A new modeling approach allowing prediction and comparison of the long-term outcomes of treatments for hemophilia B.

AIM: To develop a modeling approach to compare clinical outcomes of nonacog beta pegol to a standard-acting factor IX (FIX) product. METHODS: Regression analysis linked FIX activity to bleed rates. Pharmacokinetic parameters were used to estimate FIX activity over time. The probability of bleeds was estimated for both treatment arms. A Markov model estimated the presence of target joints and annualized bleed rates (ABRs). RESULTS: Higher FIX activity showed reduced ABRs (p < 0.001). Target joints resulted in higher bleed rates (p < 0.001). When FIX activity levels and bleed risks were applied to the Markov model, ABRs for nonacog beta pegol and its comparator were 2.40 and 6.36, respectively. CONCLUSION: This model provides a starting point for assessing the added value of new FIX products.

Prediction of human pharmacokinetics of activated recombinant factor VII and B-domain truncated factor VIII from animal population pharmacokinetic models of haemophilia.

Various experimental animal models are used in haemophilia research, however, little is known about how well the different species predict pharmacokinetic (PK) profiles in haemophilia patients. The aim of the current study was to describe the plasma concentration-time profile of recombinant activated factor VII (rFVIIa) and recombinant factor VIII (rFVIII) in several experimental animal models using population PK modelling, and apply a simulation-based approach to evaluate how well the developed animal population PK models predict human PK. PK models were developed for rFVIIa and rFVIII in mice, rats, monkeys, and dogs using nonlinear mixed-effects modelling, accounting for inter-individual variability, nonlinear kinetics and covariate effects. Three scaling principles were applied to predict human PK: proportional scaling to body weight from single species, scaling with fixed theory-based allometric exponents from single species, and allometric interspecies scaling with estimated allometric coefficients and exponents. The plasma concentration-time profile of rFVIIa and rFVIII in mice, rats, monkeys and dogs were accurately described by the developed species-specific PK models, accounting for nonlinear kinetics and gender-specific difference in clearance for rFVIII. The predictive performance of the animal population PK models of rFVIIa and rFVIII revealed significant species-variation. The developed PK models of rFVIIa and rFVIII in monkeys and dogs along with allometric interspecies scaling revealed high predictive performance for human PK, and may promote rational decision-making in future first-in-human trials for rFVIIa and rFVIII variants.