3.1 Differential mRNA expression level of LAYN in pan-cancer and its relationship with the prognosis of HNSCC

We employed TIMER database to investigate the variances of LAYN mRNA expression levels between tumor and normal tissues in a wide range of cancer types. The analysis indicated that LAYN expression was higher in HNSCC, hepatocellular carcinoma, cholangiocarcinoma, and renal clear cell carcinoma than in their respective adjacent normal tissues. In contrast, lower expression of LAYN was observed in bladder urothelial cancer, breast cancer, colorectal cancer, kidney chromophobe, lung adenocarcinoma, prostate cancer, thyroid cancer, and endometrial cancer compared with adjacent normal tissues (Fig. 1A). The increased expression of LAYN in HNSCC was also confirmed in TCGA (P < 0.0001, Fig. 1B). These results implied that the expression of LAYN in pan-cancer was heterogeneous.

Fig. 1

The differential expression of LAYN in diverse tumors and its role in the prognosis of HNSCC. A LAYN expression in different tumor types from the TCGA database was explored with TIMER (*P < 0.05, **P < 0.01, ***P < 0.001). B LAYN expression in HNSCC was explored in TCGA database. C OS survival curves of HNSCC patients with high and low LAYN expression. D DFS survival curves of HNSCC patients with high and low LAYN expression. E OS survival curves of HPV-positive and HPV-negative HNSCC patients. F Forest plot of LAYN expression and survival in common HNSCC subtypes. OS overall survival; DFS disease-free survival

We further examined whether LAYN expression was correlated with the prognosis of HNSCC patients. Survival analyses revealed that high LAYN expression worsened the overall survival (OS) of HNSCC patients (Fig. 1C) (HR = 1.3, P < 0.05), while no difference existed in disease-free survival (DFS) with two curves intersecting between the LAYN-high and LAYN-low groups (HR = 1, P = 1) (Fig. 1D). Additionally, we explored whether HPV infection status was associated with patients’ survival, and found that the survival of HPV-positive patients was significantly better than that of HPV-negative patients (HR = 0.6, P < 0.05) (Fig. 1E). Through forest plot, we found that among common head and neck tumors, the expression of LAYN was only correlated with the survival of the base of tongue cancer patients with statistical difference (HR = 1.11, 95% CI 1.01–1.21, P < 0.05) (Fig. 1F). Therefore, LAYN expression could be an independent prognostic factor that led to the poor prognosis of HNSCC, especially the base of tongue cancer.

3.2 The relationship between LAYN expression and clinicopathological factors in HNSCC

To better understand the relevance and underlying mechanisms of LAYN expression in HNSCC patients, we analyzed the relationship between LAYN expression and clinical characteristics of HNSCC. Using R software to analyze data from TCGA database, we found that no correlation existed between LAYN expression and age, gender, clinical stage, T stage, and tumor grade, respectively (Fig. 2A–E). However, LAYN expression was associated with HPV infection status with statistical significance (P < 0.05), with LAYN expression in HPV-negative patients higher than that in HPV-positive patients (P < 0.01) (Fig. 2F).

Fig. 2

A–F The relationship between LAYN expression and clinicopathological factors in HNSCC patients. ns not significant, *P < 0.05, **P < 0.01, ***P < 0.001

3.3 LAYN was significantly related to immune infiltration in HNSCC

To explore the potential mechanisms by which LAYN influenced prognosis, correlation analyses were conducted between LAYN expression and tumor immunity based on tumor purity, immune cells, and immune checkpoint molecules using TIMER database. As was shown in Fig. 3A, after adjusting for the influence of tumor purity, LAYN expression was positively associated with the infiltrating levels of B cells (P < 0.05), CD4 + T cells (P < 0.001), macrophages (P < 0.001), and dendritic cells (P < 0.001) in HNSCC tumor tissues, while the correlation was not statistically significant with CD8 + T (P = 0.79) and neutrophils (P = 0.12).

Fig. 3

The correlation of LAYN expression with immune infiltration levels in A HNSCC, B HPV-positive HNSCC and C HPV-negative HNSCC patients

We further analyzed the correlation between LAYN expression and immune cells in HPV-positive and HPV-negative HNSCC patients. Overall, the relationship between LAYN expression and immune markers in HPV-positive HNSCC group was less tightly correlated than that in HPV-negative HNSCC group (Table 1). In HPV-positive HNSCC patients, no correlation between LAYN expression and various immune cells was observed (P > 0.05) (Fig. 3B). However, in HPV-negative HNSCC patients, LAYN was positively correlated with infiltrating B cells (P < 0.01), CD4 + T cells (P < 0.001), macrophages (P < 0.001), neutrophils (P < 0.05), and dendritic cells (P < 0.001) in tumor tissues. The correlation between LAYN and CD8 + T cells in HPV-negative HNSCC patients was not statistically significant (P = 0.26) (Fig. 3C) (Table 1). These findings strongly suggested that LAYN played a specific role in immune cell infiltration in HPV-negative HNSCC patients.

Table 1 Correlation analysis between LAYN and immune cells in HPV-related HNSCC through TIMER database3.4 Correlation analysis between LAYN expression and immune marker sets

In order to further explore the role of LAYN in the tumor immune process and its potential as a target for immunotherapy, we analyzed the correlation between LAYN expression and various immune marker sets of immune cells in HPV-related HNSCC through GEPIA database. All the markers were used to characterize immune cells, including markers of B cells, CD8 + T cells, CD4 + T cells, monocytes, macrophages, neutrophils, NK cells, dendritic cells and different functional T cells (Th1, Th2, Tfh, Th17, Treg, and exhausted T cells).

We found that the expression level of LAYN was associated with 30 out of 57 immune cell markers in HNSCC, 2 out of 57 in HPV-positive HNSCC, and 31 out of 57 in HPV-negative HNSCC with statistical significance (Table 2). The elevated LAYN expression level was associated with multiple immune cell markers in HNSCC, especially HPV-negative HNSCC, including markers of monocytes, TAMs, M1 macrophages, M2 macrophages, neutrophils, dendritic cells, and exhausted T cells (P < 0.05). However, In HPV-positive HNSCC patients, nearly no correlation between LAYN expression and various immune cell markers was observed, which was in consistent with our aforementioned results. Specifically, CCR8 of Treg cells, revealed strong correlation with LAYN in HNSCC, HPV-positive HNSCC and HPV-negative HNSCC patients (P < 0.01). These results indicated that LAYN might act as an immunomodulatory gene that involved in a variety of biological functions, including immune cell activation, antigen presentation, macrophage polarization and regulation of the T cells in HPV-related HNSCC.

Table 2 Correlation analysis between LAYN expression and immune marker sets of various immune cells in HPV-related HNSCC through GEPIA database

Interestingly, our results also suggested a close relationship between LAYN expression and immune cell markers of M2 macrophage in HPV-negative HNSCC (P < 0.001). Moreover, high expression of LAYN was also associated with CTLA4 and TIM-3 (P < 0.05) (Table 2), which could act as crucial genes in regulating T cell exhaustion. These results further confirmed that LAYN played a vital role in immune escape and immunomodulation.

3.5 GO and KEGG enrichment analysis of LAYN target genes

We performed GO and KEGG analyses to identify LAYN-related pathways. As was shown in Fig. 4A, LAYN-related gene groups were predominantly enriched in exosomes, extracellular regions, cell membranes, plasma membranes, and membrane components based on analysis of cellular component (GO-CC). In terms of Biological Process (GO-BP), the most enriched processes were extracellular matrix reconstruction, collagen fiber reconstruction, cell adhesion, cell protein metabolism process, and positive regulation of cell migration (Fig. 4B). The most enriched terms for molecular function (GO-MF) were extracellular matrix construction, collagen binding, and protein binding (Fig. 4C). KEGG pathway analysis suggested that LAYN co-expression network was mainly enriched in ECM receptor interaction, adhesive plaque, PI3K-Akt signaling pathway, HPV infection, and cancer-related pathways (Fig. 4D).

Fig. 4

A–C GO and D KEGG Enrichment analysis of LAYN-related pathways

3.6 The expression and prognosis patterns of HPV and LAYN in HNSCC patients

We evaluated the expression of HPV and LAYN in a total of 61 HNSCC tissues through SP immunohistochemistry. Baseline characteristics of HNSCC patients are presented in Table 3. The judgment standard for histochemical results was based on the presence of brown-yellow granules in the cell membrane/cytoplasm of HNSCC cells. Five high-power fields of vision (× 400) were randomly selected, and 100 cells were counted per field to determine the percentage of positive cells (PP):  < 5% = 0 points, 5–25% = 1 point, 26–50% = 2 points, 51–75% = 3 points, and > 75% = 4 points. Staining intensity (SI) was scored as follows: no staining = 0 points, light yellow = 1 point, dark yellow = 2 points, and brown = 3 points. The final score was determined by multiplying the number of positive cells and the intensity of cell staining: total score of 0, 1–4 points, 5–8 points, and 9–12 points corresponded to negative, + , +  + , and +  +  + , respectively.

Table 3 Clinical and pathological features of HNSCC patients in the study

In 61 cases of HNSCC, 14 cases (22.95%) were HPV-positive, and 42 cases (68.85%) were LAYN-positive. All slides were reviewed by two investigators who recorded the percentage of cells positive for each stain (0 to + 4) and the intensity of stain (0 to + 3) (Fig. 5A).

Fig. 5

A Immunohistochemistry of the expression levels of HPV and LAYN in HNSCC patients. B OS survival curves of HPV-positive and HPV-negative HNSCC patients. C OS survival curves of LAYN-positive and LAYN-negative HNSCC patients. D Correlation analysis between HPV and LAYN expression levels

In addition, we explored whether the expression levels of LAYN and HPV were correlated with the survival of patients with HNSCC. Using Kaplan–Meier methods, we found that HPV-positive group was correlated with prolonged OS of HNSCC patients (HR = 2.364, 95% CI 1.031–5.420, P = 0.042). In the meantime, a favorable prognosis was observed in reduced LAYN expression for HNSCC patients, which was in accordance with our previous results (HR = 2.508, 95% CI 1.075–5.851, P = 0.033) (Fig. 5B, C). Correlation analysis showed that the expression level of HPV was significantly negatively related to the expression level of LAYN (r = − 0.707, p < 0.0001).