In this study, we explored the expression of the 33 PRGs and identified 22 of the 33 PRGs up-regulated and only one of them down-regulated. The expression of PRGs in gastric cancer is relatively active. This was consistent with previous results [25]. Studies have shown that GSDMB is highly expressed in gastric cancer [26–27]. The up-regulated expression of GSDMC, CASP3 and GPX4 may promote apoptosis of cancer cells [28–30]. These studies suggest that the key gene of each component of pyroptosis is differentially expressed in gastric cancer and other cancers. These PRGs have the potential to be used as biomarkers in the diagnosis and treatment of gastric cancer. Inhibition of GPX4 expression can inhibit the vitality of liver cancer cells and induce pyroptosis [30], and up-regulation of NLRP6 can inhibit gastric cancer progression [31]. By increasing or inhibiting the expression of PRGs, inhibiting the vitality of tumor cells and inducing pyroptosis, we can achieve the goal of tumor treatment.
Pyroptosis is a recently discovered programmed cell death that is associated with progression, prognosis, and response to treatment in gastric cancer. Pyroptosis can promote tumor cell death, making pyrolysis a potential prognostic and therapeutic target [6, 19, 32]. The results of the present study revealed global alterations in PRGs at the transcriptional and genetic levels in gastric cancer. Somatic mutation can lead to the occurrence of gastric cancer [33–34]. Our study found that the somatic mutation rate in gastric cancer was 27.02%. Further studies on somatic mutations, such as gene interaction and mutation sites of frequent somatic cell mutation genes, may analyze the role of somatic mutations in the occurrence and development of gastric cancer from the level of tumor genomics, identify different "subtypes" of somatic mutations in gastric cancer and provide a theoretical basis for targeted drug therapy [35–37]. Collectively, these studies reveal that the PRGS involved may provide potential biomarkers for early diagnosis and anti-tumor therapy.
We used the TCGA and GEO databases to clarify the prognostic value of PRGs in gastric cancer. The study generated a signature with 19 PRGs and found it could predict overall survival in patients with gastric cancer. Previous studies have also shown that PRGs can effectively predict the prognosis of patients with lung adenocarcinoma and ovarian cancer [15, 38]. A new signature of PRG (AIM2, PLCG1, ELANE, PJVK, CASP3, CASP6, and GSDMA) was identified to predict prognosis in ovarian cancer [15] and found that it could predict overall survival. Wang Qingqing discovered that decreased NLRP6 expression is associated with a poor prognosis in gastric cancer [31]. Based on the above research, we postulate that PRGs predict the prognosis of tumor patients, which will be the direction of future research and exploration.
The molecular mechanisms underlying the differential expression of PRGs and their potential prognostic impact in gastric cancer are still poorly understood. Unsupervised consensus clustering based on the expression of PRGs, clusters A and B associated with PRGs were identified. Furthermore, we found that patients in cluster A had worse OS compared to the cluster B cohort. No significant distribution difference was found in terms of gender, T stage and N stage among them. Compared to cluster B, cluster A has more patients under 65 years and more deaths. We boldly presume that pyroptosis may be more likely to occur in patients with gastric cancer ≤ 65 years of age, as pyroptosis promotes gastric cancer progression and worsens the prognosis in these patients. This interesting phenomenon forces us to delve deeper into the reasons to design and implement in vivo and in vitro experiments for the molecular mechanisms behind it. First, differences in immune infiltration could be one of the reasons. Compared to cluster B, PRGs cluster A showed a lower infiltration level of T cell and greater activation of B cells and mast cells. This can be attributed to the regulation of immune micro-environment cells such as CD8 T cell in the immune system. The pyroptosis process is accompanied by the secretion of inflammatory factors and immune response. Some studies have found that GSDME increases phagocytosis of tumor related macrophages and the number and function of tumor-infiltrating NK cells and CD8 + T lymphocytes through pyroptosis [39]. This further confirms the correlation between the expression level of PRGs in gastric cancer and immune infiltration, providing a new approach for clinical immunotherapy. Deeper understanding of the role of pyroptosis in regulating the tumor immune micro-environment has expanded understanding of cancer, increased understanding of pyroptosis mechanisms and pathology, and revealed its role in tumor development and treatment. These findings will contribute to the future exploration of tumor immunotherapy based on pyroptosis.
We also performed functional enrichment analysis of PRGs based on GSVA, which revealed that KEGG pathways were mainly different between cluster A and B. Activation and inhibition of different KEGG pathways lead to changes in clinical phenotype and prognosis in two groups of patients with gastric cancer. The results of the differential analysis of gene expression showed that there were 2601 DEGs between the two groups. Functional enrichment analysis of KEGG reveals that these 2601 DEGs were mainly involved in the PI3K-Akt signaling pathway, MAPK signaling pathway and focal adhesion KEGG pathways. These pathways have been correlated with the oncogenesis and progression of gastric cancer [40–42]. When the relevant signaling pathway is activated, it will affect tumor cell activities including cell proliferation, differentiation, cell vitality, cell stress and apoptosis. Significantly, our results using two different analyzes showed that they had many overlapping functional pathways such as DNA replication, RNA degradation, mismatch repair and the cell cycle. All these results do not further confirm the idea that these 33 PRGs play a vital role in the oncogenesis and progression of gastric cancer, but also reflect the consistency and stability of our research from a lateral perspective.
In order to further explore the clinical application value of PRGs, we attempted to build a clinical prognosis model based on PRGs. Reliable and workable nomogram to predict gastric cancer overall survival is clinically valuable and difficult to create. Cox regression analysis was performed to build a prognostic genetic model based on five prognostic PRGs (CASP5, CASP1, CASP8 and GPX4), which could predict the overall survival of patients with gastric cancer with medium to high accuracy. Multivariate cox analyses revealed that CASP5 was the independent factor affecting the prognosis of patients with gastric cancer. The discrimination ability of the final model for overall survival was evaluated using the C statistics, 0.651 for overall survival. Although the accuracy and discrimination of a model with a biomarker may be limited, our result showed that the proposed nomogram provided a more accurate overall survival prediction for patients with gastric cancer than the AJCC TNM-based nomogram [43–44]. A predictive nomogram suggested that the 3-year and 5-year overall survival rates could be predicted relatively well compared to an ideal model across the entire cohort. Considering the graphical results, the 3-year-old overall survival are more persuasive than 5-year overall survival. This suggests that PRGs can be used as biomarkers for assessing the prognosis of gastric cancer.
The limitations of our study included the following: first, as retrospective study has inherent defects such as selection bias. Second, gastric cancer is a complex disease and all kinds of clinical factors such as race, histology and treatment details should be considered to clarify the key role of PRGs in gastric cancer development; however, this information is absent or inconsistently available in public databases. Third, our nomograms have been internally validated using bootstrap validation and lack external validation. Finally, the present study was based on TCGA and GEO data mining; therefore, the protein level of PRGs expression could not be evaluated directly, and the signaling pathways involved in PRGs clusters in patients with gastric cancer need to be verified by in vivo and in vitro experiments.