Abstract
Abstract 526
In patients with acute myeloid leukemia (AML) in first complete remission (CR1), the indications for matched sibling donor (MSD) transplants and unrelated donor (URD) haemopoietic stem cell transplantation (HSCT) are different. We sought to determine the prognostic impact of donor type on the outcomes of AML with adverse risk karyotype in CR1, a high-risk AML population considered as a standard indication for MSD and URD HSCT. We evaluated the outcomes of 584 patients undergoing allogeneic HSCT for AML with adverse risk karyotype in CR1 between 1995 and 2006, reported to the CIBMTR. Adverse risk karyotype was defined according to SWOG/ECOG classification. Cytogenetics abnormalities were further classified as: complex karyotype (3 or more abnormalities), 32%; and Non-complex divided as abnormal chromosome 7, 25%; chromosome 5, 9%; MLL gene rearrangements, 18%; t (6;9), 5%; and others, 10%. 226 patients underwent MSD and 358 URD. URD were classified based on high resolution typing as:” well matched” [n=254 (71%)] with no known disparity at HLA A, B, C, DRB1; and, “partially matched” [n=104 (29%)] with one locus known or likely mismatched. Previous MDS was present in 19% and 14% had therapy-induced (t-AML). Conditioning regimens were myeloablative and reduced intensity in 74% and 26%, respectively. At 3 years treatment-related mortality (TRM) incidence was 28% (95% CI 24-31); relapse 36%(32-40); disease-free survival (DFS) 36%(32-41) and overall survival (OS) 39%(35-44). Multivariate analyses are summarized in the table. “Well matched” URD and MSD yielded similar DFS and OS, while outcomes were significantly inferior for “partially matched” URD. Cytogenetically defined subsets had similar outcomes. Evaluated as a time-dependent covariate, chronic GVHD had a significantly lower risk of relapse (RR 0.68, p=0.046), while acute GVHD had no effect (RR 0.99, p=0.96). “Well matched” URD and MSD lead to similar DFS and OS in AML CR1patients with adverse risk karyotype. The pool of patients who may benefit from graft-vs-leukemia effect generated with allogeneic HSCT may be considerably expanded with “well-matched” URD HSCT. If a suitable MSD is not availabel, “well-matched” URD should be strongly considered where a MSD HSCT would otherwise be undertaken.
| Donor type . | TRM . | Relapse . | DFS . | OS . | ||||
|---|---|---|---|---|---|---|---|---|
| RR (95% CI) . | P-value . | RR (95% CI) . | P-value . | RR (95% CI) . | P-value . | RR (95% CI) . | P-value . | |
| Matched sibling donor (n=212) | 1.00 | 1.00 | 1.00 | 1.00 | ||||
| “Well-matched” URD (n=249) | 1.55 (1.02-2.37) | 0.04 | 1.04 (0.76-1.42) | 0.82 | 1.10 (0.86-1.40) | 0.44 | 1.06 (0.83-1.37) | 0.63 |
| “Partially matched” URD (n=99) | 3.04 (1.94-4.77) | <0.001 | 0.78 (0.48-1.24) | 0.29 | 1.38 (1.02-1.87) | 0.04 | 1.42 (1.04-1.94) | 0.03 |
| Donor type . | TRM . | Relapse . | DFS . | OS . | ||||
|---|---|---|---|---|---|---|---|---|
| RR (95% CI) . | P-value . | RR (95% CI) . | P-value . | RR (95% CI) . | P-value . | RR (95% CI) . | P-value . | |
| Matched sibling donor (n=212) | 1.00 | 1.00 | 1.00 | 1.00 | ||||
| “Well-matched” URD (n=249) | 1.55 (1.02-2.37) | 0.04 | 1.04 (0.76-1.42) | 0.82 | 1.10 (0.86-1.40) | 0.44 | 1.06 (0.83-1.37) | 0.63 |
| “Partially matched” URD (n=99) | 3.04 (1.94-4.77) | <0.001 | 0.78 (0.48-1.24) | 0.29 | 1.38 (1.02-1.87) | 0.04 | 1.42 (1.04-1.94) | 0.03 |
No relevant conflicts of interest to declare.
Author notes
Asterisk with author names denotes non-ASH members.
