Association of ITGA 11 and ITGAV upregulation with outcomes in gastric cancer.

Abstract

332 Background: Gastric cancer (GC) is a common cancer worldwide. The integrin α (ITGA) family members play essential roles in various cancers and variants of ITGA are involved in metastatic process in gastric cancer. Thus, this study was conducted to explore the roles of ITGA family members in stomach adenocarcinoma (STAD). Methods: RNA-sequencing FPKM data and corresponding clinical information of 375 STAD tumor tissues and 32 normal tissues were retrieved from The Cancer Genome Atlas (TCGA). The ‘limma’ package was used to compare the expression differences between the normal and cancer tissues using Wilcoxon rank-sum test. Analysis of overall survival (OS) was conducted by Kaplan–Meier (K–M) method via ‘survminer’ package. Univariate Cox hazard regression analysis was applied to seven clinicopathological variables from T stage, N stage, M stage, pathologic stage, histologic grade, gender, age, and expression level of selected ITGA family members, by using ‘survival’ package. Furthermore, a nomogram was also visualized by the R ‘rms’ package and ‘survival’ package to predict the 1-, 3-, and 5-year OS and individual predictors. Results: The expression of ITGA2, ITGA3, ITGA4, ITGA5, ITGA6, ITGA11, and ITGAV were upregulated in STAD tissues compared with normal tissues (P<0.05), whereas ITGA7, ITGA8, and ITGA9 were downregulated in tumor tissues (P<0.05). In addition, no statistically significant difference was found for TGA1 and ITGA10 between tumor and normal tissues ((P>0.05). Further survival analysis found that higher expression of ITGA 11 (HR=1.46, 95%CI 1.05-2.03, P<0.026) and ITGAV (HR=1.92, 95%CI 1.37-2.70, P<0.001) were associate with worse OS. Univariate Cox hazard regression analysis showed that T3, T4, N1, N3, M1, pathologic stage III and IV, age>65, and high expression level of ITGA11 and ITGAV were associated with worse OS (all P < 0.05). The nomogram based on six clinicopathological variables (T stage, N stage, M stage, pathologic stage, age, and expression levels of ITGA11 and ITGAV) and 1-, 3-, 5-year OS probabilities were developed. The concordance index (C-index) of the nomograms was 0.673(0.647-0.700), indicating that the potential predicting role and sufficient discrimination ability of the nomogram as C-index was more than 0.5. The calibration curves of 1-, 3-, and 5-year indicated the consistency of our results and the predictive values, indicating satisfactory performance for this nomogram. However, 1-, 3-, 5-year AUCs of ITGA11- and ITGAV-based nomogram were 0.687, 0.691, and 0.687, showing a relatively acceptable accuracy as they were great than 0.5. Conclusions: ITGA2, ITGA3, ITGA4, ITGA5, ITGA6, ITGA11, and ITGAV are upregulated, while ITGA7, ITGA8, and ITGA9 are down regulated in STAD tumor samples. High expression levels of ITGA11 and ITGAV are associated with worse OS. The ITGA11- and ITGAV-based monogram developed by this study might be useful in predicting the OS outcome for STAD patients.