In Silico Assessment of Limited Blood Sampling Strategies for Individualised Pharmacokinetic-guided Dosing of Efanesoctocog Alfa in Haemophilia A Patients

Introduction
Efanesoctocog alfa is a novel factor VIII (FVIII) concentrate with a unique molecular design that enables Von Willebrand Factor-independent clearance in patients with haemophilia A. Limited sampling strategies (LSSs) are necessary to implement accurate pharmacokinetic (PK)-guided dosing for efanesoctocog alfa in clinical practice.

Aim
This in silico study aims to evaluate the predictive performance of 10 LSSs with one to three samples for estimating individual PK profiles of efanesoctocog alfa.

Methods
Monte Carlo simulations based on a published population PK model generated individual FVIII activity-time profiles for a virtual population. LSSs were applied to sample from these profiles, and PK parameters, FVIII activity peak level at 0.5 h (C0.5), FVIII activity trough level at 168 h (C168) and time above FVIII activity thresholds (5, 10 and 40 IU/dL) were estimated using Bayesian forecasting.

Results
All LSSs complied with our requirements of a relative mean prediction error of <±5% and a relative root mean square error of <25%. For prophylactic dosing advice, a LSS was considered clinically suitable if at least 80% of predicted C168 had a prediction error within −3 to 3 IU/dL. This criterion was met by one two-sample LSS and three three-sample LSSs. For pre-operative dosing advice, suitability required at least 80% of predicted C0.5 within –20 to 20 IU/dL, fulfilled by one two-sample and six three-sample LSSs.

Conclusions
Several LSSs demonstrated adequate predictive performance for PK-guided dosing of efanesoctocog alfa.