Survival analysis and prognostic factors in posterior fossa ependymomas in children and adolescents

OBJECTIVE In this study, the authors aimed to analyze the overall survival (OS) and progression-free survival (PFS) of patients younger than 18 years of age who were diagnosed with posterior fossa ependymomas, and to identify prognostic factors such as the degree of resection, tumor topography, and involvement of the lesion in the hindbrain.METHODS The authors performed a retrospective cohort study of patients younger than 18 years of age, treated beginning in 2000, with a diagnosis of posterior fossa ependymoma. Ependymomas were separated into three groups: tumors restricted to the fourth ventricle, tumors inside the fourth ventricle and exiting from the foramen of Luschka, and tumors inside the fourth ventricle and completely surrounding the hindbrain. Furthermore, the tumors were classified by molecular group using the staining method for H3K27me3. Statistical analysis was performed using Kaplan-Meier survival curves, with p < 0.05 considered statistically significant.RESULTS Of 1693 patients who underwent surgical treatment between January 2000 and May 2021, 55 patients who met the inclusion criteria were included. The median age at diagnosis was 2.98 years. The median OS was 44 months, and the survival rates at 1, 5, and 10 years were 92.5%, 49.1%, and 38.3%, respectively. The cases were assigned to two posterior fossa ependymoma molecular groups: 35 (63.6%) cases to group A and 8 (14.5%) to group B. The median ages in groups A and B were 2.94 and 2.85 years and the median OS values were 44 and 38 months, respectively (p = 0.9245). Statistical analysis was performed on multiple variables, including age, sex, histological grade, Ki-67 expression, tumor volume, extent of resection, and adjuvant therapies. The median PFS of patients with dorsal-only involvement was 28 months; for dorsolateral involvement, it was 15 months; and for total involvement, it was 9.5 months (p = 0.0464). No statistically significant difference was found for OS. There was a statistically significant difference between the proportion of patients in whom gross-total resection was achieved in the dorsal-only involvement group (73.1%, 19/26) and those in the total involvement group (0%, 0/6) (p = 0.0019). CONCLUSIONS This study confirmed that the extent of resection has an impact on OS and PFS. The authors found that adjuvant radiotherapy resulted in a higher OS but did not prevent progression, that the pattern of involvement of the brainstem in the tumor at diagnosis could elicit important information regarding the patient's prognosis regarding PFS, and that the total involvement of the rhombencephalon impaired the gross-total resection of these tumors.