Suruchi Bakshi (1), Marta Biedzka-Sarek (2), Eva-Maria Nichols (2), Fraser Cunningham (2), Stefano Zamuner (3), Piet H. van der Graaf (1,4)
(1) Certara QSP, Canterbury, UK; (2) Cytokine, Chemokine & Complement DPU, Immunoinflammation TA unit, GSK, UK; (3) Clinical Pharmacology, Modelling and Simulation, GSK, UK; (4) Cluster Systems Biomedicine and Pharmacology, LACDR, University of Leiden, The Netherlands
Objectives: The complement system (CS) is an integral part of the innate immune system. Its effector functions include pathogen clearance and inflammation. It can be activated via the classical, alternative or the lectin pathways (CP, AP and LP, respectively). The activation signals are then amplified and routed to the downstream terminal pathway by the AP. Dysregulation of the AP is implicated in several autoimmune as well as inflammatory diseases such as C3 glomerulopathy, macular degeneration and asthma [1]. The AP has been recognised as an attractive therapeutic target [2]. Certain potential AP targets exist at high concentration in vivo, while certain others display fast turnover. However, such experimental data is not available for the remaining AP targets. Quantitative comparison of suitability of various AP targets using systems pharmacology modelling may be useful in streamlining drug development efforts.
Methods: Towards this goal, we previously built and validated differential-equation-based models of the AP [3]. Two of these models were chosen for this work, namely the minimal model and the steady state model. The minimal model shows the maximal activation response and captures in vitro experimental results. The steady state model, on the other hand, simulates the physiological steady state response. We perturbed these two models using hypothetical drugs, considering both small molecule as well as antibody modalities. We used realistic affinities and a range of drug concentrations. We chose the modelled end-point of the pathway as a biomarker with which to study the effects of neutralising various AP targets. We used simulations using typical pharmacokinetic (PK) models for the drugs as well as sensitivity analyses (SA) to rank target suitability.
Results: Using the minimal and steady state models of AP allowed us to study the translatability of target rank-orders between models of an ‘in vitro’ versus an ‘in vivo’ systems dynamics. AP targets displayed differences in their suitability ranking based on the drug modality. Certain AP targets performed better with small molecule drugs while others performed better with antibodies. Systems pharmacology modelling which included typical drug PK was necessary to capture the effect of target-mediated drug disposition displayed by high abundance targets. Furthermore, we observed that SA and turnover rate on their own are insufficient to predict target rank order, particularly for antibodies. Similarly, baseline target concentration alone was insufficient to predict target rank order for small molecule drugs.
Conclusions: We believe that currently there is no single quantitative measure available, which incorporates the effects of target concentration, turnover as well as its sensitivity. Therefore, it is necessary to perform simulations of the full systems models including a typical PK model for the drugs. We have demonstrated the use of such systems pharmacology models in ranking target suitability using the alternative pathway of the complement system.
References:
[1] Thurman JM, Holers VM. The Central Role of the Alternative Complement Pathway in Human Disease. The Journal of Immunology. 2006;176(3):1305-1310.
[2] Morgan BP, Harris CL. Complement, a target for therapy in inflammatory and degenerative diseases. Nature Reviews Drug Discovery. 2015;14(12):857-877.
[3] Bakshi S, Biedzka-Sarek M, Nichols EM, Cunningham F, Petit-Frere S, et al. Bottom-up modelling of Alternative pathway of complement system. Submitted.
Reference: PAGE 27 (2018) Abstr 8430 [www.page-meeting.org/?abstract=8430]
Poster: Methodology - New Modelling Approaches