Dosing algorithm revisit for busulfan following IV infusion

• Published in: Cancer chemotherapy and pharmacology Vol. 75; no. 3; pp. 505 - 512
• Main Authors: Wang, Yanlin, Kato, Kazunobu, Le Gallo, Christopher, Armstrong, Elizabeth, Rock, Edwin, Wang, Xiaofeng
• Format: Journal Article
• Language: English
• Published: Berlin/Heidelberg Springer Berlin Heidelberg 01.03.2015; Springer Nature B.V
• Subjects: Adult; Aged; Algorithms; Antineoplastic Agents, Alkylating - administration & dosage; Antineoplastic Agents, Alkylating - pharmacokinetics; Busulfan - administration & dosage; Busulfan - pharmacokinetics; Cancer Research; Clinical Trials, Phase II as Topic; Computer Simulation; Dose-Response Relationship, Drug; Female; Hematopoietic Stem Cell Transplantation - methods; Humans; Infusions, Intravenous; Male; Medicine; Medicine & Public Health; Middle Aged; Models, Biological; Oncology; Original Article; Pharmacology/Toxicology; Young Adult; Dosing algorithm; Test PK study design; Busulfan exposure; IV infusion; Population PK modeling
• ISSN: 0344-5704; 1432-0843; 1432-0843
• DOI: 10.1007/s00280-014-2660-0

Purpose Busulfan (Bu) exposure is critical for efficacy and safety. Body weight (BW), or adjusted ideal body weight (AIBW)-based dosing (WBD) algorithm, has been used in hematopoietic stem cell transplantation (HSCT). A recently completed phase 2 study revealed that 33.6 % of the subjects were under-, or over-exposed, with this WBD algorithm. This paper was to investigate Bu dosing algorithm in an attempt to improve the suboptimal Bu exposure. Methods Population PK modeling was conducted using the data from 207 patients. Dosing algorithm was developed based on derived covariate model of CL. Model-based simulation was conducted to assist test PK study design. A simplified CL estimation method was proposed based on the PK structure model for Bu. Results A one-compartment structure model adequately described the PK profile of Bu following an IV infusion. BSA best described the inter-individual variability of CL. The proposed dosing algorithm was dose (mg) = (31.7 × BSA − 11.6) × target AUC [µM min]/1,000. With this dosing algorithm, 14.3 % patients could be under- or over-exposed. A test PK study with reduced study duration and three PK samples can provide as nearly as good an estimate of CL compared to 12 PK samples on two different occasions. Conclusion The proposed dosing algorithm can significantly improve the sub-exposure of Bu. A shortened test PK study duration with reduced PK samples can provide as near as good estimate for Bu CL. A simplified CL estimation method is valid.