Katharina Götz (1, 2), Birgit Burkhardt (3), Antje Walz (4), Dietrich Knoerzer (5), Thorsten Lehr (1, 2)
(1) Department of Clinical Pharmacy, Saarland University, Germany, (2) Saarmetrics GmbH, Germany, (3) Clinic of Paediatric Haematology and Oncology, University Hospital Münster, Germany, (4) 3T Biosciences, USA, (5) Biometrics & Epidemiology, Roche Pharma AG, Germany
Objectives: Epstein-Barr virus (EBV)-infection during immunosuppression after allogeneic hematopoietic stem cell transplantation (allo-HSCT) is a major cause for post-transplant lymphoproliferative disorders (PTLDs). Both conditions are associated with relevant morbidity. Approved for lymphoproliferative and autoimmune diseases, the anti-CD20 antibody rituximab depletes the EBV-bearing B lymphocytes. Hence, off-label usage includes the management of latent EBV-infection and the prevention of EBV-PTLD. According to the drug label, the adult dosage of the drug MabThera (rituximab) is used in children aged 6 months to 18 years (375 mg/m2 body surface area). In pediatric patients with allo-HSCT, however, the adult dosage might result in life-threatening B-cell depression and lymphocytopenia. This study was focused on the following three objectives: To gain insights in post-HSCT B-cell kinetics, to characterize the dose-response relationship of rituximab in pediatric patients after allo-HSCT, and to provide optimized dosing recommendations for rituximab in these patients.
Methods: Routine clinical data of 205 pediatric patients with allo-HSCT between 2010 and 2020 at the University Hospital Münster, Germany, was included. At the decision of treating oncologists, rituximab was administered in single or multiple doses of 375 mg/m2 for patients aged ≥12 months (12 kg), or 12.5 mg/kg for patients <12 months (12 kg). Due to the retrospective nature of this study, pharmacokinetic data for rituximab was not available and individual drug concentrations were derived from a published population pharmacokinetic model,[1] the individual dosing history, and relevant patient characteristics. NONMEM® was used for population analysis, simulation, and predictions. A systems medicine model was developed on basis of 2006 B-cell and 8995 lymphocyte measurements, model validation comprised 1031 B-cell and 4303 lymphocyte counts. Clinical data available between day -60 and day +60 or the first rituximab treatment post-HSCT was utilized to predict B-cell count-time profiles for the first three years after allo-HSCT.
Results: Median age was 9 years; 133 patients were male. An EBV-infection defined as ≥ 1000 copies in two consecutive measurements was documented in 60% of stem cell transplants. In 27% of stem cell transplants, rituximab was given after allo-HSCT, with EBV-infection being the most frequent indication by far (86%). The final model integrates one submodel for B-cell kinetics and one submodel for Non-B-cell kinetics, each consisting of two submodels, which reflect the maturation of hematopoietic stem cells (HSCs) from the patient and the transplanted graft. The observed lymphocyte counts were modelled as sum of predicted B-cell and Non-B-cell counts to provide information on B-cell kinetics whenever measurements were sparse. Rituximab increased the elimination of B-cells from the graft by factor 6.26. The developed model adequately describes and predicts the rituximab effects on the individual post-HSCT B-cell reconstitution. Simulations on basis of the final model revealed that reducing the rituximab dose by factor 10 shortens the period of B-cell depression by 25% from 6.4 to 4.8 months.
Conclusions: The results of this study suggest that B-cell counts are expected to be achieved substantially earlier with reduced rituximab doses for pediatric patients after allo-HSCT. Increasing the existing knowledge on the dose-response relationship of rituximab in these patients, model-based simulations could assist pediatric oncologists in the decision for individual and optimized rituximab dosing regimens, potentially decreasing the risk for life-threatening adverse events.
References:
[1] Li J et al. J Clin Pharmacol. (2012) 52(12), 1918-1926.
Reference: PAGE 32 (2024) Abstr 11264 [www.page-meeting.org/?abstract=11264]
Poster: Drug/Disease Modelling - Paediatrics