Abstract
Stringent complete remission (CR) in acute myeloid leukemia (AML) requires the absence of both morphologic and flow cytometric evidence of disease. We have previously shown that persistent AML detected by flow cytometry (FC+) before reduced-intensity conditioning (RIC) allogeneic hematopoietic cell transplantation (alloHCT) was associated with significantly increased relapse, shorter disease-free survival (DFS), and poorer overall survival (OS), independent of morphologic blast count. We evaluated the effect of FC status on outcomes of alloHCT for AML after either myeloablative conditioning (MAC) or RIC regimens in 203 patients (MAC, n = 80, and RIC, n = 123) with no morphologic evidence of persistent AML pretransplant on marrow biopsy. The allografts included 130 umbilical cord blood (UCB) and 73 sibling donors. We performed central review of pretransplant standard sensitivity FC to identify detectable FC+. Twenty-five patients were FC+, including 15 (18.7%) receiving MAC and 10 (8.1%) RIC alloHCT. Among RIC patients FC+ was associated with significantly inferior relapse, DFS, and OS (multiple regression HR, 3.8; 95% CI, 1.7 to 8.7; P < .01 for relapse; HR, 2.9; 95% CI, 1.4 to 5.9; P < .01 for DFS; and HR, 3.4; 95% CI, 1.7 to 7; P < .01 for OS). In contrast, FC+ status was not associated with relapse or decreased OS after MAC. These data suggest that MAC, but not RIC, overcomes the negative effect of pretransplant FC+ after sibling or UCB alloHCT. Therefore, a deeper pretransplant leukemia-free state is preferred for those treated with RIC.