c-Met Is a Marker of Pancreatic Cancer Stem Cells and Therapeutic Target

BACKGROUND & AIMS: Growth of many different tumor types requires a population of self-renewing cancer stem cells (CSCs). c-Met is a marker of normal mouse pancreatic stem and progenitor cells; we investigated whether it is also a marker of human pancreatic CSCs that might be developed as a therapeutic target. METHODS: We studied growth of primary human pancreatic adenocarcinoma in NOD SCID mice. The self-renewal capability of pancreatic cancer cells that expressed high levels of c-Met (c-Met(high)) was assessed using in vitro sphere assays and compared with those that were c-Met negative or expressed low levels of c-Met. The tumorigenicity of c-Met(high) pancreatic cancer cells was evaluated in NOD SCID mice. RESULTS: c-Met(high) cells readily formed spheres, whereas c-Met-negative cells did not. Use of the c-Met inhibitor XL184 or c-Met knockdown with small hairpin RNAs significantly inhibited tumor sphere formation. c-Met(high) cells had increased tumorigenic potential in mice; those that expressed c-Met and CD44 (0.5%-5% of the pancreatic cancer cells) had the capability for self-renewal and the highest tumorigenic potential of all cell populations studied. In pancreatic tumors established in NOD SCID mice, c-Met inhibitors slowed tumor growth and reduced the population of CSCs when given alone or in combination with gemcitabine. Administration of XL184 for 2 weeks after cardiac injection of cancer cells prevented the development of metastases. CONCLUSIONS: c-Met is a new marker for pancreatic CSCs. It is required for growth and metastasis of pancreatic tumors in mice and is a therapeutic target for pancreatic cancer.