OBJECTIVES: Human leukocyte antigen (HLA)-matched donors are often unavailable for AML patients undergoing allo-HSCT, leading to the use of mismatched grafts derived from peripheral blood (PB), bone marrow (BM) or umbilical cord blood (UCB). Limited published data exists on comparative effectiveness of allo-HSCT across graft sources. We reviewed published clinical evidence to compare outcomes of mismatched (<8/8 HLA) allo-HSCT in AML patients.

METHODS: Embase®, MEDLINE®, Cochrane databases and conference abstracts (ASCO, ASH) were systematically reviewed to identify publications of prospective clinical studies published between Jan-2008 to Nov-2018 in AML patients who received mismatched allo-HSCT. Records were reviewed by two independent analysts and inconsistencies were resolved through roundtable discussion with a third analyst. Survival curves were digitized to obtain additional numerical data, where available.

RESULTS: A total of 1,335 records were reviewed and 10 satisfied inclusion criteria (5 allo-HSCT, 1 unmanipulated-UCB, 4 manipulated-UCB). Baseline characteristics were generally comparable between studies. Patients receiving unmanipulated-UCB took longer to achieve neutrophil engraftment (defined as ≥0.5x10⁹L): 36 versus 18 days for UM171-expanded-UCB and ranged from 12-19 days with PB-allo-HSCT. Overall survival (OS) at 12-months was 63.7% with unmanipulated-UCB, compared with 95.0% in UM171-expanded-UCB, and ranged from 41.0-49.8% with other mismatched-allo-HSCTs. OS at 24-months was 83.2%, 50.8%, and 57.0% with UM171-expanded, NiCord®-expanded, and unmanipulated-UCB, respectively, and ranged from 28.0-39.1% with other mismatched-allo-HSCTs. Progression-free survival at 12-months was 76.8% for UM171-expanded-UCB, 55.0% with unmanipulated-UCB, and 34.0% with PB-allo-HSCT. Relapse at 24-months was 18.0% with UM171-expanded-UCB, 27.0% with NiCord®-expanded-UCB and ranged from 31.0-39.0% with PB-allo-HSCT.

CONCLUSIONS: Our review demonstrates that for AML patients without matched stem cells, expanded UCB may offer improved outcomes relative to other sources. Additionally, there is a significant unmet need demonstrated by the limited survival where approximately half of AML patients treated with mismatched-allo-HSCT die within a year.