Rachael Lawson1,2, Christine E Staatz1, Christopher J Fraser3, Stefanie Hennig4,5,
1. School of Pharmacy, University of Queensland, Brisbane, Queensland, Australia 2. Pharmacy Department, Queensland Children's Hospital, Brisbane, Queensland, Australia 3. Blood and Marrow Transplant Service, Queensland Children’s Hospital, Brisbane, Queensland, Australia 4. Certara, Inc., Princeton, New Jersey, USA 5. School of Clinical Sciences, Faculty of Health, Queensland University of Technology, Brisbane, QLD, Australia
Introduction: Busulfan is a chemotherapy drug used in haematopoietic stem cell transplantation. While several studies have observed a reduction in busulfan clearance (CL) over four days of therapy few non-linear-mixed-effects (NLME) models developed to describe busulfan pharmacokinetics in paediatrics have been based on a once daily dosing regimen or included samples from all four days of treatment [1]. Methods to measure drug exposure and perform dosage adjustment vary across treatment centers [1] and despite increasing evidence for time-dependent CL product information (PI) leaflets around the world still recommend an equation to calculate a new dose based on prior exposure using a linear CL assumption [2, 3]. Area-under-the-concentration-time curve (AUC) over a 24-hour period (AUC0-24) and cumulative exposure following all doses (AUCcum) are parameters used to measure busulfan exposure. AUCcum is increasingly being applied for reporting of outcomes and for individual dose targeting [4-6].
Objectives: To develop a population pharmacokinetic model of once daily busulfan to characterise changes in CL over a four day treatment course. To evaluate busulfan AUC0-24 after each dose as well as cumulative exposure (AUCcum) based on four different dosing adjustment scenarios.
Methods: Data from paediatric patients ≤18 years at the time of haematopoietic stem cell transplant who received intravenous once daily busulfan and had serial concentration-time measurements (0 [pre-dose], 3 [post-flush], 3.25, 4, 5, 6 and 8 [post-infusion start] hours) following each of four doses were included. The population PK model was developed using a NLME approach (NONMEM® v. 7.4.3) with FOCE+I and included both patient size (actual body weight with allometric scaling [7]) on all parameters and a maturation component on CL a priori. One, two and three compartment models with intra-individual variation (IIV) on all parameters were explored and intra-occasion variation (IOV) tested one by one on key parameters (CL, and volume of distribution). Changes in CL were explored by including a covariate effect to compare CL on each day to that estimated for Day 1 and alternatively using an empirical continuous model to describe time-dependant clearance [8]. Additional covariates (focusing on concurrent medications including acetaminophen, azole antifungals and fludarabine), identified based on physiological plausibility, were formally tested in a step-wise covariate model building process. Simulations to explore exposures (AUCcum, AUC0-24) achieved after different dosing adjustment strategies based on the final model were performed using the RxODE package in R (v. 4.0.5) in 1000 typical patients. A typical patient was 17 kg in weight and 4.2 years of age and the target AUCcum was 90 mg/L·h. Scenario 1 (S1) used the recommended PI dose [3] for all four doses with no dose adjustment performed. Scenario 2 (S2) used the recommended PI dose for dose 1 and then dose adjustment for Doses 2-4 were based on Dose 1 AUC0-24 and the equation from the PI, which assumes no changes in CL for Doses 2-4. Scenario 3 (S3) used the doses estimated from the final population PK model to calculate the administered dose based on typical CL for the typical patient and Scenario 4 (S4) used the recommended PI dose for dose 1 and then the final population PK model to calculate doses 2-4 considering the estimated CL from Dose 1 and Dose 1 AUC0-24 for the typical patient.
Results: Nighty-six (96) patients provided data from 383 dosing days and 2519 busulfan plasma concentration samples. A two-compartment model with linear elimination best described the data with a typical CL of 0.20 L/kg/h (IIV 20%, IOV 10.7%), a typical central volume of distribution of 0.58 L/kg (IIV 14.7%) and typical peripheral volume of distribution of 17.8 L/kg. Time-dependent CL was incorporated in the final model with CL reducing by 19.1% over four days of treatment in the typical patient, with the greatest estimated reduction being -27.9% and time to achieve 50% of maximal reduction estimated to be at 19.2 hours following the first dose. Simulations provided median AUCcum for Scenario 1 of 93.1 mg/L·h (range 51.8-166 mg/L·h), Scenario 2 of 108 mg/L·h (range 60.1-193 mg/L·h), Scenario 3 of 95.4 mg/L·h (range 53.2-170 mg/L·h) and Scenario 4 of 94.9 mg/L·h (range 52.8-169 mg/L·h) for a typical patient. Scenario 2, based on the PI resulted in a typical AUCcum value of 108 mg/L·h which is higher than the recommended target range for busulfan (78-101 mg/L·h) [4] and may put some patients at risk of toxicity.
Conclusion: A NLME model has been developed that describes time-dependent reduction in clearance for once daily busulfan in paediatric patients. The variability observed in both the exposure AUC0-24 after each dose and the AUCcum together with the reduction in CL over the course of treatment suggests that daily monitoring of exposure is required to ensure the patient achieves optimal target exposure across the full treatment course. This model can be used to target daily AUC0-24 values as well as an AUCcum value.
References:
[1] Lawson R, Staatz CE, Fraser CJ, Hennig S. Review of the pharmacokinetics and pharmacodynamics of intravenous busulfan in paediatric patients. Clin Pharmacokinet. 2020. doi:10.1007/s40262-020-00947-2.
[2] Otsuka.America.Pharmaceutical I. Busulfex (busulfan) for injection. Food and Drug Administration. 01/2015 ed2015.
[3] Otsuka.Australia.Pharmaceutical.Pty.Ltd. Australian Product Information Busulfex (Busulfan) Injection. eMIMs2018. p. 1-21.
[4] Bartelink IH, Lalmohamed A, van Reij EM, Dvorak CC, Savic RM, Zwaveling J et al. Association of busulfan exposure with survival and toxicity after haemopoietic cell transplantation in children and young adults: a multicentre, retrospective cohort analysis. The Lancet Haematology. 2016;3(11):e526-e36. doi:10.1016/s2352-3026(16)30114-4.
[5] Bartelink IH, van Reij EML, Gerhardt CE, van Maarseveen EM, de Wildt A, Versluys B et al. Fludarabine and exposure-targeted busulfan compares favorably with busulfan/cyclophosphamide-based regimens in pediatric hematopoietic cell transplantation: maintaining efficacy with less toxicity. Biology of blood and marrow transplantation : journal of the American Society for Blood and Marrow Transplantation. 2014;20(3):345-53. doi:10.1016/j.bbmt.2013.11.027.
[6] Hong KT, Kang HJ, Choi JY, Hong CR, Cheon J-E, Park JD et al. Favorable Outcome of Post-Transplantation Cyclophosphamide Haploidentical Peripheral Blood Stem Cell Transplantation with Targeted Busulfan-Based Myeloablative Conditioning Using Intensive Pharmacokinetic Monitoring in Pediatric Patients. Biology of blood and marrow transplantation : journal of the American Society for Blood and Marrow Transplantation. 2018;24(11):2239-44. doi:10.1016/j.bbmt.2018.06.034.
[7] Holford NHG, Anderson BJ. Allometric size: The scientific theory and extension to normal fat mass. Eur J Pharm Sci. 2017;109:S59-S64. doi:10.1016/j.ejps.2017.05.056.
[8] Wilkins JJ, Brockhaus B, Dai HQ, Vugmeyster Y, White JT, Brar S et al. Time-Varying Clearance and Impact of Disease State on the Pharmacokinetics of Avelumab in Merkel Cell Carcinoma and Urothelial Carcinoma. Cpt-Pharmacometrics & Systems Pharmacology. 2019;8(6):415-27. doi:10.1002/psp4.12406.
Reference: PAGE 29 (2021) Abstr 9686 [www.page-meeting.org/?abstract=9686]
Poster: Drug/Disease Modelling - Oncology