A five-lncRNA model predicting overall survival in gastric cancer compared with normal tissues

In cancer research, normal tissues adjacent to the tumor are usually defined as controls to compare with tumor samples, in order to screen out cancer-related genes. Although there is no obvious difference in pathology between normal tissues adjacent to the tumor and healthy tissues, there are significant changes at the molecular level. We aim to explore more potential tumor biomarkers using healthy tissues as controls rather than normal tissues adjacent to the tumor.

Here we combine the Genotype-Tissue Expression project and The Cancer Genome Atlas for differential gene analysis. Gene Ontology and Kyoto Encyclopedia of Genes and Genomes analyses were applied in order to predict the biological effects of related lncRNAs.

We established a 5-lncRNA prognosis model with an AUC value of 0.815. Pathway analysis indicated that 5-lncRNA mainly affected tissue carcinogenesis through PI3K-AKT signaling pathway, Focal adhesion, MAPK signaling pathway.

The 5-lncRNA prognostic model we set up is more conducive to assess the overall survival time of gastric cancer patients.