Pediatric spinal ependymomas are rare tumors of the central nervous system with limited treatment options. Successful outcomes depend primarily upon the extent of surgical resection, although significant impairment is common, particularly in children with developing nervous systems. Because these tumors are uncommon, there are no established tumor models nor are there many studies to define underlying molecular lesions. Here, we describe an innovative spinal ependymoma (SE) tumor model based on the recent discovery that primary carcinoma cells can be cultured indefinitely ex vivo on fibroblast feeder cells in the presence of a rho kinase inhibitor; these cells are denoted conditionally reprogrammed cells (CRCs). In prior studies, CRCs form spheres in vitro, replicate tumor pathology as xenografts in vivo, and predict drug responses to the primary tumor. We therefore adopted this strategy to generate a cell line from a 12 year-old girl with a spinal myxopapillary ependymoma. The cell line and primary tumor are of human origin and identical with respect to short tandem repeat (STR) analysis. Using targeted gene expression analysis, we discovered that genes encoding the of High Mobility Group Protein AT-hook 1 (HMGA1), High Mobility Group Protein AT-hook 2 (HMGA2), cMYC, HOXB13, and HOXA10 proteins are all up-regulated compared to their expression in spinal cord tissue from a normal individual. In primary pediatric tumors, HMGA1 gene expression was also higher in spinal ependymomas as compared to supratentorial or posterior fossa ependymomas. There was also a significant positive correlation between HMGA1 and HOX genes, including HOXA10, HOXD13, HOXD1, and HOXB7 in primary tumors. In addition, HMGA1 and cMYC, together with other Wnt target genes, were positively correlated in primary spinal ependymoma. Pathway analysis showed that HMGA1 expression correlated with the following pathways: 1) metabolic pathways, 2) osteoclast differentiation, 3) MAPK signaling, and, 4) Neurotropin signaling. Studies are underway to generate additional cell lines from primary tumor samples and to elucidate the mechanisms that account for up-regulation in HMGA1, HOX and Wnt genes. We will harness this information to identify novel therapeutic targets and agents for children and young adults with spinal ependymoma.

Citation Format: Li Z. Luo, Ewa Krawczyk, Anbarasu Lourdusamy, Lisa C. Storer, Lingling Xian, Kenneth J. Cohen, Richard Schlegel, Richard Grundy, Linda Resar. A novel model of pediatric spinal ependymoma using conditionally reprogrammed cells from a primary tumor demonstrates aberrant expression of HMGA, HOX, MYC and other Wnt target genes [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2017; 2017 Apr 1-5; Washington, DC. Philadelphia (PA): AACR; Cancer Res 2017;77(13 Suppl):Abstract nr LB-224. doi:10.1158/1538-7445.AM2017-LB-224