Purpose A molecular pathological grading method was tested in WHO grade 2 meningiomas to judge whether this molecular grading can more accurately evaluate meningioma biological behaviour. Methods The medical records and paraffin-embedded tissues of surgically resected WHO grade 2 meningioma patients in our department from January 1, 2010, to December 31, 2020, were collected. The molecular pathological risk grading suggested by Sahm et al. was adopted and the patients were graded as low, intermediate and high risk. Progression-free survival (PFS), malignant progression-free survival (MPFS) and overall survival (OS) were analysed. Univariate and multivariate analysis were performed to determine the relationship between molecular risk grading and patient survival. Results Of the 98 patients, 13 (13.2%) were graded as low risk, 63 patients (64.3%) were graded as intermediate risk, and 22 patients (22.4%) were graded as high risk. With increasing molecular risk grade, the rates of tumour recurrence, malignant progression and mortality increased significantly (P < 0.05). Multivariate analysis showed that molecular risk grading was negatively associated with PFS (HR 0.018, 95% CI 0.003-0.092), MPFS (HR 0.040, 95% CI 0.006-0.266) and OS (HR 0.088, 95% CI 0.016-0.472) (P < 0.01), and gross total resection (Simpson grade I-III) significantly prolonged PFS (HR 5.882, 95% CI 2.538-13.699) and OS (HR 2.611, 95% CI 1.117-7.299) (P < 0.05). Conclusion Sahm et al.'s molecular risk grading can further refine the classification of WHO grade 2 meningiomas and more accurately evaluate their biological behaviour and patient prognosis.