Developing a prognostic signature: identifying differentially expressed genes in cardia and non-cardia gastric cancer for immunity and therapeutic sensitivity analysis.

Background: Gastric cancer (GC) can be anatomically categorized into two subtypes; that is, cardia gastric cancer (CGC) and non-cardia gastric cancer (NCGC), which have distinct molecular mechanisms and prognoses. At present, the majority of pharmacological interventions for GC adhere to non-specific treatment regimens. The stratification of GC based on molecular disparities between CGC and NCGC has important clinical guidance value and could help in the development of precision therapies tailored to individual patient needs. Nevertheless, research in this specialized field remains notably limited. This study aims to investigate the molecular differences between CGC and NCGC and to leverage these differences to develop a prognostic risk scoring model (PRSM). Methods: We used patient data from The Cancer Genome Atlas (TCGA) and performed a differentially expressed gene (DEG) analysis between CGC and NCGC. A PRSM was developed from the prognosis-associated DEGs identified through Cox regression analyses and was well validated using Gene Expression Omnibus (GEO) data. Results: A total of 339 DEGs were identified between CGC and NCGC, and four prognosis-associated genes were used to construct the PRSM. Using the risk coefficients and expression levels of signature genes, a median risk score (RS) was calculated to classify patients into high- and low-risk groups. The high-risk group had a significantly worse prognosis than the low-risk group. An in-depth analysis revealed that Conclusions: This study elucidated the unique molecular characteristics of GC subtypes based on the anatomical site and provided a preliminary contribution for the development of precision medicine for GC.