• Utilizing GO analysis and KEGG enrichment analysis, we delved into the cellular functions and signaling pathways potentially implicated in the differentially expressed genes of cutaneous squamous cell carcinoma.

Cutaneous squamous cell carcinoma (cSCC) is the second most common type of non-melanoma skin cancer (NMSC), accounting for approximately 20% of all skin cancer cases (Gong et al., 2018). In the United States, cSCC affects an estimated one million new patients annually, leading to up to 9,000 deaths each year (Karia et al., 2013). Globally, the incidence of cSCC has been rising steadily, with an annual growth rate of 10%–12% (Rogers et al., 2010). This increase poses a significant public health threat, particularly in regions with high ultraviolet exposure, where the incidence is markedly higher. (Green and Olsen, 2017). While most cases of cSCC are curable through surgical excision, approximately 4% of patients experience poor outcomes due to tumor metastasis or local recurrence (Brantsch et al., 2008). The high incidence of cSCC has resulted in a death toll comparable to that of melanoma (Fitzmaurice et al., 2019), imposing substantial physical and economic burdens on patients, a challenge likely to worsen with the aging population.

Early diagnosis and treatment of cSCC remain challenging. Early-stage cSCC often presents as skin nodules or scaly plaques, which can easily be mistaken for other benign conditions, complicating clinical diagnosis (Stratigos et al., 2020a). Although several studies have explored molecular markers for cSCC, there is currently a lack of highly specific and sensitive markers for early diagnosis, limiting their clinical application (Stratigos et al., 2020b). For metastatic cSCC, traditional treatments such as surgery and radiotherapy have shown limited efficacy in advanced stages (Que et al., 2018). Despite recent advances in targeted therapies, effective therapeutic targets for metastatic cSCC remain elusive and treatment outcomes continue to require improvement (Karia et al., 2013; Burton et al., 2016). Therefore, further investigation into the molecular mechanisms underlying cSCC and the identification of reliable biomarkers for early diagnosis and targeted therapies are essential for improving patient outcomes.

Characteristic genes involved in tumorigenesis and progression is crucial for advancing cancer research. These genes, whose abnormal expression or mutations can serve as diagnostic and prognostic biomarkers, also present potential therapeutic targets. For example, the estrogen receptor (ER) is a characteristic gene in breast cancer, with approximately 70%–80% of breast cancer patients being ER-positive (ER+), who typically respond well to hormone therapies such as tamoxifen (Davies et al., 2011). In the case of cSCC, mutations in genes such as TP53, NOTCH1 and CDKN2A have been implicated in the initiation and progression of the disease (South et al., 2014). However, these mutations are not exclusive to cSCC and are observed in various other cancers. For instance, TP53 mutations are present in approximately 42% of all cancer patients and are closely associated with tumorigenesis, progression and prognosis (Kandoth et al., 2013). Notably, TP53 mutations are also detected in 20%–30% of cases of actinic keratosis, a precursor lesion to cSCC (Padilla et al., 2010). This suggests that the currently identified characteristic genes for cSCC lack both sensitivity and specificity. Therefore, further investigation is required to identify more specific and sensitive characteristic genes for cSCC, along with understanding their molecular mechanisms, to improve early diagnosis and targeted therapy.

High-throughput sequencing technologies, combined with weighted gene co-expression network analysis (WGCNA), has proven effective in identifying characteristic genes in cancer research. High-throughput sequencing enables comprehensive gene expression profiling, while WGCNA identifies disease-related gene modules and hub genes. For instance, in hepatocellular carcinoma, WGCNA and Least Absolute Shrinkage and Selection Operator (LASSO) algorithms have been employed to identify macrophage-related genes and further establish a prognostic model (Wang et al., 2022). While similar approaches have been applied to ovarian cancer to identify fibroblast-related genes and their roles in the tumor microenvironment (Feng et al., 2022). Despite these advances, the application of these techniques to cSCC remains unexplored. In this study, we utilized genome-wide mRNA transcriptomic data from cSCC and normal skin samples available in the Gene Expression Omnibus (GEO) database. By integrating WGCNA and LASSO algorithms, we identified characteristic genes for cSCC. These findings provide a deeper understanding of the molecular mechanisms underlying cSCC.

The tumor immune microenvironment plays a critical role in the initiation and progression of cSCC. Previous studies have demonstrated that tumor-infiltrating lymphocytes in cSCC are closely associated with immune evasion, tumor progression and patient prognosis (Saeidi et al., 2023). Additionally, tumor-associated neutrophils contribute significantly to cSCC growth and immune evasion, thereby facilitating tumor progression (Khou et al., 2020). Investigating the correlation between the expression patterns of characteristic genes and immune cells infiltration provides valuable insights into the molecular mechanisms underlying tumor growth and immune evasion. This approach also offers the potential to identify biomarkers with diagnostic, prognostic, or therapeutic relevance. In the present study, we further explored the relationship between cSCC characteristic genes and immune cell infiltration. Our findings reveal that cSCC characteristic genes are significantly correlated with the infiltration of various immune cell types. These results enhance our understanding of the immune landscape of cSCC.

cSCC, a major subtype of NMSC, is characterized by its potential for local invasion and distant metastasis. Although the overall metastatic rate is relatively low, with approximately 4% of patients developing lymph node metastasis and 1.5% succumbing to the disease (Karia et al., 2013), advanced cSCC remains a significant clinical challenge. Early-stage cSCC can be effectively treated with current therapies, but management of advanced cases continues to pose considerable difficulties. Understanding the characteristic genes involved in cSCC pathogenesis is critical for elucidating the mechanisms underlying its development and progression, and for advancing diagnostic and therapeutic strategies. However, research into the molecular profiles of cSCC remains limited, and systematic investigations to identify and explore characteristic genes are urgently needed.

In the present study, we identified five characteristic genes from three cSCC datasets (GSE45164, GSE42677 and GSE7553) for the first time. While the original study of GSE42677 also identified S100A9 as a DEG, our combined approach, which includes WGCNA, LASSO regression analysis and immunohistochemical staining, further validates that S100A9, along with the other four genes, collectively serves as a set of characteristic genes for cSCC. Due to significant differences in research objectives, analytical methods, sample sizes and sample compositions, the conclusions drawn in our study exhibit notable disparities from those of the original studies.

Among these characteristic genes, ADH1B, CCL27, ID4 and LRP4 were found to be significantly downregulated in cSCC (Figure 5C). Previous studies have highlighted the roles of these genes in the tumorigenesis and progression of other cancers, suggesting that their altered expression may be crucial for the development of cSCC. For instance, polymorphisms in the ADH1B gene, such as rs1229984, have been shown to correlate with a reduced risk of head and neck cancer (Imani et al., 2024). Additionally, ADH1B expression is notably reduced in hepatocellular carcinoma (HCC), with low expression levels associated with poor prognosis (Liu et al., 2020). Although our study does not provide direct evidence linking ADH1B to cSCC prognosis, its downregulation suggests that it may be involved in disease progression, warranting further investigation in clinical cohorts. CCL27, a chemokine involved in immune cell recruitment and lymphangiogenesis, has been shown to be upregulated in several cancers, including breast cancer and melanoma, where it contributes to tumor metastasis by promoting lymphatic spread (Karnezis et al., 2019). However, this study indicates that CCL27 is downregulated in cSCC. Given that cSCC exhibits significantly lower lymph node metastatic potential compared to breast cancer and melanoma, this finding suggests that CCL27 may play a role in the differential lymphatic metastasis capabilities among various cancers. Further research is needed to elucidate the molecular and biological functions of CCL27 in tumor metastasis. The expression of ID4 has been reported to be suppressed by UVB irradiation via DNA methylation, which in turn promotes cell proliferation, migration and invasion, contributing to tumor formation in mouse models of skin cancer (Li et al., 2020). In non-small cell lung cancer (NSCLC), ID4 upregulation is associated with improved prognosis, suggesting its potential role as a tumor suppressor (Rodón et al., 2019). Similarly, ID4 was found to be downregulated in cSCC in our study. It remains unclear whether this downregulation contributes to cSCC progression, and further experimental validation is needed to explore its potential role. LRP4, a member of the low-density lipoprotein receptor family, has been implicated in various cancers. In NSCLC, LRP4 downregulation promotes tumorigenesis and progression by modulating the Wnt/β-catenin signaling pathway (Zhang et al., 2024). In contrast, LRP4 is upregulated in papillary thyroid carcinoma, where it promotes cell proliferation, migration and invasion (Zhou et al., 2018). This contradictory expression pattern suggests that LRP4 may have context-dependent roles in different cancer types, potentially influenced by the tumor microenvironment. However, the prognostic significance of LRP4 in cSCC remains to be determined, and further studies are needed to explore its functional role. In summary, our findings reveal that four of the identified characteristic genes (ADH1B, CCL27, ID4 and LRP4) are consistently downregulated in cSCC, and prior studies suggest their involvement in tumor proliferation, migration and invasion. However, the specific roles of these genes in cSCC require further clinical and experimental validation.

In addition to the four downregulated characteristic genes mentioned, this study also confirmed the upregulation of S100A9 in cSCC (Figure 5C). S100A9 is a member of the S100 protein family, which binds calcium and plays a critical role in the recruitment and adhesion of immune cells such as neutrophils and monocytes, especially in inflammatory responses and inflammation-associated diseases (Wang et al., 2018). Previous studies have demonstrated that S100A9 is upregulated in various cancers and is closely associated with tumor initiation, progression and prognosis (Acharyya et al., 2012; Eisenblaetter et al., 2017). In cSCC, elevated S100A9 expression has been found to bind to Receptor for Advanced Glycation Endproducts on myeloid-derived suppressor cells (MDSCs), activating NF-κB signaling. This interaction promotes the recruitment of MDSCs to the tumor site, which suppresses the immune system’s anti-tumor response, thereby facilitating tumor growth and metastasis (Markowitz and Carson, 2013). Our study also observed a potential positive correlation between S100A9 and MDSC infiltration (Figure 7C). While the p-value was not sufficient to confirm statistical significance, the results partially support previous findings. Interestingly, downregulation of S100A9 expression in cSCC has been shown to inhibit tumor proliferation and migration (Zhang et al., 2021). Furthermore, IL-6 has been clinically associated with poor prognosis in cSCC (Mallardo et al., 2023), thereby underscoring the critical involvement of inflammatory molecules in tumor progression. Future investigations into how these inflammatory mediators collaboratively orchestrate tumor development through synergistic molecular networks will be of paramount importance for advancing therapeutic strategies.

Notably, the intricate tumor immune microenvironment involves not only MDSC-mediated immunosuppression but also dynamic equilibrium among immune cell populations. Our study revealed concurrent infiltration of functionally antagonistic CD8⁺ cytotoxic T-cell and immunosuppressive Tregs within tumor tissues, a paradoxical coexistence that has been previously documented in renal clear cell carcinoma and lung adenocarcinoma (Pan et al., 2020; Meng et al., 2024). Current research demonstrates functional heterogeneity in tumor-infiltrating CD8⁺ T-cell, where many exhibit non-reactive “bystander” phenotypes that challenge the prognostic utility of quantitative assessments alone (Simoni et al., 2018). Moreover, Treg cell-mediated immunosuppression further impairs the functional state of CD8⁺ T-cell (Sakaguchi et al., 2008). These findings collectively suggest that effective immunotherapeutic strategies should adopt dual targeting approaches: enhancing effector T-cell functionality while selectively inhibiting Treg-mediated immunosuppression to achieve durable antitumor responses.

Furthermore, in this study, we established correlations between the cSCC characteristic genes and various infiltrating immune cell populations through bioinformatics analysis (Figure 7C). These findings suggest that these genes may play a role in shaping the cSCC immune microenvironment and promoting immune evasion, a hypothesis supported by previous studies. For instance, S100A9 is highly expressed in neutrophils and monocytes and acts as a promoter by regulating tumor metabolism and the immune microenvironment, thereby facilitating tumor growth and metastasis (Chen et al., 2023). Further research has revealed that tumor-associated neutrophils can suppress CD8⁺ T-cell function and express Programmed Death-Ligand 1, contributing to immune evasion and promoting cSCC progression (Khou et al., 2020). Our study reaffirms the positive correlation between high S100A9 expression and increased neutrophil infiltration in cSCC, consistent with previous reports. While there is substantial evidence supporting the involvement of S100A9 in the tumor immune microenvironment and immune evasion, the roles of CCL27, ID4 and LRP4, genes that also showed strong correlations with infiltrating immune cells, remain inadequately explored. Future research is needed to clarify whether these genes influence the tumor immune environment and to elucidate their specific mechanisms. Such studies will deepen our understanding of cSCC and could uncover novel targets for immunotherapy.

However, this study has several limitations. First, the available datasets that met the analysis criteria were limited, resulting in a relatively small sample size that may not fully capture the heterogeneity of cSCC. Second, although multiple screening methods were employed, the stability and generalizability of the identified characteristic genes require validation in larger, multi-center cohorts. Additionally, the prognostic significance of these genes in cSCC has not been systematically evaluated, which should be addressed in future studies. Future research should focus on expanding sample sizes, incorporating diverse data sources, conducting functional validation experiments, integrating multi-omics data, and performing prospective clinical studies to improve the representativeness and clinical applicability of the findings.

In conclusion, this study employed bioinformatics techniques, including WGCNA, to identify five characteristic genes associated with cSCC: ADH1B, CCL27, ID4, LRP4 and S100A9. Additionally, we demonstrated that CCL27, ID4, LRP4 and S100A9 are correlated with the infiltration of various immune cell types in cSCC. Further investigation into the roles and mechanisms of these characteristic genes in cSCC formation, progression, migration and immune evasion will enhance our understanding of the disease and may provide a foundation for future diagnostic biomarkers and therapeutic targets.