Morris Muliaditan (1), Francesco Bellanti (1), Meindert Danhof (1), Oscar Della Pasqua (1, 2)
(1) Division of Pharmacology, Leiden University, Leiden, The Netherlands; (2) GlaxoSmithKline, London, United Kingdom
Objectives: Patients affected by transfusion-dependent diseases who suffer from iron overload resulting from life-long blood transfusion require chelation therapy for the removal of iron excess. Deferiprone (DFP) is one of the most extensively studied oral iron chelators to date. In association with the DEferiprone Evaluation in Paediatric (DEEP) project [1], the aim of this analysis was to develop a PKPD model with the purpose of characterising the course of iron overload (i.e., the natural course of the disease) and the effect of the chelation therapy on ferritin levels in the target population.
Methods: Literature search based on PubMed's database was initially performed to retrieve all pertinent publications. A population PKPD model was subsequently developed using published data. Serum ferritin was chosen as primary clinical endpoint and normal turnover of serum ferritin was modelled with an indirect response model, followed by the integration of the effect of blood transfusion to account for disease progression. Subsequently, the effect of chelation therapy was implemented to describe the effect of deferiprone, which is administered at a dose of 75 mg/kg/day. Simulations were performed and results were subsequently validated using published data from retrospective clinical trials [2]. The analysis was performed using a non-linear mixed effect approach, as implemented in NONMEM 7.2.
Results: Ferritin turnover in normal health conditions was described by an indirect response model. The relationship between the amount of blood units and serum ferritin in untreated patients was translated in terms of conversion rate and then incorporated into the initial model using a hyperbolic function. Simulation scenarios were performed under different conditions to explore model performance with regard to disease progression. Finally, DFP effect was included using an Emax model. Disease and treatment effects for treated and untreated patients with initial mean serum ferritin of 4622.5 mg/L (2607-11550 mg/L) were simulated as and compared to references.
Conclusion: This is the first time a population PKPD model has been used to describe iron overload in patients affected by transfusion-dependent diseases. The model accounts for the effect of blood transfusion and treatment response, mimicking the time course of the clinical endpoint of interest. External validation of the model is currently being performed using data from a 10-year follow-up cohort in patients undergoing deferoxamine treatment.
References:
[1] http://www.deep.cvbf.net/
[2] Oncologic Drugs Advisory Committee; Briefing Document, NDA # 21-825
Reference: PAGE 22 () Abstr 2717 [www.page-meeting.org/?abstract=2717]
Poster: Other Drug/Disease Modelling