Cancer is a disease that is characterized by uncontrolled cell growth with the ability to penetrate or develop to the other parts of the body. Various studies have shown the significance of identifying drug targets for cancer, although this process continues challenging in the field of anti-cancer drug designing. The primary purpose of this study is to design a novel approach to identify target genes for cancer. The sub-network of colorectal, pancreatic, and prostate cancers were constructed from human protein–protein interaction network. The potential genes were analyzed using hubs and centrality measures. For the identification of target genes, we retrieve those genes that had the highest score in both mutation rates and graph centrality. Moreover, gene deletion analysis revealed that MYC, TP53, and EGFR genes remained potential targets in colorectal, pancreatic, and prostate cancer respectively. Results suggest that combining network measures with mutation frequencies in cancer genes might assist in recognizing several potential drug targets. Future enhancement of this current approach is to combine these network properties with the biological insights from gene expression data, and their functions will provide a reliable method for rational drug designing.