Background: Cervical cancer (CC) is one of the most common gynecology malignancies and has a dismal survival outcome. The prognostic value of pyroptosis and its role in the regulation of immune metabolism in CC remain unclear.
Methods: Two independent CC cohorts collected from public databases were integrated for unsupervised cluster analysis. All CC cases were assigned to different subsets based on the pyroptosis-related genes (PRGs). The differentially expressed genes (DEGs) between different subclusters were included in stepwise Cox regression for the risk classifier establishment. Next, single-cell sequencing analysis was conducted to explore the cellular location of each model gene. The CIBERSORT algorithm was applied to estimate immunocytes infiltration. Finally, a series of functional experiments were performed to detect the role of CDH3 in CC.
Results: Based on the 52 PRGs, the combined CC cohort was clustered into two subsets (C1 (n = 259) and C2 (n = 242)). Survival and Cox regression methods were used to create a pyroptosis-based risk classifier including four PRGs (PEG3, FSCN1, CDH3 and SLC2A1). For the immune environment in CC, the high-risk group had a lower infiltration level of B cells, memory-activated CD4 T cells and CD8 T cells and a higher infiltration abundance of neutrophils. The expression pattern of model genes was confirmed in CC cell lines by PCR assay. Furthermore, we observed that knockdown of CDH3 could suppress CC cell proliferation.
Conclusion: Our project could offer promising reference for prognosis assessment, immune metabolism prediction and clinical decision-making of patients with CC.
Keywords: bioinformatics; cancer immunotherapy; cervical cancer; immune responses; pyroptosis; tumor biomarkers.
© 2023 John Wiley & Sons, Ltd.