Background: Quantitative assays are needed to characterize the multili
neage engraftment potential of clinical hematopoietic grafts, After we
observed a dose-response relationship between the number of human hem
atopoietic cells transplanted into nonobese diabetic-scid/scid (NOD/SC
ID) mice and the number of human CD45+ cells recovered in the chimeras
' marrows and spleens, we sought to develop a multiple linear regressi
on model that allows quantitative comparisons of human cell engraftmen
t in vivo. Methods: We used this NOD/SCID xenotransplant model to comp
are the engraftment potential of cord blood vs. adult marrow or mobili
zed blood, after either of 2 commonly used clinical graft engineering
procedures: CD34+ cell selection or T cell depletion. Results: The eng
raftment per transplanted cell was >20 fold higher for cord blood cell
s, as compared to hematopoietic cells from adults. However, there was
no difference in engraftment per CD34+ cell transplanted between marro
w and mobilized blood. Levels of human cell engraftment from all sourc
es could be increased by administration of human hematopoietic growth
factors to human/mouse chimeras after transplantation. Correlation ana
lysis of the number of human CD13+ myeloid cells and CD19+ B lymphoid
cells in the chimeras' marrows 8 weeks after transplantation provided
evidence that almost all the human myeloid and B lymphoid cells were d
erived from the same primitive precursor cells, Conclusions: These fin
dings and assay may be useful in the development of clinical hematopoi
etic cell therapies.