Geeta Sandhu (1,2), Sally Mapp (2), Christine Carrington (2), Stefanie Hennig (1)
(1) School of Pharmacy, The University of Queensland, Australia, (2) Division of Cancer Services, Princess Alexandra Hospital, Brisbane, Australia
Objectives: Describe the population pharmacokinetics (PK) of high dose methotrexate (MTX) in patients with non-Hodgkin lymphoma (NHL) undergoing Hyper-CVAD chemotherapy protocol.
Methods: Data was collected from NHL patients who received MTX as part of the Hyper-CVAD B cycle regimen and had >1 MTX concentration measured as part of clinical drug monitoring at the Princess Alexandra Hospital during 2002–2012. The lower limit of quantification (LLQ) of the MTX assay was 0.01umol/L. Additional censoring occurred due to “dropout” from patients being discharged after 48h post-dose with MTX concentration <0.05umol/L. Censored and LLQ observations were accounted for using the maximum likelihood estimation method (M3)[1]. The population PK parameters, inter-and intra-individual variability (IIV, IOV) and the influence of covariates on the PK of MTX were estimated using NONMEM 7.2 (Laplacian). Age, body size descriptors, haemoglobin (HB), liver and renal functions were tested via step-wise covariate modelling. A decrease in the objective function value (OFV) (Χ2 1,0.05 > 3.84) was considered a significant improvement. Visual predictive checks and non-parametric bootstrap were used for model evaluation.
Results: A total of 847 MTX measurements from 108 adults were collected over 120 h post-dose. Patients received up to 4 cycles of MTX (1 g/m2 over 24 h intravenously). Twelve observations were LLQ and 172 censored due to dropout. The data was best described by a 2-compartment model with a clearance (CL) of 23.3 L/h (IIV=17.1%, IOV=14.5%), inter-compartmental clearance of 2.3 L/h, volume of distribution for central (V1) and peripheral compartment of 47.7L and 26.8L, respectively with a proportional error model. Creatinine CL based on lean body weight (LBW) [2] and body size categorised by WHO criteria for underweight, normal, overweight and obese and HB on V1 reduced the OFV significantly. Explained parameter variability [3] for CL was greater than unexplained variability, suggesting clinical importance of the covariates included on CL.
Conclusions: LBW based creatinine CL and body size should be considered in optimising MTX dosing for patients receiving repeated Hyper-CVAD cycles. The influence of HB on V1 remains to be defined but may be due to fluid overload and third space accumulation.
References:
[1] Beal SL. Ways to fit a PK model with some data below the quantification limit. J Pharmacokinetic Pharmacodyn 2001;28(5):481-504.
[2] Janmahasatian S, Duffull SB, Chagnac A, Kirkpatrick CMJ, Green B. Lean body mass normalizes the effect of obesity on renal function. Br J Clin Pharmacol 2008;65(6):964-5.
[3] Hennig S, Karlsson MO. Concordance between criteria for covariate model building. J Pharmacokinet Pharmacodyn. 2014. DOI 10.1007/s10928-014-9350-8
Reference: PAGE 23 (2014) Abstr 3172 [www.page-meeting.org/?abstract=3172]
Poster: Drug/Disease modeling - Oncology