Significance of Busulfan Dose Intensity On Outcomes of Hematopoietic Cell Allografting for AML in Second Complete Remission or Beyond: A Report From the EBMT Acute Leukemia Working Party

Allogeneic hematopoietic cell transplantation (HCT) is associated with inferior outcomes in patients with more advanced stage AML. In such cases, reducing the ablative intensity of the preparative regimen could potentially compromise its efficacy. We hereby compare transplant outcomes of two preparative regimens, known as FB2 and FB4, in patients (pts) with AML transplanted in second complete remission (CR2) or beyond at EBMT participating centers.

Between 2003 and 2010, 128 (FB2=88 (69%), FB4=40 (31%)) pts with a median age of (FB2=60 (22–70) years, FB4=42 (19–65) years, p<0.0001), with AML in ≥CR2 (FB2 (CR2=85 (97%), CR3=3 (3%)); FB4 (CR2=38 (95%), CR3=2 (5%)), p=0.67, underwent allogeneic HCT. FB2 comprised intravenous (IV) busulfan cumulative dose of 6.4 ± 10% mg/kg, while FB4 cumulative dose was 12.8 ± 10% mg/kg. Cytogenetic risks groups were similar: FB2 (good=17, int=31, poor=6) and FB4 (good=6, int=7, poor=0), p=0.35. For pts treated with FB2, donor source consisted of matched-related donors (MRD)=32, matched unrelated donor (MUD)=39, mismatched unrelated donors (MMUD)=8, unknown=9. For FB4, donor source consisted of MRD=17, MUD=15, MMUD=6, unknown=2. Use of peripheral blood stem cells (PBSC) was higher in the FB2 group (92% vs. 70%, p=0.001). Administration of anti-thymocyte globulin (ATG) was also higher in FB2 conditioned patients (89% vs. 45%, p<0.0001).

Median follow-up time was 24 (3–76) months. Median time to absolute neutrophil count engraftment (days) was 16 (5–38) and 16 (9–29) days, for FB2 and FB4, respectively (p=0.45). The 2-year leukemia-free survival (LFS), cumulative incidence of relapse (CI-R), and of non-relapse mortality (NRM) was: LFS (FB2=47±5% vs. FB4=70±8%, p=0.01), CI-R (FB2=27±5% vs. FB4=19±7%, p=0.24), and NRM (FB2=25±5% vs. FB4=10±5%, p=0.06). The 2-year cumulative incidence of chronic GVHD was similar (FB2=41±6% vs. FB4=43±8%, p=0.8). On multivariable analysis, favorable impact on 2-year LFS was observed with FB4 conditioning (HR 0.66 (95%CI: 0.46–0.97), p=0.03) and with longer interval from diagnosis to transplantation (> median) (HR=0.49 (95%CI: 0.29,0.85), p=0.01). When the analysis was performed by age groups, the statistical advantage of FB4 on LFS was lost, likely because of small numbers in each subgroup. Indeed, for pts <50 yrs. (n=43; FB2=16, FB4=27), LFS was (FB2=50±13% vs. FB4=72±9%, p=0.11), whereas for pts ≥50 yrs. (n=85, FB2=72, FB4=13), LFS was (FB2=47±6% vs. FB4=67±13%, p=0.18).

Although limited by the small number of patients, these results suggest that FB4 is a reasonable preferred conditioning in patients with AML in ≥CR2.

No relevant conflicts of interest to declare.