Relationship between intravenous (IV) busulfan pharmacokinetics (PK) and patient characteristics in allogeneic stem cell transplant patients.

Abstract

e13507 Background: Busulfan is an alkylating agent commonly used for myeloablation in hematopoietic stem cell transplantation (SCT). Narrow therapeutic index and 2-3 fold interpatient PK variability underscores the need to further explore predictors of busulfan clearance. This analysis aimed to determine patient (pt) characteristics that may impact the clearance of busulfan. Methods: This retrospective analysis assessed allogeneic SCT pts who received a test dose of 0.8 mg/kg IV busulfan, dosed by adjusted body weight, over 2 hours on days -15 to -10 prior to conditioning therapy containing therapeutically dosed busulfan. All pts were treated between June 2004 and July 2012 at the University of North Carolina at Chapel Hill. Noncompartmental PK parameters were calculated from this test dose using Phoenix WinNonlin Pharsight software. A linear model with log of busulfan test dose clearance as an outcome and pt characteristics that included disease state, bone marrow donor type, gender, race, ethnicity, body surface area (BSA), body mass index, type of anticonvulsant prophylaxis used (levetiracetam or phenytoin), total bilirubin and serum creatinine was fit. Results: The study reviewed a total of 88 pts with a mean age of 41.1 (range 18-63). The analysis included 54 males, 76 Caucasians, and 5 African Americans (AA). Seizure prophylaxis with levetiracetam was given to 27 pts, and 51 received phenytoin. After adjusting for pt characteristics, a unit increase in BSA was associated with 0.69 (p = < 0.001) increase in log(clearance). Regression coefficients for phenytoin use with no seizure prophylaxis as reference and for African American race with Caucasian race as reference were significant (0.22, p = 0.033, and -0.31, p=0.13, respectively). No correlation was seen for age (p = 0.357), gender (p = 0.337), levetiracetam use (p = 0.3525), or for any other variable. Conclusions: Our data suggests that higher BSA and phenytoin use are associated with a higher busulfan clearance, while AAs have a lower clearance. These findings may help inform future dosing strategies in order to decrease interpatient PK variability and accurately predict plasma concentrations of busulfan.