Therapeutic Drug Monitoring of Factor VIII Prophylaxis Using Its Plasma Coagulant Activity and Global Hemostasis Biomarkers: A Pharmacokinetic/Pharmacodynamic Pilot Study

Background: The current clinical practice of factor VIII (FVIII) prophylaxis revolves around converting patients with severe hemophilia A, hereafter simply referred to as hemophilia, phenotype (defined as plasma factor VIII coagulant activity [FVIII:C] \u3c1 IU dL-1) to moderate hemophilia phenotype (defined as plasma FVIII:C from 1–5 IU dL-1). However, a wide inter-individual variation in bleeding tendency is observed despite changes in plasma FVIII:C (pharmacokinetic [PK] changes). Therefore, monitoring FVIII prophylaxis by global hemostasis biomarkers (pharmacodynamic [PD] response) can potentially be beneficial. Objective: To conduct appropriate PK/PD modeling using plasma FVIII:C and global hemostasis (platelet function and blood viscoelastic) biomarkers in severe hemophilia. Methods: Nine non-bleeding severe hemophiliacs (plasma FVIII:C \u3c1 IU dL-1) with variant bleeding tendency (5 frequent bleeders and 4 infrequent bleeders) were infused with a recombinant factor FVIII (rFVIII) prophylactic dose (mean = 32.1 international units per kilogram [IU kg-1]). Blood was collected at baseline and 0.5-, 1-, 2-, 4-, 8-, 12-, 24- and 48-hours (h) post-dose for plasma FVIII:C, platelet function (platelet contractile force [PCF], clot elastic modulus [CEM] and force onset time [FOT]) and blood viscoelastic (reaction-time [R], kinetics-time [K] and maximum amplitude [MA]) biomarkers and activated partial thromboplastin time (aPTT). Non-compartmental analysis (NCA) was performed using standard methods. Compartmental analysis (CA) and PK/PD modeling were performed by non-linear regression. Correlation and analysis of variance (ANOVA) were used to explore the role of clinically relevant modifiers of bleeding tendency, as appropriate. ANOVA was used to assess inter-group differences in pertinent PK and PD parameters. P value \u3c0.05 significance level was pre-specified for all statistical tests. Results: Mean (±SD) volume of distribution at steady state (Vss), total clearance (CLtot) and terminal half-life (t1/2) from NCA were 40.5 (±11.2) milliliter per kilogram (mL kg-1), 2.9 (±1.2) milliliter per hour per kilogram (mL h-1 kg-1) and 11.6 (±6.2) h, respectively. Mean (±SD) Vss and CLtot from the one-compartment body model (CA) were 39.6 (±8.9) mL kg-1 and 3.1 (±1.3) mL h-1 kg-1, respectively. The mean (±SD) baseline effect (E0) and slope from the PK/PD linear modeling were: for aPTT, 48.9 (±4.4) seconds (sec) and -0.025 (±0.009) second deciliter per international unit (sec dL IU-1), respectively; for PCF, 0.3 (±0.3) kilodynes (kdynes) and 0.008 (±0.004) kilodynes deciliter per international unit (kdynes dL IU-1), respectively; and for CEM, 0.0 (±0.0) kilodynes per square centimeter (kdynes cm-2) and 0.032 (±0.016) kilodynes deciliter per international unit per square centimeter (kdynes dL IU-1 cm-2), respectively. The mean (±SD) maximum effect (Emax) and half the maximum effective concentration (EC50) from the PK/PD sigmoidal Emax model were: for FOT, 70.1 (±16.9) % reduction and 87.8 (±31.4) IU dL-1 for FOT, respectively; for R, 74.9 (±26.0) % reduction and 68.5 (±28.4) IU dL-1, respectively; and for K, 73.2 (±36.4) % reduction and 67.2 (±29.0) IU dL-1, respectively. MA was not PK/PD modeled due to its low sensitivity. Conclusions: Plasma FVIII:C remained ≥1 IU dL-1 over the prophylactic interval. FOT and R were the most sensitive biomarkers at lower plasma FVIII:C. PCF and CEM were more sensitive than K and aPTT at lower plasma FVIII:C. MA was the least sensitive biomarker. Correlation and inter-group differences did not reach statistical significance (small sample size). These results may be used to assess risk of bleeding and dose-optimize FVIII prophylaxis in severe hemophilia