Abstract:
Objective To explore the predictive role of pathological risk score of gastric cancer (PRSGC) in the prognosis of gastric cancer patients, and to further investigate the correlation between PRSGC and tumor biological behavior.
Methods Clinicopathological data were retrospectively collected from 1 120 patients with gastric cancer who underwent surgical resection at Zhongshan Hospital, Fudan University, between April 2006 and November 2019. PRSGC was calculated using the previously established DeepRisk model, and patients were divided into high and low PRSGC groups (560 patients per group) based on the median PRSGC value. The proportions and spatial distances of immune cell subsets in gastric cancer tissues were assessed between the two groups. Data from stomach adenocarcinoma samples in The Cancer Genome Atlas was integrated to analyze PRSGC-associated genetic mutations and signaling pathways.
Results The survival rate of the high PRSGC group was lower than that of the low PRSGC group, and the recurrence rate was higher than that of the low PRSGC group (P<0.000 1). Compared with the low-PRSGC group, the high-PRSGC group exhibited increased infiltration of regulatory T cells (Tregs) and neutrophils within the tumor region (P<0.05), while the infiltration levels of natural killer T cells and cytotoxic T cells were markedly decreased (P<0.05). Spatial analysis revealed that a shortened spatial distance between CD4+ T cells and Tregs was closely associated with poor prognosis in the high-PRSGC group (P=0.03). Furthermore, in the high-PRSGC group, the number of TIM3+ cells showed a positive correlation with the infiltration levels of Tregs (r=0.69, P=0.01), CD8+ T cells (r=0.61, P=0.04), and CD4+ T cells (r=0.67, P=0.02). Gene set enrichment analysis indicated that a high PRSGC was associated with the activation of pathways related to actin cytoskeleton regulation, the cGMP-PKG pathway, the Hippo pathway, and tumor proteoglycans.
Conclusion A high PRSGC is associated with an immunosuppressive tumor microenvironment, as well as tumor invasion and metastasis. It effectively predicts the prognosis of patients with gastric cancer and, by doing so, provides a potential theoretical basis for developing individualized strategies that combine targeted therapy with immunotherapy.