Abstract
OBJECTIVE: Adjuvant radiotherapy is the main treatment modality for high grade meningioma after surgical resection; however, recurrence and survival outcomes vary. The aim of this study was to create a new "prognostic score" that allows personalized recommendations for post-operative adjuvant radiotherapy in patients with high grade meningioma. METHODS: Clinical data were collected from 115 patients with high grade meningioma treated with surgical resection and adjuvant radiotherapy. A prognostic model was built based on the hazards ratios of independent prognostic factors yielded by multivariate cox proportional analysis. Calibration and discrimination of the prognostic score was evaluated using good of fit test and Harrel's C index, respectively. RESULTS: A total of 115 high grade meningioma patients (72 atypical and 43 anaplastic meningiomas) were enrolled. Three factors were independently associated with progression-free survival (PFS): extent of resection (GTR vs. STR), recurrent status (de novo vs. recurrent), and Ki-67 labeling index (<5% vs. ≥ 5%). The respective β-coefficients were used to generate the "prognostic score". The cohort was divided into low-risk and high-risk groups based on the median prognostic score. Good of fit test showed strong calibration (P = 0.7133) and Harrel's C index 0.766 indicated a strong discrimination capability of the prognostic score. The Harrel's C index for OS was 0.60. CONCLUSIONS: Our prognostic model using three basic clinical parameters robustly separated high grade meningioma patients who benefit vs. do not benefit from adjuvant radiotherapy. External validation of our model is warranted to help improve patient selection suitable for adjuvant radiotherapy.