Identification of Novel Surface Markers and Targets In Neoplastic Stem Cells In AML and CML: a Flow-Gene-Flow Screen Approach.

Abstract 1575

The concept of leukemic stem cells (LSC) is increasingly employed to explain the biology of various myeloid neoplasms and to screen for pivotal targets, with the hope to improve drug therapy through elimination of disease-initiating cells. Although the stem cell hypothesis may apply to all neoplasms, leukemia-initiating cells have so far only been characterized in some detail in myeloid leukemias. In an attempt to identify novel cell surface markers and targets on leukemic stem cells (LSC) in acute (AML) and chronic myeloid leukemia (CML), we examined CD34+/CD38- and CD34+/CD38+ populations of leukemic cells in a cohort of patients with AML (n=55) and CML (n=20). In a first step, cell surface antigen profiles were determined by multicolor flow cytometry. In this screen, we were able to show that CD34+/CD38- LSC in AML and CML consistently express certain cytokine receptors, including G-CSFR (CD114), SCFR/KIT (CD117), and IL-3RA (CD123). The low affinity IL-2R (CD25) was detectable on CD34+/CD38- stem cells in patients with CML, and in a subset of AML patients. Other cytokine receptors (R) such as FLT3, IGF-1R, endoglin (CD105), GM-CSFRA (CD116), and OSMR were expressed variably on CD34+/CD38- progenitor cells, whereas the EPOR was not detectable on LSC. We were also able to detect several established therapeutic targets on LSC, including CD33 and CD44. Whereas CD44 was consistently expressed on all LSC in all donors, CD33 was found to be expressed variably on subpopulations of LSC in AML and CML, depending on the phase and type of disease. By using cytokine ligands (G-CSF, IL-3, SCF, EPO) and targeted drugs, we were also able to confirm that identified cytokine receptors and targets were functionally active molecules and potentially relevant targets. In a next step, highly enriched (purity >98%) sorted CD34+/CD38- cells, CD34+/CD38+ cells, and CD34- cells were collected in patients with AML and CML, and in 3 cord blood samples as controls. Purified cells were subjected to gene chip analyses, qPCR, and functional analyses. The identity of leukemic progenitors was confirmed by FISH, and expression of markers and targets in CML stem cells and AML stem cells was confirmed by qPCR. In gene chip analyses, we screened for novel LSC markers and targets. Candidate genes were selected based and their specific expression in progenitor cell fractions and surface membrane location, which was confirmed by antibody staining experiments. Novel stem cell markers identified so far include ROBO4, NPDC-1, and CXCR7. The previously described surface markers MDR-1 and CLL-1 were also identified by flow cytometry, but were also found to be expressed on more mature hematopoietic cells. By contrast, ROBO4 was found to be expressed preferentially on CD34+/CD38- stem cells, but less abundantly on CD34+/CD38+ progenitor cells in CML. Interestingly, whereas ROBO4 was expressed on CD34+/CD38- stem cells in most patients with CML, ROBO4 expression on leukemic stem cells was only found in a subset of AML patients. By contrast, CD34+/CD38- stem cells in AML frequently expressed CLL-1 and NPDC-1 on their surface. In conclusion, we have identified novel markers and targets in CD34+/CD38- progenitor cells in AML and CML. These markers may be useful for the identification and isolation of leukemic stem cells in AML and CML, and for the validation of drug effects on these cells.