HNSCC is a solid tumor that has high immune infiltration, aggressiveness and heterogeneity[15, 16]. HNSCC has conventionally been treated with surgery, radiotherapy, and chemotherapy. However, these therapies are not generating satisfactory results, and patients are in desperate need of novel therapies to improve their prognosis. Immunotherapy is a relatively recent and popular treatment option[17]. HNSCC is a heterogeneous disease, featuring differences in expression profiles, immune infiltration, and mutations, implying that immunotherapies could be successful[18]. Therefore, studying the molecular mechanisms and processes underlying HNSCC development and identifying appropriate biomarkers are critical for developing new treatment targets and improving patient outcomes.
In recent years, RNA molecules have attracted much attention due to their diverse variety of functions and crucial role in virtually every metabolic pathway and process. On the other hand, their ability to establish intricate networks of interactions makes it difficult to comprehend and characterize their modes of action. ceRNA networks are newly recognized regulatory mechanisms implicated in the initiation and progression of human malignancies that are being rapidly explored by numerous preclinical and clinical studies[19]. A complete HNSCC ceRNA network based on immunology and prognosis has not yet been revealed. We began by identifying IRGs and conducting GO and KEGG pathway analysis in this work. Receptor ligand activity, collagen-rich extracellular matrix, cell chemotaxis, and the cytokine–cytokine receptor interaction were the most prominently enriched terms in the GO and KEGG results for the IRGs. Then, we identified upstream mRNAs and lncRNAs of the IRGs to construct an immune-related ceRNA network. Finally, a prognostic regulatory axis (LINC00052/hsa-miR-148a-3p/PLAU) was found, and we further researched its correlation with immune infiltrates. The key genes of the axis have been studied in different cancers and HNSCC according to articles published in PubMed [20].
Evidence accumulated over the past decade shows that lncRNAs are widely expressed and have a significant impact on a variety of physiological and pathological processes, and their potential role as biomarkers has been widely discussed. Indeed, increased levels of LINC00052 are implicated in the formation, development, and metastasis of a variety of cancers[21, 22]. Likewise, in HNSCC, a recent study indicated that LINC00052 sponged miR-608 to control epidermal growth factor receptor (EGFR) expression, thereby encouraging advancement both in vitro and in vivo[23]. Upon further research, we found that LINC00052 was mostly located in the cytoplasm and highly expressed in HNSCC tissues with poor prognosis, in line with past studies. hsa-miR-148a-3p, whose family has already been proven to be biomarkers closely related to hepatocellular carcinoma (HCC)[24], COAD[25], acute myeloid leukemia (AML)[26], esophageal squamous cell carcinoma (ESCC)[27] and gastric cancer (GC)[28], is directly linked to vascular invasion. Ashizawa et al. indicated that Overexpression of miR-148a-3p in colorectal cancer reversed IFN-induced PD-L1 expression on the tumor cell surface, which was functionally connected to lower T-cell apoptosis in the TME[29]. Simultaneously, Zhou et al. showed that low hsa-miR-148a-3p expression was linked to naive B cells and poor prognosis in HNSCC and that hsa-miR-148a-3p may operate like a sponge to modify PI3K/Akt signaling (Zhou et al., 2021). Our findings support previous research that hsa-miR-148a-3p has tumor-suppressing properties.
PLAU encodes a serine protease that can convert plasminogen to plasmin, degrade the tumor-surrounding stroma and promote the migration of cells to distant organs, resulting in tumor cell invasion, migration, and homing to distant organs[30]. To date, numerous investigations have established that PLAU was significantly enhanced and participated in the metastasis of HNSCC[31–33]. Many human malignancies, including HNSCC, are influenced by abnormal DNA methylation, particularly promoter hypermethylation and methylation inducing transcriptional suppression of tumor suppressor genes[34]. In our study, we discovered that in HNSCC tissues, the level of PLAU methylation was much lower than that in normal tissues, implying that low DNA methylation promoted PLAU expression. Furthermore, aberrant methylation of cg04939496, cg23837109, cg00303111, and cg12858416, which mainly fell in the untranslated region (UTR) and TSS1500 region, resulted in the upregulation of PLAU in HNSCC.
Immune infiltration in the TME is a complex process that involves numerous subsets of lymphocytes infiltrating tumors and interacting with tumor cells and/or with one another via diverse networks[35]. Cancer growth, angiogenesis, and metastasis are all affected by immune cells that infiltrate the tumor, according to numerous studies. A combination of tumor-infiltrating lymphocytes (T lymphocytes, B lymphocytes, and natural killer (NK) cells) and cells of the myeloid lineage (macrophages, neutrophils, dendritic cells, and MDSCs) makes up the immunological component of the HNSCC TME[1]. In earlier research, 56% of HNSCC patients showed reactivity against at least one tumor-associated antigen (TAA). Patients with TAA responses had a higher percentage of CD4 + T cells infiltrating their tumors, compared to tumors with no TAA response, which predominantly had CD8 + T cells[36]. B cells play an important role in the adaptive immune system, responding to infected cells or tumor cells and regulating cellular immune responses both positively and negatively[37]. By using the CIBERSORT algorithm, we observed that 11 types of immune cells showed significantly different abundances in HNSCC (p < 0.05). Survival analysis showed that only B cells were associated with a positive prognosis in HNSCC. The findings of our research matched those of the prior experiments. PLAU was chosen for further investigation. In the above tests, we consistently observed a higher expression of PLAU associated with poor outcomes in HNSCC. Another important finding of this study is that PLAU expression was significantly associated with the infiltration of diverse immune cells in HNSCC. We found that PLAU is important for regulating immune cell infiltration in HNSCC, with particularly strong negative effects on CD8 + T cells, memory activated CD4 + T cells, follicular helper T cells, plasma B cells, memory B cells, and Tregs and positive effects on memory resting CD4 + T cells, M0 macrophages, resting mast cells, and NK cells resting. Numerous studies have revealed that T cells CD8+, Tregs, B cells naïve, T cells follicular helper, and T cells CD4 memory activated were associated with improved outcomes in HNSCC patients. In addition, in a pan-cancer analysis, high PLAU expression was consistently associated with monocyte infiltration in the TME[38]. These studies demonstrate that PLAU may mediate the proportions of infiltrating immune cells, indirectly leading to immune escape.
Immunological checkpoints, as regulatory signals, play a critical role in maintaining self-tolerance by modulating T-cell function and defending host tissues against indirect immune-mediated damage; therefore, tumors can evade attack by immune cells[10]. Immune checkpoint inhibitors (ICIs), such as those targeting PD-L1, PD-1, and cytotoxic T lymphocyte-associated antigen-4 (CTLA-4), have been investigated across multiple tumor types and have shown significant benefits in clinical studies[1]. ICIs, on the other hand, have limitations in HNSCC patients. The efficiency of ICIs is determined by a variety of factors, including the immunological TME, and only a few indicators correctly predict patient prognosis. It is necessary for us to find reliable predictive biomarkers. Certain biomarkers, including those found by TIDE, have been shown to be more accurate at predicting the fate of melanoma patients treated with first-line anti-PD-1 or anti-CTLA-4 antibodies than other biomarkers, such as PD-L1 levels and tumor mutation burden (TMB)[13]. We discovered that increased PLAU expression in HNSCC patients may be associated with resistance to immune checkpoint blockade therapy by TIDE analysis. In a pan-cancer study, high PLAU expression was consistently associated with important immune response checkpoints such as PD-L2 and CD276/B7-H3[38]. In our investigation, we discovered a strong correlation between high PLAU expression and important checkpoints, such as CD274 and PDCD1LG2. PD-L1 and PD-L2 are immunosuppressive proteins encoded by CD274 and PDCD1LG2, respectively, that restrict CD4 + and CD8 + T lymphocytes from generating cytokines and engaging in cytolytic activity[39, 40]. Monoclonal antibodies directed against PD-L1 and PD-L2 are currently being utilized in clinical trials for patients with advanced solid tumors with FDA permission[41]. The results of this study could further explain the mechanism underlying the relationship between poor prognosis and high PLAU expression in HNSCC.
The study and its findings are limited by several factors. First, this study relied on bioinformatics analysis and did not use clinical samples to test the signature's accuracy. It is still unclear how these genes are regulated in HNSCC, and further study will be needed to determine their biological functions both in vitro and in vivo.