Comparison of Low Dose Total Body Irradiation (TBI)-Based Reduced Intensity Conditioning (RIC) Vs. Chemotherapy-Based RIC Prior to Allogeneic Stem Cell Transplantation (allo-SCT) From An HLA Identical Sibling Donor for Acute Myeloid Leukemia (AML) in First Complete Remission (CR1): a Retrospective Analysis of 1200 Patients From the Acute Leukemia Working Party of EBMT.

Abstract 1190

Poster Board I-212

RIC is increasingly used in allo-SCT for AML patients. RIC allo-SCT represents an attempt to harness the immune graft-versus-leukemia effect while attempting to control or overcome toxicity. Most of the RIC protocols have been shown to be highly immunosuppressive, but because of the variability in the degree of myeloablation, the toxicity and efficacy profiles might vary from one protocol to another. Low dose TBI in RIC is frequently used, however it is still unknown whether outcomes are different when comparing low dose TBI-based RIC and chemotherapy-based RIC regimens. Moreover, some transplant centres do not have radiation facilities. Therefore, the aim of this study was to compare outcomes (leukemia-free survival: LFS, Non-Relapse Mortality: NRM, and relapse incidence) between patients receiving a low dose TBI-based RIC vs. chemotherapy-based RIC prior to allo-SCT from an HLA identical sibling donor for AML in CR1.

Between 2000 and 2008, 323 patients with AML in CR1 were treated with low dose TBI-based RIC (Fludarabine + TBI) and 877 received a chemotherapy-based RIC (including fludarabine in the majority of cases), and reported to the EBMT Registry. All patients received allogeneic PBSC from HLA-identical sibling donors.

Patients receiving TBI-based RIC were older (median 56 y. vs. 54 y.; p<0.0001), but the median time from diagnosis to allo-SCT was shorter (143 days vs. 163 d.; p<0.0001) in the TBI group. Both groups were comparable in terms of period (median year: 2005) of allo-SCT, donor-recipient gender distribution, and CMV serostatus. In terms of AML-related risk factors, WBC (median: 9600/μL), the cytogenetics risk groups (good: 4.1%; intermediate: 81.5%; and poor: 14.4%), and the FAB subtype (M1-M2: 48.8%; M4-M5: 36.4%; M0-M6-M7: 14.4%) at diagnosis, were also comparable between both groups.

The median time to engraftment (ANC>500/μL) was comparable between both groups (15 vs. 16 days, p=NS). With a median follow-up of 15 (range, 0.3-107) months, 2 years LFS were 50±3%, 53±2% in the TBI-based RIC and chemo based RIC groups respectively (p=0.16). Also, at 2 years, the relapse incidence was not significantly associated with the type of RIC regimen: 37±3%, 34±2% in the TBI-based RIC and chemo-based RIC groups respectively (p=0.17). Finally, NRM was also comparable between both groups [21±3% and 20±2% in the TBI-based RIC and chemo-based RIC groups respectively; p=0.60)

Despite its retrospective nature, results from this large study suggest that RIC allo-SCT is a valid option for AML patients not eligible for standard allo-SCT. The overall outcomes (LFS, NRM and relapse) appear not to be significantly different between AML patients in CR1 receiving a low dose TBI-based RIC allo-SCT versus those receiving a chemotherapy-based RIC allo-SCT using an HLA-identical sibling donor.