Editorial: Chordoma model

Abstract

Patients suffering from chordoma, a mesenchymal cancer thought to arise from the notochord, face limited treatment options.1,5 Although this bone tumor can on occasion metastasize to distant sites, a major clinical challenge is imposed by infiltration of the tumor cells into surrounding structures that can limit the extent of resection. Irradiation prolongs survival, but the relative radioresistance demands delivery of a high dose to sensitive regions. To date, there are no approved medical treatments for chordoma, and efforts to identify effective drugs face barriers. First, there are only few established chordoma cell lines. This limitation presents a challenge to broadening our understanding of signaling pathways that can be targeted, and to verifying the significance of chordoma genomic alterations. Second, consistently reproducible in vivo models are lacking. The list of promising agents obtained from high throughput in vitro screening methods can be strategically shortened prior to human trials by testing in animal models. Presence of such barriers restricts progress and contributes to the poor outcome of patients with recurrent disease. In their article in the Journal of Neurosurgery, Siu and colleagues6 describe a primary human chordoma xenograft model. The authors report that 1 of the 2 chordoma tumors implanted into athymic nude mice produced a serially transplantable xenograft that retains high genomic fidelity to the parental tumor. Additional comparative characterization of the tumors provides compelling evidence for successful establishment of a primary chordoma xenograft in that physaliphorous morphology and brachyury expression are maintained. Although this report describes successful implantation and propagation of a single specimen, the achievement is nevertheless important. Rather than dissociating the chordoma tumor tissue into single cells, culturing with animal serum on a plastic dish, and transplanting it into an in vivo environment, the investigators developed a primary xenograft that can be serially passaged. A primary xenograft better mirrors the original tumor compared with xenograft established from in vitro expanded tumor cells, and likely explains the greater than 99.9% genomic concordance observed by the investigators. UCH-1, an established chordoma cell line, can be xenografted, but it appears to require a highly immunodeficient background lacking natural killer cell activity (NOD-SCID IL-2 Receptor Gamma Receptor Null).4 The tumor reported by Siu and colleagues was derived from a patient with a clival chordoma. Most chordoma research is performed with tissues and cells derived from sacral chordomas, presumably because of the availability of a greater quantity of material. However, balanced representation will be important, considering that the development of the notochord demonstrates distinct regional morphogenetic origins, suggesting a histogenic basis for intertumoral heterogeneity.7 Modeling human diseases in animals allows for greater appreciation for pathogenesis, in addition to proving a platform to test potential drugs. The study investigators report that brachyury expression was higher in the xenograft than the primary tumor. Determining the mechanism responsible for upregulation of this important transcription factor would be of interest, and to see if it may be associated with enhanced proliferative index. Also, subsets of chordomas demonstrate loss of critical tumor suppressors PTEN or CDKN2A, disruptions potentially capable of altering the tumorigenicity of the chordoma cells.2,3 Molecular profiling of the parental tumor and xenograft may be worthwhile. Lastly, although chordomas can grow in a subcutaneous environment by direct invasion or metastasis, they are primary bone neoplasms. Establishing an orthotopic model by tumor implantation into trabecular bone would represent further progress in understanding this challenging disease. (http://thejns.org/doi/abs/10.3171/2011.10.JNS111716)