Skip to main content
Download PDF

A model of basement membrane-related regulators for prediction of prognoses in esophageal cancer and verification in vitro

BMC Cancer volume 25, Article number: 696 (2025) Cite this article

Abstract

Emerging evidence suggests the importance of basement membrane components in cancer metastasis; however, their specific roles in esophageal carcinoma remain underexplored. To investigate this, we analyzed 152 esophageal cancer and 11 normal esophageal tissue samples, identifying basement membrane-related prognostic signatures through differential gene expression profiling and Least Absolute Shrinkage and Selection Operator regression. A six-gene panel (LAMC2, GPC2, AGRN, ITGA3, LAMA3, and LOXL4) demonstrated robust predictive capacity, which we subsequently integrated with clinical features via nomogram modeling to predict overall survival. Our computational analyses revealed distinct tumor microenvironment immune cell profiles and chemotherapeutic drug sensitivities across risk strata. We performed an immunohistochemical assay to confirm increased tumor tissue expression, thereby reinforcing the clinical relevance of these biomarkers. Experimental validation using KYSE-150 esophageal squamous carcinoma cells demonstrated that while LAMC2 knockdown attenuated cellular migration, AGRN, GPC2, ITGA3, LAMA3, and LOXL4 suppression enhanced migratory capacity. Proliferation assays further revealed increased growth rates upon GPC2, ITGA3, and LAMA3 expression inhibition. Our results established a basement membrane-derived risk model for esophageal carcinoma and revealed the roles of the model genes in tumor progression regulation. This model advances prognostic stratification and provides insights into therapeutic targets.

Peer Review reports

Introduction

Globally, 85% of esophageal malignancies are squamous cell carcinomas, particularly in areas with high incidence [1]. According to 2020 data, the incidence rates of esophageal carcinoma (ESCA) remain high, especially in China [2]. Although surgical and medicinal advances have improved the overall prognosis of individuals with ESCA in recent decades, overall survival (OS) remains low [3]. The five-year survival rate of ESCA is approximately 20% [4]. Therefore, there is an urgent need to develop biomarkers that can substantially improve ESCA prognosis and management.

The basement membrane (BM), the extracellular matrix (ECM) that lies beneath epithelial and endothelial tissues, is mainly composed of laminin, collagen, and integrin and helps protect against cancer cell invasion [5, 6]. The ECM components play a crucial role in shaping the immune microenvironment of human cells, and their dysfunction is closely related to the occurrence and development of several diseases such as cancer [7, 8].

BM is essential for cell differentiation, polarity, migration, and survival[9,10,11], and abnormal production and conversion of BM proteins are the main pathogenic features of cancer [12,13,14]. The BM is an important barrier between non-invasive and invasive forms of cancer [15]. Approximately 90% of cancer-related deaths are caused by metastatic disease. Metastases, in which cancers infiltrate tissue, lymph, and blood arteries and spread to distal tissues, are thought to involve the BM [16], and recent research has identified a number of BM-related genes (BMRGs) [17]. The expression and turnover of BMRGs are closely related to bladder cancer, hepatocellular carcinoma, and breast cancer [18,19,20]. Thus, more detailed knowledge of the role of BMRGs in metastatic ESCA may lead to improved prognosis and treatment.

Notably, few studies have explored the relationship between BMRGs and the progression of ESCA. To understand the role of BMRGs in the progression of ESCA, we investigated immune cell infiltration and tumor mutational burden (TMB) status, performed drug analysis, analyzed protein expression using immunohistochemistry (IHC), and verified the function of the genes through a series of cellular experiments. The findings of this study are expected to provide novel information for developing effective treatment options for patients with ESCA and constructing prognostic models for predicting ESCA outcomes.

Materials and methods

Data collection

We downloaded the mRNA expression, clinical, and tumor mutation data of the esophageal cancer dataset from the University Of California Santa Cruz. The ESCA dataset contained 162 human esophageal tissue samples, of which 151 were cancerous (included 71 adenocarcinoma samples, 79 squamous cell neoplasms samples, and one serous neoplasm sample) and 11 were normal. Tumor samples were randomly assigned to the training and test sets in a 1:1 ratio. Table 1 show the clinical data of the samples. Through literature retrieval [17], and query BMRG pool (https://bmbasedb.manchester.ac.uk/), we obtain 223 BMRGs. The ESCA data were extracted and processed using R (Table 1).

Table 1 Clinical data of training set and validation set samples
Full size table

Selection of BMRGs

We used the"DESeq2"[21] (| log2 fold change (FC) |≥ 1, P < 0.05) package to identify the differentially expressed genes (DEGs) between esophageal cancer and control samples. We used the"pheatmap"package (https://CRAN.R-project.org/package=pheatmap) to generate the heat and volcanic maps to visualize the results. DEGs and BMRGs that intersected were selected for subsequent analysis.

Construction of the prognostic signature

Statistical significance was set at P < 0.05. COX and least absolute shrinkage and selection operator (LASSO) analyses were performed to assess the relationship between 92 BMRGs and survival outcomes in clinical data, and BMRGs with prognostic potential were identified. The"survival"package was used to perform a Cox regression analysis. The"glmnet"[22] program picked the genes, and the risk score was produced by applying the LASSO. Median risk scores of the training set were used to sort tumor samples into high and low-risk groups. The"survival"(https://CRAN.R-project.org/package=survival) and"survminer"packages (https://CRAN.R-project.org/package=survminer) were used to construct Kaplan–Meier curves. We used the"timeROC"package [23] to visualize the time-dependent receiver operating characteristic (time-ROC).

Establishment and verification of a predictive nomogram

Receiver operating characteristic (ROC) curve analysis was used to evaluate the prognostic prediction of several parameters. Furthermore, the"rms"package (https://CRAN.R-project.org/package=rms) was used to create a nomogram, and their predictive efficiency was assessed using decision curves and calibration plots. The"ggpubr"package (https://CRAN.R-project.org/package=ggpubr) was used to explore the distribution of risk scores in the clinical features.

Functional enrichment analysis

Gene ontology (GO) and the Kyoto Encyclopedia of Genes and Genomes (KEGG) were used to analyze the functional roles and pathway information [24, 25].

Establishing a protein–protein interaction network

To gather data on protein–protein interactions (PPIs), we uploaded BMRGs to the “Search Tool for Retrieval of Interacting Genes/Proteins” (https://cn.string-db.org/), a widely used database for predicting and visualizing PPI networks.

Immune infiltration analysis and TMB

CIBERSORT [26] were used to ascertain the degree of immune cell infiltration. The TMB and mutation frequencies of the samples were obtained from the ESCA dataset using a Perl script. Additionally, for analysis, all samples were split into distinct groups based on median TMB, and then coupled with information on the survival of the patient to plot the TMB survival curve. Waterfall plots were used to visualize gene mutations. Furthermore, we evaluated the prognosis of the combined risk group based on TMB to assess its impact on survival.

Medication analysis

“Genomics of Drug Sensitivity in Cancer database” (https://www.cancerrxgene.org/) was used to compare sensitivity to paclitaxel, MG132, oxaliplatin, and docetaxel of patients in different risk groups.

Cell cultures

A human esophageal squamous cell carcinoma (ESCC) cell line, KYSE- 150, was obtained from Shanghai Biological Technology Co., Ltd. Cells were cultured using complete RPMI 1640 (GIBCO, USA) medium containing 10% fetal bovine serum (FBS, VivaCell, CHINA) and placed at 5% CO2 and 37 ℃. The liposome method (M5 Super Lipo 3000 Transfection Reagent, XinKailai CHINA) was used to transfect small interfering RNA (siRNA) into the cell lines.KYSE- 150 (4 × 105) cells were cultured overnight in a six-well plate, and siRNA was transfected into the cells using Lipofectamine 3000 reagent according to the manufacturer's instructions when the cell density reached 40%. Table 2 lists the siRNA (Tsingke, CHINA) sequences.

Table 2 Sequences of the siRNA used in this study
Full size table

RNA extraction and quantitative PCR

An appropriate amount of TRIzol lysate (Genesand CHINA) was added to the cell culture dishes or pores to extract RNA, and cDNA was prepared by reverse transcription. System 2 × SYBR Green qPCR Premix (XinKailai CHINA) was used for Real-Time Quantitative Reverse Transcription PCR (RT-qPCR). The standard conditions for PCR amplification are listed in Table 3. The specific primer sequences (Tsingke, CHINA) are listed in Table 4. We used relative quantification to evaluate the relative mRNA expression by calculating 2−ΔΔCt.

Table 3 Standard conditions for the PCR amplification reactions
Full size table
Table 4 Primer sequences of all target genes and the reference gene (GAPDH)
Full size table

CCK8 assay

KYSE- 150 cells (5 × 103) were seeded into 96-well plates, CCK8 solution (Dojindo Laboratories) was added, and the cells were left for 2 h. The OD450 value was measured at 24, 48, 72, and 96 h using a microplate reader, and a proliferation curve was drawn.

Scratch assay

KYSE- 150 cells (5 × 105) were seeded into 6-well plates. When the cell fusion reached 90%, it was scraped using a pipette tip (200 µL). The scratched areas were measured at 0, 12, and 24 h under an inverted microscope.

Transwell migration assay

KYSE- 150 cells (5 × 104) were added to the upper chamber (200 μL serum-free 1640 medium) of a cell culture insert and a complete 1640 medium was added to the lower chamber (600 μL 20% FBS 1640 medium). After 48 h, the migratory cells were fixed with 4% paraformaldehyde and stained with 1% crystal violet. Subsequently, five fields of view were selected in a random manner under the inverted microscope for photo analysis.

IHC assay

Paraffin-embedded specimens of esophageal tissue were collected from patients with surgical resection at the Affiliated Panyu Central Hospital, Guangzhou Medical University. A total of 14 paraffin-embedded esophageal tissue specimens from patients undergoing surgical resection from 2023 to 2024 were selected to assess the expression levels of the six BMRGs. Pathological sections were diagnosed by pathologists and divided into two groups: esophageal squamous cell carcinoma groups (n = 7) and normal esophageal epithelial tissue groups (n = 7). All patients strictly met the diagnostic criteria recognized by international or professional societies and did not receive any radiotherapy or chemotherapy before surgery. Paraffin-embedded tissue sections were dewaxed using an environmentally friendly dewaxing agent and hydrated using an alcohol gradient. For antigen retrieval, tissue slices were immersed in ethylenediamine tetra-acetic acid antigen retrieval solution (BIOSSCI, CHINA) at pH 9.0, boiled, and placed in a pressure cooker. The pressure cooker continues to emit gas for 1 min and 30 s, then stops firing, the pressure was released, the sections were cooled to room temperature, washed three times in Tris-buffered saline, blocked with 10% goat serum, and treated overnight with primary antibodies against agrin (AGRN; 1:100, SAB, 37,092), glypican 2 (GPC2; 1:500, SAB, 47,512), laminin subunit alpha 3 (LAMA3; 1:100, SAB, 36,945), laminin subunit gamma 2 (LAMC2; 1:9600, SAB, 41,766), integrin subunit alpha 3 (ITGA3; 1:300, SAB, 47,660), and lysine oxidase-like 4 (LOXL4; 1:300, SAB, 36,389). The sections were incubated with the corresponding secondary antibodies (Goat Anti Rabbit; 1:2000, Abcam, ab205718) for 45 min on the second day and then visualized using 3,3′-diaminobenzidine as the chromogenic agent, counterstained with hematoxylin, and sealed.

Statistical analysis

The R version 4.2.3 program and associated packs were used to perform statistical analysis and graphical visualization. The GraphPad Prism (Version 10.002) was conducted to perform Unpaired Student's t-test and one-way analysis of variance. P < 0.05 was considered significant.

Results

Initial selection of BMRGs

Figure 1 shows the flowchart of the study. First, we performed a difference analysis on the ESCA dataset. As shown in the volcano and VENN plots (Fig. 2A, B), we obtained a total of 3429 DEGs, which intersected with 223 bmrgs to obtain 92 BMRGs, suggesting an association between BMRGs and ESCA.

Fig. 1
figure 1

Flow chart

Full size image
Fig. 2
figure 2

Identification of DEGs among BMRGs genes. A Volcano plots showing the differential expressed genes (DEGs) in the entire set. B VENN plot showing the intersection of DEGs and basal membrane-related genes (BMRGs)

Full size image

Construction of the prognostic signature

To further screen and obtain BMRGs with prognostic value, Cox analysis of 92 BMRGs was performed, and the results were combined with clinical information (survival status and time), which showed that six of them were risk variables affecting the survival of patients with ESCA (Fig. 3A, B). We then selected these six genes with prognostic value and performed LASSO analysis using the training set. The six genes were then used to develop the BMRG prognostic index (Fig. 3C, D). As shown in the heatmap, these six genes were highly expressed in the tumor group (Fig. 3E). The riskscore is calculated as follows: Riskscore = ExpLAMC2*Coefficient + ExpGPC2*CoefficientGPC2 + ExpAGRN*CoefficientAGRN + ExpITGA3*CoefficientITGA3 + ExpLAMA3*CoefficientLAMA3 + ExpLOXL4*CoefficientLOXL4 (Table 5).

Fig. 3
figure 3

Construction of the BMRGs and verification of their prognostic value. A Univariate Cox regression analysis. B Multivariate Cox regression analysis. C–D LASSO analysis. E. Heatmap of six BMRGs from the entire set. F Boxplot of the expression of six genes in the entire set

Full size image
Table 5 Regression coefficients of six BMRGs
Full size table

The median risk score from the training set was applied to the test set and the entire set. Figure 3F illustrates the differences in expression of the six genes in the full set in different risk groups. A total of 37, 32, and 69 high-risk patients and 38, 44, and 82 low-risk patients were identified in the training, test, and entire sets, respectively. The survival probability of the high-risk group had significantly reduced compared to the low-risk group (Fig. 4A–C), which validated the predictive capacity of the BMRG index for ESCA prognosis. A comparison was made between the risk scores and survival status distributions (Fig. 4D–I). Principal component analysis revealed strong clustering of patients with ESCA (Fig. 5A–F).

Fig. 4
figure 4

Prognostic value of the BMRG model in all training and test sets. A–C Kaplan–Meier survival analysis of patients between the high- and low-risk groups in the entire, training and test sets, respectively. D–F Risk score distributions for the entire training and test sets. G–I Scatter plot of patient survival status for the entire training and test sets

Full size image
Fig. 5
figure 5

Principal component analysis (PCA) of patients with esophageal carcinoma (ESCA) between high- and low-risk groups in the entire training and test groups. A–C 2D-PCA dimension reduction diagram. D–F 3D-PCA dimension reduction diagram

Full size image

Independent prognosis values of the risk model

Cox regression analysis illustrated that three factors (stage, N and risk score) were substantially correlated with survival in patients with ESCA (Fig. 6A). Multivariate Cox analyses revealed a substantial association among prognosis, risk score, and stage (Fig. 6B). The time-ROC curve showed that the risk score had good predictive power in the training, test, and entire sets (Fig. 6C–E). ROC analysis demonstrated that the risk score had the highest AUC among the clinical parameters (Fig. 6F–H), except in the test set. These results indicate that BMRGs are superior to sex, TN stage, and age in predicting the prognosis of patients with ESCA. These findings imply that BMRGs can serve as independent prognostic predictors for patients with ESCA.

Fig. 6
figure 6

Validation of the BMRGs prognosis and its association with clinical parameters. A Univariate Cox regression analysis; B Multivariate Cox regression analysis. C–E ROC curves for predicting survival at 1, 2, and 3 years in the entire training and test sets. F–H ROC curves comparing different clinical parameters in the entire training and test sets

Full size image

We explored the distribution of risk scores among patients with ESCA with different clinical characteristics and found that patients aged 70 to 79 years had a higher risk score than those younger than 50 years (Fig. 7A), N-N3 patients had a higher risk score than N0 patients (Fig. 7B), and stage 4 patients had a higher risk score than stage 2 patients (Fig. 7C). This indicates that the risk score has a consistent trend with other clinical indicators for the prognosis of ESCA.

Fig. 7
figure 7

Violin map of the clinical features of the TCGA-ESCA population. A Age. B Node. C Stage

Full size image

Establishment and verification of a nomogram

To better predict the survival of patients with ESCA, we combined multiple prognostic indicators to construct a nomogram to graphically evaluate the survival probability of patients, and one patient with ESCA was randomly chosen for scoring (Fig. 8A). The calibration curve demonstrated a strong predictive power (Fig. 8B). According to the decision curve analysis, the nomogram model offered the greatest net benefit, making it a viable tool for clinical decision-making (Fig. 8C).

Fig. 8
figure 8

Nomogram and assessment of the risk model. A. Nomogram. B Calibration curves. C. Decision curve analysis

Full size image

PPI and functional enrichment analyses

PPI is a complex network composed of proteins and their interactions. These interactions play a key role in biological signal transmission and gene expression regulation. We uploaded BMRGs to the STRING database and selected genes with a minimum interaction score of 0.9 to construct the PPI network. The PPI network showed that six BMRGs in the prognostic model had key interactions with 92 BMRGs (Fig. 9A).

Fig. 9
figure 9

Enrichment analyses. A PPI networks with an interaction score greater than 0.9 in 92 BMRGs. B GO analysis. C KEGG analysis

Full size image

Samples from the high- and low-risk groups were analyzed for differences. Using | log2 fold change (FC) |≥ 1, P < 0.05 | as the standard, the obtained DEGs were subjected to enrichment analysis. GO analysis showed that the identified biological process included genes that significantly participated in lipid localization and transport. The cellular component was enriched in genes involved in the collagen-containing ECM and apical part of cell, whereas molecular function included genes strongly associated with signaling receptor activator activity and receptor ligand activity (Fig. 9B). According to the KEGG pathway analysis, the following pathways were mostly involved: Neuroactive ligand − receptor interaction, Hormone signaling, and Protein digestion and absorption (Fig. 9C). The major enriched lipid metabolic pathways can support tumorigenesis and disease progression as well as treatment resistance by enhancing lipid synthesis, storage, and breakdown.

Immune Infiltration Analysis

The estimated percentages and ratio of 22 immune cell types in the tumor microenvironment for both groups were shown (Fig. 10A, B). The high-risk group had a higher proportion of regulatory T cells and neutrophils, and the low-risk group had a higher proportion of M0 macrophages. Additionally, we compared the gene expression levels of 19 immunological checkpoint-related genes (ICRGs) among the groups (Fig. 10D).

Fig. 10
figure 10

Properties of the tumor microenvironment. A, B Proportion of 22 immune cell types. C Immune checkpoint-related genes

Full size image

TMB as a prognostic feature

TMB analysis showed that the mutation rates in the high- and low-risk groups were 98.51% and 100%, respectively (Fig. 11A, B). The waterfall plot indicated that TP53, TTN, CSMD3, SYNE1, FLG, MUC16, DNAH5, RYR2, FAT3, and PCLO were the main mutant genes in both groups. The low-risk group had higher mutation rates for TTN, CSMD3, MUC16, RYR2, FAT3, and PCLO than did the high-risk group.

Fig. 11
figure 11

The TMB as a prognostic feature. A Waterfall plots of the TMB for the top 20 mutated genes in the high-risk group. B Low-risk group. C Violin plots. E Survival curves

Full size image

Although there was no significant difference with regard to the TMB among the groups (Fig. 11C), the TMB survival curve revealed that a low TMB was associated with high survival probability (Fig. 11D). Among all groups, the best prognosis was achieved when both the risk score and TMB were low (Fig. 11E).

Drug sensitivity analysis

To enhance treatment outcomes, we examined how the two groups differed in their sensitivity to widely used chemotherapeutic drugs. The high-risk group had lower IC50 s for paclitaxel, MG- 132, and docetaxel, indicating greater sensitivity, whereas the low-risk group had greater sensitivity to oxaliplatin (Fig. 12).

Fig. 12
figure 12

Patients in the low- and high-risk groups differed in their response to widely used chemotherapeutic drugs. A–D Treatment with paclitaxel (A), MG − 132 (B), Oxaliplatin (C), docetaxel (D)

Full size image

IHC assay

Immunohistochemical assay demonstrated that the expression of GPC2, AGRN, ITGA3, LAMA3, LOXL4, and LAMC2 was concentrated in the cell membrane and cytoplasm, and compared to normal tissues, tumor tissues showed high expression (Fig. 13A-F). Five visual fields were randomly selected, and the proportion of positive area under the visual field was statistically analyzed. The protein expression levels of the 6 BMRGs were different between the normal and the tumor group (Fig. 13G).

Fig. 13
figure 13

Immunohistochemical analysis of BMRGs. A AGRN. B GPC2. C LAMA3. D LAMC2. E ITGA3. F LOXL4. G Proportion of positive area in normal and tumor groups. *P < 0.05, **P < 0.01, ***P < 0.001

Full size image

Knockdown of BMRGs correlated with metastasis and prognosis in KYSE- 150 cells

To elucidate the functional significance of LAMC2, GPC2, AGRN, ITGA3, LAMA3, and LOXL4 expression on tumor progression, we knocked down LAMC2, GPC2, AGRN, ITGA3, LAMA3, and LOXL4 in KYSE- 150 cells. For each gene, we used three kinds of si-RNA to temporarily knock down the above genes in KYSE- 150. The si-RNA with the best knockdown effect was selected using RT-qPCR (Fig. S1).

The CCK- 8 results illustrated that the proliferative ability of KYSE- 150 cells significantly increased after the knockdown of GPC2, ITGA3, and LAMA3 (Fig. 14). Subsequent transwell experiments showed that the migration ability of KYSE- 150 cells was significantly enhanced after the knockdown of GPC2, AGRN, ITGA3, LAMA3, and LOXL4, whereas the migration ability was significantly weakened after LAMC2 knockdown compared to the negative control (Fig. 15A). Scratch assay showed that GPC2, AGRN, ITGA3, LAMA3, and LOXL4 promoted KYSE- 150 cell migration, whereas LAMC2 inhibited it (Fig. 15B). These findings tentatively indicate the role of BMRGs in ESCA progression.

Fig. 14
figure 14

The effect of siRNA knockdown on cell proliferation measured using the CCK- 8 method. A AGRN. B GPC2. C LAMA3. D LAMC2. E ITGA3. F LOXL4. *P < 0.05, **P < 0.01, ***P < 0.001

Full size image
Fig. 15
figure 15

Effects of GPC2, AGRN, ITGA3, LAMA3, LOXL4, and LAMC2 on KYSE- 150 cells. A Transwell migration assay. B Scratch assay. *P < 0.05, **P < 0.01, ***P < 0.001

Full size image

Discussion and conclusion

ESCA is a prevalent cancer in China with a high incidence and fatality rate [27]. According to earlier research, BMRGs is linked to a variety of malignancies, including gastric cancer [28], lung adenocarcinoma [29], hepatic carcinoma [30, 31], and WHO Grade II/III gliomas [32]. However, no research currently examines the role of BMRGs on ESCA progression. We developed a reliable prognostic signature with high predictive value based on six BMRGs identified as risk signatures and demonstrated that the OS of patients with high-risk was poor.

BMs are mostly made up of trimeric proteins called laminins, which are made up of three chains (α, β, and γ) [33]. Laminin- 332, which is also composed of laminin subunits β3 (LAMB3) and γ2 (LAMC2), has LAMA3 as its alpha subunit. LAMC2 is overexpressed in many tumors, including ESCC [34], pancreatic ductal adenocarcinoma [34, 35], cholangiocarcinoma [36], and non-small cell lung cancer [37]. A previous study demonstrated that LAMC2 overexpression promotes ESCC metastasis [34]. LAMC2 inhibits oral squamous cell carcinoma metastasis, invasion, and proliferation by interacting with autophagy through the PI3 K–AKT–mTOR pathway [38]; however, the precise mechanistic involvement of LAMC2 in ESCA is not fully understood. Higher levels of LAMC2 expression have been linked to poorer survival, tumor–node–metastasis phases, lymph node metastasis, and tumor status [33], which is consistent with our model. According to previous studies, reduced LAMA3 expression can limit lung cancer cell growth, and lead to epithelial–mesenchymal transition [39].

GPC2 is a gene in the glypican family that encodes a proteoglycan with a glycosylphosphatidylinositol anchor, and its family includes GPC1 to GPC6 [40]. The GPC2 levels are upregulated in childhood cancers, such as neuroblastoma [41]. GPC2 can positively regulate canonical Wnt signaling in neuroblastoma [40]. AGRN could potentially play a significant part in several human malignancies, such as hepatocellular carcinoma [42], pancreatic ductal carcinoma [43], lung adenocarcinoma [44], and oral cancer [45]. Down-regulation of AGRN has been found to slow colorectal cancer progression via modulation of the Wnt–catenin pathway [46]. In pancreatic duct adenocarcinoma, ITGA3 is associated with poorer overall survival and reduced T cell levels [47]. However, the role and mechanisms of GPC2, AGRN, and ITGA3 in ESCA remain unclear.

LOXL4 is involved in maintaining ECM homeostasis [48] and is highly expressed in various tumors, including melanoma, and serves as a marker and target in the therapy of head and neck squamous cell carcinomas [49]. Prior research has demonstrated that increased LOXL4 expression can slow the progression of head and neck carcinomas [50]. Conversely, low expression was linked to increased rates of worse survival following resection of hepatocellular carcinoma [51].

According to the enrichment analysis, LAMC2, LAMA3, and ITGA3 in BMRGs were significantly correlated with the PI3 K–Akt signaling pathway. When engaged, it can promote the development and survival of endothelial tubes [52] and causes abnormal tumor metabolism, such as an increase in aerobic glycolysis and anabolic pathways [53]. Simultaneously, the PI3 K–AKT pathway is involved in regulating tumor multidrug resistance, and studying this pathway may be helpful in combating tumor drug resistance [54].

Cox regression analysis demonstrated that the risk score could predict the prognosis for patients with ESCA. Compared with other clinicopathological variables (stage, T, N, and age), the risk score had a stronger predictive effect.

Regulatory T cells work in collaboration with cancer cells to produce bioactive TGFβ and create an immunosuppressive microenvironment [55,56,57]. M0 macrophages are resting macrophages that can transform into M1/M2 macrophages upon activation. Pro-inflammatory M1 macrophages can phagocytose tumor cells, whereas anti-inflammatory M2 macrophages such as tumor-associated macrophages promote tumor growth and invasion [58]. The role of neutrophils in tumors is heterogeneous. They participate in anti-tumor immune regulation, synergistically activate the immune cell network, enhance the anticancer effect, and directly support tumor cell proliferation by releasing reactive oxygen species or prostaglandin E2 and other paracrine signals [59, 60]. In the CIBERSORT, the high-risk group had a higher proportion of regulatory T cells and neutrophils, and the low-risk group had a higher proportion of M0 macrophages. The high expression of regulatory T cells indicated that the high-risk group had a worse tumor invasion environment. However, due to the double-edged nature of neutrophils and M0 macrophages in tumor progression, their specific roles in the high-risk group require further investigation. The use of immune checkpoint inhibitors is an important treatment option for patients with advanced ESCA. In an ESCA study, ICRGs, such as inhibit programmed death 1/programmed death ligand 1 and CTLA- 4, showed encouraging results [61, 62]. In our study, five ICRGs (CD40LG, HHLA2, LGALS9, TNFRSF14, and TNFRSF14.1) were highly expressed in the high-risk group. We also discovered that the TMB levels across patients in the two risk categories did not differ significantly; however, the survival of the low-risk group was better. Furthermore, in both groups, the genes with the highest mutation frequencies were TP53, TTN, CSMD3, SYNE1, FLG, and MUC16. A previous study suggested that MUC16 overexpression has predictive relevance in ESCA [63].

A series of cell function experiments were conducted in KYSE- 150 cells to explore the functions of model genes in ESCA. After knocking down GPC2, LAMA3, and ITGA3, the proliferative ability of the cells was significantly enhanced. The results of the scratch assay illustrated that the ability of cell migration was significantly augmented after AGRN, GPC2, LAMA3, ITGA3, and LOXL4 knockdown, whereas it was significantly weakened after LAMC2 knockdown. The transwell experiment further confirmed the results of the scratch assay.

However, after knocking out LAMA3, only a small number of cells migrated, which was contrary to the results of the scratch assay and may be related to the ability of the cells to pass through the porous membrane. The scratch assay was performed by preparing appropriate gaps in the cell layer and then monitoring the area of cell movement over time. The Transwell approach is based on monitoring the movement of cells through porous membranes toward chemoattractants in a 3D environment, and then to measure the migrating cells with crystal violet staining [64].

In summary, we established a prognostic model based on six BMRGs that can accurately predict the prognosis of ESCA and explored the biological and immunological characteristics of these BMRGs in ESCA using a variety of bioinformatics techniques. These results may help inform the future development of new targeted and clinical immunotherapies for ESCA. Our experimental results preliminarily confirmed the close association between AGRN, GPC2, LAMA3, LAMC2, ITGA3, and LOXL4 and ESCA metastasis and prognosis. However, this study has some limitations. A larger sample size is needed to further validate the model. The mechanisms by which AGRN, GPC2, LAMA3, LAMC2, ITGA3, and LOXL4 affect the metastasis and prognosis of esophageal cancer remain unclear, and further experiments are needed to explore these mechanisms.

Data availability

The original data of this study is available upon reasonable request from Jianfen Su (corresponding author).

Abbreviations

BM:

Basement membrane

BMRG:

BM-related genes

ECM:

Extracellular matrix

ESCA:

Esophageal carcinoma

TMB:

Tumor mutational burden

DEGs:

Differentially expressed genes

time-ROC:

Time-dependent receiver operating characteristic

ROC:

Receiver operating characteristic

GO:

Gene ontology

KEGG:

Kyoto Encyclopedia of Genes and Genomes

GSEA:

Gene Set Enrichment Analysis

PPIs:

Protein–protein interactions

ESCC:

Esophageal squamous cell carcinoma

siRNA:

Small interfering RNA

qRT-PCR:

Quantitative reverse transcription real-time PCR

ICRGS:

Immunological checkpoint-related genes

AGRN:

Agrin

GPC2:

Glypican 2

LAMA3:

Laminin subunit alpha 3

LAMC2:

Laminin subunit gamma 2

LOXL4:

Lysine oxidase-like 4

ITGA3:

Integrin subunit alpha 3

References

  1. Morgan E, et al. The Global Landscape of Esophageal Squamous Cell Carcinoma and Esophageal Adenocarcinoma Incidence and Mortality in 2020 and Projections to 2040: New Estimates From GLOBOCAN 2020. Gastroenterol. 2022;163(3):649-658 e2.

    Article  Google Scholar 

  2. Sung H, et al. Global Cancer Statistics 2020: GLOBOCAN Estimates of Incidence and Mortality Worldwide for 36 Cancers in 185 Countries. CA Cancer J Clin. 2021;71(3):209–49.

    Article  PubMed  Google Scholar 

  3. Yang H, et al. Oesophageal cancer. Lancet. 2024;404(10466):1991–2005.

    Article  PubMed  Google Scholar 

  4. DomperArnal MJ, Ferrandez Arenas A, Lanas Arbeloa A. Esophageal cancer: Risk factors, screening and endoscopic treatment in Western and Eastern countries. World J Gastroenterol. 2015;21(26):7933–43.

    Article  Google Scholar 

  5. Chang J, Chaudhuri O. Beyond proteases: Basement membrane mechanics and cancer invasion. J Cell Biol. 2019;218(8):2456–69.

    Article  PubMed  PubMed Central  Google Scholar 

  6. Banerjee S, et al. Multiple roles for basement membrane proteins in cancer progression and EMT. Eur J Cell Biol. 2022;101(2):151220.

    Article  CAS  PubMed  Google Scholar 

  7. Huang J, et al. Extracellular matrix and its therapeutic potential for cancer treatment. Signal Transduct Target Ther. 2021;6(1):153.

    Article  PubMed  PubMed Central  Google Scholar 

  8. Najafi M, Farhood B, Mortezaee K. Extracellular matrix (ECM) stiffness and degradation as cancer drivers. J Cell Biochem. 2019;120(3):2782–90.

    Article  CAS  PubMed  Google Scholar 

  9. Li S, et al. Integrin and dystroglycan compensate each other to mediate laminin-dependent basement membrane assembly and epiblast polarization. Matrix Biol. 2017;57–58:272–84.

    Article  PubMed  Google Scholar 

  10. Chang J, et al. Cell volume expansion and local contractility drive collective invasion of the basement membrane in breast cancer. Nat Mater. 2024;23(5):711–22.

    Article  CAS  PubMed  Google Scholar 

  11. Sherwood DR. Basement membrane remodeling guides cell migration and cell morphogenesis during development. Curr Opin Cell Biol. 2021;72:19–27.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  12. Randles MJ, et al. Identification of an Altered Matrix Signature in Kidney Aging and Disease. J Am Soc Nephrol. 2021;32(7):1713–32.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  13. Tsilibary EC. Microvascular basement membranes in diabetes mellitus. J Pathol. 2003;200(4):537–46.

    Article  CAS  PubMed  Google Scholar 

  14. Naba A, et al. Extracellular matrix signatures of human primary metastatic colon cancers and their metastases to liver. BMC Cancer. 2014;14:518.

    Article  PubMed  PubMed Central  Google Scholar 

  15. Chang TT, Thakar D, Weaver VM. Force-dependent breaching of the basement membrane. Matrix Biol. 2017;57–58:178–89.

    Article  PubMed  Google Scholar 

  16. Lambert AW, Pattabiraman DR, Weinberg RA. Emerging Biological Principles of Metastasis. Cell. 2017;168(4):670–91.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  17. Jayadev R, et al. A basement membrane discovery pipeline uncovers network complexity, regulators, and human disease associations. Sci Adv. 2022;8(20):eabn2265.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  18. Zhang X, et al. Basement membrane-related MMP14 predicts poor prognosis and response to immunotherapy in bladder cancer. BMC Cancer. 2024;24(1):746.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  19. Li B, et al. A novel basement membrane-related gene signature predicts prognosis and immunotherapy response in hepatocellular carcinoma. Front Oncol. 2024;14:1388016.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  20. Cai J, et al. Identification of a basement membrane-related gene signature for predicting prognosis and estimating the tumor immune microenvironment in breast cancer. Front Endocrinol (Lausanne). 2022;13:1065530.

    Article  PubMed  Google Scholar 

  21. Love MI, Huber W, Anders S. Moderated estimation of fold change and dispersion for RNA-seq data with DESeq2. Genome Biol. 2014;15(12):550.

    Article  PubMed  PubMed Central  Google Scholar 

  22. Friedman J, Hastie T, Tibshirani R. Regularization Paths for Generalized Linear Models via Coordinate Descent. J Stat Softw. 2010;33(1):1–22.

    Article  PubMed  PubMed Central  Google Scholar 

  23. Blanche P, Dartigues JF, Jacqmin-Gadda H. Estimating and comparing time-dependent areas under receiver operating characteristic curves for censored event times with competing risks. Stat Med. 2013;32(30):5381–97.

    Article  PubMed  Google Scholar 

  24. The Gene Ontology, C. The Gene Ontology Resource: 20 years and still GOing strong. Nucleic Acids Res. 2019;47(D1):D330–8.

    Article  Google Scholar 

  25. Subramanian A, et al. Gene set enrichment analysis: a knowledge-based approach for interpreting genome-wide expression profiles. Proc Natl Acad Sci U S A. 2005;102(43):15545–50.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  26. Newman AM, et al. Robust enumeration of cell subsets from tissue expression profiles. Nat Methods. 2015;12(5):453–7.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  27. Ferlay J, et al. Estimating the global cancer incidence and mortality in 2018: GLOBOCAN sources and methods. Int J Cancer. 2019;144(8):1941–53.

    Article  CAS  PubMed  Google Scholar 

  28. Wang J, et al. Identification of Basement Membrane-Related Signatures in Gastric Cancer. Diagnostics (Basel). 2023;13(11):1844.

    Article  PubMed  Google Scholar 

  29. Zhang Z, et al. A novel basement membrane-related gene signature for prognosis of lung adenocarcinomas. Comput Biol Med. 2023;154:106597.

    Article  CAS  PubMed  Google Scholar 

  30. Ding R, et al. Basement membrane-related regulators for prediction of prognoses and responses to diverse therapies in hepatocellular carcinoma. BMC Med Genomics. 2023;16(1):81.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  31. Shen J, et al. A model of basement membrane-associated gene signature predicts liver hepatocellular carcinoma response to immune checkpoint inhibitors. Mediators Inflamm. 2023;2023:7992140.

    Article  PubMed  PubMed Central  Google Scholar 

  32. Zhang Z, Lai G, Sun L. Basement-Membrane-Related Gene Signature Predicts Prognosis in WHO Grade II/III Gliomas. Genes (Basel). 2022;13(10):1810.

    Article  CAS  PubMed  Google Scholar 

  33. Fu T, et al. LAMC2 as a prognostic biomarker in human cancer: a systematic review and meta-analysis. BMJ Open. 2022;12(11):e063682.

    Article  PubMed  PubMed Central  Google Scholar 

  34. Liang Y, et al. LncRNA CASC9 promotes esophageal squamous cell carcinoma metastasis through upregulating LAMC2 expression by interacting with the CREB-binding protein. Cell Death Differ. 2018;25(11):1980–95.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  35. Okada Y, et al. LAMC2 promotes cancer progression and gemcitabine resistance through modulation of EMT and ATP-binding cassette transporters in pancreatic ductal adenocarcinoma. Carcinogenesis. 2021;42(4):546–56.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  36. Pei YF, et al. Silencing of LAMC2 Reverses Epithelial-Mesenchymal Transition and Inhibits Angiogenesis in Cholangiocarcinoma via Inactivation of the Epidermal Growth Factor Receptor Signaling Pathway. Am J Pathol. 2019;189(8):1637–53.

    Article  CAS  PubMed  Google Scholar 

  37. Liu M, et al. LAMC2 promotes the proliferation of cancer cells and induce infiltration of macrophages in non-small cell lung cancer. Ann Transl Med. 2021;9(17):1392.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  38. Shan F, et al. LAMC2 regulates proliferation, migration, and invasion mediated by the Pl3K/AKT/mTOR pathway in oral. Oncol Res. 2023;31(4):481–93.

    Article  PubMed  PubMed Central  Google Scholar 

  39. Xu SF, et al. Long Non-coding RNA LINC00628 Interacts Epigenetically with the LAMA3 Promoter and Contributes to Lung Adenocarcinoma. Mol Ther Nucleic Acids. 2019;18:166–82.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  40. Li N, et al. Glypicans as Cancer Therapeutic Targets. Trends Cancer. 2018;4(11):741–54.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  41. Bosse KR, et al. Identification of GPC2 as an Oncoprotein and Candidate Immunotherapeutic Target in High-Risk Neuroblastoma. Cancer Cell. 2017;32(3):295-309 e12.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  42. Chakraborty S, et al. An oncogenic role of Agrin in regulating focal adhesion integrity in hepatocellular carcinoma. Nat Commun. 2015;6:6184.

    Article  CAS  PubMed  Google Scholar 

  43. Ruivo CF, et al. Extracellular Vesicles from Pancreatic Cancer Stem Cells Lead an Intratumor Communication Network (EVNet) to fuel tumour progression. Gut. 2022;71(10):2043–68.

    Article  CAS  PubMed  Google Scholar 

  44. Zhang H, et al. AGRN promotes lung adenocarcinoma progression by activating Notch signaling pathway and acts as a therapeutic target. Pharmacol Res. 2023;194:106819.

    Article  CAS  PubMed  Google Scholar 

  45. Rivera C, et al. Agrin has a pathological role in the progression of oral cancer. Br J Cancer. 2018;118(12):1628–38.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  46. Li X, et al. Matrine restrains the development of colorectal cancer through regulating the AGRN/Wnt/beta-catenin pathway. Environ Toxicol. 2023;38(4):809–19.

    Article  CAS  PubMed  Google Scholar 

  47. Jorgensen C. Untangling the tumorigenic role of homotrimeric collagen I. Cancer Cell. 2022;40(8):802–4.

    Article  PubMed  Google Scholar 

  48. Xie W, et al. Clinical significance of LOXL4 expression and features of LOXL4-associated protein-protein interaction network in esophageal squamous cell carcinoma. Amino Acids. 2019;51(5):813–28.

    Article  CAS  PubMed  Google Scholar 

  49. Weise JB, et al. LOXL4 is a selectively expressed candidate diagnostic antigen in head and neck cancer. Eur J Cancer. 2008;44(9):1323–31.

    Article  CAS  PubMed  Google Scholar 

  50. Gorogh T, et al. Lysyl oxidase like-4 monoclonal antibody demonstrates therapeutic effect against head and neck squamous cell carcinoma cells and xenografts. Int J Cancer. 2016;138(10):2529–38.

    Article  PubMed  Google Scholar 

  51. Tian M, et al. LOXL4 is downregulated in hepatocellular carcinoma with a favorable prognosis. Int J Clin Exp Pathol. 2015;8(4):3892–900.

    PubMed  PubMed Central  Google Scholar 

  52. Cheng HW, et al. Cancer cells increase endothelial cell tube formation and survival by activating the PI3K/Akt signalling pathway. J Exp Clin Cancer Res. 2017;36(1):27.

    Article  PubMed  PubMed Central  Google Scholar 

  53. Park JH, Pyun WY, Park HW. Cancer metabolism: phenotype, signaling and therapeutic targets. Cells. 2020;9(10):2308.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  54. Liu R, et al. PI3K/AKT pathway as a key link modulates the multidrug resistance of cancers. Cell Death Dis. 2020;11(9):797.

    Article  PubMed  PubMed Central  Google Scholar 

  55. Laine A, et al. Regulatory T cells promote cancer immune-escape through integrin alphavbeta8-mediated TGF-beta activation. Nat Commun. 2021;12(1):6228.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  56. Tay C, Tanaka A, Sakaguchi S. Tumor-infiltrating regulatory T cells as targets of cancer immunotherapy. Cancer Cell. 2023;41(3):450–65.

    Article  CAS  PubMed  Google Scholar 

  57. Negura I, Pavel-Tanasa M, Danciu M. Regulatory T cells in gastric cancer: Key controllers from pathogenesis to therapy. Cancer Treat Rev. 2023;120:102629.

    Article  CAS  PubMed  Google Scholar 

  58. Xia Y, et al. Engineering macrophages for cancer immunotherapy and drug delivery. Adv Mater. 2020;32(40):e2002054.

    Article  PubMed  Google Scholar 

  59. Hedrick CC, Malanchi I. Neutrophils in cancer: heterogeneous and multifaceted. Nat Rev Immunol. 2022;22(3):173–87.

    Article  CAS  PubMed  Google Scholar 

  60. Huang X, et al. Neutrophils in Cancer immunotherapy: friends or foes? Mol Cancer. 2024;23(1):107.

    Article  PubMed  PubMed Central  Google Scholar 

  61. Jiao R, et al. Immune checkpoint inhibitors in esophageal squamous cell carcinoma: progress and opportunities. Onco Targets Ther. 2019;12:6023–32.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  62. Baba Y, et al. Tumor immune microenvironment and immune checkpoint inhibitors in esophageal squamous cell carcinoma. Cancer Sci. 2020;111(9):3132–41.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  63. Zhao H, et al. Clinical significance of preoperative serum tumor markers in esophageal squamous cell carcinoma. J Cancer Res Ther. 2014;10(Suppl):C179–85.

    PubMed  Google Scholar 

  64. Bouchalova P, Bouchal P. Current methods for studying metastatic potential of tumor cells. Cancer Cell Int. 2022;22(1):394.

    Article  PubMed  PubMed Central  Google Scholar 

Download references

Funding

This work was supported by the Graduate Education Innovation Program of Guangdong Provincial Department of Education [grant number 2023 ANLK_046]; Guangdong Basic and Applied Basic Research Fund Enterprise Joint Fund [grant number 2023 A1515220024]; Medical Research Foundation of Guangdong Province [grant number 2024070]; Guangdong Hospital Pharmaceutical Research Fund project funding [grant number 2024 A31]; and Research Funds from Guangzhou Panyu Central Hospital [grant numbers PY- 2023–007, PY- 2023–010, PY- 2023–026].

Author information

Author notes

  1. Lang Xu and Bingna Wang are co-first authors.

Authors and Affiliations

  1. The Affiliated Panyu Central Hospital, Guangzhou Medical University, Guangzhou, 511400, China

    Lang Xu, Bingna Wang, Chen Wang, Nan Mao, Yating Huang, Xihua Fu, Yang Zhang & Jianfen Su

  2. School of Pharmacy, Guangdong Medical University, Dongguan, 523808, China

    Lang Xu, Tao Feng, Qiming He, Yang Zhang, Xinsheng Peng & Jianfen Su

  3. Affiliated Qingyuan Hospital, Guangzhou Medical University, Qingyuan People’s Hospital, Qingyuan, 511518, China

    Lang Xu

  4. School of Pharmaceutical Science, Guangzhou Medical University, Guangzhou, 511436, China

    Bingna Wang & Jianfen Su

  5. School of Biomedical and Pharmaceutical Sciences, Guangdong University of Technology, Guangzhou, 510006, China

    Guoxing You & Xiaojun Ma

Authors
  1. Lang Xu
    View author publications

    You can also search for this author inPubMed Google Scholar

  2. Bingna Wang
    View author publications

    You can also search for this author inPubMed Google Scholar

  3. Chen Wang
    View author publications

    You can also search for this author inPubMed Google Scholar

  4. Nan Mao
    View author publications

    You can also search for this author inPubMed Google Scholar

  5. Yating Huang
    View author publications

    You can also search for this author inPubMed Google Scholar

  6. Xihua Fu
    View author publications

    You can also search for this author inPubMed Google Scholar

  7. Tao Feng
    View author publications

    You can also search for this author inPubMed Google Scholar

  8. Qiming He
    View author publications

    You can also search for this author inPubMed Google Scholar

  9. Yang Zhang
    View author publications

    You can also search for this author inPubMed Google Scholar

  10. Guoxing You
    View author publications

    You can also search for this author inPubMed Google Scholar

  11. Xiaojun Ma
    View author publications

    You can also search for this author inPubMed Google Scholar

  12. Xinsheng Peng
    View author publications

    You can also search for this author inPubMed Google Scholar

  13. Jianfen Su
    View author publications

    You can also search for this author inPubMed Google Scholar

Contributions

Conceptualization and Writing-original draft: LX. Writing-review and editing: JFS and XSP. Data curation and Validation: LX, BNW, CW, NM, and YTH. Supervision: JFS and XHF. Visualization: TF, QMH, and YZ. Resources: GXY and XJM. All authors approved the final version of the manuscript.

Corresponding authors

Correspondence to Xinsheng Peng or Jianfen Su.

Ethics declarations

Ethics approval and consent to participate

The esophageal tissue samples identified originated from patients who had signed informed consent forms at the Affiliated Panyu Central Hospital, Guangzhou Medical University. All studies involving human subjects were reviewed and approved by the Ethics Committee of the Affiliated Panyu Central Hospital, Guangzhou Medical University. All experiments were conducted in accordance with relevant guidelines and regulations.

Competing interests

The authors declare no competing interests.

Additional information

Publisher’s Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Supplementary Information

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License, which permits any non-commercial use, sharing, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if you modified the licensed material. You do not have permission under this licence to share adapted material derived from this article or parts of it. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by-nc-nd/4.0/.

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Xu, L., Wang, B., Wang, C. et al. A model of basement membrane-related regulators for prediction of prognoses in esophageal cancer and verification in vitro. BMC Cancer 25, 696 (2025). https://doiorg.publicaciones.saludcastillayleon.es/10.1186/s12885-025-14081-4

Download citation

  • Received:

  • Accepted:

  • Published:

  • DOI: https://doiorg.publicaciones.saludcastillayleon.es/10.1186/s12885-025-14081-4

Keywords

BMC Cancer

ISSN: 1471-2407

Contact us