Glioblastoma (GBM) is the most aggressive glioma with a 5-year survival rate of only 6.8% and a median overall survival of 8 months. Accurate prediction of recurrence is essential for personalized treatment. Current imaging-based approaches face two major limitations: 1) reliance on single-time point images fails to capture tumor dynamics, and 2) the limited receptive fields of CNN and the quadratic complexity of Transformer hinder effective 3D MRI processing. To address these limitations, we leverage Mamba’s linear computational complexity and global modeling capabilities, and our model employs the following techniques: 1) a dual-stream framework for prediction of recurrence; 2) the BiSMamba module for multi-scale feature extraction; 3) the multi-stage fusion module for capturing dynamic changes. Extensive experiments on two public datasets (RHUH-GBM and LUMEIERE) have shown that our approach achieves impressive results in prediction of glioblastoma recurrence.