A new way to modulate tumor therapy: artificial intelligence predicts nanoshape efficacy

Cancer remains one of the leading causes of death worldwide, and its treatment continues to present significant challenges. Nanomedicines have shown remarkable potential in cancer therapy; however, research on their delivery still faces several limitations. Studies have revealed that different nanoparticle morphologies during delivery can result in variations in delivery efficiency, cellular uptake, circulation time, and tumor targeting, ultimately leading to inconsistent therapeutic outcomes. Therefore, the shape of nanoparticles is a critical factor influencing their in vivo transport behavior. In recent years, advances in artificial intelligence have enabled computational prediction to emerge as a high-throughput screening tool that effectively reduces both time and economic costs. A key question is how simulation techniques can be leveraged to predict the impact of nanoparticle shape on interactions with biological systems. This review examines the effects of various nanoparticle shapes on tumor therapy and their underlying mechanisms, outlines computational methods for predicting the impact of shape, analyzes the advantages and disadvantages of different computational approaches, and interprets considerations related to scale and implementation strategies based on computational methods and shape parameters. Finally, we discuss major challenges in computationally predicting therapeutic outcomes and highlight future directions for research on shape effect prediction.Literature Search Methods [PubMed database 2007-2025].