Identification of specific role of SNX family in gastric cancer prognosis evaluation

Abstract

AbstractWe here perform a systematic bioinformatic analysis to uncover the role of sorting nexin (SNX) family in clinical outcome of gastric cancer (GC). Comprehensive bioinformatic analysis were realized with online tools such as TCGA, GEO, String, Timer, cBioportal and Kaplan–Meier Plotter. Statistical analysis was conducted with R language or Perl, and artificial neural network (ANN) model was established using Python. Our analysis demonstrated that SNX4/5/6/7/8/10/13/14/15/16/20/22/25/27/30 were higher expressed in GC, whereas SNX1/17/21/24/33 were in the opposite expression profiles. GSE66229 was employed as verification of the differential expression analysis based on TCGA. Clustering results gave the relative transcriptional levels of 30 SNXs in tumor, and it was totally consistent to the inner relevance of SNXs at mRNA level. Protein–Protein Interaction map showed closely and complex connection among 33 SNXs. Tumor immune infiltration analysis asserted that SNX1/3/9/18/19/21/29/33, SNX1/17/18/20/21/29/31/33, SNX1/2/3/6/10/18/29/33, and SNX1/2/6/10/17/18/20/29 were strongly correlated with four kinds of survival related tumor-infiltrating immune cells, including cancer associated fibroblast, endothelial cells, macrophages and Tregs. Kaplan–Meier survival analysis based on GEO presented more satisfactory results than that based on TCGA-STAD did, and all the 29 SNXs were statistically significant, SNX23/26/28 excluded. SNXs alteration contributed to microsatellite instability (MSI) or higher level of MSI-H (hyper-mutated MSI or high level of MSI), and other malignancy encompassing mutation of TP53 and ARID1A, as well as methylation of MLH1.The multivariate cox model, visualized as a nomogram, performed excellently in patients risk classification, for those with higher risk-score suffered from shorter overall survival (OS). Compared to previous researches, our ANN models showed a predictive power at a middle-upper level, with AUC of 0.87/0.72, 0.84/0.72, 0.90/0.71 (GSE84437), 0.98/0.66, 0.86/0.70, 0.98/0.71 (GSE66229), 0.94/0.66, 0.83/0.71, 0.88/0.72 (GSE26253) corresponding to one-, three- and five-year OS and recurrence free survival (RFS) estimation, especially ANN model built with GSE66229 including exclusively SNXs as input data. The SNX family shows great value in postoperative survival evaluation of GC, and ANN models constructed using SNXs transcriptional data manifesting excellent predictive power in both OS and RFS prediction works as convincing verification to that.