The early stage of cervical cancer has no obvious symptoms, and most patients are in the advanced stage when they see a doctor with symptoms such as vaginal discharge and bleeding [12]. Nowadays, human papillomavirus (HPV) infection is considered a necessary condition for cervical cancer, followed by multiple sexual partners, smoking, early sexual life, sexually transmitted diseases, low economic status, oral contraceptives and immunosuppression [13, 14]. Despite intense screening and various treatment options such as surgery, radiotherapy and chemotherapy, prognosis in CESC remains poor due to high risk of metastasis and recurrence. Additionally, mounting evidence suggests that altered expression of many genes contributes to CESC pathogenesis [15–17]. Therefore, we expect to identify some sensitive and novel biomarkers for diagnosis and prognosis in CESC.
In the complex tumor microenvironment, conventional single-gene assessments are not capable of revealing the complex signal transduction network of tumor-related genes. Meanwhile, WGCNA uses data of thousands of genes to determine interesting gene modules, performing association analyses with various phenotypes. WGCNA results are largely available [18, 19]. Here, gene expression profiling in GSE63514 (28 CESC and 24 noncancerous specimens) was assessed by WGCNA. As shown above, the green module had a significant correlation with CESC. Next, GO and KEGG analyses were performed for this module, and oncogenic signaling pathways were assessed to functionally characterize green module genes.
KEGG enrichment analysis demonstrated that green module genes were enriched in epidermis development, peptide cross-linking, establishment of skin barrier, fatty acid derivative metabolic process, desmosome, serine-type peptidase activity and phosphatidylcholine acyl-chain remodeling signaling pathways. In addition, they contributed to many biological processes, including cell adhesion, proliferation, differentiation and fatty acid metabolism [20–22], which may be related to the fact that cervical squamous cell carcinoma is mostly advanced at diagnosis. The majority of tumours show abnormally induced lipid metabolism enabling them to biosynthesize, elongate and desaturate fatty acids for increased growth [23–25]. Several reports have shown that lipid metabolism in cancer cells and immune cells in the tumor microenvironment highly contributes to immunosuppression, and antitumor immunity could be enhanced by targeting such metabolic pathways [22, 26]. This study suggested that genes in the green module might affect cervical cancer occurrence and progression by interfering with fatty acid metabolism.
Tumor immunotherapy is an anticancer therapy with the goal of activating the immune system in the hope that its own immune function would inhibit tumor tissues [27]. At present, tumor immunotherapy has been shown to have strong antitumor activities in some tumor types, including melanoma and non-small cell lung cancer, and tumor immunotherapy drugs are clinically applied [28–30]. This study confirmed that ALOX12B, KRT78, RHOD, and ZNF750, which are related to tumor progression, affected the prognosis of CESC patients. As shown above, ALOX12B expression was highly associated with macrophage infiltration in cervical cancer, and ZNF750 was correlated with CD4 + T cell infiltration and dendritic cell infiltration, suggesting that these two factors may influence the immune response of patients by controlling immune infiltration.
Next, to investigate the biological functions of hub genes, GSVA was performed. Both P53 and hypoxia pathways have important roles in tumor survival and recurrence [31–33]. The above results showed that these four core genes were mostly related to the P53 and hypoxia signaling pathways. Similarly, many immune-related pathways, such as the IL2 and interferon response pathways, were also enriched in the high-expression groups of these hub genes, suggesting their potential contributions to immune response in CESC.
To our knowledge, this is the first study assessing the relationship between RHOD and CESC. Interestingly, the results showed that RHOD is sensitive to most CESC drugs, especially Dasatinib (a tyrosine kinase inhibitor used to treat leukemia) [34, 35], BI − 2536 (an efficacious Polo-like kinase inhibitor inducing apoptosis in multiple human cancers) [36, 37], and clofarabine (second-generation nucleoside analog potently inhibiting leukemia) [38, 39]. These findings suggest that RHOD may be a new target for developing CESC drugs. However, the clinical significance of RHOD in CESC has not been reported to date, and more research is needed for confirmation.
In conclusion, four hub genes (ALOX12B, KRT78, RHOD, and ZNF750) were shown to be closely related to CESC survival by combining WGCNA, PPI, and other bioinformatics tools. Tumor microenvironment and drug sensitivity analyses examined the potential of immunotherapy and targeted therapy, and GSVA further demonstrated their significant impacts on CESC progression. More detailed studies are needed in the future to fully reveal their roles in CESC pathogenesis and usefulness as diagnostic and/or prognostic markers.