Abstract
Reduced intensity conditioning regimend have successfully extended the use of HCT to older individuals and in those with co-morbidities. Cyclophosphamide (CY) is a commonly used alkylating agent in RIC regimens. It is an inactive pro-drug that is hepatically metabolized to the active form, PM. It exerts its cytoreductive action by further non-enzymatic conversions leading to formation of DNA adducts and cellular death. There is wide inter-patient variability in toxicity of CY, which may be due to variability in systemic exposure. The aim of our study was to characterize pharmacokinetics of PM and to identify clinical factors associated with pharmacokinetic variability.
Forty-one adults undergoing allogeneic RIC HCT with CY, fludarabine and TBI were prospectively studied for PM pharmacokinetics from March 2013 to May 2014. CY 50mg/kg x one dose was administered intravenously over 2 hr at constant rate on days -6 and pharmacokinetic sampling was conducted at 2, 4, 6, 21, 24 and 45 hrs after the end of infusion. PM was derivatized with diethyldithiocarbamate and measured by a validated HPLC assay with ultraviolet detection. The lower limit of quantification was 50ng/ml. A population pharmacokinetic analysis was conducted using non-linear mixed effects model to obtain typical value of apparent clearance (Cl/fm), apparent volume of distribution (V/fm) of the metabolite and conversion rate constant (kf) of conversion to the metabolite. Clinical covariates such as weight, ideal body weight, age, gender, CrCl, total bilirubin, albumin, previous transplant, SCr, ALT, AST, and alkaline phosphatase were tested to explain the observed variability in kf, Cl/fm and V/fm. A step-wise covariate model building strategy of forward inclusion and backward elimination was used to identify the effect of clinical covariates on PM pharmacokinetics.
A first order one-compartment absorption model described PM kinetics. The typical kf from parent CY to PM was 0.189 hr-1, with inter-individual variability of 42%. The typical Cl/fm and V/fm in the central comapartment were 39.1 L/hr and 264 L, respectively. The inter-individual variability in Cl/fm was 22.4% and 33.2% for V/fm. Gender was a significant covariate affecting kf, where females had 59% higher conversion rate than males, thus showing higher metabolite concentrations. Cl/fm and V/fm were allometrically scaled using total body weight. CrCl was also an important covariate affecting Cl/fm where 32% of PM was renally cleared. No other tested covariates were important within the range of our data. Model evaluations performed using visual predictive checks and non-parametric bootstrap determined that the pharmacokinetic model adequately described the observed data.
Gender and CrCl significantly affected PM pharmacokinetics. Females had a significantly higher formation of PM than males. PM appears to be approximately 32% renally cleared and is affected by CrCl. These data may ultimately guide dose reduction decisions in patient with organ dysfunction.
No relevant conflicts of interest to declare.
This icon denotes a clinically relevant abstract
