Investigation for genomic and molecular biomarkers which possess prognostic importance, including for the prognosis of HPV-related cancer, is still underway. The need for the identification of specific prognostic factors has emerged in HPV-related OPSCC, which includes a subgroup of patients with aggressive behavior, although the majority show favorable treatment responses (18). In this study, risk stratification of HPV-related HNC and CC was performed using the HPPI scoring system developed in our previous study (17). Of the identified DEGs, the expression of the four upregulated and five downregulated target genes was compared with the normal and HPV (-) cohort. A particularly noteworthy finding was that both HPV (-) and HPV (+) high-risk groups in HNC showed similar patterns of expression of the target genes, and different expression patterns only appeared in the HPV (+) low-risk group (Fig. 4). In the case of de-escalating treatment trials in HPV-related OPSCC, as part of an effort to reduce the toxicity caused by chemoradiation and to perform less invasive surgery for improving the quality of life (18, 19), it is important to accurately select the low-risk group by risk stratification of HPV (+) HNC. Our study can be instrumental in these instances as it can provide genomic data for the development of patient-specific treatment, which are important in both cancer biology and other clinical applications.
The extracellular matrix (ECM) is an essential milieu of tissues, consisting of a dynamic network of biochemical components (20, 21). ECM contributes to the interaction of cells with their microenvironments, which plays a crucial role in regulating cellular behavior and maintaining organ homeostasis. It is known that the disruption of these control processes can cause diseases such as cancer, thus, the expression of ECM-related genes has been studied (21, 22). It has been recently reported that the expression of ECM genes differs according to HPV association in OPSCC and with high-grade dysplasia and invasive cancer (23). In this current study, we focused on laminin genes LAMB1 and LAMC1, and their receptors, integrin genes ITGA5 and ITGB1 among 30 upregulated DEGs that participate in ECM-receptor interaction in the KEGG pathway enrichment analysis (Table 2) (22). Laminin is a major component of the basement membrane and is essential for the adhesive interaction between the cell and the basement membrane; laminins bind laminin-binding integrin heterodimers (24). LAMB combines with LAMC and together they initiate cell assembly to promote cancer invasion and gene expression in various invasive cancers, including LAMB1 gene expression in head and neck, liver, and uterine endometrial cancers, and LAMC1 gene expression in gastric cancer (25–27). Integrin subunits α5 and β1 primarily conjoin to form α5β1 heterodimers, and ITGA5 and ITGB1, the genes encoding α5 and β1, respectively have been suggested to be associated with tumor aggressive behavior, such as cancer progression and treatment resistance (28, 29). In our study, an increase in ITGA5 and ITGB1 expression was characteristically identified in high-risk HPV (+) and HPV (-) groups in HNC, and was associated with the worsening of overall survival in both HPV-related HNC and CC. Therefore, we postulate that LAMB1, LAMC1, ITGA5, and ITGB1 genes may be potential biomarkers for high risk and poor prognosis in HNC and CC, especially in HPV-related cancers.
Other components of the tumor microenvironment, including the expression of genetic markers, and distribution of immune cells and molecules that regulate cellular function, have been studied for targeted immunotherapy (30, 31). Of 144 downregulated DEGs in our study, five were CD3D and CD3E as T cell markers, CD8 + T cell-specific marker CD8B, LCK, and ZAP70, which participates in primary immunodeficiency and the T cell-receptor signaling pathway in the KEGG pathway analysis (Table 2) (30). LCK and ZAP70 are protein tyrosine kinases, and ZAP70 activated by LCK plays an important role in initiating intracellular signaling pathways related to T cell antigenic receptor in the early stages of T cell activation (32, 33). A meta-analysis of several solid tumors revealed that greater infiltration of CD3 + and CD8 + T cells had a positive effect on survival, and a recent report on HNC revealed that more tumor-infiltrating immune cells were associated with better prognosis (31, 34). An increase in T cell infiltrates, which emphasizes the role of immune system, was suggested as an important prognostic determinant contributing to tumor suppression, especially in HPV-related OPSCC (35). In our study, contrary to ECM genes, the expression of T cell-related genes was significantly higher in the low-risk HPV (+) group and lower in the high-risk HPV (+) and HPV (-) groups in HNC, and higher expression improved overall survival in HPV-related HNC and CC. Therefore, in addition to ECM genes, T cell-related genes CD3D, CD3E, CD8B, LCK, and ZAP70 could be useful biomarkers for risk stratification of HPV-related cancer.
We also performed correlation analyses between the selected target genes in the HPV-related HNC and CC groups. A positive correlation was observed between DEGs with the same expression pattern, while a negative correlation was observed between upregulated ECM genes and downregulated T cell-related genes. It has been proposed that cellular components, including immune cells in the tumor microenvironment, contribute to the regulation of ECM dynamics (21). A study on HNC that investigated the tumor ecosystem with a single-cell transcriptome proved that heterogeneous cellular components, including T cells, influence the expression of ECM markers related to tumor migration, such as laminin, by active interaction (36). This study also suggests the possibility that different ECM pathways and immunity within the tumor microenvironment are complexly intertwined with cancer biology, and require further study.
In summary, this study contributes to our understanding of the tumor biology of virus-related cancers, and suggests biomarkers for risk stratification in the HPV (+) group by conducting an integrated bioinformatics analysis on the association between DEGs and prognosis of HPV infection in CC and HNC. However, we only analyzed the upregulation and downregulation of genes and their association in selected pathways using the cohort in the TCGA database, and the lack of experimental verification of the expression patterns of the target genes in cancer tissues requires future investigation.