Abstract Meningiomas are the most common primary tumors of the central nervous system (CNS). Atypical meningioma (AM) recurs in 40% of patients despite total resection and radiotherapy (RT). No consensus on optimal adjuvant management was found, and it is difficult to identify patients insensitive to RT in real-life clinical practice. A promising group of biomarkers represent microRNAs (miRNAs), short non-coding RNAs that regulate most biological processes, including cell proliferation, differentiation, and apoptosis. The study aims to identify tissue miRNAs capable of predicting patients with AM who could benefit from the indicated adjuvant RT. The study includes 80 patients with AM in the exploratory phase and 400 patients with meningioma in the validation phase. Total RNA enriched with miRNAs was isolated from FFPE tissue using the mirVana miRNA Isolation Kit (TF Scientific). Subsequently, RNA quantity and quality controls were measured using NanoDrop 2000 (TF Scientific) and Qubit 2.0 (TF Scientific) instruments. A global miRNA expression profile was generated using the TaqMan Array Human MicroRNA Cards (TF Scientific), which allow the detection of up to 754 miRNAs simultaneously. Obtained data were processed and integrated through bioinformatics algorithms with clinicopathological data of patients with AM.The study identified significantly dysregulated miRNAs among AM patients with and without recurrence (p < 0.05). Results also suggest dysregulated miRNA expression profile in AM patients with indicated RT who did/did not develop a recurrence (p < 0.05). Lists of individual miRNAs and detailed graphical analyses will be included in the conference presentation.The results will help predict the prognosis of surgically intervened patients more accurately and can help determine which patients will benefit from the adjuvant RT. This research is supported by the AZV grant from the Ministry of Health of the Czech Republic (reg. No. NV19-03-00559) and Conceptual development of research organization (FNBr, 65269705). All rights reserved.