MS4A3 as a potential prognostic biomarker for colon cancer: integrated analysis of expression patterns and immune cell infiltration

Introduction

Colon cancer (CC) represents one of the most prevalent malignancies within the digestive system (1). According to the latest global cancer statistics, the incidence of CC remains among the highest, with new cases projected to reach 3.2 million worldwide by 2040 (2). Notably, the most significant increases in incidence have been observed in transitional countries and among younger populations, imposing a substantial socioeconomic burden (3). Although certain countries have made progress through early intervention strategies and by lowering the age for screening initiation, the high mortality rate of CC continues to pose a formidable challenge in clinical practice.

Currently, the standardized clinical management of CC encompasses primary tumor resection, regional radiotherapy, and fluorouracil (5-FU)-based adjuvant chemotherapy (4). However, these traditional therapeutic modalities are associated with multiple clinical limitations (5). Due to the insidious nature of early symptoms, most patients are diagnosed at advanced stages, frequently with adjacent organ invasion or regional lymph node metastasis, rendering complete surgical resection challenging (6). Radiotherapy is often compromised by acquired resistance, leading to treatment failure and recurrence (7). Chemotherapeutic agents such as 5-FU, despite potent anti-tumor activity, are associated with severe adverse effects including neurotoxicity, bone marrow suppression, and gastrointestinal reactions, substantially impairing patients’ quality of life (8). Additionally, the high molecular and anatomical heterogeneity of CC—exemplified by distinct biological characteristics between left-sided and right-sided tumors—contributes to marked variability in responses to standard chemotherapy (9). Given these challenges, the identification of novel molecular targets and therapeutic strategies has become imperative.

Membrane Spanning 4-Domains A3 (MS4A3), a member of the MS4A superfamily, regulates cell cycle progression in hematopoietic progenitor cells (10). In myeloid leukemia, the transcription factor EVI-1 promotes tumor proliferation by suppressing MS4A3 expression, suggesting a tumor-suppressive role for MS4A3 (11). The functional roles of MS4A family members in tumorigenesis and tumor progression exhibit considerable heterogeneity (12). Among them, MS4A7 is highly expressed in gastric cancer and correlates with poor prognosis (13); MS4A8B is upregulated in prostate cancer and CC, where it facilitates cell cycle progression and tumor metastasis (14,15); TMEM176A and TMEM176B, structurally distinct members of the MS4A family, are aberrantly expressed in multiple tumor types including breast cancer and hepatocellular carcinoma, and promote tumor progression through pathways such as AKT/mTOR signaling and epithelial-mesenchymal transition (16,17). Additionally, similar to MS4A3, MS4A12 is specifically expressed in colorectal epithelial cells, and its loss of expression is associated with poor prognosis, indicating a tumor-suppressive potential (18). Collectively, MS4A family members exhibit dual tumor-suppressive or pro-tumorigenic functions in a manner highly dependent on molecular subtypes and the tumor microenvironment (TME) context.

Despite the established roles of several MS4A family members in tumor progression and prognosis across various cancer types, the functional significance and clinical relevance of MS4A3 in CC remain largely unexplored. In this study, we aimed to investigate the expression pattern, prognostic value, and biological function of MS4A3 in CC through integrated bioinformatics analyses, with the goal of identifying MS4A3 as a novel potential therapeutic target. We present this article in accordance with the TRIPOD and MDAR reporting checklists (available at https://jgo.amegroups.com/article/view/10.21037/jgo-2026-0374/rc).

Methods

Data acquisition and preprocessing

Transcriptomic sequencing data and corresponding clinical follow-up information for the Colon Adenocarcinoma (COAD) project were obtained from The Cancer Genome Atlas (TCGA) database (https://portal.gdc.cancer.gov/). The TCGA-COAD cohort comprised 488 samples, including 449 tumor tissues and 39 adjacent normal tissues. A total of 449 patients with complete survival data were included in the prognostic analysis. For external validation, four independent datasets (GSE17538, GSE39582, GSE44076, and GSE38832) were downloaded from the Gene Expression Omnibus (GEO) database. GSE39582 included 566 tumor and 19 normal tissues, with survival data available for 562 patients. GSE44076 contained 98 tumor and 148 normal samples, while GSE17538 consisted of 232 tumor samples with complete survival information. GSE38832 comprised 122 tumor samples and was used for survival analysis.

Clinical sample collection

This study collected 15 pairs of CC tissues and their corresponding adjacent normal tissues from patients who underwent surgical resection between January 1, 2026, and March 20, 2026. The inclusion criteria were: (I) pathologically confirmed CC; (II) radical surgical resection; (III) no prior antitumor therapy before surgery; (IV) availability of tumor and adjacent normal tissues; (V) complete clinical data; (VI) signed informed consent. The sample size of 15 pairs was not determined based on statistical power calculation, but on the following practical considerations: this study is exploratory in nature, aiming to preliminarily observe the expression trends of target molecules in CC tissues; a sample size of 15–30 pairs is widely accepted as reasonable in similar exploratory studies. The study protocol was approved by the Ethics Committee of Peking University Cancer Hospital (Inner Mongolia Campus), Affiliated Cancer Hospital of Inner Mongolia Medical University (approval No. KY2025193), and was conducted in accordance with the Declaration of Helsinki and its subsequent amendments. Written informed consent was obtained from all participants.

Identification of differentially expressed genes (DEGs)

DEGs between tumor and normal tissues were identified using the DESeq2 package in R (19). The statistical thresholds for significance were defined as an absolute fold change |log₂FC| > 1 and a false discovery rate (FDR) <0.05.

Functional enrichment and pathway analysis

Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway enrichment analyses were performed using the clusterProfiler package (version 4.8.3) (20). GO analysis encompassed biological processes (BPs), cellular components (CCs), and molecular functions (MFs). Additionally, gene set enrichment analysis (GSEA) and gene set variation analysis (GSVA) were performed using the clusterProfiler R package (20). For all enrichment analyses, statistical significance was determined using a FDR-adjusted P value threshold of FDR <0.05.

Survival analysis and prognostic evaluation

Univariate Cox proportional hazards regression analysis was performed using the survival package in R to evaluate the prognostic significance of candidate genes. Hazard ratios (HRs) and 95% confidence intervals (CIs) were calculated to assess the impact of gene expression levels on the overall survival (OS) of patients. Genes with P<0.05 were considered potential prognostic biomarkers and were subsequently included in multivariable Cox regression analysis to evaluate whether MS4A3 expression served as an independent prognostic factor. Survival rates were assessed using the Kaplan-Meier method, with differences between survival curves compared using the log-rank test, implemented via the survival and survminer R packages. Patients were stratified into high- and low-expression groups based on the median MS4A3 expression level within each cohort. Receiver operating characteristic (ROC) curves were generated using the pROC package to calculate the area under the curve (AUC) (21). Additionally, time-dependent ROC curves were constructed using the timeROC package to assess predictive accuracy at 1, 3, and 5 years (22).

Tumor immune infiltration analysis

The relative abundance of 22 immune cell types was estimated using the CIBERSORT algorithm, based on a validated leukocyte gene signature matrix (23). Additionally, the ESTIMATE algorithm was applied to quantify the infiltration levels of stromal and immune cells within the tumor tissues (24). Stromal Scores, Immune Scores, and ESTIMATE Scores were calculated to evaluate the composition of the TME.

Prediction of immunotherapy response

The Tumor Immune Dysfunction and Exclusion (TIDE) algorithm (http://tide.dfci.harvard.edu/) was utilized to predict the potential response to immune checkpoint blockade. Cytotoxic T lymphocyte (CTL) levels were initially calculated to stratify patients into high- and low-CTL groups. Subsequently, Pearson correlation analysis was employed to calculate Dysfunction scores for the high-CTL group and Exclusion scores for the low-CTL group, ultimately generating a comprehensive TIDE score for each patient.

Cell culture and transfection

The human colon cell lines CCD-18Co (JY273), HT-29 (JY-Y14675), and SW480 (JY-T153) were obtained from Shanghai Jinyuan Biotechnology Co., Ltd. (Shanghai, China). These cells were cultured in DMEM supplemented with 10% fetal bovine serum (FBS) and 1% penicillin-streptomycin. The Caco-2 cell line (ml096032; Shanghai Enzyme-linked Biotechnology Co., Ltd.) was maintained in IMDM containing 10% FBS and 1% penicillin-streptomycin. All cell lines were incubated at 37 ℃ in a humidified atmosphere with 5% CO₂. Subculturing was performed upon reaching 70–80% confluence.

For MS4A3 overexpression in SW480 cells, transfection was carried out using Lipofectamine 2000 reagent (Invitrogen, USA) with pcDNA3.1 negative control (OE-NC) and pcDNA3.1-MS4A3 (OE-MS4A3) plasmids.

HT-29 cells were transfected with MS4A3-targeting siRNAs (si-MS4A3-1/2/3) and si-NC using a transfection kit (GoldenTran-R, GoldenTran Biotechnology) according to the manufacturer’s protocol. The siRNA sequences were: si-MS4A3-1, 5'-ACAAGAACATGGATACAGA-3'; si-MS4A3-2, 5'-GAACATTGCCAGTGCTACA-3'; and si-MS4A3-3, 5'-GCAGTTAATATCCAGTCAT-3'; si-NC, TTCTCCGAACGTGTCACGT.

Quantitative real-time polymerase chain reaction (qRT-PCR)

Total RNA was extracted using TriQuick Reagent (R1100, Solarbio), and complementary DNA (cDNA) was synthesized using the SureScript™ First-Strand cDNA Synthesis Kit (QP056T, GeneCopoeia) according to the manufacturer’s instructions. Quantitative PCR was performed with 2× SYBR Green qPCR Master Mix (None ROX) (G3320-05, Servicebio) on a real-time PCR system. GAPDH served as the internal control, and the relative expression levels of target genes were calculated using the 2^-ΔΔCt method. The primer sequences used were as follows: GAPDH: F: 5'-CAGGAGCGAGACCCCACTAA-3', R: 5'-ATCACGCCACAGCTTTCCAG-3'; MS4A3: F: 5'-ATTGCACTAGTGGGGACTGC-3', R: 5'-GCAGTTTGCATTGCACCACA-3'. Each experiment was performed with three biological replicates, and each biological replicate was analyzed in technical triplicate. Data are presented as mean ±standard deviation (SD).

Western blot analysis

Total proteins were extracted using RIPA lysis buffer (BN25011-A-100mL, Biorigin) and quantified with a BCA Protein Assay Kit (SL201, UtiBody). Equal amounts of protein were separated and transferred onto membranes, which were then incubated overnight at 4 ℃ with the following primary antibodies: anti-MS4A3 (PR15091, Abmart), anti-beta actin (ab8226, Abcam). Following washing, the membranes were incubated with HRP-conjugated Goat Anti-Rabbit IgG H&L secondary antibody (ab205718, Abcam) for 1.5 hours at room temperature. Protein bands were visualized using an Enhanced Chemiluminescence Detection Kit (BN16009-100 mL, Biorigin) and captured with the ChemiScope 6100 imaging system. Band intensities were quantified using ImageJ software. All western blot experiments were repeated independently three times.

Cell viability assay

SW480 cells transfected with OE-NC or OE-MS4A3 were seeded into 96-well plates at a density of 2×10³ cells per well. Cell viability was quantitatively assessed at 0, 24, 48, and 72 hours using a CCK-8 assay kit (C0039, Beyotime) according to the manufacturer’s instructions. The absorbance at 450 nm was measured using a microplate reader. All experiments were performed in triplicate and repeated independently three times. Data are presented as mean ± SD.

Transwell assay

Cell motility was evaluated using Transwell chambers. For migration assays, SW480 cells (5×10⁵ cells/mL) were resuspended in 200 µL of medium and seeded into the upper chamber. For invasion assays, the upper chamber was pre-coated with Matrigel (356234, BD Biosciences) to simulate the extracellular matrix, while the remaining procedures were the same as those for the migration assay. In both assays, 500 µL of medium containing 10% FBS was added to the lower chamber as a chemoattractant. Following incubation at 37 ℃ for 24 h, cells on the upper membrane surface were removed using cotton swabs. Cells that had migrated or invaded to the lower surface were fixed with a cell fixative solution (BN20094, Biorigin), stained with crystal violet, and visualized under a light microscope. Cell counts were obtained from five randomly selected fields per chamber. All experiments were performed in triplicate and repeated independently three times. Data are presented as mean ± SD.

EdU staining assay

Cell proliferation was assessed using the EdU Detection Kit (CA1174, Solarbio). After being fixed with 4% paraformaldehyde and permeabilized with 0.5% Triton X-100, cells were stained with the Click reaction mixture and counterstained with Hoechst 33342. EdU-positive cells were visualized and photographed using a fluorescence microscope. All EdU assays were performed in triplicate and repeated independently three times. Data are presented as mean ± SD.

TUNEL staining assay

Cell apoptosis was evaluated using a TUNEL Assay Kit (C1086, Beyotime) according to the manufacturer’s instructions. Briefly, SW480 cells were fixed with immunostaining fixative (P0098, Beyotime) for 30 min at room temperature, followed by permeabilization with a strong immunostaining permeabilization solution (P0097, Beyotime) for 5–10 min. Subsequently, the cells were incubated with the TUNEL reaction mixture at 37 ℃ for 60 min in the dark. After counterstaining with DAPI, apoptotic cells were visualized and captured using a fluorescence microscope. All TUNEL assays were performed in triplicate and repeated independently three times. At least three random microscopic fields per sample were selected for analysis, and the apoptotic rate was quantified as the percentage of TUNEL-positive cells relative to the total number of DAPI-stained cells. Quantification was performed using ImageJ software (version 1.53k). Data are presented as mean ± SD.

Cell apoptosis analysis

Cell apoptosis was assessed using the Annexin V-FITC Apoptosis Detection Kit (A5001-02P-L, Tianjin Simu Biotechnology Co., Ltd.) according to the manufacturer’s protocol. Briefly, cells were resuspended in 1× Binding Buffer at a density of 1×10⁶ cells/mL, and 100 µL of cell suspension (1×10⁵ cells) was transferred to a flow cytometry tube. Subsequently, 5 µL of Annexin V-FITC and 5 µL of PI Solution were added, gently mixed, and incubated at room temperature (25 ℃) for 15 min in the dark. Following the addition of 400 µL of 1× Binding Buffer, samples were analyzed by flow cytometry within 1 hour, with a minimum of 10,000 events recorded per sample. All experiments were performed in triplicate and repeated independently three times.

Statistical analysis

All statistical analyses were conducted using R software (Version 4.3.3). Continuous variables, such as gene expression levels and immune cell infiltration abundance, were compared between two groups using the Wilcoxon rank-sum test. Correlations between variables were assessed using Pearson correlation analysis. All statistical tests were two-sided, and a value of P<0.05 was considered statistically significant. For experimental data, the Student’s t-test was used for comparisons between two groups, while one-way analysis of variance (ANOVA) was applied for comparisons among three or more groups. Data are shown as mean ± SD, with P<0.05 considered significant.

Results

Identification and clinical significance of the prognostic gene MS4A3 in CC

Differential expression analysis was performed utilizing the TCGA-COAD cohort, identifying a total of 4,040 DEGs based on the criteria of |log₂FC| >1 and FDR <0.05 (Figure 1A,1B; see table online: https://cdn.amegroups.cn/static/public/jgo-2026-0374-1.xlsx). Subsequently, univariate Cox survival analysis of these DEGs within the TCGA-COAD cohort yielded 475 genes significantly associated with patient prognosis. By intersecting these 475 genes with prognostic genes identified from the GSE38832 dataset, 17 candidate genes with consistent prognostic significance across both cohorts were ultimately identified (Figure 1C). Among these, MS4A3, which has been reported to exert tumor-suppressive effects in various malignancies but remains understudied in CC, was selected as the core gene for further investigation (25). Expression analysis based on the TCGA-COAD cohort revealed that MS4A3 levels were significantly downregulated in tumor tissues compared to normal tissues (Figure 1D). Further Kaplan-Meier survival analysis using the TCGA-COAD cohort demonstrated that, using the median MS4A3 expression as a cutoff, patients in the low-expression group had a significantly shorter OS than those in the high-expression group (Figure 1E). Additionally, ROC curve analysis based on the TCGA-COAD cohort yielded an AUC of 0.683 (Figure 1F), indicating a moderate discriminatory capacity of MS4A3 in distinguishing CC tissues from normal counterparts.

Figure 1 Identification and clinical significance of MS4A3 in the TCGA-COAD cohort. (A,B) Volcano plot and heatmap displaying DEGs in the TCGA-COAD cohort. (C) Venn diagram illustrating the intersection of prognostic DEGs from the TCGA-COAD and GSE38832 datasets. (D) Comparison of MS4A3 expression levels between tumor and normal tissues in the TCGA-COAD cohort. (E) Kaplan-Meier survival curves for OS in patients stratified by median MS4A3 expression. (F) ROC curve evaluating the diagnostic performance of MS4A3. ****, P<0.0001. AUC, area under the curve; DEGs, differentially expressed genes; FC, fold change; OS, overall survival; ROC, receiver operating characteristic; TCGA-COAD, The Cancer Genome Atlas-Colon Adenocarcinoma.

Validation of the prognostic and diagnostic value of MS4A3 using external datasets

Multiple independent GEO cohorts were integrated to perform external validation of the findings. Initial expression analysis based on the GSE44076 cohort revealed that MS4A3 levels were significantly lower in tumor tissues than in normal control tissues (Figure 2A). Furthermore, the ROC curve analysis for the diagnosis of CC yielded an AUC of 0.933, demonstrating superior diagnostic performance (Figure 2B). Subsequently, the prognostic predictive capability of MS4A3 was evaluated across two independent survival cohorts. In the GSE39582 cohort, Kaplan-Meier survival analysis indicated that patients in the low MS4A3 expression group had a significantly lower survival probability compared to those in the high-expression group (Figure 2C). This trend was further corroborated by the GSE17538 cohort, which demonstrated that low MS4A3 expression was closely associated with a poor prognosis (Figure 2D).

Figure 2 External validation of the diagnostic and prognostic value of MS4A3 in multiple independent GEO cohorts. (A) Comparison of MS4A3 expression levels between tumor tissues and normal control tissues in the GSE44076 cohort. (B) ROC curve analysis of MS4A3 in the GSE44076 cohort. (C) Kaplan-Meier survival curves for OS of patients in the GSE39582 cohort. (D) Kaplan-Meier survival curves for OS of patients in the GSE17538 validation cohort. (E) qRT-PCR analysis of MS4A3 mRNA levels in CC and adjacent normal tissues (n=15). (F) Representative Western blot images (left) and quantitative analysis (right) of MS4A3 protein levels in CC and adjacent normal tissues. β-actin served as the loading control. Data are shown as mean ± SD. ***, P<0.001; ****, P<0.0001. AUC, area under the curve; CC, colon cancer; GEO, Gene Expression Omnibus; OS, overall survival; qRT-PCR, quantitative real-time polymerase chain reaction; ROC, receiver operating characteristic; SD, standard deviation.

Based on the above observations, we subsequently validated the expression of MS4A3 in clinical specimens. In paired tumor tissues and adjacent normal tissues, both qRT-PCR and western blot analyses demonstrated that MS4A3 was significantly down-regulated in CC tissues compared with the adjacent normal tissues (Figure 2E,2F).

Pan-cancer expression profiles and genomic alteration characteristics of MS4A3

Pan-cancer expression analysis revealed that, compared with corresponding normal tissues, MS4A3 exhibited a consistent pattern of low expression across multiple solid tumors, including COAD (P<0.001) (Figure S1). Beyond CC, MS4A3 was also significantly downregulated in lung adenocarcinoma (LUAD) and lung squamous cell carcinoma (LUSC). Subsequently, the genomic alteration landscape of MS4A3 was evaluated using the cBioPortal database. The results indicated that the overall mutation frequency of MS4A3 across various cancers remained low, ranging from 1% to 4%. Specifically, the alteration frequency in COAD was only approximately 0.5%, with the primary mutation types encompassing somatic mutations and deep deletions (Figure 3A). As shown in Figure 3B, the vast majority of tumor samples harbored no genomic alterations in MS4A3. Among the few instances of variation, the alterations were predominantly sporadic missense mutations and shallow deletions. These findings suggest that the significant downregulation of MS4A3 in CC is likely driven by transcriptional or epigenetic dysregulation rather than structural genomic instability.

Figure 3 Pan-cancer genomic alteration landscape of MS4A3. (A) Genomic alteration frequencies and mutation types of MS4A3 across various cancer types. (B) Detailed distribution of genomic alterations of MS4A3 in tumor samples. CNA, NOS, not otherwise specified; TCGA, The Cancer Genome Atlas; VUS.

MS4A3 as an independent prognostic biomarker

To evaluate the independent prognostic value of MS4A3, univariate Cox regression analysis was performed in the TCGA cohort incorporating MS4A3 expression levels along with clinicopathological variables including age, gender, stage, T classification, N classification, M classification, MSI status, and tumor location (Figure 4A), with factors reaching P<0.05 entered into multivariate models. To avoid multicollinearity arising from the hierarchical relationship between overall stage and T, N, M classifications, two independent multivariate Cox regression models were constructed: Model 1 included MS4A3 expression, age, and stage (Figure 4B), while Model 2 included MS4A3 expression, age, and T, N, M classifications (Figure 4C). Both models consistently identified MS4A3 as an independent prognostic factor in CC (both P<0.05), with concordant results confirming the robustness of this finding.

Figure 4 Evaluation of MS4A3 as an independent prognostic biomarker and construction of a nomogram prediction model. (A) Univariate Cox regression analysis of MS4A3 expression and clinical features in the TCGA-COAD cohort. (B) Multivariate Cox regression Model 1 incorporating MS4A3 expression, age, and stage as covariates. (C) Multivariate Cox regression Model 2 incorporating MS4A3 expression, age, and T, N, M classifications as covariates. (D) Nomogram integrating MS4A3 expression, age, and clinical stage for predicting 1-, 3-, and 5-year survival probabilities. (E-G) Calibration curves comparing the predicted and actual survival probabilities at 1-, 3-, and 5-year intervals. Model 1: ; Model 2:. *, P<0.05; **, P<0.01; ***, P<0.001. AIC, Akaike information criterion; CI, confidence interval; HR, hazard ratio; M, metastasis; MSI-H, microsatellite instability-high; MSI-L, microsatellite instability-low; MSS, microsatellite stable; N, node; NOS, not otherwise specified; OS, overall survival; T, tumor; TCGA-COAD, The Cancer Genome Atlas-Colon Adenocarcinoma.

Building upon these findings, a nomogram prediction model was constructed by integrating the aforementioned independent prognostic factors to quantitatively predict the survival probability of patients at 1, 3, and 5 years (Figure 4D). Bootstrap internal validation (B=1,000) yielded corrected C-indices of 0.733, 0.734, and 0.732 at 1, 3, and 5 years, respectively (apparent C-index: 0.745), indicating no substantial overfitting. The calibration curves revealed that the predicted survival probabilities at 1, 3, and 5 years were in close agreement with the ideal 45° reference line, indicating that the model possesses high accuracy and significant predictive value (Figure 4E-4G).

Functional enrichment analysis

To further elucidate the potential mechanisms by which MS4A3 regulates the pathogenesis and progression of CC, a comprehensive enrichment analysis was performed on the DEGs between the MS4A3 high- and low-expression groups. GO enrichment analysis revealed that these DEGs were significantly enriched in 1,443 GO terms, including 1,114 BP terms (e.g., small GTPase-mediated signal transduction, positive regulation of cell adhesion, and endosomal transport), 173 MF terms (e.g., GTPase binding, small GTPase binding, and nucleoside-triphosphatase regulator activity), and 156 CC terms (e.g., cell-substrate junction, focal adhesion, and Golgi apparatus subcompartment) (Figure 5A; see table online: https://cdn.amegroups.cn/static/public/jgo-2026-0374-2.xlsx). Furthermore, KEGG pathway enrichment analysis identified 8 significantly enriched pathways (P<0.05), including the TNF signaling pathway, regulation of the actin cytoskeleton, and autophagy-animal (Figure 5B; see table online: https://cdn.amegroups.cn/static/public/jgo-2026-0374-2.xlsx). GSEA further indicated that 125 pathways were significantly enriched in the high-expression group compared to the low-expression group. Specifically, signaling pathways such as the FoxO signaling pathway, natural killer cell mediated cytotoxicity and the TGF-beta signaling pathway were activated in the high-expression group, while metabolic pathways like the pentose phosphate pathway were suppressed (Figure 5C,5D; see table online: https://cdn.amegroups.cn/static/public/jgo-2026-0374-2.xlsx). Finally, GSVA was performed to further validate the differences in pathway activities. The results demonstrated significant disparities between the high- and low-expression groups across 186 KEGG pathways, notably including Ribosome, Oxidative phosphorylation, and Pyruvate metabolism (Figure 5E; see table online: https://cdn.amegroups.cn/static/public/jgo-2026-0374-2.xlsx).

Figure 5 Functional enrichment and pathway analysis of MS4A3-related DEGs in colon cancer. (A) GO enrichment analysis of DEGs between the MS4A3 high-expression and low-expression groups. (B) KEGG pathway enrichment analysis. (C,D) GSEA. (E) GSVA. BP, biological process; CC, cellular component; DEGs, differentially expressed genes; GO, Gene Ontology; GSEA, gene set enrichment analysis; GSVA, gene set variation analysis; KEGG, Kyoto Encyclopedia of Genes and Genomes; MF, molecular function.

Correlation analysis between MS4A3 expression and immune microenvironment charActeristics in CC

The CIBERSORT algorithm was first employed to calculate the relative abundance of 22 types of immune cells in the TCGA-COAD cohort (Figure 6A). Differential analysis revealed significant differences between the MS4A3-high and MS4A3-low groups across seven immune cell types, including M2 macrophages and neutrophils (Figure 6B). Pearson correlation analysis further confirmed that MS4A3 expression was positively correlated with regulatory T cells (Tregs) and memory B cells, while negatively correlated with M1/M2 macrophages, resting mast cells, neutrophils, and eosinophils (Figure 6C). Consistently, xCell analysis demonstrated that the MS4A3 expression level is intimately linked to diverse immune cell infiltrates. Specifically, the low MS4A3 expression group was significantly associated with the enrichment of pro-tumor cells, including Tregs, M2 macrophages, and Th2 cells (Figure 6D,6E). To resolve the expression pattern of MS4A3 at single-cell resolution, we analyzed two independent CC single-cell RNA-sequencing datasets (EMTAB8107 and GSE16655). t-SNE clustering and quantitative expression analysis demonstrated that MS4A3 was predominantly localized in and highly expressed by mast cells (Figure S2).

Figure 6 Analysis of immune infiltration and single-cell expression patterns of MS4A3. (A) Quantification of the relative abundance of 22 immune cell types in the TCGA-COAD cohort using the CIBERSORT algorithm. (B) Comparison of immune cell infiltration levels between MS4A3-high and MS4A3-low expression groups based on CIBERSORT. (C) Pearson correlation analysis between MS4A3 expression and infiltrating immune cell populations. (D) Evaluation of immune cell infiltration across MS4A3 expression groups using the xCell algorithm. (E) Pearson correlation analysis between MS4A3 expression and specific immune cell subsets derived from xCell. *, P<0.05; **, P<0.01; ***, P<0.001; ****, P<0.0001. aDC, activated dendritic cell; cDC, conventional dendritic cell; CMP, DC, dendritic cell; HSC, hematopoietic stem cell; iDC, immature dendritic cell; NK, natural killer cell; TCGA-COAD, The Cancer Genome Atlas-Colon Adenocarcinoma.

Subsequently, assessment via the ESTIMATE algorithm showed that the Immune Score, Stromal Score, and overall ESTIMATE Score were significantly higher in the MS4A3-low group compared to the MS4A3-high group (Figure 7A-7C). Furthermore, TIDE analysis indicated that cancer-associated fibroblasts (CAFs) were significantly enriched in the MS4A3-high group, whereas myeloid-derived suppressor cells (MDSCs) and TAM.M2 were significantly elevated in the MS4A3-low group (Figure 7D-7F). Immune checkpoint analysis revealed that key molecules—including CD274, CD80, CD86, CTLA4, LAG3, PDCD1, PDCD1LG2, and TIGIT—exhibited significantly higher expression levels in the MS4A3-low group (Figure 7G). These findings suggest that low MS4A3 expression may be associated with an immunosuppressive state in CC tissues.

Figure 7 Evaluation of the immune microenvironment and checkpoint expression associated with MS4A3. (A-C) Comparison of Immune Score, Stromal Score, and ESTIMATE Score between MS4A3-high and MS4A3-low expression groups using the ESTIMATE algorithm. (D-F) TIDE analysis evaluating the enrichment of CAFs, MDSCs, and TAM.M2. (G) Expression levels of key immune checkpoint molecules in MS4A3-high and MS4A3-low expression groups. *, P<0.05; ****, P<0.0001. CAF, cancer-associated fibroblast; MDSC, myeloid-derived suppressor cell; TIDE, Tumor Immune Dysfunction and Exclusion.

Construction of the MS4A3-related ceRNA regulatory network

To explore the potential upstream regulatory mechanisms underlying the downregulation of MS4A3 in CC, an lncRNA-miRNA-MS4A3 regulatory network was constructed based on the ceRNA hypothesis. By integrating multiple database predictions, four key miRNAs potentially targeting MS4A3, including hsa-miR-150-5p and hsa-miR-27a-3p, were identified, along with numerous candidate lncRNAs such as XIST, OIP5-AS1, and NEAT1 (Figure 8). The Sankey diagram illustrated a complex interactive regulatory network among these non-coding RNAs, suggesting that the low expression of MS4A3 in CC may be precisely orchestrated by specific regulatory axes.

Figure 8 Construction of the lncRNA-miRNA-MS4A3 ceRNA regulatory network in CC. CC, colon cancer.

Low expression of MS4A3 in CC cell lines and its tumor-suppressive function in vitro

To further validate the biological function of MS4A3 in CC, its expression levels were initially evaluated across several CC cell lines. The results revealed significantly low expression of MS4A3 in all three cancer cell lines compared with the normal colonic epithelial cell line CCD-18Co, with the most pronounced reduction observed in SW480 cells (Figure 9A,9B). To further investigate its functional role, MS4A3-overexpressing SW480 cells were established, and the overexpression efficiency was validated via qRT-PCR and western blot analysis (Figure 9C,9D). Functional assays demonstrated that MS4A3 overexpression markedly suppressed the viability, proliferation, migration, and invasion of SW480 cells (Figure 10A-10D). Consistently, two independent methods, TUNEL staining and Annexin V/PI double-staining flow cytometry, both confirmed that MS4A3 overexpression significantly promoted cell apoptosis, with the apoptotic proportion in the OE-MS4A3 group being significantly higher than that in the control group (Figure 10E,10F). To validate the robustness and generalizability of these findings, MS4A3 overexpression experiments were simultaneously conducted in HT-29 cells. CCK-8 assays demonstrated that MS4A3 overexpression significantly reduced HT-29 cell viability, consistent with the results observed in SW480 cells (Figure S3A-S3C).

Figure 9 Expression and overexpression efficiency of MS4A3 in CC cell lines. (A,B) MS4A3 mRNA and protein levels in three CC cell lines (HT-29, Caco-2, SW480) and a normal colonic cell line (CCD-18Co) were determined by qRT-PCR (A) and Western blot (B). (C,D) The overexpression efficiency of MS4A3 in SW480 cells was validated at both mRNA (C) and protein levels (D). β-actin was used as the loading control. Data are expressed as mean ± SD. ns, not significant; ***, P<0.001. CC, colon cancer; NC, negative control; OE, overexpression; qRT-PCR, quantitative real-time polymerase chain reaction; SD, standard deviation.

Figure 10 MS4A3 inhibits CC cell viability, proliferation, migration, and promotes apoptosis. (A,B) The effects of MS4A3 overexpression on SW480 cells were assessed by CCK-8 viability assay (A) and EdU proliferation assay (B). (C,D) Representative images (left) and quantitative analysis (right) of SW480 cell migration (C) and invasion (D) following MS4A3 overexpression. Staining method: . (E) Cell apoptosis was evaluated by TUNEL assay, with representative fluorescence images (left) and relative quantification of TUNEL-positive cells (right). (F) Cell apoptosis was further evaluated by Annexin V/PI double-staining flow cytometry, with representative scatter plots and quantification of the apoptotic proportion. Data are presented as mean ± SD of three independent experiments (n=3). ns, not significant; *, P<0.05; **, P<0.01; ***, P<0.001. CC, colon cancer; CCK-8, Cell Counting Kit-8; NC, negative control; OE, overexpression; PI; SD, standard deviation.

To further confirm the specificity of MS4A3-mediated suppression of cell viability, HT-29 cells, which exhibit relatively higher endogenous MS4A3 expression and are therefore more suitable for knockdown experiments than SW480 cells with their negligible endogenous MS4A3 levels, were selected for siRNA screening. Three siRNAs targeting MS4A3 were designed and screened, and following validation of knockdown efficiency at both the mRNA and protein levels, si-MS4A3-1 was identified as the most effective and selected for subsequent functional experiments (Figure 11A,11B). CCK-8 assays demonstrated that si-MS4A3-1-mediated MS4A3 knockdown significantly enhanced HT-29 cell viability (Figure 11C). Furthermore, rescue experiments conducted in SW480 cells confirmed that si-MS4A3-1 reversed the reduction in cell viability induced by OE-MS4A3, demonstrating that the inhibitory effect of MS4A3 on cell viability is target-specific (Figure 11D). Collectively, these findings suggest that MS4A3 exerts a tumor-suppressive effect in CC.

Figure 11 Loss-of-function and rescue experiments confirm the tumor-suppressive role of MS4A3 in CC. (A,B) Knockdown efficiency of three MS4A3-targeting siRNAs in HT-29 cells was validated by qRT-PCR (A) and Western blot (B). (C) The effect of si-MS4A3-1-mediated MS4A3 knockdown on HT-29 cell viability was assessed by CCK-8 assay. (D) Rescue experiments in SW480 cells showing that si-MS4A3-1 reverses the reduction in cell viability induced by OE-MS4A3, as assessed by CCK-8 assay. Data are presented as mean ± SD of three independent experiments (n=3). ns, not significant; ***, P<0.001; ****, P<0.0001. CC, colon cancer; CCK-8, Cell Counting Kit-8; NC, negative control; OD, optical density; OE, overexpression; qRT-PCR, quantitative real-time polymerase chain reaction; SD, standard deviation.

Discussion

MS4A3 is a key member of the MS4A family, which is typically characterized by four transmembrane domains and plays a pivotal role in cell signal transduction and hematopoiesis (10). Previous studies have demonstrated that MS4A3 exhibits tumor-suppressive potential in hematological malignancies and various solid tumors, with its loss of expression frequently associated with disease progression (25,26). However, the specific expression profile and prognostic value of MS4A3 in CC have not yet been systematically investigated.

By integrating data from TCGA-COAD and multiple independent GEO cohorts, this study identified MS4A3 as a gene closely associated with CC. Our results showed that MS4A3 is significantly downregulated in CC tissues. Low expression of MS4A3 was found to be a risk factor for significantly shortened OS and was identified as an independent poor prognostic factor for CC. Furthermore, functional analysis demonstrated that overexpression of MS4A3 in CC cells reduced cell viability, proliferation, and migration while promoting apoptosis. These findings provide evidence that MS4A3 exerts a tumor-suppressive effect in CC.

Through GSEA, we found that high MS4A3 expression is correlated with the activation of the transforming growth factor-beta (TGF-β) signaling pathway. Previous studies have demonstrated that MS4A3 serves as a positive regulator of the TGF-β pathway, and its deficiency leads to a significant attenuation of TGF-β-mediated cell cycle arrest (27). Given that TGF-β signaling functions as a tumor suppressor in the early stages of malignancy by inhibiting cell proliferation and inducing apoptosis (28,29), we hypothesize that the low expression of MS4A3 in CC may result in the dysfunction or dysregulation of the TGF-β pathway. The loss of this synergistic tumor-suppressive effect impairs the braking capacity on the cell cycle, thereby accelerating malignant progression. Furthermore, GSEA revealed a significant negative correlation between MS4A3 expression and the pentose phosphate pathway, suggesting that the loss of MS4A3 is frequently coupled with the aberrant activation of the PPP. As a critical metabolic route for maintaining energy supply and redox homeostasis under stress, the activation of the PPP has been confirmed to promote tumor progression and induce multidrug resistance by remodeling biosynthetic patterns (30-33). Notably, Zhang et al. explicitly reported that PPP activation can directly drive CC growth via the TFEB-PGD axis (34). In summary, reduced MS4A3 expression may contribute to tumor progression by attenuating the tumor-suppressive TGF-β signaling pathway while concurrently activating pro-tumorigenic pentose phosphate pathway metabolism. Nevertheless, the precise molecular mechanisms underlying these processes warrant further investigation.

This study reveals that MS4A3 expression levels are closely linked to the remodeling of the immune microenvironment in CC. In the MS4A3-low expression group, we observed a significant enrichment of M2 macrophages and neutrophils. As critical immunosuppressive components within the TME, M2 macrophages play a pivotal role in inducing tumor angiogenesis and immune evasion by secreting pro-tumorigenic cytokines and remodeling the extracellular matrix (35). Concurrently, the enriched neutrophils exhibit pro-tumorigenic activity by interfering with the function of natural killer cells and can further suppress T-cell-mediated anti-tumor immune responses through the formation of neutrophil extracellular traps (36,37). More importantly, our results demonstrate that the expression levels of key immune checkpoint molecules—including PDCD1, CD274, CTLA-4, LAG3, TIGIT, CD80 and CD86—are significantly upregulated in the MS4A3-low group. This phenotype of "high immune infiltration accompanied by high inhibitory molecule expression" characterizes a classic state of immune exhaustion (38). In this state, although T cells may infiltrate the tumor tissue, their effector functions are severely impaired due to the synergistic effects of multiple inhibitory signaling gradients (39). Consequently, the low expression of MS4A3 may be associated with an immunosuppressive TME, potentially characterized by the upregulation of multiple immune checkpoints and the enrichment of immunosuppressive cell populations, which may in turn contribute to immune evasion in CC. Notably, our single-cell analysis revealed that MS4A3 is predominantly expressed in mast cells. The role of mast cells in cancer is complex and context-dependent. Some studies have reported that mast cell infiltration in tumors is associated with enhanced tumor growth and metastasis (40), whereas others have demonstrated that mast cells can exert anti-tumor effects by augmenting anti-tumor immune responses and inducing tumor cell apoptosis (41). These findings suggest that MS4A3 may indirectly modulate the tumor immune microenvironment through regulation of mast cell function, thereby contributing to its tumor-suppressive role, although this hypothesis warrants further experimental validation.

To further elucidate the potential molecular mechanisms underlying the downregulation of MS4A3 in CC, an MS4A3-based ceRNA regulatory network was constructed. Among the identified key candidates, hsa-miR-27a-3p exhibited significant oncogenic potential. As a classic oncogenic miRNA, hsa-miR-27a-3p is aberrantly overexpressed in various malignancies and has been proven to activate core oncogenic pathways, such as Wnt/β-catenin, by antagonizing multiple tumor suppressor genes, thereby accelerating tumor progression (42-44). Moreover, this miRNA plays a pivotal role during the early stages of CC transformation from normal mucosa to malignant tissue and is regarded as a central driver of tumor initiation (45). Our results suggest that hsa-miR-27a-3p may represent a potential upstream miRNA targeting MS4A3. Aberrant expression of hsa-miR-27a-3p might potentially reduce MS4A3 expression levels through induction of mRNA degradation or translational repression, which could in turn partially attenuate the tumor-suppressive effects of MS4A3. However, these regulatory relationships have not yet been experimentally validated, and their precise roles in the malignant progression of CC warrant further investigation.

This study has several limitations. First, the nomogram has undergone Bootstrap internal validation only, and its generalizability and clinical applicability require further evaluation in larger independent cohorts. Second, the proposed associations between MS4A3 and the TGF-β signaling pathway, pentose phosphate pathway, and immune regulation are inferred solely from enrichment analyses and lack direct experimental support; these mechanistic conclusions represent hypothesis-driven observational findings that will be systematically validated through assessment of key TGF-β pathway protein expression, pentose phosphate pathway-related enzyme activity, and intracellular ROS levels in future studies. Similarly, the immune infiltration analyses are based entirely on computational algorithms, and the resulting conclusions have not been directly validated through immunohistochemistry, flow cytometry, or spatial transcriptomics, future studies will systematically validate these computational predictions through experimental means. Finally, the potential regulatory relationship between hsa-miR-27a-3p and MS4A3 remains a bioinformatic prediction that has not been experimentally confirmed through luciferase reporter assays, miRNA overexpression/inhibition experiments, or RIP assays, and its precise role in CC progression warrants further investigation.

Conclusions

In summary, this study indicates that MS4A3 is significantly downregulated in CC and may serve as a critical predictor of poor prognosis and an immunosuppressive microenvironment. Functional experiments reveal that MS4A3 overexpression significantly inhibited the proliferation, invasion, and migration of CC cells. These findings highlight the significant potential of MS4A3 as a biomarker for the diagnosis and prognostic assessment of CC.

Acknowledgments

None.

Footnote

Reporting Checklist: The authors have completed the TRIPOD and MDAR reporting checklists. Available at https://jgo.amegroups.com/article/view/10.21037/jgo-2026-0374/rc

Data Sharing Statement: Available at https://jgo.amegroups.com/article/view/10.21037/jgo-2026-0374/dss

Peer Review File: Available at https://jgo.amegroups.com/article/view/10.21037/jgo-2026-0374/prf

Funding: This study was supported by the Science and Technology Project of Joint Fund for Scientific Research in Public Hospitals (grant Nos. 2024GLLH0379, 2025GLLH0208), Discipline Construction Project of Inner Mongolia Medical University (grant No. YKD2023XK012), High-level Clinical Specialty Construction Science and Technology Project of Public Hospitals in the Capital Area of Inner Mongolia Autonomous Region Health and Wellness Committee in 2023 (grant Nos. 2023SGGZ070, 2023SGGZ074, 2023pp001, and 2023pp004), Joint Project of Inner Mongolia Medical University (grant No. YKD2021LH032), Demonstration Project on Public Hospital Reform and High-Quality Development (Gastrointestinal Tumors) (grant Nos. 2023SGGZ070-02 and 2023SGGZ070-05), the Science and Technology Innovation Team Project of Inner Mongolia Medical University (grant No. YKD2024TD014), 2025 Annual Science Popularization Empowerment for Rural Revitalization Project of Inner Mongolia Association for Science and Technology (grant No. ZD-044), and The Inner Mongolia Autonomous Region Science and Technology Planning Project (grant No. 2023KJHZ0023).

Conflicts of Interest: All authors have completed the ICMJE uniform disclosure form (available at https://jgo.amegroups.com/article/view/10.21037/jgo-2026-0374/coif). The authors have no conflicts of interest to declare.

Ethical Statement: The authors are accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved. This study protocol was approved by the Ethics Committee of Peking University Cancer Hospital (Inner Mongolia Campus), Affiliated Cancer Hospital of Inner Mongolia Medical University (approval No. KY2025193), and was conducted in accordance with the Declaration of Helsinki and its subsequent amendments. Written informed consent was obtained from all participants.

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

References

1. Wang K, Wang A, Wang S, et al. The best anthropometric indices to predict colon cancer. BMC Gastroenterol 2025;25:242. [Crossref] [PubMed]
2. Abbaspour E, Karimzadhagh S, Monsef A, et al. Application of radiomics for preoperative prediction of lymph node metastasis in colorectal cancer: a systematic review and meta-analysis. Int J Surg 2024;110:3795-813. [Crossref] [PubMed]
3. He Q, Xiang L, Luo Y, et al. Tumor-associated macrophages in colon cancer immunotherapy: mechanisms, natural product interventions, and microenvironment remodeling. Front Immunol 2025;16:1642091. [Crossref] [PubMed]
4. Xiang J, Wang J, Xiao H, et al. Targeting tumor-associated macrophages in colon cancer: mechanisms and therapeutic strategies. Front Immunol 2025;16:1573917. [Crossref] [PubMed]
5. Morton D, Seymour M, Magill L, et al. Preoperative Chemotherapy for Operable Colon Cancer: Mature Results of an International Randomized Controlled Trial. J Clin Oncol 2023;41:1541-52. [Crossref] [PubMed]
6. Chuang JP, Tsai HL, Chen PJ, et al. Comprehensive Review of Biomarkers for the Treatment of Locally Advanced Colon Cancer. Cells 2022;11:3744. [Crossref] [PubMed]
7. Galeaz C, Totis C, Bisio A. Radiation Resistance: A Matter of Transcription Factors. Front Oncol 2021;11:662840. [Crossref] [PubMed]
8. Panjeta A, Kaur K, Sharma R, et al. Human Intestinal Defensin 5 Ameliorates the Sensitization of Colonic Cancer Cells to 5-Fluorouracil. Arch Med Res 2024;55:102966. [Crossref] [PubMed]
9. Castria TB, Lenz G, Valagni G, et al. Role of neoadjuvant therapies in locally advanced colon cancer. Chin Med J (Engl) 2025;138:2091-101. [Crossref] [PubMed]
10. Kutok JL, Yang X, Folkerth R, et al. Characterization of the expression of HTm4 (MS4A3), a cell cycle regulator, in human peripheral blood cells and normal and malignant tissues. J Cell Mol Med 2011;15:86-93. [Crossref] [PubMed]
11. Heller G, Rommer A, Steinleitner K, et al. EVI1 promotes tumor growth via transcriptional repression of MS4A3. J Hematol Oncol 2015;8:28. [Crossref] [PubMed]
12. Luo X, Luo B, Fei L, et al. MS4A superfamily molecules in tumors, Alzheimer's and autoimmune diseases. Front Immunol 2024;15:1481494. [Crossref] [PubMed]
13. Sun L, Zhang Y, Zhang C. Distinct Expression and Prognostic Value of MS4A in Gastric Cancer. Open Med (Wars) 2018;13:178-88. [Crossref] [PubMed]
14. Ye L, Yao XD, Wan FN, et al. MS4A8B promotes cell proliferation in prostate cancer. Prostate 2014;74:911-22. [Crossref] [PubMed]
15. Zhou H, Shen Y, Zheng G, et al. Integrating single-cell and spatial analysis reveals MUC1-mediated cellular crosstalk in mucinous colorectal adenocarcinoma. Clin Transl Med 2024;14:e1701. [Crossref] [PubMed]
16. Wang Y, Zhang Y, Herman JG, et al. Epigenetic silencing of TMEM176A promotes esophageal squamous cell cancer development. Oncotarget 2017;8:70035-48. [Crossref] [PubMed]
17. Kang C, Rostoker R, Ben-Shumel S, et al. TMEM176B Regulates AKT/mTOR Signaling and Tumor Growth in Triple-Negative Breast Cancer. Cells 2021;10:3430. [Crossref] [PubMed]
18. Dalerba P, Kalisky T, Sahoo D, et al. Single-cell dissection of transcriptional heterogeneity in human colon tumors. Nat Biotechnol 2011;29:1120-7. [Crossref] [PubMed]
19. Love MI, Huber W, Anders S. Moderated estimation of fold change and dispersion for RNA-seq data with DESeq2. Genome Biol 2014;15:550. [Crossref] [PubMed]
20. Yu G, Wang LG, Han Y, et al. clusterProfiler: an R package for comparing biological themes among gene clusters. OMICS 2012;16:284-7. [Crossref] [PubMed]
21. Robin X, Turck N, Hainard A, et al. pROC: an open-source package for R and S+ to analyze and compare ROC curves. BMC Bioinformatics 2011;12:77. [Crossref] [PubMed]
22. Blanche P, Dartigues JF, Jacqmin-Gadda H. Estimating and comparing time-dependent areas under receiver operating characteristic curves for censored event times with competing risks. Stat Med 2013;32:5381-97. [Crossref] [PubMed]
23. Newman AM, Liu CL, Green MR, et al. Robust enumeration of cell subsets from tissue expression profiles. Nat Methods 2015;12:453-7. [Crossref] [PubMed]
24. Yoshihara K, Shahmoradgoli M, Martínez E, et al. Inferring tumour purity and stromal and immune cell admixture from expression data. Nat Commun 2013;4:2612. [Crossref] [PubMed]
25. Duan Z. MS4A3 Promotes the Chemosensitivity of Lung Cancer via THAP1/EGFR Pathways. Crit Rev Eukaryot Gene Expr 2024;34:1-11. [Crossref] [PubMed]
26. Lu Q, Guo P, Wang X, et al. MS4A3-HSP27 target pathway reveals potential for haematopoietic disorder treatment in alimentary toxic aleukia. Cell Biol Toxicol 2023;39:201-16. [Crossref] [PubMed]
27. Dong Y, Yue T, Wang X, et al. MS4A3 regulates hematopoietic myeloid differentiation through ROS/TGF-β/p38MAPK pathway. Int Immunopharmacol 2024;143:113578. [Crossref] [PubMed]
28. Goenka A, Khan F, Verma B, et al. Tumor microenvironment signaling and therapeutics in cancer progression. Cancer Commun (Lond) 2023;43:525-61. [Crossref] [PubMed]
29. Peng D, Fu M, Wang M, et al. Targeting TGF-β signal transduction for fibrosis and cancer therapy. Mol Cancer 2022;21:104. [Crossref] [PubMed]
30. TeSlaa T, Ralser M, Fan J, et al. The pentose phosphate pathway in health and disease. Nat Metab 2023;5:1275-89. [Crossref] [PubMed]
31. Meng Q, Sun H, Zhang Y, et al. Lactylation stabilizes DCBLD1 activating the pentose phosphate pathway to promote cervical cancer progression. J Exp Clin Cancer Res 2024;43:36. [Crossref] [PubMed]
32. Ghanem N, El-Baba C, Araji K, et al. The Pentose Phosphate Pathway in Cancer: Regulation and Therapeutic Opportunities. Chemotherapy 2021;66:179-91. [Crossref] [PubMed]
33. Gong Y, Gao D, Shi Y, et al. SRC enhanced cisplatin resistance in bladder cancer by reprogramming glycolysis and pentose phosphate pathway. Commun Biol 2025;8:36. [Crossref] [PubMed]
34. Zhang F, Wu Z, Yu B, et al. ATP13A2 activates the pentose phosphate pathway to promote colorectal cancer growth though TFEB-PGD axis. Clin Transl Med 2023;13:e1272. [Crossref] [PubMed]
35. Chen S, Saeed AFUH, Liu Q, et al. Macrophages in immunoregulation and therapeutics. Signal Transduct Target Ther 2023;8:207. [Crossref] [PubMed]
36. Khan U, Chowdhury S, Billah MM, et al. Neutrophil Extracellular Traps in Colorectal Cancer Progression and Metastasis. Int J Mol Sci 2021;22:7260. [Crossref] [PubMed]
37. Zhang Y, Wang Z, Lu Y, et al. Region-Specific CD16(+) Neutrophils Promote Colorectal Cancer Progression by Inhibiting Natural Killer Cells. Adv Sci (Weinh) 2024;11:e2403414. [Crossref] [PubMed]
38. Rezazadeh-Gavgani E, Majidazar R, Lotfinejad P, et al. Immune Checkpoint Molecules: A Review on Pathways and Immunotherapy Implications. Immun Inflamm Dis 2025;13:e70196. [Crossref] [PubMed]
39. Naimi A, Mohammed RN, Raji A, et al. Tumor immunotherapies by immune checkpoint inhibitors (ICIs); the pros and cons. Cell Commun Signal 2022;20:44. [Crossref] [PubMed]
40. Alda S, Ceausu RA, Gaje PN, et al. Mast Cell: A Mysterious Character in Skin Cancer. In Vivo 2024;38:58-68. [Crossref] [PubMed]
41. Qiao T, Ding C, Yu S, et al. Diversity of mast cell subpopulations in the tumor microenvironment of colorectal cancer and their prognostic implications. Cancer Immunol Immunother 2025;74:255. [Crossref] [PubMed]
42. Chen L, Li H, Gao H, et al. Hsa_circ_0004194 suppresses colorectal cancer progression via hsa-miR-27a-3p. Heliyon 2024;10:e39549. [Crossref] [PubMed]
43. Salmani T, Ghaderian SMH, Hajiesmaeili M, et al. Hsa-miR-27a-3p and epidermal growth factor receptor expression analysis in glioblastoma FFPE samples. Asia Pac J Clin Oncol 2021;17:e185-90. [Crossref] [PubMed]
44. An T, Fan T, Zhang XQ, et al. Comparison of Alterations in miRNA Expression in Matched Tissue and Blood Samples during Spinal Cord Glioma Progression. Sci Rep 2019;9:9169. [Crossref] [PubMed]
45. Huang JC, Li MC, Huang IC, et al. Gene Coexpression and miRNA Regulation: A Path to Early Intervention in Colorectal Cancer. Hum Gene Ther 2024;35:855-67. [Crossref] [PubMed]