BuCy is considered as a standard conditioning regimen for allo-HSCT in AML patients. The introduction of i.v. busulfan decreases the intra- and inter-individual variability in systemic busulfan exposure, improving the safety of BuCy regimen. However, early RRT was still substantial, especially HVOD. Busulfan and cyclophosphamide are mainly metabolized in liver. Both toxic cyclophosphamide metabolites and busulfan can decrease the levels of glutathione (GSH) [4, 22]. Combination of these two alkylating agents may result in an exacerbated risk for serious hepatic injuries . Recently, to limit RRT and TRM, non-myeloablative and reduced-intensity conditioning regimens are increasingly used, however, relapse becomes more prevalent. Therefore, the challenge is to develop myeloablative regimens associated with low TRM and sufficient anti-leukemic effect.
Fludarabine is widely used in chemotherapy of acute leukemia (especially refractory leukemia) and in conditioning regimens for allo-HSCT. Fludarabine has a synergistic interaction with busulfan through inhibition of DNA ligase and DNA primase, and prevention of DNA polymerization, impairing alkylator-induced damage repair. In addition, fludarabine does not depend on hepatic GSH-stores for its detoxification. Thus, there are non-overlapping organ toxicities between fludarabine and busulfan. A growing body of evidence suggests that fludarabine plus busulfan as myeloablative or non-myeloablative conditioning regimen results in a favorable balance between anti-malignancy efficacy and reduced toxicities [5–12]. In this study, we observed that the incidence of RRT and regimen-related death were significantly higher in BuCy regimen compared with BuFlu regimen in patients with AML-CR1 undergoing allo-HSCT. BuCy regimen showed higher incidence of bladder, mucosa and gut adverse events compared with BuFlu regimen. These results were consistent with previous studies [8–10]. Surprisingly, we observed that 2 patients died from heart toxicities and autopsy of one case showed limited myocardial fibrosis, suggesting that BuCy might be more toxic to the heart than BuFlu, especially to those with a history of heart disease. Although total hepatic toxicities were similar between the two regimens, one patient died of HVOD in BuCy regimen in our study. No differences in TRM between the two regimens might be attributed to the young age and lack of comorbidities of patients in our cohort. We also observed that the median duration of neutrophil count below 0.1 × 109/L and platelet count below 20 × 109/L in BuCy regimen were significantly longer than that in BuFlu regimen. Meanwhile, BuCy regimen required more red blood cells and platelet concentrates transfusions compared with BuFlu regimen. From these data, it can be suggested that BuFlu regimen might possess less RRT and bone marrow toxicities as well as safer than BuCy regimen. In addition, Russell et al. reported that single daily intravenous busulfan would be more convenient and could be achieved with acceptable toxicity compared with traditional 4-times-daily dosing. In this report, busulfan was administered every 12 hours and associated with higher peak blood concentration after the first day and similar toxicity compared with 4-times-daily dosing (data not shown). Whether this 2-times-daily dosing is associated with increased anti-leukemic activity remains to be discussed.
Other than relapse of malignancy, GVHD and infections remain as the two main causes of death after allo-HSCT. A shorter duration of neutropenia and faster hematopoietic recovery may reduce risk of infection after transplants [23, 24]. In the present study, the bone marrow suppression toxicities were lower in BuFlu regimen than BuCy regimen. However, the incidence of early infectious post-transplantation (including bacterial and fungal infections) did not differ between the two regimens, probably as a consequence of almost all patients being treated in a sterile ward for 100 days after transplantation and the short duration of neutropenia. Some studies suggested that the conditioning regimens containing fludarabine were associated with higher incidence of opportunistic infections such as CMV and EBV after transplantation because of the immunosuppressive effect of fludarabine [10, 25, 26]. Here, we observed that the incidence of CMV and EBV viremia was similar between BuFlu and BuCy regimens within 6 months after transplantation. During the follow-up period, there were also no differences in the incidence of CMV and EBV-associated diseases between the two regimens.
GVHD is a complex pathological process mediated by allo-reactive donor T cells recognizing the disparate HLA antigens and involving tissue-specific immune cells and inflammatory cytokines . The incidence and severity of GVHD can be influenced by many factors including age of recipient and donor, HLA typing, source of donor and stem cells, and conditioning regimens. The conditioning can provoke the release of inflammatory factors, which play critical roles in aGVHD . Chae et al. reported that the incidence and severity of acute and chronic GVHD were lower in BuFlu regimen compared with BuCy regimen . However, in a randomized trial, Lee et al. suggested that the incidence and severity of acute and chronic GVHD were similar between the two regimens . In this report, the incidence of I-II° and III-IV° aGVHD as well as total and extensive cGVHD were also similar between BuFlu and BuCy regimens.
Some single arm and retrospective comparison studies suggested that myeloablative BuFlu regimen did not increase the risk of disease relapse [9, 10]. However, Shimoni et al. reported a non-statistically significant trend for higher incidence of relapse after BuFlu myeloablative regimen . Lee et al. reported that BuFlu regimen had lower relapse-free survival than BuCy regimen . However, these two studies enrolled patients with different diseases and disease status before transplantation. In this study, we found no significant difference in relapse rate between BuCy and BuFlu regimens for AML-CR1 patients. Based on these results, we concluded that BuFlu and BuCy had equivalent anti-leukemic activity to AML-CR1 patients undergoing allo-HSCT.
The survival of patients with AML after allo-HSCT is influenced by many factors, such as the response to induction therapy, white blood cell count at diagnosis, cytogenetics, the status of disease during transplantation, and conditioning regimens [29, 30]. Several single arm and retrospective comparisons showed that myeloablative regimen based on busulfan and fludarabine might be associated with fewer RRT, lower TRM, and higher survival rates compared with BuCy [8–10]. However, a recent randomized prospective trial from Lee et al. reported that BuCy regimen had better 2-year OS and DFS than BuFlu in patients with leukemia and MDS . In the present randomized prospective trial, the 5-year cumulative OS and DFS were 81.9 ± 7.0% and 75.3 ± 7.2% in AML-CR1 patients receiving BuFlu, and were not different from patients receiving BuCy (72.3 ± 7.5% and 67.4 ± 7.6%). We used similar conditioning regimen as Lee et al’s study except the schedule of busulfan. The different results might be attributed to the different disease composition and disease status in these two studies.