A Quantitative Systems Pharmacology Model for the Coagulation Network Describes Biomarker Changes Observed in a Clinical Study with FXa Variant and Predicts Age-Associated Biomarker Variations

Abstract

It is known that the coagulation factor levels change with age. In this work, we have applied a quantitative systems pharmacology (QSP) model of the coagulation network to predict whether age related changes in coagulation factor levels will impact dose response for PF-05320907 on various pharmacodynamic endpoints. PF-05230907 (PF-907) is a variant of FXa, in which the conformational transition from zymogen to active protease is impaired. Binding to activated factor V facilitates its transition to the active conformation, rescues procoagulant activity and is hypothesized to localize PF-907 hemostatic effect to sites of hemorrhage. It is currently in development for the indication of intracerebral hemorrhage (ICH). Pharmacokinetic (PK) and pharmacodynamic (PD) data are available from the healthy subjects phase 1 study of single dose escalation of intravenous bolus infusion of PF-907. The next study for PF-907 will be conducted in ICH patients, who will have a much higher median age than the median age in the healthy subjects in the Phase 1 study.Our group has implemented a QSP model for the coagulation network to enable integrated understanding of all the data and underlying pharmacology1. The model has been optimized to describe in vitro biomarker changes including; thrombin generation assay (TGA), activated partial thromboplastin time (aPTT) and prothrombin time (PT) as well as in vivo biomarker changes including prothrombin fragment 1+2 (PF1+2), thrombin-anti-thrombin III complex (TAT) and D-dimer. In this simulation study, we used the model to first describe biomarker changes with treatment of FXa variant in hemostatic normal subjects and then used model simulations to predict the behavior of important biomarkers in an older ICH population. A single compartment PK model for PF-907 was first established to describe the PK data obtained from the phase 1 study. The PK model was then combined with the QSP model to predict biomarker changes following PF-907 treatment. Comparison with observed clinical data showed that the model adequately predicted dose-dependent change in TGA parameters, aPTT, PF1+2, TAT and D-dimer. In addition, the model also predicted that there would be no change in PT, which was consistent with observed first in human results with the PF-907 treatment. After model validation using FIH data, the model was then used to predict biomarker changes for older subjects using literature reported changes in baseline levels of coagulation factors for subjects over a period of 40 years. The simulation predicted minimal shifts in the PD responses suggesting that the dose-response to PF-907 may not change significantly between young and older populations. The model, however, did not consider other characteristics beyond coagulation factor level changes in older populations, which may impact the safety profile of PF-907 treatment.In summary, this study indicates that it is possible to predict the response of a hemostatic agent with a QSP model. Following validation, the model can also extrapolate from a standard subject to new patient populations and indicates that no dose adjustment due to age is required.Reference1. Nayak, S., Lee, D., Patel-Hett, S., Pittman, D., Martin, S., Heatherington, A., Vicini, P. and Hua, F. (2015), Using a Systems Pharmacology Model of the Blood Coagulation Network to Predict the Effects of Various Therapies on Biomarkers. CPT: Pharmacometrics & Systems Pharmacology. doi:10.1002/psp4.50 DisclosuresNayak:Pfizer Inc: Employment. Lee:Pfizer Inc.: Employment. Arkin:Pfizer Inc: Employment. Martin:Pfizer Inc: Employment. Heatherington:Pfizer Inc.: Employment. Denney:Pfizer Inc.: Employment. Vicini:Pfizer Inc.: Employment. Hua:Pfizer Inc: Employment.