Myriam Briki1,2, Paul Thoueille1, Markus Joerger3, Etienne Chatelut4, Fabienne Thomas4, Florent Puisset4, Joseph Standing5, Thierry Buclin1, Sandro Carrara2, Monia Guidi1,6,7
1Service of Clinical Pharmacology, Lausanne University Hospital and University of Lausanne, 2Bio/CMOS Interfaces Laboratory, Department of Engineering, Swiss Federal Institute of Technology in Lausanne - EPFL, 3Department of Oncology and Hematology, Cantonal Hospital St. Gallen, 4Oncopole Claudius-Regaud, Institut Universitaire du Cancer de Toulouse–Oncopole, Cancer Research Center of Toulouse, Inserm U1037, 5Department of Pharmacy, Great Ormond Street Hospital for Children NHS Foundation Trust, 6Institute of Pharmaceutical Sciences of Western Switzerland, University of Geneva, University of Lausanne, 7Centre for Research and Innovation in Clinical Pharmaceutical Sciences, Lausanne University Hospital and University of Lausanne
Introduction: Numerous population pharmacokinetic (popPK) studies of MTX have been published, characterizing MTX disposition in the context of oncology among both pediatric and adult patients [1]. This amount of MTX popPK knowledge has paved the way for model-informed precision dosing (MIPD), which promises to identify patients at risk of delayed MTX clearance and to grant them proactive care for toxicity prevention [2]. However, popPK models, by definition, remain dependent on the dataset they were built on [3], not necessarily representative of the heterogeneity prevailing in a large population. A model-based meta-analysis (MBMA) using individual patient data is an available approach to make a summary model based on different patient populations [4], which could potentially accommodate a broader target patient group. This study aims at characterising the PK of intravenous (iv) MTX through an MBMA, to summarize the available individual patient data. Methods: Authors of published MTX models in oncology settings were contacted based on a recently published systematic review of available models [1]. A pooled popPK analysis was performed (NONMEM) following a stepwise procedure testing multi-compartment disposition with linear elimination, as per standard workflow. Residual variability was evaluated using a single error model or an error model per study. Moreover, a nested analysis was performed (using $LEVEL blocks) to assess both inter-study and inter-individual variability (ISV, IIV in CV%) on PK parameters. Sex, age, bodyweight (BW), body surface area (BSA) and estimated glomerular filtration rate (eGFR) were tested as covariates. As the meta-dataset comprises both adults and children, prior allometric scaling using BW was applied to the PK parameters. Results: A total of 3’146 MTX concentrations measured in 472 patients were obtained from three independent datasets (i.e. Joerger et al. (2006)[5], Johansson et al. (2011)[6], and Gallais et al. (2021)[7]). A three-compartment model with linear elimination from the central compartment best described the overall MTX disposition. The base model parameter estimates with IIV and/or ISV (CV%) were a clearance (CL) of 19.2 L/h (IIV, 29%; ISV, 20%), intercompartmental clearances of 0.265 L/h (Q2) (IIV, 52%; ISV, 0%) and 0.466 L/h (Q3) (IIV and ISV 0%), and volumes of distribution 71.7 L (central V1) (IIV, 0%; ISV, 28%), 84.6 L (peripheral V2) (IIV, 29%; ISV, 19%) and 5.08 L (peripheral V3) (IIV, 0%; ISV, 25%). A proportional error model stratified per SID performed better than keeping a single overall mixed error model. Age significantly impacted CL and Q2, with an 80-year-old patient having a CL decrease of 26% and a near 2-fold increase in Q2 compared to an 18-year-old patient. Conclusion: Harmonizing data from different sources into one overall model allows diversifying the studied population and building a suitably generalized model. In the case of MTX, differences in studied patient populations appear significant, thus needing to be handled with two model features: nested variability and stratified residual error. MBMA based on individual data enables identification of covariates, such as BW and age if the merged populations include both adults and children. The TDM of MTX, already widely adopted for leucovorin rescue, might benefit from an MIPD approach for improving pre-emptive patient care. When multiple popPK models compete as priors for MIPD, algorithms such as model selection and/or averaging [8] are gaining interest; the elaboration of a meta-model based on individual data, like the present one for MTX, offers a valuable alternative worth evaluating.
1. Zhang, Y.; Sun, L.; Chen, X.; Zhao, L.; Wang, X.; Zhao, Z.; Mei, S. A Systematic Review of Population Pharmacokinetic Models of Methotrexate. Eur J Drug Metab Pharmacokinet 2022, 47, 143-164, doi:10.1007/s13318-021-00737-6. 2. Olivo, L.B.; de Oliveira Henz, P.; Wermann, S.; Dias, B.B.; Porto, G.O.; Pinhatti, A.V.; Martins, M.D.; Gregianin, L.J.; Costa, T.D.; de Araujo, B.V. Anticipating Leucovorin Rescue Therapy in Patients with Osteosarcoma through Methotrexate Population Pharmacokinetic Model. Pharmaceutics 2024, 16, doi:10.3390/pharmaceutics16091180. 3. Taylor, Z.L.; Poweleit, E.A.; Paice, K.; Somers, K.M.; Pavia, K.; Vinks, A.A.; Punt, N.; Mizuno, T.; Girdwood, S.T. Tutorial on model selection and validation of model input into precision dosing software for model-informed precision dosing. CPT Pharmacometrics Syst Pharmacol 2023, 12, 1827-1845, doi:10.1002/psp4.13056. 4. van Wijk, R.C.; Imperial, M.Z.; Savic, R.M.; Solans, B.P. Pharmacokinetic analysis across studies to drive knowledge-integration: A tutorial on individual patient data meta-analysis (IPDMA). CPT Pharmacometrics Syst Pharmacol 2023, 12, 1187-1200, doi:10.1002/psp4.13002. 5. Joerger, M.; Huitema, A.D.; van den Bongard, H.J.; Baas, P.; Schornagel, J.H.; Schellens, J.H.; Beijnen, J.H. Determinants of the elimination of methotrexate and 7-hydroxy-methotrexate following high-dose infusional therapy to cancer patients. Br J Clin Pharmacol 2006, 62, 71-80, doi:10.1111/j.1365-2125.2005.02513.x. 6. Johansson, A.M.; Hill, N.; Perisoglou, M.; Whelan, J.; Karlsson, M.O.; Standing, J.F. A population pharmacokinetic/pharmacodynamic model of methotrexate and mucositis scores in osteosarcoma. Ther Drug Monit 2011, 33, 711-718, doi:10.1097/FTD.0b013e31823615e1. 7. Gallais, F.; Oberic, L.; Faguer, S.; Tavitian, S.; Lafont, T.; Marsili, S.; Brice, A.; Chatelut, E.; Puisset, F. Body Surface Area Dosing of High-Dose Methotrexate Should Be Reconsidered, Particularly in Overweight, Adult Patients. Ther Drug Monit 2021, 43, 408-415, doi:10.1097/FTD.0000000000000813. 8. Uster, D.W.; Stocker, S.L.; Carland, J.E.; Brett, J.; Marriott, D.J.E.; Day, R.O.; Wicha, S.G. A Model Averaging/Selection Approach Improves the Predictive Performance of Model-Informed Precision Dosing: Vancomycin as a Case Study. Clin Pharmacol Ther 2021, 109, 175-183, doi:10.1002/cpt.2065.
Reference: PAGE 33 (2025) Abstr 11356 [www.page-meeting.org/?abstract=11356]
Poster: Drug/Disease Modelling - Oncology