AML engraftment in the NOD/SCID assay reflects the outcome of AML: implications for our understanding of the heterogeneity of AML.

The nonobese diabetic/severe combined immunodeficient (NOD/SCID) assay is the current model for assessment of human normal and leukemic stem cells. We explored why 51% of 59 acute myeloid leukemia (AML) patients were unable to initiate leukemia in NOD/SCID mice. Increasing the cell dose, using more permissive recipients, and alternative tissue sources did not cause AML engraftment in most previously nonengrafting AML samples. Homing of AML cells to the marrow was the same between engrafters and nonengrafters. FLT3 internal tandem duplication (ITD) and nucleophosmin mutations occurred at a similar frequency in engrafters and nonengrafters. The only variable that was related to engraftment ability was the karyotypically defined risk stratification of individual AML cases. Of interest, follow-up of younger patients with intermediate-risk AML revealed a significant difference in overall survival between NOD/SCID engrafting and nonengrafting AMLs. Hence, the ability of AML to engraft in the NOD/SCID assay seems to be an inherent property of AML cells, independent of homing, conditioning, or cell frequency/source, which is directly related to prognosis. Our results suggest an important difference between leukemic initiating cells between engrafting and nonengrafting AML cases that correlates with treatment response.