Identification of Shared Biomarkers in HCC and NSCLC: An Integrative Machine Learning and Bioinformatics Approach

Non-small cell lung cancer (NSCLC) and hepatocellular carcinoma (HCC) are two prevalent malignancies associated with high morbidity and mortality worldwide. Despite differences in organ origin and clinical presentation, there exist common molecular pathways that could potentially serve as therapeutic targets across these cancers. In this study, an integrative analysis was conducted to identify shared biomarkers between HCC and NSCLC. Initially, two microarray datasets were utilized to identify differentially expressed genes (DEGs) in each cancer type. Subsequently, SVM was applied to identify SVM-DEGs within each dataset. By intersecting the SVM-DEGs from both datasets, 43 common SVM-DEGs were identified. To delineate the key genes within these shared SVM-DEGs, a protein-protein interaction (PPI) network was constructed using the STRING database, followed by comprehensive network analysis in Cytoscape. Through various network centrality measures, including MCC, MNC, and Degree, eight common hub genes were identified, namely TOP2A, CDK1, BUB1B, PBK, ZWINT, RRM2, NEK2, and BIRC5. The robustness and relevance of these hub genes were further validated using two independent test datasets, where their Area Under the Curve (AUC) scores from Receiver Operating Characteristic (ROC) curves were calculated. Additionally, their mRNA expression levels were evaluated in normal and tumor samples using Gene Expression Profiling Interactive Analysis (GEPIA), confirming their dysregulation in both cancer types. These hub genes could be potential biomarkers and lead to better treatments for both HCC and NSCLC, improving outcomes for patients with these aggressive cancers.