Cancer continues to pose a formidable global health challenge, with conventional chemotherapies often resulting in significant collateral damage to healthy cells and severe side effects. In response to these limitations, anticancer peptides (ACPs) have emerged as a promising alternative due to their selective targeting and elimination of cancer cells, which can potentially enhance treatment efficacy and improve patient quality of life. However, the experimental identification of effective ACPs is typically labor-intensive and time-consuming. To address these challenges, we have developed ACP-ETBLCA, a deep learning framework designed to predict ACPs directly from protein sequences. This model synergistically integrates Evolutionary Scale Modeling 2 (ESM-2) with conventional feature extraction techniques, and it utilizes a cross-attention mechanism to fuse multimodal features effectively. By leveraging these advanced computational strategies, ACP-ETBLCA represents a significant step forward in the rapid and accurate identification of novel ACP candidates, thereby contributing to the advancement of precision oncology.