Identification of Novel Molecular Panel as Potential Biomarkers of PAN-Gastrointestinal Cancer Screening:...

PAN-gastrointestinal cancers (PAN-GI cancers), including the oral, esophageal, gastric, hepatocellular, pancreatic=, and colorectal cancers, are the leading cause of cancer-related mortality. Despite recent advances in identifying the molecular mechanisms driving these malignancies, the high incidence and recurrence of the PAN-gastrointestinal cancers and the low survival rates of patients indicate the need to introduce biomarkers for early diagnosis to improve diagnostic and therapeutic approaches. In the present study, using integrated transcriptomics, RNA-Seq and microarray data, from the TCGA and GEO databases, respectively, were combined to discover and validate a global biomarker panel for PAN-gastrointestinal cancers. In order to validate the bioinformatics data, the expression levels of genes in the molecular panel were evaluated using real-time quantitative polymerase chain reaction (qPCR) in tumor tissues of 21 patients with early diagnosis of gastric cancer and colorectal cancer (Stage I and II). By examining the transcriptomic profiles of six types of PAN-gastrointestinal cancers, a network of closely related hub genes (n = 167) with biomarker potential (p value < 0.05) was identified. Also, using ROC curve analysis and the Youden index, a molecular panel consisting of AURKA, CEP55, DTL, and TTK was presented (95% confidence interval and p value < 0.05), which showed exceptional sensitivity and specificity in differentiating malignant tissue from normal tissue (AUC > 80%). The diagnostic efficacy of these markers was confirmed by further validation using qPCR in colorectal and gastric tumor samples (p value < 0.05). In conclusion, a novel molecular signature panel including the AURKA, CEP55, DTL, and TTK genes could improve early cancer detection and diagnostic accuracy, and it may contribute to the treatment outcomes of PAN-gastrointestinal cancer patients.