Meningiomas account for 30 – 36% of all brain tumors in adults and occur approximately three times more frequently in women [1]. Most meningiomas are benign and slow-growing and can cause different symptoms depending on their location. For meningiomas at surgically inaccessible sites or in case of incomplete resection a single-fraction stereotactic radiosurgery (SRS) is treatment option, providing the tumor is small and well defined [2]. Stereotactic irradiation is a radiation therapy technique in which a precisely localized volume is irradiated with a prescribed dose of ionizing radiation. The SRS is characterized by a very steep dose fall-off at a distance of 2 – 4 mm from the edge of the treated volume, which guarantees protection of surrounding tissue [3]. SRS achieves high local control of meningiomas. Five-year tumor control rate is reported to be greater than 95% for a therapeutic dose of 12 Gy [4, 5]. Multivariable models are an effective tool for assessment of the relationship between the outcome (incidence of complications  or tumor progression after SRS) and selected predictors – SRS parameters and clinical characteristics of patients. Models are most often used in the form of nomograms as a tool for assessing odds or risk, usually in the form of a ratio comparing the odds (risks) of two groups in question.

We analyzed 196 patients with benign meningioma treated at the St. Elisabeth Cancer Institute between 2004 – 2019. In our work we retrospectively evaluated the relationship between the incidence of postirradiation edema and tumor progression after SRS and predictors. Bivariate analysis revealed an association of therapeutic dose (p = 0.020), tumor volume (p = 0.020), maximum tumor diameter (p = 0.050), gradient index (p = 0.030) and tumor extirpation before SRS (p = 0.003) with postirradiation edema. Tumor progression after SRS in our cohort was associated with volume of tumor treated with prescribed dose (%) (p = 0.020), tumor volume (p = 0.010), maximum tumor diameter (p = 0.030) and tumor extirpation before SRS (p = 0.030). The multivariable logistic models were built using manual variable selection method on the basis of bivariate analysis results. Selection of final model for each outcome was based on the comparison of the area under the ROC curve and the Akaike information criterion. The resulting full model for edema consists of 11 predictors with the only statistically significant predictor being tumor volume (OR = 1.200; p = 0.044). The resulting model for progression consists of 10 predictors with statistically significant predictors being volume of tumor treated with prescribed dose (%) (OR = 0.670; p = 0.033), tumor volume (OR = 1.330; p = 0.008) and tumor extirpation before SRS (OR = 6.620; p = 0.013).

Our research was conducted in cooperation with St. Elisabeth Cancer Institute in Bratislava. The use of patient data for the purposes of our work was approved by the institutional ethics committee. This contribution was supported by the grant KEGA 041UK-4/2020.

[1] LOUIS et al. WHO Classification of Tumours of the Central Nervous System. 2016. ISBN 9789283224303.
[2] GOLDBRUNNER et al. EANO guidelines for the diagnosis and treatment of meningiomas. B.m.: Lancet Publishing Group. 1. září 2016. ISSN 14745488.
[3] CHIN, Lawrence S. and William F. REGINE. Principles and practice of stereotactic radiosurgery [online]. B.m.: Springer New York, 2015. ISBN 9781461483632.
[4] KOLLOVÁ et al. Gamma Knife surgery for benign meningioma. Journal of neurosurgery [online]. 2007, 107(2), 325–36.
[5] MILANO et al. Radiation-Induced Edema After Single-Fraction or Multifraction Stereotactic Radiosurgery for Meningioma: A Critical Review. B.m.: Elsevier Inc. 2018. ISSN 1879355X.