Jeffrey S Barrett1, Robin Norris1, Sundararajan Vijayakumar2, John Mondick1, Kalpana Vijayakumar2, Bhuvana Jayaraman1, Erin Cummins1, Mahesh Narayan1, Athena Zuppa1, Jeffrey M. Skolnik1
1Laboratory for Applied PK/PD, Clinical Pharmacology & Therapeutics Division, The Children’s Hospital of Philadelphia; Pediatrics Department, School of Medicine, University of Pennsylvania; 2Intek Partners, Bridgewater, NJ
Objectives: The use of pharmacokinetic (PK) and pharmacodynamic (PD) measurements in the clinical care of patients is often limited to clinical pharmacists and pharmacologists; with the exception of few medications, treating physicians and caregivers are often unfamiliar with the value of such assessment to guide the day-to-day management of their patients. Methotrexate (MTX) is an anti-folate chemotherapeutic agent used in the therapy of several childhood cancers, including acute lymphoblastic leukemia, non-Hodgkin lymphoma, and osteosarcoma. MTX dosing is protocol-driven, however rescue from MTX toxicity with leucovorin can be modified by drug exposure over time. While there is no precise relationship between methotrexate serum levels and antineoplastic efficacy, levels below approximately 0.02 uM/L are seen as necessary for resumption of DNA synthesis. The correlation between serum methotrexate drug concentration and duration of tumor cell exposure in predicting methotrexate toxicity has been demonstrated. While a well-defined MTX therapeutic window has not been established, there are exposure targets which generally inform the caregiver on whether a patient is being effectively managed. Our objectives were to design an interface to the hospital’s electronic medical records system which facilitates the management of MTX therapy, develop a decision support system (DSS) that provides early assessment of high dose MTX renal toxicity and recommendation for leucovorin rescue, verify the outcomes of the DSS against historical controls and current best practices, and design a testing strategy for this system’s ultimate implementation.
Methods: Patient data used to develop the MTX population pharmacokinetic model and prototype dashboard application were obtained from source medical records from the Chartmaxx and Sunrise Clinical Manager (SCM) systems. Dosing histories and the visit-based demographics (e.g., weight, height, etc) were ultimately hand entered in Excel. Patient records including methotrexate TDM concentrations, laboratory values and medical record number were abstracted from Oracle tables in SCM. From these two sources, the joined data was generated in NONMEM and SAS dataset format and ultimately loaded into the Oracle database supporting the MTX dashboard using SQL loader. MTX disposition is described by a two-compartment model with first-order elimination. Although MTX clearance changes over time in patients with renal dysfunction, clearance is approximated with a simple model defined by two different clearance distributions for the two populations. The Bayesian forecasting model utilizes the NONMEM PRIOR subroutine to incorporate population priors into the model. Fixed effects parameters obtained from the final pop PK model were implemented for the initial Bayesian model. Prior distributions of the fixed effects parameters were obtained from the variance-covariance matrix from the final pop PK model as well. Prior distributions for random effects parameters were specified as an inverse Wishart distribution. Clearance was implemented as a mixture model, where a patient is assigned to a population (normal or impaired clearance) based on the probability of that patient belonging to either population given their MTX plasma concentrations. The Bayesian forecasting model was evaluated using MTX plasma concentrations that were not used during model construction. The model reliably predicts future MTX plasma concentrations from two prior concentrations in all patients except a small number who develop renal toxicity at delayed times (> 48 hours). In these patients, the addition of a third concentration after 48 hours increases the precision of the prediction of concentrations at later times.
The MTX dashboard was developed based on a three-tier architecture comprising a back end database tier, a business logic middle tier and a data presentation/user interface tier at the front. The database tier consists of patient records from our electronic medical records system (SCM) merged with data from patient registration system (IDX), lab data management system (Clarity) and adverse event management system. Data fields are processed for gaps and can be manually entered (from patient charts) when missing data is critical for functionality. Views and summary tables are created from the relational tables for quick retrieval by the application. Predictions are conducted in an external computational platform – the modeling and simulation (M&S) workbench which can execute code in a variety of languages provided they can run in a batch mode (e.g., NONMEM, SAS, SPLUS and R). All analytics are gated in the middle tier through logic to ensure that minimally required data sets are available for each patient or sets of patients. The user interface is web-based and utilizes a combination of HTML, JavaScript and XML content.
Validation of the MTX dashboard contains three distinct components: (1) qualification of the population pharmacokinetic (PPK) model and the forecasting algorithm derived from this model, (2) assessment of the clinical performance of the decisions and decision logic derived from the forecasting routine and interface and (3) the system validation of the dashboard integration with the existing electronic medical records system.
Results: The MTX pop-PK model has been validated and is generalizable across a broad range of paediatric patients (age, size, renal function, etc). The clinical validation of the forecasting tool confirms that predictions of MTX exposure and guidance for leucovorin rescue. Dashboard views can toggle between the most recent MTX dose event with the complementary monitored MTX plasma concentrations and safety markers, the MTX exposure projected after the dosing guidance menu button is selected, a view of the individual patient projection overlaid against a nomogram used to assess the potential for MTX toxicity with consideration for drug rescue with leucovorin and an update of the model fit when the additional blood collection time points were added to the patient data set. Histoical pharmacotherapy summarizations within and across patients are also available. A working demo of the application will be shown at the meeting.
Conclusion: We have created a prototype web-based tool that utilizes MTX PK/PD defined in paediatric patients with cancer to forecast individual patient response to MTX therapy and provide guidance with respect to rescue therapy. In addition, this application provides real-time views of complementary data related to the clinical care of these patients that is also essential for the management of MTX therapy (e.g., urine pH, hydration, serum creatinine, etc). Future development of this tool will provide prediction of increased risk of MTX toxicity and drug interaction potential. Integration of the production application for the clinical management of patients at the Children’s Hospital of Philadelphia is expected within the year; additional international test sights are being sought to provide additional feedback on the system.
References:
[1] Barrett JS, Mondick JT, Narayan M, Vijayakumar K, Vijayakumar S. Integration of Modeling and Simulation into Hospital-based Decision Support Systems Guiding Pediatric Pharmacotherapy. BMC Medical Informatics and Decision Making 8:6, 2008.
[2] Barrett JS, Vijayakumar K, Krishnaswami S, Gupta M, Mondick J, Jayaraman B, Muralidharan A, Santhanam S, Vijayakumar S. iClinical: NONMEM Workbench. PAGE 15, Belgium, 2006, PAGE 15 (2006) Abstr 1016 [www.page-meeting.org/?abstract=1016]
[3] Skolnik JM, Vijayakumar S, Vijayakumar K, Narayan M, Patel D, Mondick J, Paccaly D, Adamson PC, and Barrett JS. The creation of a clinically useful prediction tool for methotrexate toxicity using real-time pharmacokinetic and pharmacodynamic modeling in children with cancer. J. Clin. Pharmacol 46: 1093 (Abstr. 135), 2006
Reference: PAGE 17 () Abstr 1264 [www.page-meeting.org/?abstract=1264]
Poster: Oral Presentation: Applications