Identifying multi-target drugs for prostate cancer using machine learning-assisted transcriptomic analysis

Prostate cancer is a leading cause of cancer death in men, and the development of effective treatments is of great importance. This study explored to identify the candidate drugs for prostate cancer by transcriptomic data and CMap database analysis. After integrating the results of omics analysis, bisoprolol is confirmed as a promising drug. Moreover, cell experiment reveals its potential inhibitory effect on the proliferation of prostate cancer cells. Importantly, machine learning methods are employed to predict the targets of bisoprolol, and the dual-target ADRB3 and hERG are explored by dynamic simulation. The findings of this study demonstrate the potential of bisoprolol as a multi-target drug for prostate cancer treatment and the feasibility of using beta-adrenergic receptor inhibitors in prostate cancer treatment. In addition, the proposed research approach is promising for discovering potential drugs for cancer treatment by leveraging the concept of drug side effects leading to anticancer effects. Further research is necessary to investigate the pharmacological action, potential toxicity, and underlying mechanisms of bisoprolol in treating prostate cancer with ADRB3.