A population pharmacokinetic study of high dose methotrexate treatment in infants with acute lymphoblastic leukemia: The importance of modeling inter-occasion variability across treatment cycles
Ryan J. Beechinor (1), Patrick A. Thompson (2), Michael F. Hwang (1), Ryan C. Vargo (3), Lisa R. Bomgaars (4), Daniel Gonzalez (1)
(1) Division of Pharmacotherapy and Experimental Therapeutics, UNC Eshelman School of Pharmacy, The University of North Carolina at Chapel Hill, Chapel Hill, NC, USA (2) University of North Carolina Lineberger Comprehensive Cancer Center, Chapel Hill, NC, USA; (3) Pharmacokinetics, Pharmacodynamics, and Drug Metabolism, Merck & Co., Inc., Kenilworth, NJ, USA; (4) Texas Children's Cancer and Hematology Center, Baylor College of Medicine, Houston, TX, USA
Introduction: Infants with acute lymphoblastic leukemia (ALL) treated with high dose methotrexate (MTX) may have reduced MTX clearance (CL) due to renal immaturity present at birth, which may predispose them to toxicity [1-2].
Objectives:
- Develop a population pharmacokinetic (PK) model of MTX in infants with ALL
- Characterize the impact of interoccasion variability (IOV) on this model
- Apply the model with simulations to evaluate the impact of various doses on MTX exposure
Methods: MTX concentrations were obtained from infants enrolled in the Children’s Oncology Group (COG) Clinical Trial P9407. A MTX PK database was developed by combining MTX PK data from 18 infants enrolled in a PK substudy with MTX PK data collected from 56 infants as part of routine care. All subjects were screened for adequate organ function prior to MTX treatment, including a creatinine CL >70 mL/min/1.73 m2. Each infant received MTX 4 g/m2 intravenously (IV) for four cycles during weeks 4, 5, 11, and 12 of intensification chemotherapy. A total of 711 MTX plasma concentrations were available from 74 infants treated with 234 cycles of high dose MTX. Median (range) baseline demographics of infants enrolled in this study included a post-natal age (PNA) of 7.5 months (2-12), total body weight of 8.8 kg (4.5-11.9), height of 69 cm (51-78), and body surface area of 0.40 m2 (0.24 – 0.48). A population PK analysis was performed using NONMEM® version 7.4 [3]. The impact of patient demographics was explored in a covariate analysis, and interindividual (IIV), IOV, and residual variability were estimated. Covariates tested on CL and volume of distribution (Vd) parameters included total body weight, post-natal age, height, and body surface area. The final model was evaluated using a nonparametric bootstrap analysis and a visual predictive check using Perl-speaks-NONMEM (version 3.6.2) [4-5]. Simulations were performed to assess the frequency of subtherapeutic and supratherapeutic MTX concentrations with doses ranging from 2 – 8 g/m2 of MTX given IV over 24 hours. For these simulations, a virtual population of 1000 infants was created, and MTX concentrations at 24, 48, and 72 hours were simulated. Target MTX concentrations were defined based on previous literature suggesting that MTX concentrations <16 µM at 24 hours are associated with relapse and subthe rapeutic, and MTX concentrations >1.0 µM at 48 hours or >0.1 µM at 72 hours are associated with renal toxicity and supratherapeutic [6-7].
Results: MTX plasma concentrations were best described by a two compartment model with linear elimination. After allometrically scaled total body weight was incorporated into CL and Vd terms, no other covariates were found to be significant. The final model included IIV on CL and central Vd, and IOV was included on CL. The mean (% relative standard error) final model parameter estimates allometrically scaled to a 70 kg adult were as follows: a CL of 10.9 L/h (3.0%), a central Vd of 62.6 L (4.9%), a peripheral Vd of 12.7 L (8.2%), and an intercompartmental CL of 0.128 L/hr (7.2%). The coefficient of variation for IOV was relatively high at 26.1%, compared to the IIV for CL and central Vd, 10.6% and 11.9% respectively. Simulations revealed that only the 2 g/m2 dosing regimen resulted in simulated 24 hour MTX concentrations <16 µM, and this occurred at a low frequency (2.4%). Additionally, compared to the study does of 4 g/m2 , escalating doses of 6 g/m2 and 8 g/m2 resulted in a greater percentage of simulated MTX concentrations in the supratherapeutic range at 72 hours, with 29%, 44%, and 57% of infants, respectively.
Conclusions: This is first population PK model of high dose MTX in infant ALL to include subjects as young as 2 months PNA. Infants in this study demonstrated a similar magnitude of IOV in their CL of MTX compared to previous studies performed in children [8]. Variation in the CL of MTX across cycles may be explained by changes in disease progression, drug-drug interactions, or unmeasured covariates which vary across cycles. The magnitude of IOV in the CL of MTX suggests that Bayesian adaptive dosing algorithms may be optimized by obtaining MTX concentrations during the current cycle infusion rather than concentrations measured from previous cycles. Our simulations revealed that dosing regimens of 2-8 g/m2 are likely to provide sufficient MTX exposure, however, increasing doses of high dose MTX above 4 g/m2 may result in untoward renal toxicity.
References:
[1] Lönnerholm G, Valsecchi MG, De Lorenzo P, et al. Pharmacokinetics of high-dose methotrexate in infants treated for acute lymphoblastic leukemia. Pediatr Blood Cancer. 2009;52(5):596-601.
[2] Thompson PA, Murry DJ, Rosner GL, et al. Methotrexate pharmacokinetics in infants with acute lymphoblastic leukemia. Cancer Chemother Pharmacol. 2007;59(6):847-853.
[3] Beal SL, Sheiner LB, Boeckmann AJ & Bauer RJ (Eds.) NONMEM Users Guides. 1989-2011. Icon Development Solutions, Ellicott City, Maryland, USA.
[4] Baverel PG, Savic RM, Karlsson MO. Two bootstrapping routines for obtaining imprecision estimates for nonparametric parameter distributions in nonlinear mixed effects models. J Pharmacokinet Pharmacodyn. 2011;38(1):63-82.
[5] Lindbom L, Pihlgren P, Jonsson EN, Jonsson N. PsN-Toolkit--a collection of computer intensive statistical methods for non-linear mixed effect modeling using NONMEM. Comput Methods Programs Biomed. 2005;79(3):241-257.
[6] Evans WE, Relling M V., Rodman JH, Crom WR, Boyett JM, Pui C-H. Conventional Compared with Individualized Chemotherapy for Childhood Acute Lymphoblastic Leukemia. N Engl J Med. 1998;338(8):499-505.
[7] Widemann BC, Adamson PC. Understanding and managing methotrexate nephrotoxicity. Oncologist. 2006;11(6):694-703.
[8] Aumente D, Buelga DS, Lukas JC, Gomez P, Torres A, García MJ. Population pharmacokinetics of high-dose methotrexate in children with acute lymphoblastic leukaemia. Clin Pharmacokinet. 2006;45(12):1227-1238.