Construction and analysis of a postoperative recurrence risk modelfor meningioma patients based on histological features

Ye Zhongwei; Wu Xuelian; Huang Xuecai

doi:10.11850/j.issn.1009-122X.2024.11.004

Abstract：Objective To explore the construction of a postoperative recurrence risk model for meningioma patients based on histological features and Ki-67. Methods The clinical and pathohistological data of 178 meningioma patients were analyzed retrospectively,including age,gender,WHO grade of meningioma,and specific pathological manifestations of the tumor. The patients were divided into a recurrence group (n=52) and a non-recurrence group (n=126) based on postoperative recurrence. Multivariate logistic regression analysis was performed to investigate the relationship between various histological features and postoperative recurrence risk. A prediction model for postoperative recurrence risk in meningioma patients based on histological features was constructed,and the predictive performance of the model was evaluated using the receiver operating characteristic (ROC) curve and decision curve analysis. Results Compared with the non-recurrence group,the recurrence group showed statistically significant differences in localized necrosis,brain tissue infiltration,Ki-67,mitotic index,and tumor grade (all P<0.05). Multivariate logistic regression analysis revealed that localized necrosis,brain tissue infiltration,Ki-67,mitotic index,and tumor grade were risk factors for postoperative recurrence in meningioma patients (all P<0.05). The ROC curve was used for goodness-of-fit testing,with postoperative recurrence as the status variable and the model's prediction probability value as the test variable. The area under the curve (AUC) of the prediction model was 0.837 (95％CI:0.775-0.888),with a specificity of 80.16％and a sensitivity of 76.92％. The decision curve results indicated that the model for predicting postoperative recurrence risk in meningioma patients based on histological features had a good net benefit rate. Conclusions The model constructed based on histological features and Ki-67 in meningioma patients has high application value in predicting postoperative recurrence risk.