O., Rigel, D.F., Webb, R.L., Feldman, D., et al. (2013). PKPD modelling of the interrelationship between mean arterial BP, cardiac output and total peripheral resistance in conscious rats. Br. J. Pharmacol. 169: 1510–1524.
[3] Ligon, T.S., Fröhlich, F., Chis, O.T., Banga, J.R., Balsa-Canto, E., and Hasenauer, J. (2018). GenSSI 2.0: Multi-experiment structural identifiability analysis of SBML models. Bioinformatics 34: 1421–1423.
[4] Baker, J.G. (2005). The selectivity of β-adrenoceptor antagonists at the human β1, β2 and β3 adrenoceptors. Br. J. Pharmacol. 144: 317–322.
Reference: PAGE 29 (2021) Abstr 9635 [www.page-meeting.org/?abstract=9635]
Poster: Drug/Disease Modelling - Other Topics