Identification of genes associated with gastric cancer survival and construction of a nomogram to improve risk stratification for patients with gastric cancer.

Abstract

The present study aimed to identify genes associated with gastric cancer survival and improve risk stratification for patients with gastric cancer. Transcriptomic and clinicopathological data from 443 gastric cancer samples were retrieved from The Cancer Genome Atlas database. The DESeq R package was applied to screen for differentially expressed genes between Tumor-Node-Metastasis (TNM) stage (I vs. IV) and histological grade (G3 vs. G1 and G2). A total of seven genes were common to both comparisons; spondin 1 (SPON1); thrombospondin 4 (THBS4); Sushi, Von Willebrand factor type A, EGF and pentraxin domain containing 1 (SVEP1); prickle planar cell polarity protein 1 (PRICKLE1); ATP binding cassette subfamily A member 8 (ABCA8); Slit guidance ligand 2 (SLIT2); and EGF containing fibulin extracellular matrix protein 1 (EFEMP1), were selected as candidate survival-associated genes for further analysis. The prognostic value of these genes was assessed according to a literature review and Kaplan-Meier survival analysis. In addition, a multivariate Cox regression analysis revealed PRICKLE1 expression to be an independent prognostic factor for patients with gastric cancer. Furthermore, a predictive nomogram was generated using PRICKLE1 expression, patient age and TNM stage to assess overall survival (OS) rate at 1, 3 and 5 years, with an internal concordance index of 0.65. External validation was conducted in an independent cohort of 59 patients with gastric cancer, and high consistency between the predicted and observed results for OS was exhibited. Overall, the current findings suggest that PRICKLE1 expression may serve as an independent prognostic factor that can be integrated with age and TNM stage in a nomogram able to predict OS rate in patients with gastric cancer.