Identification of biomarkers associated with diagnosis and prognosis of colorectal cancer patients based on integrated bioinformatics analysis

doi:10.1016/j.gene.2019.01.001

Gene

Volume 692, 15 April 2019, Pages 119-125

https://doi.org/10.1016/j.gene.2019.01.001 Get rights and content

Highlights

•
Colorectal cancer related differentially expressed genes were identified based on GEO database and TCGA database.
•
The top 10 hub genes with relatively high diagnostic values for the patients of CRC were identified.
•
A 9-gene prognostic signature was identified as a potential prognostic predictor for CRC patients.

Abstract

Background

The current study aimed to identify potential diagnostic and prognostic gene biomarkers for colorectal cancer (CRC) based on the Gene Expression Omnibus (GEO) datasets and The Cancer Genome Atlas (TCGA) dataset.

Methods

Microarray data of gene expression profiles of CRC from GEO and RNA-sequencing dataset of CRC from TCGA were downloaded. After screening overlapping differentially expressed genes (DEGs) by R software, functional enrichment analyses of the DEGs were performed using the DAVID database. Then, the STRING database and Cytoscape were used to construct a protein–protein interaction (PPI) network and identify hub genes. The receiver operating characteristic (ROC) curves were conducted to assess the diagnostic values of the hub genes. Cox proportional hazards regression was performed to screen the potential prognostic genes. Kaplan-Meier curve and the time-dependent ROC curve were used to assess the prognostic values of the potential prognostic genes for CRC patients.

Results

Integrated analysis of GEO and TCGA databases revealed 207 common DEGs in CRC. A PPI network consisted of 70 nodes and 170 edges were constructed and top 10 hub genes were identified. The area under curve (AUC) of the ROC curves of the hub genes were 0.900, 0.927, 0.869, 0.863, 0.980, 0.682, 0.903, 0.790, 0.995, and 0.989 for CCL19, CXCL1, CXCL5, CXCL11, CXCL12, GNG4, INSL5, NMU, PYY, and SST, respectively. A prognostic gene signature consisted of 9 genes including SLC4A4, NFE2L3, GLDN, PCOLCE2, TIMP1, CCL28, SCGB2A1, AXIN2, and MMP1 was constructed with a good performance in predicting overall survivals of CRC patients. The AUC of the time-dependent ROC curve was 0.741 for 5-year survival.

Conclusion

The results in this study might provide some directive significance for further exploring the potential biomarkers for diagnosis and prognosis prediction of CRC patients.

Introduction

Colorectal cancer (CRC) is one of the most common cancers with high morbidity and mortality worldwide. Nearly 1.4 million new cases of CRC and 700,000 CRC-related deaths were reported each year in the world (Torre et al., 2015). Brenner et al. (2014) found that the 5-year survival rate of CRC patients was >90% when diagnosed at early stages. Due to the lack of adequate diagnostic methods, CRC is often diagnosed at an advanced stage. Despite the significant improvements in diagnosis and treatment, the 5-year survival rate for CRC patients diagnosed with metastatic is still low at approximately 12% (Siegel et al., 2015). Thus, it's urgently needed for understanding the molecular mechanisms of CRC development and identification of novel biomarkers are used for the early detection and prognosis evaluation of CRC.

Over the past decades, developing of molecular biology has increased our understanding of the pathogenesis of CRC. Previous researches have indicated that CRC is a genetic disease, which depends on alteration of numerous of oncogenes and tumor suppressor genes (Bogaert and Prenen, 2014). A growing number of genes and their coding proteins related to CRC have been explored. Previous studies revealed that they play crucial roles in a large number of physiological and pathological processes including cell proliferation, differentiation, apoptosis, and metastasis (Hisamuddin and Yang, 2006; Testa et al., 2018). However, the precise molecular mechanisms of CRC are still far from being deep understood. Recently, several studies have discovered a group of CRC related candidate genes by bioinformatics analysis. For example, Huang et al. (2018) identified hundreds of CRC associated differentially expressed genes (DEGs) based on the Gene Expression Omnibus (GEO) and The Cancer Genome Atlas (TCGA) database, and five genes of which can used as diagnostic biomarkers for CRC patients. Hou et al. (2018) found a collection of DEGs and DNA methylation aberrations in CRC. Their results indicated that the combination of DEGs, DNA methylation aberrations, and tumor stages result in more effective prognostic evaluation of patients with CRC. In addition, lncRNAs and the associated regulatory network consisting of transcription factors, microRNAs, mRNAs, and RNA-binding proteins could also be identified using bioinformatics analysis (Zhang et al., 2018). But it is still paramount to find more potential biomarkers for effective diagnosis and prognosis assessment of CRC patients.

In this study, we downloaded large-scale gene datasets regarding CRC from GEO and TCGA databases. After integrated analysis of both two databases, we identified 10 hub genes from the common DEGs by constructing protein-protein interaction (PPI) network. The results of receiver operating characteristic (ROC) curves showed that the top 10 hub genes had high diagnostic values for patients with CRC. Then, we conducted a gene signature for prognosis of CRC patients by univariate and multivariable Cox regression analyses, which performed well in predicting overall survivals of CRC patients.

Section snippets

Data collection

Series matrix files of GSE32323, GSE74602, and GSE113513 were downloaded from the GEO (http://www.ncbi.nlm.nih.gov/geo/) database. The platforms they based on were GPL570 (Affymetrix Human Genome U133 Plus 2.0 Array), GPL6104 (Illumina humanRef-8 v2.0 expression beadchip), and GPL15207 (Affymetrix Human Gene Expression Array), respectively. The datasets of GSE32323 contained 17 cancer tissues and paired adjacent non-cancerous tissues. The datasets of GSE74602 comprised 30 paired normal and

Identification of DEGs in CRC

In the present study, 1445 DEGs in GSE32323, 1484 DEGs in GSE74602, 1284 DEGs in GSE113513, and 2150 DEGs in TCGA were identified. Among the DEGs, 700, 815, 511, and 1205 genes were upregulated while 745, 669, 773, and 945 genes were downregulated in GSE32323, GSE74602, GSE113513, and TCGA, respectively (Fig. 1A). We used Volcano Plots to visualize the DEGs in different studies. Red dots represent the upregulated genes and green dots represent the downregulated genes. The consistently

Discussion

In this study, we integratedly analyzed three microarray datasets from GEO and RNA sequencing data from TCGA. A total of 207 DEGs consisting of 57 upregulated DEGs and 150 downregulated DEGs were identified between CRC tissues and normal tissues. The functional enrichment analyses demonstrated that the DEGs were enriched in some biological processes such as EMC organization, angiogenesis, cell adhesion, cell differentiation, and cell migration. The results consistent with previous knowledge

Acknowledgments

This study was supported by the Zhejiang Provincial Natural Science Foundation of China (No. LY16H160004) and Natural Science Foundation of Ningbo (No. 2018A610381). The authors thank the contributors of the TCGA (https://tcga-data.nci.nih.gov/) database and GEO (http://www.ncbi.nlm.nih.gov/geo/) database for sharing their data on open access.

Conflict of interest

The authors declare that they have no conflict of interest.

References (44)

H. Brenner et al.
Colorectal cancer
Lancet
(2014)
F. Hohla et al.
Targeted cytotoxic somatostatin analog AN-162 inhibits growth of human colon carcinomas and increases sensitivity of doxorubicin resistant murine leukemia cells
Cancer Lett.
(2010)
M. Kawamura et al.
CXCL5, a promoter of cell proliferation, migration and invasion, is a novel serum prognostic marker in patients with colorectal cancer
Eur. J. Cancer
(2012)
X. Liu et al.
Identification of potential key genes associated with the pathogenesis and prognosis of gastric cancer based on integrated bioinformatics analysis
Front. Genet.
(2018)
P. Przygodzka et al.
Neuromedin U is upregulated by snail at early stages of EMT in HT29 colon cancer cells
Biochim. Biophys. Acta
(2016)
S. Pyronnet et al.
Antitumor effects of somatostatin
Mol. Cell. Endocrinol.
(2008)
A. Reaux-Le Goazigo et al.
Current status of chemokines in the adult CNS
Prog. Neurobiol.
(2013)
C. Susini et al.
Rationale for the use of somatostatin analogs as antitumor agents
Ann. Oncol.
(2006)
W.W. Tseng et al.
Peptide YY and cancer: current findings and potential clinical applications
Peptides
(2002)
J. Vandercappellen et al.
The role of CXC chemokines and their receptors in cancer
Cancer Lett.
(2008)

Y. Zhang et al.

The regulatory network analysis of long noncoding RNAs in human colorectal cancer

Funct. Integr. Genom.

(2018)

J. Bogaert et al.

Molecular genetics of colorectal cancer

Ann. Gastroenterol.

(2014)

S. Cheng et al.

Crk-like adapter protein regulates CCL19/CCR7-mediated epithelial-to-mesenchymal transition via ERK signaling pathway in epithelial ovarian carcinomas

Med. Oncol.

(2015)

C.H. Chin et al.

cytoHubba: identifying hub objects and sub-networks from complex interactome

BMC Syst. Biol.

(2014)

J. Hamanishi et al.

Activated local immunity by CC chemokine ligand 19-transduced embryonic endothelial progenitor cells suppresses metastasis of murine ovarian cancer

Stem Cells

(2010)

P.J. Heagerty et al.

Survival model predictive accuracy and ROC curves

Biometrics

(2005)

I.M. Hisamuddin et al.

Molecular genetics of colorectal cancer: an overview

Curr. Color. Cancer Rep.

(2006)

X. Hou et al.

Genome-wide network-based analysis of colorectal cancer identifies novel prognostic factors and an integrative prognostic index

Cell. Physiol. Biochem.

(2018)

W. Huang da et al.

Systematic and integrative analysis of large gene lists using DAVID bioinformatics resources

Nat. Protoc.

(2009)

Z. Huang et al.

Identification of critical genes and five prognostic biomarkers associated with colorectal cancer

Med. Sci. Monit.

(2018)

M. Itakura et al.

High CC chemokine receptor 7 expression improves postoperative prognosis of lung adenocarcinoma patients

Br. J. Cancer

(2013)

J. Lu et al.

Antitumor efficacy of CC motif chemokine ligand 19 in colorectal cancer

Dig. Dis. Sci.

(2014)

Cited by (134)

Linearithmic and unbiased implementation of DeLong's algorithm for comparing the areas under correlated ROC curves
2024, Expert Systems with Applications
Receiver operating characteristic (ROC) analysis is a widespread framework in two-group classification problems. The area under the ROC curve (AUC) has been a popular figure of merit to summarize the performance of binary classifiers. Because of its simplicity of implementation, DeLong’s algorithm (DLA) has been widely used in practice. However, DLA suffers two drawbacks of time inefficiency and biasedness. The authors’ previous work has overcome the first drawback. Specifically, the time complexity of DLA is reduced from the quadratic order down to a linearithmic one, by exploiting the relationship between the Heaviside function and the mid-ranks of samples. Unfortunately, the second drawback of DLA remains unaddressed, to the best of our knowledge. Motivated by such unsatisfactory situation, in this paper, the authors correct the bias of DLA with the assistance of an efficient implementation of Kendall’s tau (KT). The overall computational load of the improved DLA (being strictly unbiased) is in linearithmic time. Theoretical analyses and numerical results verified the proposed algorithmic findings in this work.
Exploring salivary gene expression clusters: A bioinformatics approach for advanced diagnosis and prognosis in head and neck squamous cell carcinoma
2024, Oral Oncology Reports
This study postulates that distinct patterns in salivary gene expression clusters among HNSCC patients provide valuable insights into diagnostic and prognostic markers for the disease. The aim is to identify critical genes and their protein-protein interactions through bio-computational analyses, contributing to a deeper understanding of HNSCC progression. Ultimately, the hypothesis anticipates the discovery of novel molecular signatures in saliva, potentially enhancing early detection and prognostic precision in head and neck squamous cell carcinoma.
A comprehensive keyword search was conducted to extract 74 salivary gland secretion genes from the Human Protein Atlas. GeneMANIA and STRING databases were employed to analyse protein-protein interactions and predict functional partners. UALCAN facilitated the visualization of salivary gland secretion gene expression patterns in HNSCC, providing insights into survival analysis.
In HNSCC compared to normal tissue, CHRM3, GYLTL1B, MMP3, and WDR91 were identified as statistically upregulated genes, while the downregulated genes included AGFG2, CCDC64B, CD24, CRISP3, CXCL17, ELF5, KRT7, MANSC1, MYO5C, PRR4, SCGB3A1, SH3YL1, TCN1, TRPV6, and ZG16B. Kaplan–Meier survival curve analysis revealed that high expression of salivary genes such as AGFG2, AMY1A, CCDC64B, CCL28, C4ORF7, GYLTL1B, TMEM211, LMX1B, ODAM, SCGB3A1, and SLC26A9 was significantly associated with poorer overall survival outcomes in HNSCC patients.
This study identifies distinct salivary gene expression patterns in HNSCC, highlighting potential diagnostic and prognostic markers. Specific upregulated genes, such as CHRM3 and GYLTL1B, and downregulated genes, including AGFG2 and CCDC64B, offer insights into HNSCC progression. Elevated expression of particular genes correlates with poorer overall survival, emphasising their prognostic significance.
Characterization and verification of CAF-relevant prognostic gene signature to aid therapy in bladder cancer
2024, Heliyon
Cancer-associated fibroblasts (CAFs) are significantly involved in determining the patient's prognosis and response to bladder cancer (BLCA) therapy. CAFs can induce epithelial-mesenchymal transformation (EMT) as well as complex interaction with immune cells. Hence, it is imperative to identify potential markers for enhancing our understanding of CAFs in BLCA progression and immune regulation. A variety of algorithms and analyses were employed in the study, leading to the development of a novel prognostic feature for CAFs-Stromal-EMT (CSE)-prognostic feature. This feature was constructed based on the genes MFAP5, PCOLCE2, and JUN. Furthermore, we revealed that patients with higher CSE risk scores responded to immunotherapy better compared to those with lower. Finally, we verified two CSE-related genes using in vitro experiments. Our results suggested that the CSE-prognostic feature could predict the prognosis and evaluate the response of patients to immune and chemotherapies. This would aid clinicians in designing treatment strategies for patients with BLCA.
In vivo differentiation of adenoma and carcinoma in CRC progression by PAR2-M/DiD ratiometric fluorescence and its microstructures
2023, Sensors and Actuators B: Chemical
Colorectal cancer (CRC) is the third leading cause of morbidity and mortality for malignant cancer globally. Accurate differentiation of colorectal lesions can provide information for precise treatment and minimize postoperative complications. Specific biomarkers for adenoma and carcinoma classification were compellingly needed. TCGA datasets were applied for identifying differential genes. Particularly, protease activated receptor 2 (PAR2) coded by F2RL1 gene displayed significant difference between clinical colorectal normal, adenoma and carcinoma tissues by IHC. Based on PAR2 targeting agonist, high affinity NIR probe PAR2-M was designed and validated both in vitro and in vivo, with membrane tag DiD as internal reference. In vivo fluorescence ratio of PAR2-M/DiD was absolutely higher in carcinoma than adenoma on CRC mouse models and confocal laser endomicroscopy (CLE) images exhibited their microstructures along with different PAR2 expression. The conclusive differentiation of adenoma and carcinoma was firmly supported by Convolutional Neural Networks (CNN) algorithm, with specificity of 95%.
Clinical importance and PI3K/Akt pathway-dependent anti-proliferative role of PALMD and DPT in breast cancer
2023, Pathology Research and Practice
This study aimed to identify novel differentially expressed genes in breast cancer and to explore the clinical value and the anti-tumor or oncogenic effects of the identified genes using bioinformatics analysis and in vitro experiments. The differentially expressed genes in breast cancer patients were identified using Gene Expression Omnibus (GEO) database with the cut-off criteria p < 0.05 and |logFC| > 1. The expression levels of palmdelphin (PALMD) and dermatopontin (DPT) in normal tissues and breast cancer tissues were evaluated based on GEPIA and UALCAN databases. PALMD and DPT expression levels in clinical subgroups of patients with breast cancer were analyzed to assess the association of PALMD and DPT expression with clinical characteristics. The prognostic and diagnostic values of PALMD and DPT in breast cancer were evaluated from Kaplan-Meier (K-M) survival curves and receiver operating characteristic (ROC) curves. Pearson’s correlation coefficient was performed using LinkedOmics. KEGG pathway enrichment analysis was performed using DAVID. The protein levels were evaluated using western blot analysis. Cell proliferation was assessed using MTT and EdU assays. Two important genes, PALMD and DPT, were identified in breast cancer. The expression levels of PALMD and DPT were significantly lower in breast cancer tissues. The expression levels of PALMD were closely related to age, histological type, and T stage of breast cancer patients. The expression levels of DPT were closely related to age, histological type, T stage, N stage, estrogen receptor status, and progesterone receptor status of breast cancer patients. The K-M survival curves showed that PALMD or DPT was not an independent prognostic factor for breast cancer. The ROC curves showed that both PALMD and DPT had good diagnostic potential for breast cancer. KEGG pathway enrichment results showed that PI3K/Akt pathway was an important overlapping signaling for PALMD and DPT. Further studies proved that overexpression of PALMD and DPT inhibited proliferation in MCF-7 and MDA-MB-231 cells by suppressing the PI3K/Akt pathway. PALMD and DPT knockdown promoted proliferation in MCF-7 and MDA-MB-231 cells by activating the PI3K/Akt pathway. These results collectively suggested that PALMD and DPT might serve as potential diagnostic biomarkers and therapeutic targets for breast cancer.
Integrated analysis of multi-omics data for the discovery of biomarkers and therapeutic targets for colorectal cancer
2023, Computers in Biology and Medicine
The considerable burden of colorectal cancer and the rising trend in young adults emphasize the necessity of understanding its underlying mechanisms, providing new diagnostic and prognostic markers, and improving therapeutic approaches. Precision medicine is a new trend all over the world and identification of novel biomarkers and therapeutic targets is a step forward towards this trend. In this context, multi-omics data and integrated analysis are being investigated to develop personalized medicine in the management of colorectal cancer. Given the large amount of data from multi-omics approach, data integration and analysis is a great challenge. In this Review, we summarize how statistical and machine learning techniques are applied to analyze multi-omics data and how it contributes to the discovery of useful diagnostic and prognostic biomarkers and therapeutic targets. Moreover, we discuss the importance of these biomarkers and therapeutic targets in the clinical management of colorectal cancer in the future. Taken together, integrated analysis of multi-omics data has great potential for finding novel diagnostic and prognostic biomarkers and therapeutic targets, however, there are still challenges to overcome in future studies.

View all citing articles on Scopus

¹: Feng Xu and Xianpeng Li are co-corresponding authors of this work.

View full text

Research paperIdentification of biomarkers associated with diagnosis and prognosis of colorectal cancer patients based on integrated bioinformatics analysis

Highlights

Abstract

Background

Methods

Results

Conclusion

Introduction

Section snippets

Data collection

Identification of DEGs in CRC

Discussion

Acknowledgments

Conflict of interest

Lancet

Cancer Lett.

Eur. J. Cancer

Front. Genet.

Biochim. Biophys. Acta

Mol. Cell. Endocrinol.

Prog. Neurobiol.

Ann. Oncol.

Peptides

Cancer Lett.

Funct. Integr. Genom.

Molecular genetics of colorectal cancer

Ann. Gastroenterol.

Crk-like adapter protein regulates CCL19/CCR7-mediated epithelial-to-mesenchymal transition via ERK signaling pathway in epithelial ovarian carcinomas

Med. Oncol.

cytoHubba: identifying hub objects and sub-networks from complex interactome

BMC Syst. Biol.

Activated local immunity by CC chemokine ligand 19-transduced embryonic endothelial progenitor cells suppresses metastasis of murine ovarian cancer

Stem Cells

Survival model predictive accuracy and ROC curves

Biometrics

Molecular genetics of colorectal cancer: an overview

Curr. Color. Cancer Rep.

Genome-wide network-based analysis of colorectal cancer identifies novel prognostic factors and an integrative prognostic index

Cell. Physiol. Biochem.

Systematic and integrative analysis of large gene lists using DAVID bioinformatics resources

Nat. Protoc.

Identification of critical genes and five prognostic biomarkers associated with colorectal cancer

Med. Sci. Monit.

High CC chemokine receptor 7 expression improves postoperative prognosis of lung adenocarcinoma patients

Br. J. Cancer

Antitumor efficacy of CC motif chemokine ligand 19 in colorectal cancer

Dig. Dis. Sci.

Research paper
Identification of biomarkers associated with diagnosis and prognosis of colorectal cancer patients based on integrated bioinformatics analysis