Cong Shen
Cong Luo2,3†
Zhijie Xu
Qiuju Liang
Yuan Cai
Yuanliang Yan
Fada Xia- 1Department of Thyroid Surgery, Xiangya Hospital, Central South University, Changsha, China
- 2National Clinical Research Center for Geriatric Disorders, Xiangya Hospital, Central South University, Changsha, China
- 3Department of Urology, Xiangya Hospital, Central South University, Changsha, China
- 4Department of Pathology, Xiangya Hospital, Central South University, Changsha, China
- 5Department of Pharmacy, Xiangya Hospital, Central South University, Changsha, China
Background: Colon cancer is an aggressive and heterogeneous disease associated with high morbidity and mortality. The immune system is intimately involved in tumorigenesis and can influence malignant properties at the protein, epigenetic, and even genomic levels by shaping the tumor immune microenvironment (TIM). However, immune-related molecules that can effectively predict the prognosis of colon cancer remain under exploration.
Methods: A total of 606 patients from TCGA and GEO databases were employed in our study, in which 429 cases were set as the training cohort and 177 were defined as the validation cohort. The immune infiltration was evaluated by ESTIMATE, TIMER, and CIBERSORT algorithms. The risk signature was constructed by LASSO Cox regression analysis. A nomogram model was generated subsequent to the multivariate Cox proportional hazards analysis to predict 1-, 3-, and 5-year survival of patients with colon cancer.
Results: Infiltrating immune cell profiling identified two colon cancer clusters (Immunity_L group and Immunity_H group). The abundances of immune cells were higher in the Immunity_H group, which indicated a better prognosis. Through further statistical analysis, we identified four genes which were highly correlated with prognosis and representative of this gene set, namely ARL4C, SERPINE1, BST2, and AXIN2. When the patients were divided into low- and high-risk groups based on their risk scores, we found that patients in the high-risk group had shorter overall survival time. Moreover, a nomogram including clinicopathologic features and the established risk signature could robustly predict 1-, 3-, and 5-year survival in patients with colon cancer.
Conclusion: We identified two distinct immune patterns by analyzing clinical and transcriptomic information from colon cancer patients. A subsequently constructed immune-related gene-based prognostic model as well as a nomogram model can be used to predict the prognosis of colon cancer, thereby guiding risk stratification and treatment regimen development for colon patients.
Introduction
Recent cancer statistics reveals a high incidence of 10.2% for human colon cancer, while the mortality rate is up to 9.2%, rising from the fourth to the second place in the oncological field, seriously threatening human life health (Bray et al., 2018; Siegel et al., 2020a). Surgical removal of cancerous tissue, combined with radiotherapy and chemotherapy (if necessary), has been the mainstay of combating colon malignancies (Nie et al., 2020; Liao et al., 2021). Due to advances in systemic drug targeting and surgical techniques, the prognosis of colon cancer patients will be significantly improved if they are diagnosed at an early stage (Labianca et al., 2010). Therefore, accurate grading/staging of colon cancer is helpful for the development of treatment options as well as the prognosis of patients.
Current prognostic prediction mainly relies on the Tumor, Nodes, Metastases (TNM) classification system (clinical level) and histopathological criteria (histological level) (Pagès et al., 2018). Unfortunately, patients still demonstrated an unpleasing survival outcome due to the recurrence, metastasis, and resistance to the agents. Clinicians and researchers have been searching for novel treatment strategies in the hope for achieving better results, and they focused on the cellular and molecular levels to identify valuable markers and tumor-cell differentiation events. A growing number of studies have demonstrated the role of gene mutation status, gene expression levels, and signaling pathway alterations in tumor initiation and progression, but accurately identifying prognostic factors that can provide targets for therapy remains difficult at the genomic level (Beane et al., 2009; Nagy et al., 2018). The immune system is intimately involved in tumorigenesis and can influence malignant properties at the protein, epigenetic, and even genomic levels by shaping the tumor immune microenvironment (TIM) (Deng et al., 2021). Recently, a large body of evidence has shown that immune-related molecules are of great value in predicting prognosis and assessing therapeutic efficacy (Binnewies et al., 2018).
The TIM is a complex system composed of multiple immune cells infiltrating into tumor tissue and various cytokines and chemokines secreted by them (Mallmann-Gottschalk et al., 2019). In there, natural killer (NK) cells can secrete cytokines, such as interferon (IFN)-gamma and tumor necrosis factor (TNF)-α, to exert an immunosuppressive phenotype via inhibiting tumor cell proliferation and tumor angiogenesis (Liu et al., 2018). Moreover, tumor associated macrophages (TAMs) and regulatory T cells (Tregs) mediate a suppressed tumor microenvironment which helps tumor cells achieve immune escape, and promotes the development of malignancy (Pan et al., 2020).
To address these suppressive immune phenotypes, targeting immune-tumor cell interactions has become more intensively studied, and immunotherapy has emerged as a promising area of cancer treatment and has demonstrated its impressive clinical value in colon cancer. This is mostly attributed to the face of immune checkpoint inhibitors as antitumor agents, such as programmed death 1 receptor (PD1) and cytotoxic T lymphocyte antigen 4 (CTLA-4) inhibitors (Yaghoubi et al., 2019). However, immunotherapy for colon cancer is still imperfect, for example, the effect of immunotherapy cannot be evaluated in advance. In conclusion, immunogenomic classification will help to guide the identification and effective treatment of early colon cancer and improve the accuracy of prognosis evaluation.
In this study, we performed an immunogenomic profiling of patients with colon cancer and divided them into two distinct subtypes: high immunity (Immunity_H) and low immunity (Immunity_L). We focused on analyzing two independent cohorts of colon cancer patients to identify genes highly associated with prognosis; based on their expression levels, the patients were allocated risk scores. We then combined clinicopathological characteristics and the risk scores to establish a model to accurately predict the survival rate of patients with colon cancer. This analysis is of great significance for the survival prediction of patients with colon cancer and provides a potential target for its treatment.
Materials and methods
Data Source and Extraction
The data we used for analysis were obtained from The Cancer Genome Atlas (TCGA, https://portal.gdc.cancer.gov/) (The Cancer Genome Atlas Network, 2012). In the dataset, 429 colon adenocarcinoma (COAD) patients had complete clinicopathological and transcriptomic expression data. Thus, they were enrolled as the training set. In addition, we also downloaded three datasets, including GSE17536 (n = 177), GSE17537 (n = 55), and GSE103479 (n = 156) in the Gene Expression Omnibus (GEO, https://www.ncbi.nlm.nih.gov/geo/) to validate the results (Smith et al., 2010; Dienstmann et al., 2019). The crude RNA expression data Fragments Per Kilobase of exon model per Million mapped fragments (FPKM) were transformed into Transcripts Per Kilobase of exon model per Million mapped reads (TPM) for a better statistical evaluation.
Clustering for Distinct Immune Patterns in Colon Cancer
We screened and quantified 28 immune cell types in the TCGA-COAD dataset using the single sample gene set enrichment analysis (ssGSEA) algorithm (Barbie et al., 2009). For each independent dataset, an enrichment score was calculated to represent the enrichment level of 28 immune cell types for each tumor sample. Based on these ssGSEA scores, we performed consensus clustering on TCGA-COAD. In brief, cluster analysis was performed using “ConsensusClusterPlus” (Wilkerson and Hayes, 2010), using agglomerative k-means clustering with a 1-Pearson correlation distance and resampling 80% of the samples for 10 repetitions. The optimal number of clusters was determined using the empirical cumulative distribution function plot. A principal component analysis (PCA) was conducted to analyze the distinguishing ability of the clustering.
Quantification of Immunotherapy Response, Major Histocompatibility Complex, and Immune Microenvironment
We used the “ESTIMATE” package to calculate the Immune, Stromal, and ESTIMATE scores for each sample to show the component fractions and tumor purity (Yoshihara et al., 2013). ImmuCellAI (Immune Cell Abundance Identifier) is a tool to estimate the abundance of 24 immune cells from gene expression dataset including RNA-Seq and microarray data (Miao et al., 2020). We applied it to predict patients’ response to immune checkpoint blockade therapy. The gene expression of the major histocompatibility complex (MHC), human leukocyte antigen (HLA), has also been explored in different clusters. Moreover, the abundances of diverse immune and stromal components were calculated using the “IOBR” R package (https://github.com/IOBR/IOBR), which is designed for multi-omics immuno-oncology biological research to decode tumor microenvironment and signatures (Zeng et al., 2021). Specifically, five algorithms built into the tool, including CIBERSORT (Newman et al., 2015), immunophenoscore (IPS) (Charoentong et al., 2017), MCP-counter (Becht et al., 2016), xCell (Aran et al., 2017), and EPIC (Racle et al., 2017), were used to calculate the scores of 51 infiltrating (immune and stromal) cells in each sample.
Profiling of the Differentially Expressed Prognostic Genes Related to Colon Cancer Subtype-specific Immunity
We performed difference analysis between the Immunity_Low (Immunity_L) group and Immunity_High (Immunity_H) group using the “limma” package (Ritchie et al., 2015). The absolute values of differential expression multiples >1.5 and p < 0.05 were used as the criteria for screening differentially expressed genes. The up- and down-regulated immune related genes (IRGs) in colon cancer were displayed in volcano plot. For functional analysis, we used the “org.Hs.eg.db” package (version 3.1.0) to perform the gene ontology (GO; including biological process, molecular function, and cellular component) and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway annotation. A Benjamini–Hochberg false discovery rate (BH-FDR) <0.05 was considered statistically significant. The enrichment results were displayed as histogram and lollipop charts.
Identification of Prognostic Genes and Construction of Immune Signature
Univariate Cox regression analysis based on differentially expressed genes was used to screen immune related genes significantly related to the prognosis of colon cancer with p < 0.05 as the threshold. Subsequently, a Least Absolute Shrinkage and Selection Operator (LASSO) Cox regression analysis with “glmnet” package (Tibshirani, 1996) was used to further identify essential genes and allocate coefficients for them. The risk score of each sample was calculated using the following formula:
where Coefi is the risk coefficient of each factor and xi is the mRNA expression value (logarithmic transformed TPM) of each factor. After determining the optimal cut-off value of risk score through “survival” and “survminer” packages, patients were divided into low risk and high risk groups correspondingly. Survival curves were used to show the differences in survival time and survival probability between high risk and low risk patients based on the Kaplan–Meier method. The area under curve (AUC) of receiver operating characteristic (ROC) curves represented the predictive accuracy (Heagerty et al., 2000). Univariate and multivariate Cox regression models were used to analyze whether the risk score was able to independently predict survival in patients with colon cancer.
Establishment of Nomogram Prognosis Prediction Model
Nomograms are widely used to predict the prognosis of the disease, so we drew a nomogram based on all independent prognostic factors identified by multivariate Cox regression analysis to predict the survival probability of patients within 1, 3, and 5 years by using the “rms” package. For practical application, we created a dynamic nomogram through “DynNom” package and built an interactive web-based tool with Shiny (https://shiny.rstudio.com/) (Supplementary Material S1). A nomogram calibration curve was plotted to judge nomogram accuracy by observing the relationship between predicted probability and actual incidence. The practicability of 1-, 3-, and 5-years OS was evaluated by ROC curves. The prognostic ability of the nomogram and other predictors (risk score, N stage, and M stage) for survival were evaluated by decision curve analysis (DCA) curves using the “rmda” R package.
Cell Culture
Normal human colon epithelial cells (NCM-460) and two human colon cancer cell lines (HCT116, HCT8) were obtained from the cell bank at the Chinese Academy of Sciences (Shanghai, China). All cells were authenticated by short tandem repeat (STR) profiling upon receipt and were propagated for less than 6 months after resuscitation. All cell lines were cultured in Dulbecco’s modified eagle medium (DMEM) medium with 10% fetal bovine serum (FBS; Thermo Fisher Scientific, Waltham, MA, USA). These cell lines were maintained in a humidified chamber containing 5% CO2 at 37°C.
RNA Extraction and qPCR
Total RNA from cultured cells and fresh tissues was extracted with Trizol regent (Thermo Fisher Scientific). We used NanoDrop and an Agilent 2,100 bioanalyzer (Thermo Fisher Scientific) to determine the concentration of extracted total RNA. cDNA was obtained by reverse transcription using a reverse transcription kit (Hiscrip II Q RT SuperMix for qPCR; Vazyme, Nanjing, China) according to the manufacturer’s protocol. Quantitative real-time polymerase chain reaction amplification was performed with SYBR Green PCR master mix (Takara, Japan) according to the manufacturer’s protocol. Primers were designed as follows: SERPINE1, forward, 5′-AGTGGACTTTTCAGAGGTGGA-3′, reverse, 5′-GCCGTTGAAGTAGAGGGCATT-3′; ARL4C, forward, 5′-CCAGTCCCTGCATATCGTCAT-3′, reverse, 5′-TTCACGAACTCGTTGAACTTGA-3′; BST2, forward, 5′-CACACTGTGATGGCCCTAATG-3′, reverse, 5′-GTCCGCGATTCTCACGCTT-3′; AXIN2, forward, 5′-TACACTCCTTATTGGGCGATCA-3′, reverse, 5′-TTGGCTACTCGTAAAGTTTTGGT-3′. GAPDH was used as an endogenous control, and relative gene expression was determined by the comparative 2−ΔΔCT method.
Immunohistochemistry
The Human Protein Atlas (HPA) (https://www.proteinatlas.org/) is a program for mapping human proteins in cells, tissues, and organs using integration of various omics technologies (Uhlen et al., 2017; Sun et al., 2018). We obtained representative immunohistochemistry results of the four target proteins in colon cancer and normal colon tissues from the tissue atlas and pathology atlas in HPA database, respectively.
Statistical Analysis
In this study, Kaplan–Meier method was used to estimate the overall survival rate of different groups, and log-rank was used to test the significance. The inter-group comparisons were achieved by using Wilcoxon rank sum test. The chi-square test was used to compare the clinicopathologic features (age, gender, TNM stage, and AJCC stage) between the low risk group and high risk group. Univariate and multivariate Cox regression analyses were utilized to evaluate the independent prognostic value of the risk signature regarding OS. Statistical analyses were made using R software (version 4.0.3). Most visualizations were achieved by “ggplot2” package. In most situations, p < 0.05 was used as a significant threshold if not otherwise specified Figure 1.
FIGURE 1. Flow chart of this study. The square element represents the research process, and the oval element represents the specific research method and result.
Results
Infiltrating Immune Cell Profiling Identified Two Colon Cancer Clusters
We screened and analyzed 28 immune cell types by ssGSEA for every tumor sample (Supplementary Table S2). Then, we performed consensus clustering on TCGA-COAD dataset based on the ssGSEA scores, which represented the activity or infiltration levels of immune cells in the tumor sample. Finally, the patients were divided into two unique immune clusters (C1 and C2; Figures 2A–C). We found that the ssGSEA scores of all the immune cell types were higher in C2 than those in C1 (Figure 2D). The C1 and C2 were thus defined as the Immunity_Low (Immunity_L) group and Immunity_High (Immunity_H) group, respectively (Supplementary Table S3). Furthermore, the heatmap clearly demonstrated that the Immunity_H group possessed higher stromal scores, immune scores, and ESTIMATE scores when compared with the Immunity_L group (Figure 2D). Correspondingly, the Immunity_H group exhibited lower tumor purity.
FIGURE 2. Two patterns of colon cancer based on immune cell clustering. (A) Cumulative distribution function (CDF) under different clustering numbers. (B) Relative change in area under CDF curve with different clustering numbers. (C) Clustering heatmap when k = 2. (D) Heatmap showing the differences in immune cell ssGSEA scores between two clusters. (E) PCA regarding two clusters. (F) The differences in response rates to immune checkpoint blockades between two clusters. (G) The expressions of HLA genes between two clusters.
PCA result further confirmed the reliability of this clustering, as the subgroups could be significantly distinguished (Figure 2E). More importantly, the Immunity_H group showed a higher response rate to immune checkpoint blockers than the Immunity_L group (51 vs. 15%; p < 0.001; Figure 2F). At the same time, we found that most HLA genes were expressed at higher levels in the Immunity_H group than in the Immunity_L group (Figure 2G). These results collaborated that our clustering significantly distinguished colon cancers into two groups with distinct immune landscapes.
Next, we examined the relationship between individual immune cell ssGSEA score and COAD patient’s overall survival and found that 10/28 cell types were significantly correlated with patient prognosis. Specifically, activated B cell (p = 0.03; Figure 3A), effector memory CD4+ T cell (p = 0.02; Figure 3B), eosinophil (p = 3.4e-3; Figure 3C), immature B cell (p = 0.04; Figure 3D), neutrophil (p = 0.04; Figure 3E), and type 17 T helper cell (Th17; p = 7.7e-3; Figure 3F) were beneficial to patient survival, as a patient with higher scores of these cell types would have a higher survival probability, whereas CD56dim NK cell (p = 0.02; Figure 3G), myeloid-derived suppressor cell (MDSC; p = 0.02; Figure 3H), natural killer T cell (p = 0.05; Figure 3I), and T follicular helper cell (Tfh; p = 8.2e-3; Figure 3J) were unfavorable to patient survival, as a patient with higher scores of these cell types would have a worse survival. We then compared the abundances of these immune cells between normal and tumor tissues. Except for the natural killer T cell and type 17 T helper cell, the remaining eight immune cells had generally lower abundance in tumor tissues than in normal tissues (Figure 3K). We used five algorithms to demonstrate differences in the infiltrating cells between the two groups. Overall, the abundances of immune cells were higher in the Immunity_H group, especially for those calculated by CIBERSOFT algorithm (Figure 3L).
FIGURE 3. (A–J) Kaplan–Meier survival curves of individual cell types with significant prognostic values. (K) Differences in abundance of 10 immune cells between normal and tumor tissues. (L) Immune cell abundances estimated by five algorithms (CIBERSORT, EPIC, IPS, MCPcounter, and TIMER) between two clusters.
Differentially Expressed Genes Related to Colon Cancer Cluster-Specific Immunity Were Identified and Verified
We analyzed the differentially expressed genes between the two colon cancer immune subtypes in TCGA-COAD cohort. A total of 18 genes were up-regulated in the immunty_H group and down-regulated in the Immunity_L group. Conversely, 259 genes were down-regulated in the immunty_H group and up-regulated in the Immunity_L group (Supplementary Table S4; Figure 4A). Through GO enrichment analysis, we found these differentially expressed genes were mostly located on collagen-containing extracellular matrix, involved in biological processes including response to interferon-gamma and extracellular matrix organization. In terms of molecular functions, these genes were mainly involved in extracellular matrix structural constituent and chemokine activity (Figure 4B). The results were consistent to the well-established immune processes. Furthermore, KEGG enrichment analysis concluded that these genes mainly participated in the cellular processes of phagosome, environmental information processing such as cytokine-cytokine receptor interaction. Besides, they were enriched in the organismal systems including hematopoietic cell lineage, Th17 cell differentiation, and chemokine signaling pathway (Figure 4C). These results largely indicated that these differentially expressed genes derived from immunogenomic clusters were closely linked to immune-related pathways.
FIGURE 4. The biological functions of differentially expressed genes between two clusters. (A) The differentially expressed genes between the Immunity_L and Immunity_H groups. Red triangles represent up-regulated genes and green triangles represent down-regulated genes. (B) Enrichment results of GO terms. Orange: biological process, green: cellular component, purple: molecular function. (C) Enrichment results of KEGG pathways. Blue: cellular process, green: environmental information processing, red: organismal systems.
A Four-Gene Risk Signature was Constructed in the Training Set
These differentially expressed genes showed a completely distinct expression patterns between the immunty_H group and the Immunity_L group (Figure 5). By univariate Cox regression analysis, we screened 11 differentially expressed genes that were significantly associated with prognosis (Figure 6A), of which AXIN2 was a protective factor because it harbored a hazard ratio (HR) of less than 1 (HR = 0.861, p = 0.0481), while the remaining 10 genes were risk factors: SERPINE1 (HR = 1.24, p = 0.0076), SFRP2 (HR = 1.12, p = 0.0203), APOE (HR = 1.15, p = 0.0230), ARL4C (HR = 1.27, p = 0.0236), BST2 (HR = 1.15, p = 0.0283), TGFB1 (HR = 1.25, p = 0.0302), SLC2A3 (HR = 1.23, p = 0.0369), VSIG4 (HR = 1.2, p = 0.0389), C1QA (HR = 1.18, p = 0.0455), and BGN (HR = 1.18, p = 0.0466). We performed further screening of these genes by LASSO regression analysis and identified four genes which were highly correlated with prognosis and representative of this gene set, namely ARL4C, SERPINE1, BST2, and AXIN2 (Figure 6B).
FIGURE 5. The heatmap of differentially expressed genes.
FIGURE 6. Construction of risk signature based on differentially expressed genes. (A) The forest plot of differentially expressed genes with prognostic significance. (B) The screening of coefficients and likelihood deviance under LASSO analysis. (C) The coefficients of model genes. (D) Best cutoff value selected by the log-rank test. (E) The relationship between risk score, survival status, and model gene expression. (F) The Kaplan–Meier survival curve showing the difference between high risk and low risk groups in TCGA-COAD. (G) The ROC curves regarding 1-, 3-, and 5-year survival outcomes in TCGA-COAD. (H) The differences in response rates to immune checkpoint blockades between two groups.
The LASSO analysis also allocated coefficients for these four genes (Figure 6C), thus facilitating the assignment of risk score to each patient. Under the best cut-off value of 0.56 detected by log-rank test (Figure 6D), we displayed risk scores and survival status distributions as Figure 6E. The high risk group had more mortality events. And it clearly showed that ARL4C, SERPINE1, and BST2 expression increased with increasing risk score, whereas the expression of the AXIN2 decreased with elevating risk score. The clinicopathological characteristics between the high and low risk groups are summarized in Table 1. It shows that the risk signature was an independent indicator of prognosis as the characteristics were not significantly different between the two groups, including age (p = 0.097), gender (p = 0.36), AJCC pathologic stage (p = 0.22), T stage (p = 0.052), N stage (p = 0.46), M stage (p = 0.48), and disease type (p = 0.40). Kaplan-Meier survival curves depicted that colon cancer patients with higher risk scores significantly had worse clinical outcomes (HR = 2.17, 95% CI 1.41–3.35, p < 0.001; Figure 6F). The ROC curves demonstrated that the risk signature harbored a promising ability to predict OS in the TCGA-COAD cohort (AUC: 1 year = 0.60, 3 years = 0.61, 5 years = 0.60; Figure 6G). Besides, we also analyzed the response to immune checkpoint blockers in the high risk and low risk groups. The result showed that the low risk group showed higher response rate to immune checkpoint blockers than the high risk group (35 vs. 22%; p = 0.017; Figure 6H).
TABLE 1. Clinicopathological characteristics between the low risk and high risk groups
Validation Cohort Demonstrated Stability of the Risk Signature
To externally validate the prognostic ability of the established risk signature, we calculated risk scores for patients in another three independent cohorts (GSE17536, GSE17537, and GSE103479) using the same formula (Figures 7A–C). Consistently, colon cancer patients with higher risk scores had lower OS rate and shorter OS time in the validation cohorts. The ROC analysis also indicated that the risk signature had a promising prognostic value for patients with colon cancer in the validation cohort (AUC for GSE17536: 1 year = 0.65, 3 years = 0.61, 5 years = 0.58; AUC for GSE17537: 1 year = 0.68, 3 years = 0.73, 5 years = 0.58; AUC for GSE103479: 1 year = 0.56, 3 years = 0.58, 5 years = 0.61). These results showed that the risk signature had an effective and stable OS-predictive ability for colon cancer patients.
FIGURE 7. External validation. (A) The risk distribution, the Kaplan–Meier survival curve, and the ROC curves in GSE17536. (B) The risk distribution, the Kaplan–Meier survival curve, and the ROC curves in GSE17537. (C) The risk distribution, the Kaplan–Meier survival curve, and the ROC curves in GSE103479.
In Vitro Validation Demonstrated High Expression of Model Genes in Colon Cancer at Transcriptional and Protein Levels
To confirm that these four genes are indeed highly expressed in colon cancer, the expressions of these four genes were detected by quantitative PCR (qPCR) in normal colon epithelial cells (NCM-460) and colon cancer cells (HCT116 and HCT8). The results showed that the expression levels of these four genes in colon cancer cells were obviously higher than those in normal colon epithelial cells (Figure 8A). Furthermore, we confirmed the protein expression profiles of these four genes in human tissues. As showed in immunohistochemistry results (Figure 8B), these four proteins were mainly distributed in the cytoplasm or membrane and were upregulated in colon cancer tissues compared with corresponding normal tissues. These results evidently demonstrated that ARL4C, SERPINE1, BST2, and AXIN2 were upregulated in colon cancer cells and tissues at the transcriptional and protein levels, implying the importance of these four genes in colon cancer pathogens.
FIGURE 8. The expression of these four genes in colon cancer cells and tissues. (A) The relative mRNA levels of the four genes in normal colon epithelial cells (NCM-460) and colon cancer cells (HCT116, HCT8) by qPCR. (B) Representative IHC images of the four genes in colon cancer and normal colon tissues.
The Risk Signature-based Nomogram Had Better Prediction Ability and Practical Value
Firstly, we used univariate and multivariate Cox analyses to assess whether the established risk signature was an independent prognostic factor for patients with colon cancer (Table 2). Based on the data of colon cancer samples in the TCGA data set, univariate Cox analysis indicated that N stage, M stage, AJCC stage, and risk score were remarkably associated with OS (p < 0.05). Subsequent multivariate Cox analysis further showed that N, M stages, and risk score were independent predictors of OS (p < 0.05). These results indicated that our risk signature, as an independent prognostic indicator, might be useful for clinical prognosis evaluation.
TABLE 2. Univariate and multivariate analyses in TCGA-COAD cohort.
To create a clinically applicable quantitative tool to predict the OS of colon cancer patients, we constructed a nomogram model including the risk score, N stage, and M stage in the TCGA data set (Figure 9A), which was available online (https://scxiangya.shinyapps.io/DynNom/) as screenshot in Figure 9B. The Sankey diagram exhibited the distribution of the clinicopathological features of the patients in different groups (Figure 9C). Calibration plots using 1,000 booted resampling revealed perfect concordance regarding the observed vs. predicted rates of 1-, 3- and 5-year OS in the TCGA-COAD cohort (Figure 9D). The ROC analysis also indicated that the nomogram had a stable and robust power in predicting the OS for colon cancer patients (AUC: 1 year = 0.775, 3 years = 0.766, 5 years = 0.717; Figure 9E). The DCA result indicated that the model combining prognosis-related clinicopathologic characteristics and risk signature conferred a better predictive potency than the three-factor model alone (Figure 9F).
FIGURE 9. Construction and validation of the nomogram model. (A) Construction of the nomogram model combining risk score and prognostic clinicopathological indicators in TCGA-COAD. (B) Online dynamic nomogram accessible at https://scxiangya.shinyapps.io/DynNom/, depicting an example for predicting the survival probability of a patient with TxN0M0 colon cancer and a risk score of 0.448. (C) Sankey diagram showing the distribution of patients in the high risk and low risk groups. (D) The calibration curve judging nomogram accuracy by comparing the relationship between predicted and observed overall survival. (E) Time dependent ROC curves showing the predictive ability of the nomogram. (F) Decision curve analysis curves and clinical impact curves evaluating the benefit when using the nomogram model.
Discussion
Colon cancer is one of the major malignant tumors of the gastrointestinal tract, and approximately 600,000 people die from it every year (Bray et al., 2018; Siegel et al., 2020b), although 5-year survival for colon cancer has approached 65% with improved surgical methods and subsequent treatments in developed countries (Miller et al., 20192019). For colon cancer patients who present with local invasion or distant metastasis, the mortality rate is very high (Misale et al., 2012; Edwards et al., 2014; Fang et al., 2017). Therefore, there is an urgent need to find some new predictive parameters or therapeutic targets that are highly correlated with prognosis. This helps us to establish an early warning system in advance to rapidly identify patients with more critical conditions in clinical work to guide the development of subsequent treatment regimens and the prediction of survival outcomes.
Although it has long been recognized that the immune cells play an important role in tumor initiation and development (Fridman et al., 2012), these insights have not made a major influence on routine clinical practice. Moreover, the transcriptomic correlation of immune infiltration in cancer tissues on diagnosis and prognosis has attracted substantial interest. However, very few of these studies focused on the association between the difference of the immune cell composition and prognosis in colon cancer.
In this study, we first analyzed 28 immune cell types to divide TCGA-COAD into two unique immune patterns: the Immunity_H group and Immunity_L group. The two groups showed significant differences in anti-tumor immune activity, immune cell infiltration, and response to immune checkpoint blockades. Regarding the specific cell types, the complex and diverse immune cells in the TIM include T lymphocytes (70–80%), B lymphocytes (10–20%), macrophages (5–10%), NK cells (<5%), and dendritic cells (1–2%) (Frankel et al., 2017). Additionally, Tregs and TAMs contribute to tumor escape with immune suppressive activity and inhibit anti-tumor responses. Immune cells infiltrating tumors mediate the TIM and thereby influence tumor prognosis (Wu et al., 2013). In this study, we found that effector memory CD4 T cells, activated B cells, eosinophils, and Th17 cells were positively correlated with patient survival prognosis, while the immunosuppressive cells MDSCs were negatively correlated with survival prognosis. Tumor immunotherapy is known to act as a tumor suppressor by acting through these immune cells, so we could screen out colon cancer patients who could benefit from immunotherapy based on different expression levels of immune cells.
It is of great significance to find which immune-related genes that play important roles in the development of the colon cancer and the prognosis of the patient. By specific algorithms, we identified a group of immune-related genes that predict the prognosis of colon cancer patients. With further screening and model construction, four genes including ARL4C, SERPINE1, BST2, and AXIN2 were singled out to be highly associated with prognosis. It has been found that overexpression of ARL4C might contribute to the tumorigenesis and lead to worse prognosis in colorectal cancer (Chen et al., 2016), which supported the result of our study. A recent study has found that SERPINE1 participated in colon cancer microenvironment remodeling and immune cell infiltration (Wang et al., 2021). This explained why patients with overexpression of SERPINE1 had poor prognosis in our study. Moreover, the prognostic significance of BST2 in colon cancer has been put forward as early as 2015 (Chiang et al., 2015). And finally, AXIN2 has been consistently classified as a tumor suppressor gene in colorectal cancers both in vivo and in vitro (Church and Fazio, 2005; Waaler et al., 2012). However, it should be noted that AXIN2 was previously identified as a potent tumor promoter instead of a suppressor, as it was found to exert global control over gene expression networks which were critical for tumor-invasive and metastatic behavior (Wu et al., 2012). Its precise function in the carcinomatous state may require further studies. In general, these four genes are promising prognostic molecules in colon cancer. The constructed model can well distinguish colon cancer patients and predict prognosis, thereby helping to develop individualized treatment options based on survival risk.
The aim of this study is to construct a model composed of prognostic immune related genes, which can robustly predict prognosis. The multivariate Cox regression analysis result showed that the survival time of the high risk group was significantly lower than that of the low risk group. This shows that our model can be used as an independent prognostic factor for colon cancer patients. According to the nomogram model, the survival rate of colon cancer patients is consistent with the actual situation. This indicates that the model can well distinguish colon cancer patients and outperform clinical parameters alone. Combining the fact that there was a significant difference in the immune cell constitutions between the Immunity_H and Immunity_L groups, we hypothesized that the immune-related genes may affect the tumor prognosis by affecting the immune infiltration. Consequently, the responses rates to immune checkpoint blockade were significantly different between the groups with distinct immune landscapes. It is thus suggested that the poor prognosis of patients in the Immunity_L group may be due to the immunosuppressive microenvironment.
However, there are still some limitations in our study. Due to insufficient clinical information in the three GEO cohorts, the nomogram model failed to be validated, and the AUC values of ROC curves for the risk signature were not high due to the limited sample size. Therefore, we will further validate this prognostic model in other independent large cohorts to ensure the reliability of our model. Moreover, functional experiments are also needed to further reveal the interplay between immune related genes and tumors.
Conclusion
We identified two distinct immune patterns by analyzing clinical and transcriptomic information from colon cancer patients, which exhibited distinct tumor purity and immune composition. A subsequently constructed immune-related gene-based prognostic model as well as a nomogram model was closely related to the prognosis of colon cancer patients to predict prognosis more precisely, thereby guiding risk stratification and treatment regimen development for colon patients.
Data Availability Statement
The original contributions presented in the study are included in the article/Supplementary Material, further inquiries can be directed to the corresponding authors.
Author Contributions
Acquisition of Data: CS and CL. Analysis and Interpretation of Data: ZX and QL. Conception and Design: YY and FX. Data Curation: YY and FX. Development of Methodology: YC and BP. Writing the Manuscript: CS and CL.
Funding
This work was supported by grants from the Science and Technology Innovation Program of Hunan Province (No. 2021RC3029), the China Postdoctoral Science Foundation (Nos. 2021T140754 and 2020M672521), and the Natural Science Foundation of Hunan Province (No. 2020JJ5934).
Conflict of Interest
The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.
Publisher’s Note
All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.
Acknowledgments
All authors thank TCGA and GEO databases for providing the worthy data and the Sangerbox tools which is a free online platform for data analysis.
Supplementary Material
The Supplementary Material for this article can be found online at: https://www.frontiersin.org/articles/10.3389/fbioe.2022.820092/full#supplementary-material
References
Aran, D., Hu, Z., and Butte, A. J. (2017). xCell: Digitally Portraying the Tissue Cellular Heterogeneity landscapePubMed Central PMCID: PMCPMC5688663 Authors Declare that They Have No Competing Interests. PUBLISHER'S NOTE: Springer Nature Remains Neutral with Regard to Jurisdictional Claims in Published Maps and Institutional Affiliations. Genome Biol. 18 (1), 220. Epub 2017/11/17PubMed PMID: 29141660;. doi:10.1186/s13059-017-1349-1
Barbie, D. A., Tamayo, P., Boehm, J. S., Kim, S. Y., Moody, S. E., Dunn, I. F., et al. (2009). Systematic RNA Interference Reveals that Oncogenic KRAS-Driven Cancers Require TBK1. Nature 462 (7269), 108–112. Epub 2009/10/23PubMed PMID: 19847166; PubMed Central PMCID: PMCPMC2783335. doi:10.1038/nature08460
Beane, J., Spira, A., and Lenburg, M. E. (2009). Clinical Impact of High-Throughput Gene Expression Studies in Lung Cancer. J. Thorac. Oncol. 4 (1), 109–118. Epub 2008/12/20PubMed PMID: 19096318; PubMed Central PMCID: PMCPMC2731413. doi:10.1097/JTO.0b013e31819151f8
Becht, E., Giraldo, N. A., Lacroix, L., Buttard, B., Elarouci, N., Petitprez, F., et al. (2016). Estimating the Population Abundance of Tissue-Infiltrating Immune and Stromal Cell Populations Using Gene Expression. Genome Biol. 17 (1), 218. Epub 2016/10/22PubMed PMID: 27765066; PubMed Central PMCID: PMCPMC5073889. doi:10.1186/s13059-016-1070-5
Binnewies, M., Roberts, E. W., Kersten, K., Chan, V., Fearon, D. F., Merad, M., et al. (2018). Understanding the Tumor Immune Microenvironment (TIME) for Effective Therapy. Nat. Med. 24 (5), 541–550. Epub 2018/04/25PubMed PMID: 29686425; PubMed Central PMCID: PMCPMC5998822. doi:10.1038/s41591-018-0014-x
Bray, F., Ferlay, J., Soerjomataram, I., Siegel, R. L., Torre, L. A., and Jemal, A. (2018). Global Cancer Statistics 2018: GLOBOCAN Estimates of Incidence and Mortality Worldwide for 36 Cancers in 185 Countries. CA: A Cancer J. Clinicians 68 (6), 394–424. Epub 2018/09/13PubMed PMID: 30207593. doi:10.3322/caac.21492
Charoentong, P., Finotello, F., Angelova, M., Mayer, C., Efremova, M., Rieder, D., et al. (2017). Pan-cancer Immunogenomic Analyses Reveal Genotype-Immunophenotype Relationships and Predictors of Response to Checkpoint Blockade. Cel Rep. 18 (1), 248–262. Epub 2017/01/05PubMed PMID: 28052254. doi:10.1016/j.celrep.2016.12.019
Chen, X., Su, Z., Wang, S., and Xu, H. (2016). Clinical and Prognostic Significance of Arl4c Expression in Colorectal Cancer. Cbm 16 (2), 253–257. Epub 2016/01/13PubMed PMID: 26756615. doi:10.3233/cbm-150562
Chiang, S.-F., Kan, C.-Y., Hsiao, Y.-C., Tang, R., Hsieh, L.-L., Chiang, J.-M., et al. (2015). Bone Marrow Stromal Antigen 2 Is a Novel Plasma Biomarker and Prognosticator for Colorectal Carcinoma: A Secretome-Based Verification Study. Dis. Markers 2015, 1–10. Epub 2015/10/27PubMed PMID: 26494939; PubMed Central PMCID: PMCPMC4606116. doi:10.1155/2015/874054
Church, J. M. (2005). “Molecular Genetics of Colorectal Cancer,” in Current Therapy in Colon and Rectal Surgery. Editor V. W. Fazio. Second Edition (Hamilton, Canada: Decker), 343–347. doi:10.1016/b978-1-55664-480-1.50064-2
Deng, D., Luo, X., Zhang, S., and Xu, Z. (2021). Immune Cell Infiltration-Associated Signature in colon Cancer and its Prognostic Implications. Aging 13 (15), 19696–19709. Epub 2021/08/06PubMed PMID: 34349038; PubMed Central PMCID: PMCPMC8386549. doi:10.18632/aging.203380
Dienstmann, R., Villacampa, G., Sveen, A., Mason, M. J., Niedzwiecki, D., Nesbakken, A., et al. (2019). Relative Contribution of Clinicopathological Variables, Genomic Markers, Transcriptomic Subtyping and Microenvironment Features for Outcome Prediction in Stage II/III Colorectal Cancer. Ann. Oncol. 30 (10), 1622–1629. Epub 2019/09/11PubMed PMID: 31504112; PubMed Central PMCID: PMCPMC6857614. doi:10.1093/annonc/mdz287
Edwards, B. K., Noone, A. M., Mariotto, A. B., Simard, E. P., Boscoe, F. P., Henley, S. J., et al. (2014). Annual Report to the Nation on the Status of Cancer, 1975‐2010, Featuring Prevalence of Comorbidity and Impact on Survival Among Persons with Lung, Colorectal, Breast, or Prostate Cancer. Cancer 120 (9), 1290–1314. Epub 2013/12/18PubMed PMID: 24343171; PubMed Central PMCID: PMCPMC3999205. doi:10.1002/cncr.28509
Fang, C., Fan, C., Wang, C., Huang, Q., Meng, W., Yu, Y., et al. (2017). Prognostic Value of CD133+CD54+CD44+circulating Tumor Cells in Colorectal Cancer with Liver Metastasis. Cancer Med. 6 (12), 2850–2857. Epub 2017/11/07PubMed PMID: 29105339; PubMed Central PMCID: PMCPMC5727299. doi:10.1002/cam4.1241
Frankel, T., Lanfranca, M. P., and Zou, W. (2017). The Role of Tumor Microenvironment in Cancer Immunotherapy. Adv. Exp. Med. Biol. 1036, 51–64. Epub 2017/12/25PubMed PMID: 29275464. doi:10.1007/978-3-319-67577-0_4
Fridman, W. H., Pagès, F., Sautès-Fridman, C., and Galon, J. (2012). The Immune Contexture in Human Tumours: Impact on Clinical Outcome. Nat. Rev. Cancer 12 (4), 298–306. Epub 2012/03/16PubMed PMID: 22419253. doi:10.1038/nrc3245
Heagerty, P. J., Lumley, T., and Pepe, M. S. (2000). Time-dependent ROC Curves for Censored Survival Data and a Diagnostic Marker. Biometrics 56 (2), 337–344. Epub 2000/07/06PubMed PMID: 10877287. doi:10.1111/j.0006-341x.2000.00337.x
Labianca, R., Beretta, G. D., Kildani, B., Milesi, L., Merlin, F., Mosconi, S., et al. (2010). Colon Cancer. Crit. Rev. Oncology/Hematology 74 (2), 106–133. Epub 2010/02/09PubMed PMID: 20138539. doi:10.1016/j.critrevonc.2010.01.010
Liao, Z., Nie, H., Wang, Y., Luo, J., Zhou, J., and Ou, C. (2021). The Emerging Landscape of Long Non-coding RNAs in Colorectal Cancer Metastasis. Front. Oncol. 11, 641343. Epub 2021/03/16PubMed PMID: 33718238; PubMed Central PMCID: PMCPMC7947863. doi:10.3389/fonc.2021.641343
Liu, X., Hu, J., Li, Y., Cao, W., Wang, Y., Ma, Z., et al. (2018). Mesenchymal Stem Cells Expressing Interleukin-18 Inhibit Breast Cancer in a Mouse Model. Oncol. Lett. 15 (5), 6265–6274. Epub 2018/05/05PubMed PMID: 29725393; PubMed Central PMCID: PMCPMC5920279. doi:10.3892/ol.2018.8166
Mallmann-Gottschalk, N., Sax, Y., Kimmig, R., Lang, S., and Brandau, S. (2019). EGFR-specific Tyrosine Kinase Inhibitor Modifies NK Cell-Mediated Antitumoral Activity against Ovarian Cancer Cells. Ijms 20 (19), 4693. Epub 2019/09/25PubMed PMID: 31546690; PubMed Central PMCID: PMCPMC6801374. doi:10.3390/ijms20194693
Miao, Y. R., Zhang, Q., Lei, Q., Luo, M., Xie, G. Y., Wang, H., et al. (2020). ImmuCellAI: A Unique Method for Comprehensive T‐Cell Subsets Abundance Prediction and its Application in Cancer Immunotherapy. Adv. Sci. 7 (7), 1902880. Epub 2020/04/11PubMed PMID: 32274301; PubMed Central PMCID: PMCPMC7141005. doi:10.1002/advs.201902880
Miller, K. D., Nogueira, L., Mariotto, A. B., Rowland, J. H., Yabroff, K. R., Alfano, C. M., et al. (20192019). Cancer Treatment and Survivorship Statistics, 2019. CA A. Cancer J. Clin. 69 (5), 363–385. Epub 2019/06/12PubMed PMID: 31184787. doi:10.3322/caac.21565
Misale, S., Yaeger, R., Hobor, S., Scala, E., Janakiraman, M., Liska, D., et al. (2012). Emergence of KRAS Mutations and Acquired Resistance to Anti-EGFR Therapy in Colorectal Cancer. Nature 486 (7404), 532–536. Epub 2012/06/23PubMed PMID: 22722830; PubMed Central PMCID: PMCPMC3927413. doi:10.1038/nature11156
Nagy, Á., Lánczky, A., Menyhárt, O., and Győrffy, B. (2018). Validation of miRNA Prognostic Power in Hepatocellular Carcinoma Using Expression Data of Independent Datasets. Sci. Rep. 8 (1), 9227. Epub 2018/06/17PubMed PMID: 29907753; PubMed Central PMCID: PMCPMC6003936. doi:10.1038/s41598-018-27521-y
Newman, A. M., Liu, C. L., Green, M. R., Gentles, A. J., Feng, W., Xu, Y., et al. (2015). Robust Enumeration of Cell Subsets from Tissue Expression Profiles. Nat. Methods 12 (5), 453–457. Epub 2015/03/31PubMed PMID: 25822800; PubMed Central PMCID: PMCPMC4739640. doi:10.1038/nmeth.3337
Nie, H., Wang, Y., Liao, Z., Zhou, J., and Ou, C. (2020). The Function and Mechanism of Circular RNAs in Gastrointestinal Tumours. Cell Prolif 53 (7), e12815. Epub 2020/06/10PubMed PMID: 32515024; PubMed Central PMCID: PMCPMC7377939. doi:10.1111/cpr.12815
Pagès, F., Mlecnik, B., Marliot, F., Bindea, G., Ou, F. S., Bifulco, C., et al. (2018). International Validation of the Consensus Immunoscore for the Classification of colon Cancer: a Prognostic and Accuracy Study. Lancet 391 (10135), 2128–2139. Epub 2018/05/15PubMed PMID: 29754777. doi:10.1016/s0140-6736(18)30789-x
Pan, Y., Yu, Y., Wang, X., and Zhang, T. (2020). Tumor-Associated Macrophages in Tumor Immunity. Front. Immunol. 11, 583084. Epub 2020/12/29PubMed PMID: 33365025; PubMed Central PMCID: PMCPMC7751482. doi:10.3389/fimmu.2020.583084
Racle, J., de Jonge, K., Baumgaertner, P., Speiser, D. E., and Gfeller, D. (2017). Simultaneous Enumeration of Cancer and Immune Cell Types from Bulk Tumor Gene Expression Data. Elife 6, e26476. Epub 2017/11/14PubMed PMID: 29130882; PubMed Central PMCID: PMCPMC5718706. doi:10.7554/eLife.26476
Ritchie, M. E., Phipson, B., Wu, D., Hu, Y., Law, C. W., Shi, W., et al. (2015). Limma powers Differential Expression Analyses for RNA-Sequencing and Microarray Studies. Nucleic Acids Res. 43 (7), e47. Epub 2015/01/22PubMed PMID: 25605792; PubMed Central PMCID: PMCPMC4402510. doi:10.1093/nar/gkv007
Siegel, R. L., Miller, K. D., Goding Sauer, A., Fedewa, S. A., Butterly, L. F., Anderson, J. C., et al. (2020). Colorectal Cancer Statistics, 2020. CA A. Cancer J. Clin. 70 (3), 145–164. Epub 2020/03/07PubMed PMID: 32133645. doi:10.3322/caac.21601
Siegel, R. L., Miller, K. D., and Jemal, A. (2020). Cancer Statistics, 2020. CA A. Cancer J. Clin. 70 (1), 7–30. Epub 2020/01/09PubMed PMID: 31912902. doi:10.3322/caac.21590
Smith, J. J., Deane, N. G., Wu, F., Merchant, N. B., Zhang, B., Jiang, A., et al. (2010). Experimentally Derived Metastasis Gene Expression Profile Predicts Recurrence and Death in Patients with colon Cancer. Gastroenterology 138 (3), 958–968. Epub 2009/11/17PubMed PMID: 19914252; PubMed Central PMCID: PMCPMC3388775. doi:10.1053/j.gastro.2009.11.005
Sun, B. B., Maranville, J. C., Peters, J. E., Stacey, D., Staley, J. R., Blackshaw, J., et al. (2018). Genomic Atlas of the Human Plasma Proteome. Nature 558 (7708), 73–79. Epub 2018/06/08PubMed PMID: 29875488; PubMed Central PMCID: PMCPMC6697541. doi:10.1038/s41586-018-0175-2
The Cancer Genome Atlas Network (2012). Comprehensive Molecular Characterization of Human colon and Rectal Cancer. Nature 487 (7407), 330–337. Epub 2012/07/20PubMed PMID: 22810696; PubMed Central PMCID: PMCPMC3401966. doi:10.1038/nature11252
Tibshirani, R. J. (1996). Regression Shrinkage and Selection via the LASSO. J. R. Stat. Soc. Ser. B: Methodological 73 (1), 273–282. doi:10.1111/j.2517-6161.1996.tb02080.x
Uhlen, M., Zhang, C., Lee, S., Sjöstedt, E., Fagerberg, L., Bidkhori, G., et al. (2017). A Pathology Atlas of the Human Cancer Transcriptome. Science 357 (6352), eaan2507. Epub 2017/08/19PubMed PMID: 28818916. doi:10.1126/science.aan2507
Waaler, J., Machon, O., Tumova, L., Dinh, H., Korinek, V., Wilson, S. R., et al. (2012). A Novel Tankyrase Inhibitor Decreases Canonical Wnt Signaling in colon Carcinoma Cells and Reduces Tumor Growth in Conditional APC Mutant Mice. Cancer Res. 72 (11), 2822–2832. Epub 2012/03/24PubMed PMID: 22440753. doi:10.1158/0008-5472.Can-11-3336
Wang, S., Pang, L., Liu, Z., and Meng, X. (2021). SERPINE1 Associated with Remodeling of the Tumor Microenvironment in colon Cancer Progression: a Novel Therapeutic Target. BMC Cancer 21 (1), 767. Epub 2021/07/04PubMed PMID: 34215248; PubMed Central PMCID: PMCPMC8254339. doi:10.1186/s12885-021-08536-7
Wilkerson, M. D., and Hayes, D. N. (2010). ConsensusClusterPlus: a Class Discovery Tool with Confidence Assessments and Item Tracking. Bioinformatics 26 (12), 1572–1573. Epub 2010/04/30PubMed PMID: 20427518; PubMed Central PMCID: PMCPMC2881355. doi:10.1093/bioinformatics/btq170
Wu, F. Y., Fan, J., Tang, L., Zhao, Y. M., and Zhou, C. C. (2013). Atypical Chemokine Receptor D6 Inhibits Human Non-small Cell Lung Cancer Growth by Sequestration of Chemokines. Oncol. Lett. 6 (1), 91–95. Epub 2013/08/16PubMed PMID: 23946783; PubMed Central PMCID: PMCPMC3742824. doi:10.3892/ol.2013.1358
Wu, Z.-Q., Brabletz, T., Fearon, E., Willis, A. L., Hu, C. Y., Li, X.-Y., et al. (2012). Canonical Wnt Suppressor, Axin2, Promotes colon Carcinoma Oncogenic Activity. Proc. Natl. Acad. Sci. 109 (28), 11312–11317. Epub 2012/06/30PubMed PMID: 22745173; PubMed Central PMCID: PMCPMC3396472. doi:10.1073/pnas.1203015109
Yaghoubi, N., Soltani, A., Ghazvini, K., Hassanian, S. M., and Hashemy, S. I. (2019). PD-1/PD-L1 Blockade as a Novel Treatment for Colorectal Cancer. Biomed. Pharmacother. 110, 312–318. Epub 2018/12/07PubMed PMID: 30522017. doi:10.1016/j.biopha.2018.11.105
Yoshihara, K., Shahmoradgoli, M., Martínez, E., Vegesna, R., Kim, H., Torres-Garcia, W., et al. (2013). Inferring Tumour Purity and Stromal and Immune Cell Admixture from Expression Data. Nat. Commun. 4, 2612. Epub 2013/10/12PubMed PMID: 24113773; PubMed Central PMCID: PMCPMC3826632. doi:10.1038/ncomms3612
Zeng, D., Ye, Z., Shen, R., Yu, G., Wu, J., Xiong, Y., et al. (2021). IOBR: Multi-Omics Immuno-Oncology Biological Research to Decode Tumor Microenvironment and Signatures. Front. Immunol. 12, 687975. Epub 2021/07/20PubMed PMID: 34276676; PubMed Central PMCID: PMCPMC8283787. doi:10.3389/fimmu.2021.687975
Keywords: immune, colon cancer, signature, risk score, prognosis
Citation: Shen C, Luo C, Xu Z, Liang Q, Cai Y, Peng B, Yan Y and Xia F (2022) Molecular Patterns Based on Immunogenomic Signatures Stratify the Prognosis of Colon Cancer. Front. Bioeng. Biotechnol. 10:820092. doi: 10.3389/fbioe.2022.820092
Received: 22 November 2021; Accepted: 11 January 2022;
Published: 14 February 2022.
Edited by:
Giulia Russo, University of Catania, ItalyReviewed by:
Debashis Sahoo, University of California, San Diego, United StatesQian Chen, Guangxi Medical University Cancer Hospital, China
Copyright © 2022 Shen, Luo, Xu, Liang, Cai, Peng, Yan and Xia. This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) and the copyright owner(s) are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms.
*Correspondence: Yuanliang Yan, eWFueXVhbmxpYW5nQGNzdS5lZHUuY24=; Fada Xia, eGlhZmFkYUBjc3UuZWR1LmNu
†These authors contributed equally to this work and shared first authorship