High-risk neuroblastoma (NB) remains a major challenge in pediatric oncology accounting for 15% of all pediatric cancer mortality with an overall 5-year survival of less than 50% for this aggressive embryonal malignancy. As with 90% of all solid tumors, the primary cause of death for NB is relapsed drug-resistant disease. Significant improvements in cure rates for NB will require novel biologically specific approaches since chemotherapy at maximal tolerated doses still fails in over half of theses cases. Models of cancer stem cells (CSCs) now suggest that treatment failure and relapse is often driven by a small drug-resistant subpopulation of self-renewing tumor initiating cells. Our therapeutic goal isto combine anti-CSC therapies with current chemotherapy approaches to both improve initial drug responses and prevent recurrent metastatic disease. Preliminary Data: We have recently demonstrated the isolation and characterization of a highly tumorigenic subpopulation based on the expression of CD114 (the surface receptor for granulocyte colony-stimulating factor (G-CSF)), which fulfills the major phenotypic requirements for tumor initiating CSCs. This tumor subpopulation is highly enriched (up to 10-fold) after chemotherapy in NB tumors resected after 3 cycles of chemotherapy and has gene expression, microRNA expression, and epigenetic profiles which recapitulate those of iPSCs (induced pluripotent stem cells) and ESCs (embryonic stem cells). STAT3 activation through phosphorylation (pSTAT3) is the canonical downstream effector of G-CSF ligand binding to its receptor (CD114). pSTAT3 has a central role in regulating the maintenance of normal and malignant stem cell populations in part through transcriptional regulation of microRNAs involved in reprogramming and epithelial-mesenchymal transitions (EMT). We demonstrate within the CD114+ CSC-like cells, G-CSF dependent pSTAT3 activation results in increased expression of STAT3 target microRNAs. We also demonstrate that the CSFR3 gene (encoding CD114) is a direct transcriptional target of pSTAT3, suggesting a feedback loop for maintenance of expression of this receptor. We therefore hypothesize the following: a) the G-CSF/STAT3 signaling axis promotes the maintenance, drug resistance and metastatic potential of CD114+ NB cells b) Targeting CD114+ cells, either directly or via G- CSF/STAT3 pathway inhibition will limit tumor proliferationand the development of metastasis. C) Concomitant targeting of both the CD114+ and CD114- subpopulations in neuroblastoma should significantly augment chemotherapies by preventing regeneration of CD114- cells from the CD114+ precursors. In Specific Aims 1 and 2, we will block the G-CSF/STAT3 axis in CD114+ NB cells using by several independent approaches (anti-receptor antibody, JAK/STAT small molecule inhibitors) to determine the role of G-CSF signaling in CD114+ driven NB tumorigenesis and metastasis. In Aim 3 we will determine how STAT3 target microRNAs contribute to the stemness phenotype of CD114+ cells.