Improving Dosing Precision in Buslfan (BU) Based Preparative Regimens for Hematopoietic Stem Cell Transplantation (HSCT).

Precision dosimetry with TBI has facilitated definition of maximal and minimal limits for dose intensity in HSCT preparative regimens assuring more predictable toxicity and therapeutic benefit than high dose chemotherapy preparative regimens, where 3–5 fold variation in drug level achieved with a given dose directly results in more variable outcome in both toxicity and disease control. BU is the most widely used drug in preparative regimens for allogeneic (allo) HSCT and is used in autologous (auto) HSCT as well, despite the fact that at standard doses of the drug given po the range of levels achieved include levels associated with fatal toxicity at the high end and relapse at the low end. It has long been recognized that BU pharmacokinetics (PK) are simple; the drug is active only as the parent compound, it is rapidly distributed in total body water, it’s metabolism has a single metabolic pathway, and there are no active metabolites. Thus its PK are fully described as single compartment, first order elimination. However, PK directed therapy has been difficult because the oral preparation has of highly variable absorption between patients and between doses. Use of the iv formulation has permitted the development of limited sampling strategies and improved intra-patient predictability. We have developed a specific limited sampling strategy which provides excellent fit to model and has been validated against historical data. We have identified significant concomitant medication effects. We have now studied 51 consecutive patients (pt) receiving BU based preparative regimens adjusting dose 2 of the standard 16 dose regimen based upon the AUC for the first dose given as a test dose 48 hr prior to the remaining 15 given in the standard Q6h intervals. The pt ranged in age from 18–69 (med 44) and in weight from 47–166 Kg (med 84). The preparative regimens consisted of Bu plus cyclophosphamide (Cy) and etoposide for 18 auto HSCT, and 2 haplo-identical allo HSCT, Bu plus fludarabine for 15 matched related donor allo HSCT, and BuCy for 16 matched unrelated donor allo HSCT. Using a single laboratory, careful monitoring for outliers due to concomitant medications and other sources of error, and a fixed dose in mg for all pt, we have reduced the AUC range from 34–164% of target level for dose 1 to 71–130% of target level for dose 2. After removal of 7 predictable outliers due to concomitant meds (4 pt), capping of dose adjustment at 130% of 0.8mg/Kg (2 pt), and sampling error (1 pt), we achieved a range of 73–121% for dose 2. Thus, excluding outliers, 82% of pt could be adjusted to within 90% of target dose and 95% of pt could be adjusted to within 80% of target dose for dose 2. No pt, including the outliers, had dose 2 Bu level in either the range associated with high risk of regimen related mortality or associated with increased risk of relapse. These studies demonstrate that in a routine clinical practice environment, with careful control of concomitant medications, pt can be treated with Bu based preparative regimens achieving levels within 10% of target for over 80% of pt and within 20% of target for over 95%. We conclude that dosing precision with iv busulfan and HSCT preparative regimens is feasible in routine clinical practice and no patient needs to be out of the acceptable range. This dosing precision will improve safety and greatly facilitate dose intensity research in Phase 2 and comparative trials to better define maxi-dose, mini-dose, and “right dose” Bu for use in HSCT preparative regimens for various transplant situations.