Application of trial simulation in the design of a clinical study of recombinant factor (RFVIIA) in cirrhotic patients with active varicose bleeding

Purpose To develop a population PK (popPK) model for rFVIIa, based on multiple dose (MD) data from one study (STD A), to simulate the single dose (SD) required to reach the same maximum plasma concentration (C max ) in a new study (STD B), and to compare simulated multiple dose (MD) ‐scenarios for these. Methods: A popPK model was developed, using MD data from STD A on 243 patients with upper gastro‐intestinal bleeding. Both 1‐ and 2–compartment models with baseline were tested. Influence of body weight on model parameters was assessed. MD simulations were performed to sample distributions of C max , distribution volume (V c ), equivalent single dose (ESD, single dose resulting in same C max as in MD study), and PK profiles for the two studies. Results: Estimated parameters (between‐subject %CV) in the final, one‐compartment model with baseline were: CL, 37.7 mL/(kg*h) (55.1); V c , 179.5 mL/kg (67.2); baseline, 0.28 U/mL (N/A). No effect of body weight was found. For STD A, estimates of parameter median (95% confidence interval, 95%CI) from 10000 simulated subjects were: C max , 38.6 U/mL (9.5‐108.6 U/mL); V c , 179.2 mL/kg (36.9‐537.2 mL/kg); ESD, 239.4 mcg/kg (114.9‐351.5 mcg/kg). MD regimens of STD A and STD B were simulated for 1000 patients, estimates of C max median (95% CI) were: 39.7 U/mL (14.9‐105.4 U/mL) (STD A) and 39.2 U/mL (15.4‐99.5 U/mL) (STD B). Profiles with 95%CI are shown in Figure 1. Conclusion: A popPK model for rFVIIa was used in the design of a new clinical study. Clinical Pharmacology & Therapeutics (2004) 75 , P40–P40; doi: 10.1016/j.clpt.2003.11.152 Simulated profiles for STUDY A and STUDY B. See text for details