Abstract
BackgroundIntratumoral oxidative stress (OS) has been associated with the progression of various tumors. However, OS has not been considered a candidate therapeutic target for pancreatic cancer (PC) owing to the lack of validated biomarkers.MethodsWe compared gene expression profiles of PC samples and the transcriptome data of normal pancreas tissues from The Cancer Genome Atlas (TCGA) and Genome Tissue Expression (GTEx) databases to identify differentially expressed OS genes in PC. PC patients’ gene profile from the Gene Expression Omnibus (GEO) database was used as a validation cohort.ResultsA total of 148 differentially expressed OS-related genes in PC were used to construct a protein-protein interaction network. Univariate Cox regression analysis, least absolute shrinkage, selection operator analysis revealed seven hub prognosis-associated OS genes that served to construct a prognostic risk model. Based on integrated bioinformatics analyses, our prognostic model, whose diagnostic accuracy was validated in both cohorts, reliably predicted the overall survival of patients with PC and cancer progression. Further analysis revealed significant associations between seven hub gene expression levels and patient outcomes, which were validated at the protein level using the Human Protein Atlas database. A nomogram based on the expression of these seven hub genes exhibited prognostic value in PC.ConclusionOur study provides novel insights into PC pathogenesis and provides new genetic markers for prognosis prediction and clinical treatment personalization for PC patients.
Highlights
Pancreatic cancer (PC) is one of the most common tumors worldwide and is a severe threat to human health (Kamisawa et al, 2016)
We aimed to identify candidate oxidative stress (OS) genes that are significantly differentially expressed between PC and normal pancreatic tissues based on publicly available data obtained from The Cancer Genome Atlas (TCGA) and Genome Tissue Expression (GTEx) databases
Gene ontology analysis showed that, with respect to the upregulated differentially expressed OS-related genes (DEOGs), the most enriched biological processes included the response to lipopolysaccharide, leukocyte migration, and extracellular structure organization (Figure 3A), whereas relative to the downregulated DEGs, intrinsic apoptotic signaling pathway, cellular oxidant detoxification, and cellular detoxification were most enriched terms (Figure 3B)
Summary
Pancreatic cancer (PC) is one of the most common tumors worldwide and is a severe threat to human health (Kamisawa et al, 2016). Surgical resection of cancer tissues remains the most common choice for PC treatment, which effectively increases patients’ 5-year overall survival rate to 20– 30%; less than 20% of PC patients are eligible for surgical treatment because of advanced-stage diagnoses, at which point cancer has already metastasized (Kamisawa et al, 2016). Many studies have focused on constructing more effective prediction models that could better clarify the factors contributing to the prognosis and progression of PC, aiming to provide more evidence for individual treatment strategies. Despite these efforts, few screening biomarkers and tools have shown sufficient significance for widespread clinical application in PC. OS has not been considered a candidate therapeutic target for pancreatic cancer (PC) owing to the lack of validated biomarkers
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
