Allogeneic Stem Cell Transplantation (allo-SCT) from a HLA Identical Sibling for CR1 AML Patients Aged above 55 Years Old after a Reduced Intensity Conditioning: A Survey from the Société Française de Greffe de Moelle et de Thérapie Cellulaire (SFGM-TC).

Background: AML incidence increases with age. Allogeneic stem cell transplantation (Allo-SCT) offers the best leukemic control but is associated with high non-relapse mortality (NRM). Recent advances using non-myeloablative strategies have established the feasibility of allo-SCT in elderly patients. However in the context of AML, an important reduction of the intensity of the preparative regimen may be also associated with a loss of leukemic control, offsetting the impact of toxicity reduction in elderly.

Methods: We retrospectively analyzed AML patients in first complete remission (CR1) reported to the SFGM-TC, aged above 55 and transplanted from a HLA identical sibling donor prior to 01/01/05.

Results: 62 patients prepared with a non-myeloablative conditioning (NMAC) (fludarabine (75mg/m²) + TBI (2Gy)) (N=14) or a reduced intensity conditioning (RIC) (busulfan (4 to 8 mg/m²) + fludarabine (150 to 180 mg/m²) + thymoglobulin (2.5 to 12.5 mg/m²)) (N=48). GVHD prophylaxis used CSA+MMF for NMAC, CSA +/− MMF for RIC. With few exceptions, all patients in a given centre were treated identically. Major characteristics were: age 58 (55–67), M/F= 27/35. Time between diagnosis and allo-SCT: 171 days (101–467). BMT/PBSCT: 7/55. Pts have been considered for allo-SCT because of poor prognosis factors: no favorable cytogenetics (100%), secondary AML (15%), 2 chemotherapy course to achieve CR1 (24%) or M0, M6 or M7 FAB (20%). All pts engrafted. Acute GVHD occurred in 16 pts (grade 1: 6; grade 2: 6; grade 3–4: 4) and chronic GVHD in 18 pts (limited: 8; extensive: 10) with no difference in the 2 groups. The cumulative incidences of grade 2–4 aGVHD and cGVHD were 16% (95%CI: 7–25) and 29% (95%CI: 18–40) respectively. 15 pts relapsed and 8 died from NRM. Relapse and NRM cumulative incidences were 24% (95%CI: 13–35) and 13% (95%CI: 5–21) respectively. 49 pts were evaluable for cGVHD: of the 18 expressing cGVHD none have relapsed as compared to 12 of the 31 who did not present cGVHD (39%, 95%CI: 22–56) (p=.007). The 2 year KM survival and DFS probabilities are 63% (47–76) and 56% (42–69). In a landmark analysis investigating patients alive on day 100, 2 year DFS are 94% and 38% respectively with or without cGVHD (p=0.001) and 2 year KM OS are respectively 94% and 45% (p=0.01).

Conclusions: We conclude that such a strategy in elderly can afford a high relapse control with low NRM conducting to high survival probability at 2 years. Results are achieved through the allogeneic effect as expressed by cGVHD inviting to a careful immunomodulation in the early post transplant period to further improve results.