Evaluating the Extent of Chemotherapeutic Contamination from Central Venous Catheters in Children with Cancer and Providing Guidance for Accurate Reporting of PK Parameters
A.Y. Zhang (1), J.M. Skolnik (1, 2), J.S. Barrett (1)
(1) Laboratory for Applied PK/PD, Division of Clinical Pharmacology and Therapeutics; (2) Division of Oncology, The Children’s Hospital of Philadelphia, Philadelphia, Pennsylvania, United States
Objectives: To support the use of a clearing procedure that minimizes catheter contamination when a single central venous line (CVL) is used to administer and sample chemotherapy in children, modeling and simulation (M&S) strategies have been employed to estimate the magnitude of residual bias, develop an algorithm to ensure the accurate reporting of PK results and evaluate the efficiency of the process for an ongoing BPCA trial with Actinomycin-D (AMD) and vincristine (VCR) in pediatric cancer patients.
Methods: An in vivo evaluation of the proposed catheter clearance procedure (four blood-draw return cycles) was performed in 3 pediatric cancer patients receiving AMD and VCR. Paired PK samples were obtained from CVL and peripheral IV (PIV) from 5 min to 24 hr post infusion. Using a 3 CPM structural model for AMD and combining our dataset with the published studies [1], three approaches were evaluated to assess the effects of catheter contamination on drug PK: (1) catheter type as a covariate on PK parameters; (2) CVL contamination as a fixed baseline effect on drug concentration response; (3) catheter binding compartment and blood-draw return cycle-dependent rate constant as components of drug input function. All M&S were performed using NONMEM VI. Comparison of model performance was conducted by objective function and goodness of fit diagnostics, predictive checks and via simulation with the clinical trial design serving as the primary metric of evaluation for the proposed correction algorithm.
Results: AMD dataset combined three pediatric studies containing 36 patients and 199 plasma concentrations. Covariate analysis yielded the best model when CVL sampling catheter was applied as a power function on central volume (0.84) and clearance (0.37). Sensitivity analyses based on time and concentration indicated that a baseline drug contamination factor in an exponential function (THETA=16.4 ng/mL, ETA=0.28) should be applied to individual prediction when CVL plasma concentration ≥ 25 ng/mL. The mechanism-based working model for catheter drug binding leveraged parameter estimates from the in vitro study, including initial percentage of drug adsorption, drug dissociation constants in the presence or absence of blood-draw return cycles. Further model refinement is ongoing.
Conclusions: Catheter clearance procedures can efficiently reduce AMD and VCR contamination during PK sampling from a single central catheter. M&S approaches support their use in prospective pediatric trials where PIV sampling is often a deterrent to enrollment.
References:
[1] Mondick JT, Gibiansky L, Gastonguay MR, et al: Population Pharmacokinetic Investigation of Actinomycin-D in Children and Young Adults. J Clin Pharmacol 48(1):35-42, 2008
