Rotational thromboelastometry can predict the probability of bleeding events in a translational rat model of haemophilia A following gene‐based FVIIa prophylaxis

Abstract

Introduction: Monitoring of clinical effectiveness of bypassing agents in haemophilia patients is hampered by the lack of validated laboratory assays. Thromboelastography (TEG) and rotational thromboelastometry (ROTEM) have been evaluated for predicting clinical effectiveness of bypassing agents, however, with limited success. Aim: Application of a longitudinal model-based approach may allow for a quantitative characterization of the link between ROTEM parameters and the probability of bleeding events. Methods: We analyse longitudinal data from haemophilia A rats receiving gene-based FVIIa prophylaxis in terms of total circulatory levels of FVII/FVIIa, clotting time (CT) measured using ROTEM and the probability of bleeding events. Results: Using population pharmacokinetic-pharmacodynamic (PKPD) modelling, a PK-CT-repeated time-to-event (RTTE) model was developed composed of three submodels (a) a FVII/FVIIa PK model, (b) a PK-CT model describing the relationship between predicted FVIIa expression and CT and (c) a RTTE model describing the probability of bleeding events as a function of CT. The developed PK-CT-RTTE model accurately described the vector dose-dependent plasma concentration-time profile of total FVII/FVIIa and the exposure-response relationship between AAV-derived FVIIa expression and CT. Importantly, the developed model accurately described the occurrence of bleeding events over time in a quantitative manner, revealing a linear relationship between predicted change from baseline CT and the probability of bleeding events. Conclusion: Using PK-CT-RTTE modelling, we demonstrated that ROTEM parameters can accurately predict the probability of bleeding events in a translational animal model of haemophilia A. K E Y W O R D S haemophilia A, pharmacometrics, rat model, rotational thromboelastometry, time-to-event analysis, translational pharmacology

FVIII activity following FVIII protein infusion or FVIII gene transfer predicts the bleeding risk in hemophilia A rats

“…A similar approach (RTTE modeling) has previously been used in HA rats expressing rat FVIIa to correlate a clotting assay parameter with the probability of bleeds. 54 Using this approach, we showed that a reduction in bleeding risk by 80% or 95% compared to untreated rats, required 2.1 and 10 IU/dL of hFVIII (or 0.021 and 0.1 IU/mL), respectively. At a concentration of 10 IU/ dL (0.1 IU/mL or 10% in humans), the probability of experiencing one bleed during the full 12-week study period was 9%.…”

“…However, both AAV‐ and protein infusion‐based prophylaxis data were used for an in silico model‐based approach. A similar approach (RTTE modeling) has previously been used in HA rats expressing rat FVIIa to correlate a clotting assay parameter with the probability of bleeds 54 . Using this approach, we showed that a reduction in bleeding risk by 80% or 95% compared to untreated rats, required 2.1 and 10 IU/dL of hFVIII (or 0.021 and 0.1 IU/mL), respectively.…”