Different Master Regulators Define Proximal and Distal Gastric Cancer: Insights into Prognosis and Opportunities for Targeted Therapy

Abstract

Background: Gastric cancer (GC) represents a significant global health burden with considerable heterogeneity in clinical and molecular behavior. The anatomical site of tumor origin-proximal versus distal-has emerged as a determinant of prognosis and response to therapy. The aim of this paper is to elucidate the transcriptional and regulatory differences between proximal gastric cancer (PGC) and distal gastric cancer (DGC) through master regulator (MR) analysis. Methods: We analyzed RNA-seq data from TCGA-STAD and microarray data from GEO (GSE62254, GSE15459). Differential gene expression and MR analyses were performed using DESeq2, limma, corto, and RegEnrich pipelines. A harmonized matrix of 4785 genes was used for MR inference following normalization and batch correction. Functional enrichment and survival analyses were conducted to explore prognostic associations. Results: Among 364 TCGA and 492 GEO patients, PGC was associated with more aggressive clinicopathological features and poorer outcomes. We identified 998 DEGs distinguishing PGC and DGC. PGC showed increased FOXM1 (a key regulator of cell proliferation), STAT3, and NF-κB1 activity, while DGC displayed enriched GATA6, CDX2 (a marker of intestinal differentiation), and HNF4A signaling. Functional enrichment highlighted proliferative and inflammatory programs in PGC, and differentiation and metabolic pathways in DGC. MR activity stratified survival outcomes, reinforcing prognostic relevance. Conclusions: PGC and DGC are governed by distinct transcriptional regulators and signaling networks. Our findings provide a biological rationale for location-based stratification and inform targeted therapy development.

Keywords: CDX2; FOXM1; gastric cancer; master regulators; survival analysis; transcriptomics; tumor localization.

Figure 1

Survival curves of patients stratified by master regulator activity. Kaplan–Meier survival curves illustrating overall survival probabilities over time for patients stratified by master regulator (MR) activity levels. Elevated FOXM1 (HR = 1.82, p = 0.001) and STAT3 (HR = 1.56, p = 0.007) activities were associated with reduced survival, while higher GATA6 (HR = 0.61, p = 0.009) and CDX2 (HR = 0.68, p = 0.014) activities predicted improved outcomes. Baseline curve represents low MR activity.

Figure 2

Forest plot of hazard ratios for key master regulators. Summary of forest plot illustrating the hazard ratios (HRs) and 95% confidence intervals (CIs) for master regulators significantly associated with overall survival in the TCGA cohort. FOXM1, STAT3, and NF-κB1 were associated with poor prognosis (HR > 1), while GATA6 and CDX2 were linked to favorable outcomes (HR < 1). These findings were adjusted for confounding clinical variables in multivariate Cox regression models. The red dashed line indicates the null value (HR = 1), where no effect is observed.