Bayesian pharmacokinetic-guided prophylaxis with recombinant factor VIII in severe or moderate haemophilia A

Personalised pharmacokinetics (PK) using Bayesian analysis with limited sampling is assumed to help to optimise prophylaxis in haemophilia A (HA) patients.

Our prospective, observational study analysed the influence of PK parameters on clinical variables (bleeding rates, joint status, adherence, and consumption) using myPKFiT® in a cohort of twenty-one severe and moderate HA patients on prophylaxis with recombinant FVIII (Advate®) in two periods of one year, the first before PK-based tailoring and the second after PK-guided prophylaxis. Intra-individual and inter-individual coefficients of variation (CV) of half-life (t_1/2) were calculated.

A total of 73 PK estimations were performed in both periods, resulting in 17.2% inter-individual CV in mean t_1/2, and 4.9% intra-individual CV. Before PK-based tailoring a significant association between joint bleeds and t_1/2 was found (P = 0.010), especially in patients with short t_1/2. This finding was reproduced (P = 0.013) after withdrawal of two patients with bleeding phenotype related to their advanced arthropathy but normal t_1/2 and trough levels. Patients with joint bleeds weighed less (P = 0.039) and required higher doses (P = 0.032) than patients with zero joint bleeds. These associations were not observed in the second period after the adoption of PK-guided prophylaxis. There were no differences between the two periods, although a tendency to fewer spontaneous bleeds was suggested after PK-based tailoring.

PK-guided prophylaxis facilitates an adequate level of bleeding control in patients with HA, maintaining clinical variables and patient convenience in an integrative manner, without increasing FVIII consumption.