The exploration of non-cancer medications with potential anti-cancer activity offers a promising avenue for drug repurposing, accelerating the development of new oncological therapies. This study employs Quantitative Structure-Property Relationship (QSPR) modeling to identify and predict the anti-cancer efficacy of various non-cancer drugs, utilizing topological indices as key descriptors. Topological indices, which capture the molecular structure’s geometric and topological characteristics, provide critical insights into the pharmacological interactions relevant to anti-cancer activity. By analyzing a comprehensive dataset of non-cancer medications, this research establishes robust QSPR models that correlate topological indices with anti-cancer activity. The models demonstrate significant predictive power, highlighting several non-cancer drugs with potential anti-cancer properties. Further, we will use linear, quadratic and logarithmic regression to understand the structures of anti-cancer drugs and strengthen our ability to manipulate the molecular structures. The findings underscore the utility of topological indices in drug repurposing strategies and pave the way for further experimental validation and clinical trials. This integrative approach enhances our understanding of drug action mechanisms and offers a cost-effective strategy for expanding the repertoire of anti-cancer agents.