Objectives: Model-informed precision dosing (MIPD) is crucial for optimizing busulfan therapy and leveraging a population pharmacokinetic model with MAP Bayesian methods to estimate AUC provides superior precision in dose individualization compared with traditional noncompartmental analysis.[1,2] We previously developed a population pharmacokinetic model using data from 188 children and young adults treated at a single center[3]; however, after implementation, some biases were observed which were potentially related to the covariate effects identified in the original model. To address reported biases, expand the population informing the model structure and covariate relationships, and improve model performance, we aimed to refit the Shukla model and assess its forecasting performance against published models.
Methods: Data were collected from 1062 children and adults and 9123 busulfan concentration samples across 11 US medical centers using the InsightRX Nova clinical decision support software from 2017-2024. Individual clinical sites were responsible for all clinical care decisions. The data were modeled using NONMEM and R, and the model performance was compared to that of published models using the PsN tool proseval.[3-7]
Results: For individuals undergoing busulfan MIPD, 84.4% achieved a cAUC within 80 to 120% of the target, and mean percent error of cAUC target attainment revealed bias towards achieving cAUCs under the user-selected target. Model refinement included adding a peripheral compartment and an exponential decay function for clearance. The previously reported regimen effect on clearance now estimated an opposite effect on busulfan clearance and was removed. The refined model outperformed all available models in forecasting prediction with a mean percent error of -2.9% (Bartelink: -4.1%, Bognar: -3.5%, Langenhorst: -6.7%, McCune: -7.31%, and Shukla: -6.8%), suggesting less bias towards underexposure than other available models, and normalized root mean square error of 23.2% (Bartelink: 26.2%, Bognar: 25.6%, Langenhorst: 26.1%, McCune 28.0%, Shukla: 27.8%).
Conclusions: MIPD of busulfan results in high target achievement and continuous refinement of models is essential to improve predictive performance. By leveraging a large, diverse, multi-center dataset and incorporating key structural and covariate improvements, this model demonstrates superior predictive accuracy and generalizability compared to previously published models.
Citations: [1] Hughes JH, Long-Boyle J, Keizer RJ. Maximum a posteriori Bayesian methods out-perform non-compartmental analysis for busulfan precision dosing. J Pharmacokinet Pharmacodyn. 2024 Jun;51(3):279-288. doi: 10.1007/s10928-024-09915-w. Epub 2024 Mar 23. PMID: 38520573; PMCID: PMC11136738. [2] Bognàr T, Garcia-Rosa M, Lalmohamed A, Güngör T, Hauri-Hohl M, Prockop S, Oram L, Pai SY, Brooks J, Savic RM, Dvorak CC, Long-Boyle JR, Krajinovic M, Bittencourt H, Teyssier AC, Théorêt Y, Martinez C, Egberts TCG, Morales E, Slatter M, Cuvelier GDE, Chiesa R, Wynn RF, Coussons M, Cicalese MP, Ansari M, Long SE, Ebens CL, Lust H, Chaudhury S, Nath CE, Shaw PJ, Keogh SJ, van der Stoep MYEC, Bredius R, Lindemans CA, Boelens JJ, Bartelink IH. Association of busulfan exposure and outcomes after HCT for patients with an inborn error of immunity. Blood Adv. 2024 Oct 8;8(19):5137-5145. doi: 10.1182/bloodadvances.2024013275. PMID: 39074263; PMCID: PMC11470247. [3] Shukla P, Goswami S, Keizer RJ, et al. Assessment of a Model-Informed Precision Dosing Platform Use in Routine Clinical Care for Personalized Busulfan Therapy in the Pediatric Hematopoietic Cell Transplantation (HCT) Population. Front Pharmacol. 2020;11:888. Published 2020 Jul 2. doi:10.3389/fphar.2020.00888 [4] Bartelink IH, van Kesteren C, Boelens JJ, Egberts TC, Bierings MB, Cuvelier GD, Wynn RF, Slatter MA, Chiesa R, Danhof M, Knibbe CA. Predictive performance of a busulfan pharmacokinetic model in children and young adults. Ther Drug Monit. 2012 Oct;34(5):574-83. doi: 10.1097/FTD.0b013e31826051bb. PMID: 22972539. [5] Bognár T, Bartelink IH, Egberts TCG, Rademaker CMA, Versluys AB, Slatter MA, et al. Association between the magnitude of intravenous busulfan exposure and development of hepatic veno-occlusive disease in children and young adults undergoing myeloablative allogeneic hematopoietic cell transplantation. Transplant Cell Ther. 2022 Apr;28(4):196-202. doi: 10.1016/j.jtct.2022.01.013. [6] Langenhorst JB, Admiraal R, van Kesteren C, van der Straaten T, de Jonge R, de Vries A, et al. A semi-mechanistic model based on glutathione depletion to describe intra-individual reduction in busulfan clearance. Br J Clin Pharmacol. 2020 Aug;86(8):1820-1834. doi: 10.1111/bcp.14256. Epub 2020 Feb 18. PMID: 32067250; PMCID: PMC7373715. [7] McCune JS, Bemer MJ, Barrett JS, Scott Baker K, Gamis AS, Holford NH. Busulfan in infant to adult hematopoietic cell transplant recipients: a population pharmacokinetic model for initial and Bayesian dose personalization. Clin Cancer Res. 2014 Feb 1;20(3):754-63. doi: 10.1158/1078-0432.CCR-13-1960. Epub 2013 Nov 11. PMID: 24218510; PMCID: PMC3946385.
