Fibulin-5 (FBLN5) as a context-dependent pan-cancer regulator: integrated multi-omics analysis reveals dual roles in tumor progression and therapeutic resistance

Introduction

Cancer is a leading cause of morbidity and mortality globally, with a complex pathogenesis driven by heterogeneous genetic and molecular alterations across tumor types (1). Traditional research has largely focused on individual tumor types, aiming to identify disease-specific biomarkers or therapeutic targets; however, this approach often overlooks conserved oncogenic mechanisms that transcend tissue origins (2).

The advent of pan-cancer analysis has revolutionized tumor genomics by integrating multi-omics data from diverse malignancies, revealing shared pathways (e.g., cell cycle dysfunction and immune evasion) and context-dependent molecular features (3,4). This approach systematically compares molecular profiles (e.g., mutations, epigenetic changes, and immune microenvironments) across tens of thousands of tumors, as exemplified by The Cancer Genome Atlas (TCGA) Pan-Cancer Atlas (5).

Fibulin-5 (FBLN5), also known as embryonic vascular epidermal growth factor (EGF)-repeat-containing protein (EVEC) or developmental arteries and neural crest EGF-like (DANCE) protein, is a member of the extracellular matrix (ECM) protein fibulin family. It was discovered in 1999 by two research teams attempting to isolate a new vascular growth regulator (6). It is a glycoprotein containing 448 amino acids, mainly composed of six calcium binding EGF-like repeats and a circular C-terminal domain. The first EGF-like repeat region includes a proline rich insertion sequence that interacts with a large number of ECMs and secreted proteins through its structural domain (7). In addition, FBLN5 also has a unique arginine-glycine-aspartic (Arg-Gly-Asp) acid (RGD) sequence, which mediates endothelial cell adhesion by binding to α v β 3, α v β 5, and α 9 β 1 integrins (8).

Previous studies found that FBLN5 is lowly expressed in many tumor tissues, including prostate, bladder, liver, colorectal cancer and breast cancer tissues. The regulation of ECM protein expression changes the function of the matrix, which in turn inhibits tumor cell proliferation and metastasis (9-14). FBLN5 has also been shown to be highly expressed in gastric cancer and pancreatic cancer, facilitating tumor occurrence and development by promoting tumor cell proliferation and migration (15,16). Resent researches showed that cancer-associated fibroblast-derived FBLN5 promoted radioresistance and epithelial-mesenchymal transition (EMT) in lung cancer and gastric cancer (17,18). These results suggest that as a key member of the ECM, FBLN5 plays an important role in the occurrence and development of tumors.

Despite its potential importance, the role of FBLN5 in driving tumor initiation and metastasis remains controversial across different tumor types. Thus, the molecular profiles of FBLN5 across different tumor types needs to be examined. Additionally, in-depth mechanistic studies are required to determine how FBLN5 intrinsically influences immune regulation in the tumor microenvironment (TME).

Given the current paucity of pan-cancer investigations on FBLN5, this study systematically examined its multifaceted roles in oncogenesis and its prognostic significance. Using integrative bioinformatics approaches, we characterized FBLN5’s genomic alterations, expression patterns, associations with malignant phenotypes, and clinical relevance across malignancies. Leveraging established public databases and computational tools, including R-based analyses, we found that FBLN5 was significantly associated with critical oncological features, tumor immune microenvironment modulation, therapeutic resistance patterns, and patient survival outcomes. These comprehensive findings not only provide insights into the critical role of FBLN5 in tumor biology, but also provide a foundation for subsequent mechanistic investigations and the development of targeted therapies. We present this article in accordance with the REMARK reporting checklist (available at https://tcr.amegroups.com/article/view/10.21037/tcr-2026-0623/rc).

Methods

Data acquisition

Multi-omics data [e.g., gene expression, copy number alterations (CNAs), and DNA methylation] were retrieved from TCGA pan-cancer samples via the Firehose database, including both tumor and normal tissue data. Proteomic [The Cancer Proteome Atlas (TCPA)], mutational, and clinical data for the analyzed cohorts were obtained from the University of California, Santa Cruz (UCSC) Xena Browser platform, which provides harmonized multi-omics datasets for tumor research.

The FBLN5 mutational landscape was characterized through the cBioPortal by analyzing somatic mutation frequency in pan-cancer TCGA samples. Quantitative protein expression profiling of FBLN5 was performed using mass spectrometry data from the Clinical Proteomic Tumor Analysis Consortium (CPTAC) database, with normalization to internal reference standards.

To characterize the relationship between FBLN5 expression and tumor immunology, we employed multiple deconvolution algorithms available via the Tumor Immune Estimation Resource 2.0 (TIMER2.0) to quantify immune cell infiltration patterns across tumor samples. To validate immune infiltration patterns identified through bulk transcriptomics, we analyzed 99 single-cell RNA sequencing datasets from the Tumor Immune Single-cell Hub (TISCH) database.

Additionally, drug sensitivity data were obtained from three pharmacogenomic repositories: CellMiner for Food and Drug Administration (FDA)-approved compounds; the Genomics of Drug Sensitivity in Cancer (GDSC) for small-molecule screening; and the Cancer Therapeutics Response Portal (CTRP) for tumor dependency mapping. The study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments.

Analysis of the prognostic and diagnostic value of FBLN5

Using TCGA survival data, we systematically evaluated the diagnostic and prognostic potential of FBLN5 through computational analyses. Survival outcomes were analyzed using the “survival” and “survminer” packages in R. Multivariate Cox proportional hazards models were implemented to assess prognostic significance, and the results were visualized using the “forestplot” package.

Genetic alteration analysis

The cBioPortal platform serves as an integrated bioinformatics resource for the computational analysis and biological interpretation of tumor genomics data, enabling researchers to derive meaningful insights from both histopathological and cytogenetic studies. It provides information on somatic mutations, DNA CNAs, and DNA methylation, along with tools for mutation analysis and its visualization.

“TCGA Pan-Cancer Atlas Studies” data were used for the analysis. Mutation frequency, mutation type, and CNA data were obtained from the “Cancer Type Summary” section. DNA methylation differences in FBLN5 between malignant and normal tissues were assessed using the Wilcoxon signed-rank test; a P value <0.05 was considered statistically significant. Significant hypomethylation or hypermethylation events were identified based on this threshold. Further, we examined the association between FBLN5 expression levels and promoter methylation using Spearman’s rank correlation coefficients; correlations with P values <0.05 were considered statistically significant.

Exploring pathways and functional mechanisms

To identify FBLN5-associated signaling pathways, tumor samples were stratified based on FBLN5 expression levels (with the top and bottom 30th percentiles defined as high- and low-expression groups, respectively). Subsequently, a gene set enrichment analysis (GSEA) was performed to systematically evaluate pathway activity differences between the high- and low-expression groups. The CistromeDB database (http://dbtoolkit.cistrome.org/) was then used to identify potential regulatory upstream factors for FBLN5.

FBLN5 expression and immune correlation

Using TCGA data, we identified significant correlations between FBLN5 expression levels and various immune-related genes, including immunostimulatory genes, immunoinhibitory genes, chemokines, chemokine receptors, and major histocompatibility complex (MHC) genes.

Drug sensitivity analysis

The Gene Set Cancer Analysis (GSCA) website integrates 750 small-molecule drugs from the GDSC and CTRP databases. Additionally, it leverages gene expression data to identify potential therapeutic compounds associated with specific gene profiles. It was used to analyze the correlation between gene expression and drug sensitivity.

Immunohistochemistry

Paraffin sections were obtained from the Ningbo Pathology Center. The sections were dewaxed in xylene and dehydrated through an alcohol series, and then incubated at 62 °C for 2 hours. Antigen retrieval was performed by heating the samples in ethylenediaminetetraacetic acid buffer (pH 7.4) at 120 °C for 3 minutes, followed by gradual cooling to room temperature.

To inhibit endogenous peroxidase activity, the sections were incubated in 0.3% hydrogen peroxide in methanol for 30 minutes. After three 10-minute rinses in phosphate-buffered saline, the slides were blocked with goat serum at room temperature for 1 hour, and the serum was then discarded. A 1:100 dilution of the rabbit polyclonal FBLN5 antibody was then applied, and the slides were incubated overnight at 4 °C (≤16 hours).

On the following day, the sections were incubated with secondary antibodies in a humidified chamber at room temperature for 40 minutes. They were then subjected to color development using diaminobenzidine. Hematoxylin counterstaining was performed, followed by differentiation and bluing in ammonia water. Scanning was performed at ×200 total magnification using a microscope.

Statistical analysis

All the data analyses were performed using the designated web tools and R software (version 4.3.0, R Foundation for Statistical Computing, Vienna, Austria). Wilcoxon rank-sum tests were employed to analyze FBLN-5 expression levels according to clinical characteristics using combined TCGA and Genotype-Tissue Expression (GTEx) data. Survival analyses, including Kaplan-Meier curves and logistic regression, were performed using the R packages “survival” and “survminer”. All the statistical tests were two-tailed, with significance thresholds defined as follows: *, P<0.05 (significant); **, P<0.01 (highly significant); ***, P<0.001 (very highly significant); and ****, P<0.0001 (extremely significant).

Results

FBLN5 expression aberrations in human cancers

Using TCGA and GTEx data, FBLN5 was found to be differentially expressed in most tumors (Figure 1A). Notably, using TCGA data, FBLN5 was found to be downregulated in 17 different tumor types, including bladder urothelial (BLCA), breast cancer (BRCA), cholangiocarcinama (CHOL), colon adenocarcinoma (COAD), esophageal cancer (ESCA), head and neck squamous cell carcinoma (HNSC), kidney chromophobe carcinoma (KICH), kidney renal clear cell carcinoma (KIRC), kidney renal papillary cell carcinoma (KIRP), liver hepatocellular carcinoma (LIHC), lung adenocarcinoma (LUAD), lung squamous cell carcinoma (LUSC), prostate adenocarcinoma (PRAD), rectum adenocarcinoma (READ), stomach adenocarcinoma (STAD), thyroid carcinoma (THCA), and uterine corpus endometrial carcinaoma (UCEC), while it was found to be upregulated in glioblastoma multiforme (GBM) (Figure 1B). Further, FBLN5 expression was significantly decreased in BLCA, BRCA, CHOL, COAD, ESCA, HNSC, KICH, KIRC, KIRP, LIHC, LUAD, LUSC, PRAD, READ, STAD, THCA, and UCEC, compared to matched normal tissues (Figure 1C). Among these results, the expression of FBLN5 in gastric cancer is different from the findings of previous studies (10), which found the expression of FBLN5 was overpressed in the gastric cancer with poor prognosis. But the researchers chose the Gene Expression Omnibus (GEO) databases that was different from what we used (15).

Figure 1 FBLN5 expression levels in tumor and normal tissues. (A) FBLN5 was differentially expressed between tumor and normal tissues across various tumors using TCGA and GTEx data. (B) FBLN5 mRNA expression levels in tumor and normal tissues in TCGA. (C) FBLN5 mRNA expression levels in paired samples grouped by tumor type in TCGA. Each point represents one sample. (D-K) Differential protein expression between tumor and normal tissues in brain cancer, breast cancer, kidney cancer, lung adenocarcinoma, lung squamous cell carcinoma, ovarian cancer, pancreatic ductal adenocarcinoma, and uterine cancer based on the CPATC database. (L) IHC results showing that FBLN5 exhibited negative to weak cytoplasmic staining in most malignant cells. Different colors represent different staining intensities. *, P<0.05; **, P<0.01; ***, P<0.001. CPATC, Clinical Proteomic Tumor Analysis Consortium; FBLN5, fibulin-5; GTEx, Genotype-Tissue Expression; IHC, immunohistochemical; mRNA, messenger RNA; TCGA, The Cancer Genome Atlas.

The CPTAC database was used to validate the findings at the protein level (Figure 1D-1K). The higher protein expression of FBLN5 in brain cancer in these results was similar with the mRNA expression as mentioned in the previous text. Besides, the overexpression of FBLN5 in pancreatic cancer was also consistent with previous study (16).

The immunohistochemical (IHC) results showed FBLN5 was not expressed in most tumors. While weak staining was observed in glioma, melanoma, lung cancer, and urothelial cancer, and weak-to-moderate staining was observed in testicular cancer (Figure 1L). Thus, our results demonstrated that FBLN5 expression was downregulated in 92% of tumor types compared to both paired and unpaired normal tissues.

Prognostic value of FBLN5 across cancers

To investigate the association between FBLN5 expression and clinical outcomes, we conducted survival analyses across multiple tumor types, assessing overall survival (OS), disease-specific survival (DSS), the disease-free interval (DFI), and the progression-free interval (PFI).

The Cox proportional hazards regression analysis revealed a significant association between FBLN5 expression levels and OS in adrenocortical carcinoma (ACC), BRCA, GBM, brain lower grade glioma (LGG), LIHC, sarcoma (SARC), STAD, and thymoma (THYM) (Figure 2A). FBLN5 was associated with a poor prognosis in ACC, GBM, LGG, and STAD. Conversely, it was associated with a good prognosis in BRCA, LIHC, and SARC. The results were consistent with those of previous studies in both gastric cancer and breast cancer (12,15).

Figure 2 FBLN5 expression and survival outcomes. (A) Association between FBLN5 expression levels and OS analyzed using the Cox proportional hazards model. (B-G) Kaplan-Meier models showing the relationship between FBLN5 expression and OS. (H-J) Forest plot of DSS, DFI, and PFI. (K) Summary analysis of FBLN5 expression and multiple survival outcomes (OS, DSS, DFI and PFI). Only P values <0.05 are shown. CI, confidence interval; DFI, disease-free interval; DSS, disease-specific survival; FBLN5, fibulin-5; HR, hazard ratio; OS, overall survival; PFI, progression-free interval.

Similarly, the Kaplan-Meier survival analysis results also indicated that increased FBLN5 expression was correlated with reduced OS in ACC, BLCA, LGG, and STAD, while increased FBLN5 expression was correlated with longer OS in LIHC and LUAD (Figure 2B-2G). The results indirectly indicates that in different types of tumors it changed its identity of promoting tumor and inhibiting tumor as previous studies (10,11,15,17). Further, the forest plots showed that FBLN5 was associated with poor DSS in ACC and LGG, while it was associated with good DSS in PRAD and SARC (Figure 2H).

The DFI analysis revealed that high FBLN5 expression was associated with a poor prognosis in CESC, KIRC, PAAD, and UCEC, but was not associated with a good prognosis in any tumor (Figure 2I). The PFI results revealed that high FBLN5 expression was associated with a poor PFI in LGG and STAD, but a good PFI in HNSC, SARC, and HJCA (Figure 2J). The summary analysis of FBLN5 expression showed diverse and tumor-specific correlations with multiple survival endpoints (OS, DSS, DFI and PFI), highlighting its context-dependent prognostic roles (Figure 2K).

Genetic alteration analysis

We conducted integrated genomic analyses of TCGA data, examining genetic variations, somatic copy number alterations (SCNAs), and messenger RNA (mRNA) expression in tumor and normal tissues. In most tumors, the rate of exceeded 6%, of which the most common types were “mutations”, “deep deletions”, and “amplifications”. Skin cutaneous melanoma (SKCM) exhibited the highest rate of genetic variations, mainly in the form of mutations. The second- and third-highest frequencies of FBLN5 alterations were observed in LUSC and UCEC, respectively, and were also mainly in the form of mutations (Figure 3A).

Figure 3 FBLN5 expression and genetic alterations. (A) Frequency and types of FBLN5 mutations across different tumor types. (B) Locations and counts of FBLN5 genetic alterations across different tumor types. (C) 3D structure showing FBLN5 mutation sites. (D) Histogram showing the frequency of SCNAs for FBLN5 in each tumor type. (E) Spearman’s correlation between SCNAs and FBLN5 expression. (F) Relationship between FBLN5 mRNA expression and different SCNA patterns. (G) Correlation between FBLN5 mRNA expression and DNA methylation status. (H) Comparison of DNA methylation status across tumor and corresponding normal tissues. ¹, complete methylation; ***, P<0.001. CNA, copy number alterations; CNV, copy number variations; FBLN5, fibulin-5; FDR, false discovery rate; mRNA, messenger RNA; RSEM, RNA-seq by expectation-maximization; SCNAs, somatic copy number alterations; TCGA, The Cancer Genome Atlas; VUS, variants of uncertain significance.

Figure 3B presents a two-dimensional graphic of the FBLN5 mutation site, showing that four samples harbored missense mutations, resulting in the replacement of arginine (R) at position 77 with glutamate (Q). The mutation site was also visualized in a 3D graphic (Figure 3C).

We also assessed the relationship between FBLN5 CNAs and its mRNA expression levels in pan-cancer samples. Further, we analyzed SCNAs in tumors, as they act as key regulators of gene expression in malignant processes. The results indicated that SCNAs occurred at high frequency in most types of tumors (Figure 3D). However, the relationship between SCNVs and mRNA expression appeared to be weak (Figure 3E).

To evaluate the regulatory effect of SCNAs on FBLN5 expression, we performed a Spearman correlation analysis between FBLN5 mRNA expression levels and masked copy number segments across TCGA datasets. Our results revealed that FBLN5 expression was associated with SCNA patterns in the majority of tumors analyzed (Figure 3F). We also examined the relationship between DNA methylation and FBLN5 mRNA expression, and observed a negative correlation between FBLN5 mRNA expression and DNA methylation status (Figure 3G). Therefore, it can be concluded that CNAs of FBLN5 are common across various tumor types and can influence FBLN5 expression levels. Notably, the DNA methylation levels were significantly higher in the tumor tissues than the corresponding normal tissues (Figure 3H). This is because response to significantly variated FBLN5 expression, the body attempts to restore balance by increasing promoter methylation as a compensatory mechanism.

FBLN5 and cancer-related pathways

A GSEA was conducted to identify key cell signaling pathways in different tumor types. An enrichment analysis of FBLN5 across 50 HALLMARK and 83 metabolism-related gene sets showed that tumors with higher FBLN5 expression were consistently enriched in the ultraviolet (UV) response pathway, peroxisome metabolism pathway, Notch-signaling pathway, heme metabolism pathway, and other metabolism-related pathways across multiple tumor types. The results indicated that FBLN5 is involved in the aforementioned processes in tumors (Figure 4A,4B).

Figure 4 FBLN5 expression and associated pathways. (A,B) Enrichment differences in FBLN5 across 50 HALLMARK and 83 metabolism-related gene sets. (C) Correlation analyses between FBLN5 mRNA expression and 14 malignant features across all tumors. (D,E) Correlation of FBLN5 mRNA expression with EMT and stemness across all tumors. (F) Results of the CHIP-seq data and expression analyses. CHIP-seq, chromatin immunoprecipitation sequencing; EMT, epithelial-mesenchymal transition; FBLN5, fibulin-5; FDR, false discovery rate; mRNA, messenger RNA; NES, normalized enrichment score.

Correlation analyses were performed between the FBLN5 expression z-scores, and the z-scores of 14 functional hallmarks of tumor, including angiogenesis, metastasis, proliferation, and other critical pathways. Notably, EMT and stemness had the highest and second-highest correlation coefficients, respectively, among all functional phenotypes (R=0.51, P<0.001), indicating a significant positive association with FBLN5 expression (Figure 4C). Among them, COAD and PRAD had the highest coefficients for the two pathways, respectively (Figure 4D,4E). Previous studies have also confirmed that FBLN5 promote the progression of tumors through EMT and stemness (12,18). While the relationship between the two pathways with COAD and PRAD needed to be studied further.

By combining chromatin immunoprecipitation sequencing (ChIP-seq) data with expression analyses, we demonstrated that the transcription factor EP300 binds directly to FBLN5 and may upregulate its expression (Figure 4F).

Using the Compartmentalized Protein-Protein Interaction (ComPPI) database, we explored the FBLN5-interacting proteins, and found that they were positioned in the cytosol, nucleus, or membrane, and were functionally associated with tumor occurrence, antioxidant activity, DNA damage repair, etc. (Figure 5A).

Figure 5 FBLN5 expression and gene interactions. (A) Network of genes interacting with FBLN5. (B) KEGG pathway enrichment analysis results showing highly expressed genes. ECM, extracellular matrix; FBLN5, fibulin-5; IL, interleukin; KEGG, Kyoto Encyclopedia of Genes and Genomes; TGF, transforming growth factor; TNF, tumor necrosis factor.

Using the Kyoto Encyclopedia of Genes and Genomes (KEGG), we also analyzed the pathways related to FBLN5, and found that the calcium signaling pathway, cell adhesion molecules, and the chemokine signaling pathway were the three most relevant pathways (Figure 5B). It could be inferred from former study that the structure with calcium-binding and elastin-binding associated domain of FBLN5 may have relationship with calcium signaling pathway (7).

FBLN5 expression is associated with immunity in cancers

We analyzed the relationship between FBLN5 and immune infiltration by examining immune-related gene mRNA and immune cells. The heatmap showed that FBLN5 was positively related to transforming growth factor-β1 (TGF-β1) and interleukin-10 (IL-10), while it was negatively related to C-X-C motif chemokine ligand 9 (CXCL9) and C-X-C motif chemokine ligand 10 (CXCL10) (Figure 6A). And this echoed previous results which demonstrated that FBLN5 worked via TGFβ signalling (6).

Figure 6 FBLN5 expression and immune infiltration. (A) Heatmap showing correlations between FBLN5 mRNA expression, and the expression of chemokines, chemokine receptors, immune inhibitors, immune stimulators, and MHC genes. (B) Identification of significant correlations (FDR <0.05) between FBLN5 expression and tumor-infiltrating immune cell subsets, as well as stromal compartments, using the TIMER2.0 platform. (C) Single-cell RNA sequencing analysis showing the relationship between FBLN5 expression and proliferating T cells. *, P<0.05; **, P<0.01; ***, P<0.001. FBLN5, fibulin-5; FDR, false discovery rate; MHC, major histocompatibility complex; mRNA, messenger RNA; TIMER2.0, Tumor Immune Estimation Resource 2.0.

Using the TIMER2.0 platform, we identified significant correlations [false discovery rate (FDR) <0.05] between FBLN5 expression and the differential abundance of tumor-infiltrating immune subsets (including CD8⁺ T cells and macrophages) and stromal compartments (including cancer-associated fibroblasts and endothelial cells) (Figure 6B). Due to the varying proportions of immune infiltration and distinct TMEs across pan-cancers, the observed correlations exhibited varied patterns. The single-cell RNA sequencing analysis revealed that while FBLN5 expression was generally low across tumor populations, its primary cellular sources comprised malignant cells and proliferating T-cell subsets (Figure 6C). This infer that FBLN5 may foster immunological rejection or a state marked by an “immunological desert”and thus serve as a key driver in immunity-cancer crosstalk, with a particular emphasis on facilitating immune escape.

Drug sensitivity analysis

Three different databases (CellMiner, CTRP, and GDSC) were used to analyze the relationship between FBLN5 expression and drug sensitivity. The analysis of the CellMiner database revealed significant negative correlations between FBLN5 expression and sensitivity to multiple chemotherapeutic agents (Pearson r>0.3, FDR <0.05) (Figure 7A). The CTRP database results also suggested that FBLN5 expression is associated with chemotherapy drug resistance, and an increased sensitivity to targeted drugs and immunotherapeutic drugs (Figure 7B). The GDSC database revealed that the elevated expression of the FBLN5 gene may lead to resistance to pazopanib (Figure 7C). Therefore, our findings suggest that FBLN5 does not appear to be a potential chemotherapy-sensitive gene.

Figure 7 FBLN5 expression and drug sensitivity. (A) Analysis of the relationship between FBLN5 expression and sensitivity to multiple chemotherapeutic agents using the CellMiner database. (B) The analysis of FBLN5 expression and sensitivity to multiple antitumor drugs with CTRP database. (C) The analysis of FBLN5 expression and sensitivity to multiple antitumor drugs with GDSC database. (D) Prediction of potential compounds targeting FBLN5 with CMap database. CTRP, Cancer Therapeutics Response Portal; FBLN5, fibulin-5; FDR, false discovery rate; GDSC, Genomics of Drug Sensitivity in Cancer; mRNA, messenger RNA.

To explore the relationship between the FBLN5 gene and drug interactions, we used the CMap database to analyze changes in FBLN5 gene expression following drug treatment. The results showed that fumonisin-b1 was downregulated in multiple tumor types, suggesting potential broad-spectrum anti-cancer effects. BX-01338 and PD-0325901 were upregulated in multiple types of tumor, indicating potential broad-spectrum oncogenic effects. Afatinib was downregulated in lung cancer, which is consistent with clinical practice. Benzylpenicillin appeared red in some tumor types, potentially indicating an unexpected pro-cancer effect of the antibiotic (Figure 7D). These findings lend substantial support to the validity of our predictions, although further research is required to gain a deeper understanding of the underlying mechanisms.

Expression of FBLN5 in tissue samples

IHC staining was performed on tissue samples from lung cancer patients to assess FBLN5 expression. Weak diffuse cytoplasmic staining of FBLN5 was observed in the LUAD samples (Figure 8A) and strong staining was observed in the normal tissue (Figure 8B). The results were similar in the LUSC samples and the normal tissue (Figure 8C,8D).

Figure 8 IHC staining of FBLN5 in tumor and normal tissues. (A) IHC staining in lung adenocarcinoma tissue. (B) IHC staining in the adjacent non-tumor lung tissue from a lung adenocarcinoma patient. (C) IHC staining in the squamous cell lung carcinoma tissue. (D) IHC staining in the adjacent non-tumor lung tissue from a squamous cell lung carcinoma patient. Scale bar =200 µm. FBLN5, fibulin-5; IHC, immunohistochemical.

Discussion

Cancer research remains a central yet challenging frontier in biomedical science. The ECM is increasingly recognized as a critical regulator of tumor progression and the therapeutic response. In this study, we employed a multi-omics integrative approach, leveraging data from diverse platforms to systematically investigate the pan-cancer roles of FBLN5. Our comprehensive pan-cancer analysis identified FBLN5 as a multifaceted ECM component with context-dependent roles across different malignancies, influencing prognosis, immune modulation, and drug sensitivity.

The ECM is a dynamic and complex network of proteins that provides structural support to tissues and regulates critical cellular processes, including proliferation, migration, and signaling (19). Dysregulation of ECM remodeling is a hallmark of tumor progression, contributing to tumor invasion, metastasis, and therapy resistance (20). Among the ECM components, the fibulin family of glycoproteins has emerged as key players in tumor biology, with FBLN5 exhibiting context-dependent roles in tumor suppression and promotion (21). FBLN5, also known as DANCE, is a secreted ECM protein that mediates cell-matrix interactions by binding to integrins, elastin, and other ECM proteins (22).

Initially studied for its role in vascular development and elastic fiber assembly (23), FBLN5 has since been implicated in tumor progression, though its functions appear tissue-specific and stage-dependent. In breast cancer, FBLN5 acts as a tumor suppressor by inhibiting angiogenesis and EMT (24). In hepatocellular cancer and ovarian cancer, FBLN5 promotes metastasis by enhancing ECM stiffness and activating integrin-mediated signaling (11,25). The expression of FBLN5 was found to be significantly reduced in KIRC tissue; thus, FBLN5 could serve as a potential independent prognostic biomarker of KIRC in children and young adults (26). In gastric cancer, the expression level of FBLN5 was upregulated, which was associated with a poor prognosis (15).

Despite these advances, a systematic pan-cancer analysis of FBLN5’s clinical relevance, mechanistic underpinnings, and therapeutic potential had not previously been conducted. Our initial investigations quantified FBLN5 expression at both the mRNA and protein levels across multiple tumor types. At the mRNA level, FBLN5 expression varied widely across different tumor types with some showing elevated levels in tumor tissues and cells, and others exhibiting reduced expression compared to normal tissues. In the limited relevant studies with FBLN5 in tumors showed that except for gastric cancer and pancreatic cancer, the expression of FBLN5 were lower in tumor tissues than in normal tissues which demonstrated its function of tumor suppressor genes. While in gastric cancer and pancreatic caner, it was the opposite. In our study, the expression of FBLN5 was lower in gastric cancer especially the expression of mRNA, that was different from the past studies. But the database they used before was a little different from the one we used and we have not performed IHC or western-blot of gastric cancer tissue so that it was impossible to determine the expression within the tumor tissue. And further studies may be conduct in the future. Besides, our study demonstrated that FBLN5 protein overexpressed in the pancreatic cancer that was similar with past research (9-16).

The IHC results demonstrated that FBLN5 is expressed in most normal organs in both male and females. However, its expression is significantly decreased in tumor tissues. In testicular cancer, FBLN5 exhibited moderate staining, representing the strongest staining among all tumor types. The inconsistencies in mRNA and protein expression suggests that FBLN5 may undergo epigenetic changes such as transcriptional and post-translational modifications.

We also investigated the relationship between FBLN5 expression and prognosis, and found that it varies across different tumors. High FBLN5 expression was associated with longevity in certain tumors, while in other tumors, it was associated with a shorter survival time. The relationship between FBLN5 expression and OS, DSS, DFI, and PFI is similar. While in gastric cancer, FBLN5 was associated with a poor prognosis, and in breast cancer, it was associated with a good prognosis. Combined with the expression results mentioned above, it may conclude that FBLN5 still was an oncogene in gastric cancer but an antioncogene in breast cancer like previous researches (12,15). Consistent with former findings, these results suggest that FBLN5 exhibits heterogeneity in different tumors, exerting pro- or anti-tumor effects.

We performed comprehensive mutational profiling across pan-cancer cohorts to characterize mutation types, frequencies, and genomic loci. The genomic landscape of FBLN5 reveals significant alterations across tumors, with important implications for tumor progression and the therapeutic response. Mutations, amplifications, and deep deletions were the most common types of genetic variations identified in our study. Using integrated 2D linear domain mapping and 3D protein structural modeling, we comprehensively visualized the spatial distribution of FBLN5 mutation hotspots across tumor types. Previous research had found FBLN5 gene deletion in bladder cancer, while functional experiments have demonstrated that restoring FBLN5 expression can inhibit tumor cell proliferation and invasion (10). Another researcher suggested that the RGD-to-arginyl-glycyl-glutamate (RGE) change of FBLN5 may enhance tumor invasion (12). And previous researchers also found that calcium-binding EGF-like domains and the C-terminal integrin-binding motif are associated with the disruption of ECM integrity and impaired cell-matrix adhesion, respectively (27,28). We may conclude that mutation and gene deletion of FBLN5 were quite common in tumors and associated with the progression.

At the same time, we found for the first time that copy number variations are commonly observed in tumors in our study. Especially the SCNAs, was associated with FBLN5 expression in the majority of tumors which could affect FBLN5 expression. Methylation was also found to be significantly increased in many types of tumors in our research, and consistent with previous studies, a close correlation between methylation and FBLN5 mRNA expression was observed (29). Also, we found the DNA methylation levels were significantly higher in the tumor tissues than the corresponding normal tissues, which may variate according to the FBLN5 expression level and indicate its potential role in the progress.

We also performed pathway enrichment analyses in the research, and found that the co-expression of FBLN5 and pathways varied slightly across different tumors; however, EMT and stemness-related pathways were the most significantly enriched. Consistent with previous studies, EMT as well as stemness may increase the risk of tumor progression and metastasis (18,30). In intrahepatic cholangiocarcinoma (ICC), LOXL1 interacts with FBLN5 to regulate angiogenesis by binding to the α v β 3 integrin via an Arg-Gly-Asp domain-dependent mechanism, thereby activating the focal adhesion kinase (FAK)-mitogen-activated protein kinase (MAPK) signaling pathway in vascular endothelial cells (31). In pancreatic ductal adenocarcinoma (PDA), hypoxia triggers FBLN5 expression in a TGF-β- and PI3K-dependent manner, promoting PDA progression (32). In breast cancer, miR-370-3p inhibits FBLN5 expression and activates the nuclear factor kappa-light-chain-enhancer of activated B cells (NF-κB) signaling pathway, promoting breast cancer cell proliferation, migration, and stemness (12).

Previous studies examining the relationship between FBLN5 and immune infiltration are limited. Our study found that several immune-related factors, including TGF-β1, CD8⁺ T cells, and macrophages, were significantly and negatively correlated with FBLN5 expression and across multiple tumor types for the first time, suggesting that FBLN5 overexpressing tumors may be resistant to immunotherapy. Furthermore, the results suggest that FBLN5 is functionally adverse, facilitating immune escape, which in turn blunts immune responses to tumor treatments and ultimately results in unfavorable clinical outcomes.

Unlike past studies, we further explored chemotherapeutic drug resistance at a more mechanistic level to elucidate the principle underlying diminished responses to medications. From various angles, we analyzed correlations between FBLN5 expressions and drug sensitivity. We found that FBLN5 was associated with resistance to various chemotherapy drugs, targeted therapies, and immunotherapeutic drugs, thereby addressing a gap in the current literature. Thus, we think that FBLN5 may represent a potential target for overcoming treatment resistance. We can use this as a target to conduct in-depth research on drug resistance in the future.

This study systematically analyzed the expression, prognostic significance, and biological roles of FBLN from a bioinformatics perspective. By conducting comprehensive pan-cancer bioinformatics analyses of multi-platform genomic datasets, we identified the significant differential expression of FBLN5 between malignant and normal tissues across diverse tumor types. FBLN5 RNA expression exhibited polarization across tumors, while protein levels were significantly decreased. Meanwhile, the FBLN5 expression was highly related to the prognosis. Therefore, FBLN5 may be a good factor for diagnosis and prognosis assessment in certain types of tumor. FBLN5 expression was closely associated with methylation, suggesting that the low expression of the FBLN5 protein may be associated with methylation. Meanwhile, FBLN5 shaped the immunosuppressive TME across tumors and caused resistance to immunotherapy. FBLN5 expression was highly correlated with chemotherapy and targeted drug resistance, potentially due to FBLN5-mediated ECM stiffness. Considering its insensitivity to immunotherapy and drug treatment as well as the significant role in the occurrence and development of tumors, FBLN5 will be a very promising target for precise treatment both in monotherapy and combination therapy with immunotherapy.

This study had several limitations. First, no analysis was performed on individual tumor types to explore the patterns and reasons for survival differences across tumors. Second, the data analysis and the investigation of the mechanisms related to immunity and drug resistance were limited. Third, clinically relevant data and sample collections were lacking. We intend to conduct further and more comprehensive research in the future to address these issues.

Conclusions

FBLN5 is a key member of the ECM, that plays an important role in the occurrence and development of tumors. FBLN5 expressed differently between malignant and normal tissues across diverse tumor types and the expression was closely associated with methylation, chemotherapy and targeted drug resistance. Meanwhile, FBLN5 shaped the immunosuppressive TME across tumors and caused resistance to immunotherapy.

Acknowledgments

The authors gratefully acknowledge the reviewers for their insightful and constructive comments.

Footnote

Reporting Checklist: The authors have completed the REMARK reporting checklist. Available at https://tcr.amegroups.com/article/view/10.21037/tcr-2026-0623/rc

Data Sharing Statement: Available at https://tcr.amegroups.com/article/view/10.21037/tcr-2026-0623/dss

Peer Review File: Available at https://tcr.amegroups.com/article/view/10.21037/tcr-2026-0623/prf

Funding: This study was funded by grants from the Ningbo Natural Science Foundation (No. 2023J225) and the Zhejiang Provincial Medical and Health Science and Technology Project (No. 2025KY1268).

Conflicts of Interest: All authors have completed the ICMJE uniform disclosure form (available at https://tcr.amegroups.com/article/view/10.21037/tcr-2026-0623/coif). The authors have no conflicts of interest to declare.

Ethical Statement: The authors are accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved. The study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments.

Open Access Statement: This is an Open Access article distributed in accordance with the Creative Commons Attribution-NonCommercial-NoDerivs 4.0 International License (CC BY-NC-ND 4.0), which permits the non-commercial replication and distribution of the article with the strict proviso that no changes or edits are made and the original work is properly cited (including links to both the formal publication through the relevant DOI and the license). See: https://creativecommons.org/licenses/by-nc-nd/4.0/.

References

1. Sung H, Ferlay J, Siegel RL, et al. Global Cancer Statistics 2020: GLOBOCAN Estimates of Incidence and Mortality Worldwide for 36 Cancers in 185 Countries. CA Cancer J Clin 2021;71:209-49. [Crossref] [PubMed]
2. Hoadley KA, Yau C, Hinoue T, et al. Cell-of-Origin Patterns Dominate the Molecular Classification of 10,000 Tumors from 33 Types of Cancer. Cell 2018;173:291-304.e6. [Crossref] [PubMed]
3. Cancer Genome Atlas Research Network. The Cancer Genome Atlas Pan-Cancer analysis project. Nat Genet 2013;45:1113-20.
4. Sanchez-Vega F, Mina M, Armenia J, et al. Oncogenic Signaling Pathways in The Cancer Genome Atlas. Cell 2018;173:321-337.e10. [Crossref] [PubMed]
5. Hutter C, Zenklusen JC. The Cancer Genome Atlas: Creating Lasting Value beyond Its Data. Cell 2018;173:283-5. [Crossref] [PubMed]
6. Zhou JJ, Qiao E, He ZY, et al. Fibulin 5 mediates pulmonary artery smooth muscle cell dysfunction and vascular remodelling in congenital heart disease associated pulmonary arterial hypertension via TGF-β1/PI3K/AKT signalling. Respir Res 2025;26:276. [Crossref] [PubMed]
7. Okuyama T, Tsuno T, Inoue R, et al. The matricellular protein Fibulin-5 regulates β-cell proliferation in an autocrine/paracrine manner. iScience 2025;28:111856. [Crossref] [PubMed]
8. Suzuki Y, Nampei M, Kawakita F, et al. The effect of Fibulin-5 on early brain injury after subarachnoid hemorrhage in mice. Neurochem Int 2025;187:105989. [Crossref] [PubMed]
9. Wlazlinski A, Engers R, Hoffmann MJ, et al. Downregulation of several fibulin genes in prostate cancer. Prostate 2007;67:1770-80. [Crossref] [PubMed]
10. Hu Z, Ai Q, Xu H, et al. Fibulin-5 is down-regulated in urothelial carcinoma of bladder and inhibits growth and invasion of human bladder cancer cell line 5637. Urol Oncol 2011;29:430-5. [Crossref] [PubMed]
11. Tang JC, Liu JH, Liu XL, et al. Effect of fibulin-5 on adhesion, migration and invasion of hepatocellular carcinoma cells via an integrin-dependent mechanism. World J Gastroenterol 2015;21:11127-40. [Crossref] [PubMed]
12. Mao J, Wang L, Wu J, et al. miR-370-3p as a Novel Biomarker Promotes Breast Cancer Progression by Targeting FBLN5. Stem Cells Int 2021;2021:4649890. [Crossref] [PubMed]
13. Gao C, Wang Y, Zhou G, et al. KIF1Bβ suppresses hepatocellular carcinoma by transporting and secreting FBLN5 to attenuate the integrin pathway. Gut 2026;75:1030-42. [Crossref] [PubMed]
14. Chen J, Bai Y, Liu S, et al. A combined computational and functional approach identifies FBLN5 as an amino acid metabolism-related gene and suppressing colorectal cancer progression. Mol Genet Genomics 2026;301:28. [Crossref] [PubMed]
15. Bian X, Yin S, Yin X, et al. Clinical and Biological Significances of FBLN5 in Gastric Cancer. Cancers (Basel) 2023;15:553. [Crossref] [PubMed]
16. Wang M, Topalovski M, Toombs JE, et al. Fibulin-5 Blocks Microenvironmental ROS in Pancreatic Cancer. Cancer Res 2015;75:5058-69. [Crossref] [PubMed]
17. Zhang R, Yan W, Yuan J, et al. Cancer-associated fibroblast-derived fibulin-5 promotes radioresistance in non-small-cell lung cancer. Cell Rep 2025;44:116018. [Crossref] [PubMed]
18. Choi J, Kwak Y, Park M, et al. Cancer-associated fibroblast-derived fibulin-5 promotes epithelial-mesenchymal transition in diffuse-type gastric cancer via cAMP response element-binding protein pathway, showing poor prognosis. Exp Mol Med 2025;57:1106-19. [Crossref] [PubMed]
19. Wang D, Brady T, Santhanam L, et al. The extracellular matrix mechanics in the vasculature. Nat Cardiovasc Res 2023;2:718-32. [Crossref] [PubMed]
20. Sleeboom JJF, van Tienderen GS, Schenke-Layland K, et al. The extracellular matrix as hallmark of cancer and metastasis: From biomechanics to therapeutic targets. Sci Transl Med 2024;16:eadg3840. [Crossref] [PubMed]
21. Li R, Wu H, Jiang H, et al. FBLN5 is targeted by microRNA 27a 3p and suppresses tumorigenesis and progression in high grade serous ovarian carcinoma. Oncol Rep 2020;44:2143-51. [Crossref] [PubMed]
22. Thomas C, Bouezzedine F, Bonnier D, et al. Proteomic analysis of liver fibrosis reveals EFEMP1 as a new modulator of focal adhesion and migration of hepatic stellate cells. FASEB J 2025;39:e70515. [Crossref] [PubMed]
23. Ramachandra AB, Mikush N, Sauler M, et al. Compromised Cardiopulmonary Function in Fibulin-5 Deficient Mice. J Biomech Eng 2022;144:081008. [Crossref] [PubMed]
24. Xie L, Palmsten K, MacDonald B, et al. Basement membrane derived fibulin-1 and fibulin-5 function as angiogenesis inhibitors and suppress tumor growth. Exp Biol Med (Maywood) 2008;233:155-62. [Crossref] [PubMed]
25. Heo JH, Song JY, Jeong JY, et al. Fibulin-5 is a tumour suppressor inhibiting cell migration and invasion in ovarian cancer. J Clin Pathol 2016;69:109-16. [Crossref] [PubMed]
26. Zhang M, Chen F, Feng S, et al. FBLN5 as One Presumably Prognostic Gene Potentially Modulating Tumor Immune Microenvironment for Renal Clear Cell Carcinoma in Children and Young Adults. Pharmgenomics Pers Med 2024;17:27-40. [Crossref] [PubMed]
27. Shi J, Gong P, Ye K, et al. Elevated Serum Fibulin-5 Levels Correlate with Coronary Artery Disease Progression. Int J Gen Med 2025;18:4625-34. [Crossref] [PubMed]
28. Gharesouran J, Hosseinzadeh H, Ghafouri-Fard S, et al. New insight into clinical heterogeneity and inheritance diversity of FBLN5-related cutis laxa. Orphanet J Rare Dis 2021;16:51. [Crossref] [PubMed]
29. Yan B, Hu Y, Ma T, et al. IDH1 mutation promotes lung cancer cell proliferation through methylation of Fibulin-5. Open Biol 2018;8:180086. [Crossref] [PubMed]
30. Tian W, Luo Y, Tang Y, et al. Novel Implication of the Basement Membrane for Breast Cancer Outcome and Immune Infiltration. Int J Biol Sci 2023;19:1645-63. [Crossref] [PubMed]
31. Yuan R, Li Y, Yang B, et al. LOXL1 exerts oncogenesis and stimulates angiogenesis through the LOXL1-FBLN5/αvβ3 integrin/FAK-MAPK axis in ICC. Mol Ther Nucleic Acids 2021;23:797-810. [Crossref] [PubMed]
32. Topalovski M, Hagopian M, Wang M, et al. Hypoxia and Transforming Growth Factor β Cooperate to Induce Fibulin-5 Expression in Pancreatic Cancer. J Biol Chem 2016;291:22244-52. [Crossref] [PubMed]

(English Language Editor: L. Huleatt)