Prognosis by tumor location in adults with spinal ependymomas

Abstract

Object: Ependymomas are primary central nervous system tumors that occur more frequently in the spines of adults than they do there in children. Previous studies consist mainly of retrospective single-institutional experiences or case studies. In this study, a comprehensive literature review was performed on reported cases of spinal ependymoma treated with resection to determine whether tumor location along the spinal axis conveys important prognostic information.

Methods: A PubMed search was performed to identify all papers that included data on patients with spinal ependymoma. Only cases involving adult patients who underwent ependymoma resection with a clearly reported tumor location were included for analysis. Tumor locations were separated into 6 groups: cervicomedullary, cervical, cervicothoracic, thoracic, thoracolumbar, and conus + cauda equina. Kaplan-Meier survival and Cox regression analyses were performed to determine the effect of tumor location on progression-free survival (PFS) and overall survival (OS).

Results: A total of 447 patients who underwent resection of spinal ependymomas with clearly indicated location of tumor were identified. The most common locations of spinal ependymomas were the cervical (32.0%) and conus + cauda equina (26.8%) regions. The thoracolumbar and cervicomedullary regions had the fewest tumors (accounting for, respectively, 5.1% and 3.4% of the total number of cases). The conus + cauda equina and thoracolumbar regions had the highest percentage of WHO Grade I tumors, while tumors located above these regions consisted of mostly WHO Grade II tumors. Despite the tendency for benign grades in the lower spinal regions, PFS for patients with spinal ependymomas in the lower 3 regions (thoracic, thoracolumbar, conus + cauda equina) was significantly shorter (p < 0.001) than for those with tumors in the upper regions (cervicomedullary, cervical, cervicothoracic), but the difference in OS did not achieve statistical significance (p = 0.131).

Conclusions: Spinal ependymomas along different regions of spinal axis have different characteristics and clinical behaviors. Tumor grade, extent of resection, and PFS varied by tumor location (upper vs lower spinal regions), while OS did not. Recurrence rates were higher for the lower spinal cord tumors, despite a greater prevalence of lower WHO grade lesions, compared with upper spinal cord tumors, suggesting that tumor location along the spinal axis is an important prognostic factor.