Matched Pairs Analysis of IV vs. PO Busulfan as a Conditioning Agent Prior to Transplantation.

The Center for International Blood and Marrow Transplant Research (CIBMTR) is an international study group involved in ongoing investigations of allogeneic (allo), syngeneic and autologous (auto) transplantations. They have collected data from over 550 institutions worldwide. With the CIBMTR providing information on oral busulfan (Bu) usage from their clinical database, a matched pairs analysis was conducted comparing these data against clinical data obtained from patients receiving intravenous busulfan (IV Busulfex, IVBu) in four clinical studies and two clinical amendments.

The primary objective of the analysis was to compare two key clinically important variables in patients receiving IVBu or oral Bu, i.e., overall survival to day 100, and the incidence of hepatic veno-occlusive disease (HVOD) and mortality through post-transplant day +28 (HVOD28). All patients received the BuCy2 conditioning regimen. Primary matching criteria included disease, disease stage/status, stem cell source, and performance score at time of transplant; a goal of three oral-Bu matches per each IVBu recipient was sought. A total of 216 patients (161 allo, 55 auto) were identified in the CIBMTR database that matched criteria for 101 of the 138 IVBu patients. No matches could be found for 37 IVBu patients. Of the 101 IVBu patients (70 allo, 31 auto), 47 had three, 21 had two, and 33 had one CIBMTR oral Bu match(es). There were no graft failures among the patients receiving IVBu; six (2.9%) oral Bu patients failed to engraft (p=0.19). Overall incidence of HVOD28 was 4.6% (4/83) with IVBu and 20.3% (38/149) with oral Bu (p<0.001). Among autotransplant recipients, 100-day mortality was 0% for those receiving IVBu and 9.3% for those receiving oral Bu (p=0.16). Among allotransplant recipients, 100-day mortality was 8.7% with IVBu and 22.5% with oral Bu patients (p=0.015). Logistic regression analysis showed that only the mode of Bu administration was a significant factor for the risk of HVOD28, with IVBu associated with a greatly reduced risk (p=0.004) compared to oral Bu. Bayesian analyses provided the same conclusion, and indicated that there was a >99% probability that IVBu was superior to oral Bu with regard to the probability of HVOD28 and 100-day mortality. Logistic regression analyses by treatment group indicated that IVBu was associated with a lower probability of 100-day mortality compared to oral Bu for all patients combined (p=0.005) and for allogeneic transplant recipients only (p=0.021), but not for autotransplant recipients. In conclusion, based on these analyses of controlled case-matched data, there appears to be a beneficial effect of IVBu compared to oral Bu on the outcome of hematopoietic stem cell transplantation (HSCT), with lower early mortality associated with IVBu administration. These findings are consistent with results of other controlled and uncontrolled studies comparing IVBu to oral Bu when either is given as a component of an HSCT regimen.