In an attempt to reduce toxicity in patients not eligible for standard myeloablative allo-SCT, different RIC regimens aiming to induce an allogeneic graft-vs-tumor (GVT) effect, have been investigated and could result in durable donor cell engraftment. However, as already established in the myeloablative setting, specific studies are needed to identify the type of disease that is more likely to benefit from this RIC-related putative GVT effect. Specifically, the role and exact benefit of RIC allo-SCT for AML patients is still under considerable investigations. Our current policy is to propose an HLA-identical RIC allo-SCT to AML patients with high risk leukemic features and with age (>50 years) and/or comorbidities precluding the use of standard myeloablative allo-SCT. We report here a single center retrospective study that aimed to evaluate this treatment strategy, and to determine whether or not all AML patients with high risk leukemic features and high risk clinical features with a suitable HLA-identical sibling should receive RIC allo-SCT. Between 1999 and 2003, 74 AML (85% de novo and 15% secondary) patients were considered as potential candidates for RIC allo-SCT. Median age was 50 (range, 24–65) years. Of the 74 patients, 25 (34%; “transplant” group) had an HLA-identical donor (24 related and 1 unrelated) and could effectively receive the RIC allo-SCT (mainly a fludarabine, busulfan and ATG-based RIC regimen; n=22). The remaining 49 patients (66%; “no transplant” group) were treated according to standard procedures. With a median follow-up of 688 (range, 173–1595) days, leukemia-free survival was significantly higher in the RIC allo-SCT group as compared to the group of patients that did not receive a RIC allo-SCT (P=0.002; 87% vs. 41% respectively at 4 years). 24 patients (49%) relapsed in the “no transplant” group during the follow-up period as compared to only 3 patients (12%) in the “transplant” group (P=0.001). There were no significant differences between the two groups with respect to age, gender, FAB subtype, disease origin (secondary vs. de novo) cytogenetic risk group, history of prior high dose cytarabine or autologous transplantation and number of chemotherapy induction courses. In addition, in the “transplant” group, little procedure-related toxicities were observed, and the cumulative incidence of transplant-related mortality was impressively low (12%; 95%CI, 3–32%) for this group of elderly high risk AML patients. We conclude that a potent graft-vs.-leukemia effect can be induced in AML patients after RIC allo-SCT, with a significant benefit in term of leukemia-free survival. Thus, if a related donor is identified, RIC allo-SCT should be performed since it represents a valid option for AML patients with poor risk leukemic features and not eligible for standard myeloablative allo-SCT.