Outcome Following Hematopoietic Cell Transplantation for Patients with AML-CR1: Comparison between Matched-Sibling and Unrelated Allografts.

Although allogeneic hematopoietic cell transplantation (HCT) often offers the best, and sometimes only chance for cure for AML patients, the procedure too often fails to eradicate the patient’s malignancy or is associated with fatal toxicities. Graft-vs-leukemia effects are felt to be associated with improved survival in the HCT setting, and recipients of unrelated allografts generally experience higher rates of graft-vs-host disease (GvHD) as compared to patients transplanted from a matched-sibling donor. With these observations in mind, we have evaluated outcome among patients transplanted for acute myeloid leukemia (AML) in first complete remission (CR1), assessing the differences between recipients of a matched unrelated donor allograft (MUD) who were matched for 10 of 10 alleles at HLA-A, -B, -C, -DRB1, and -DQB1 and allografts from an HLA-identical sibling donor (MRD). Between 1992 and 2004, 183 patients greater than 18 years of age with AML-CR1 and a donor as described above (n = 47 MUD; n = 136 MRD) received an ablative HCT at our Center. Median age was 41 (range, 19–67) years. Conditioning included use of total body irradiation (TBI) (38% MUD; 29% MRD), busulfan and cyclophosphamide (45% MUD; 39% MRD), an anti-CD45 radiolabeled antibody (9% MUD; 29% MRD), or other regimens (11% MUD; 4% MRD). Most received cyclosporine and methotrexate for GvHD prophylaxis (81% in MUD; 92% in MRD). Source of stem cells was bone marrow in 30% MUD, 63% MRD. Patients were further classified by cytogenetic risk as favorable (2% MUD; 8% MRD), intermediate (81% MUD; 70% MRD), or unfavorable (17% MUD; 22% MRD). Multivariable Cox regression models were fit for the endpoints mortality, relapse, and transplant-related mortality (TRM) and logistic regression was used for acute GvHD. Eighty percent of MUD patients developed grades 2–4 acute GVHD compared to 68% of MRD patients (adjusted odds ratio (OR) = 2.9; 95% CI 1.2–7.1; p = 0.02). Surprisingly however, the probability of grades 3–4 was not higher (11% in MUD group, 20% in MRD group; adjusted OR = 0.6; 95% CI 0.2–1.7; p = 0.31). Nineteen percent of MUD patients relapsed compared to 22% of MRD patients, leading to a 5-year estimated probability of relapse of 17% and 22%, respectively (adjusted hazard ratio (HR) = 0.9; 95% CI 0.4–2.0; p = 0.82). The estimated 5-year overall survival was 56% in the MUD group and 64% in the MRD group (adjusted mortality HR = 1.4; 95% CI 0.8–2.5; p = 0.21). Twenty-one percent of patients in the MUD group experienced transplant-related mortality (TRM) compared to 18% in the MRD group (HR = 2.0; 95% CI 0.9–4.2; p = 0.09). While the number of patients with unfavorable cytogenetics was relatively low (n = 38), there is some suggestion that the donor effect on TRM exists primarily among patients with intermediate-risk cytogenetics and not so in the group with unfavorable cytogenetics (HR = 2.9 in intermediate-risk and HR = 0.9 in unfavorable group). While larger numbers of patients are required to confirm these findings, these data suggest that in AML-CR1 patients, MUD HCT increases the risk of grades 2–4 acute GvHD relative to MRD. The impact of donor source on relapse is minimal, but the risk of mortality, mainly transplant-related, may be increased, particularly among patients with intermediate-risk cytogenetics.