Establishing the role of the neurotransmitter receptor-related gene GABRD in the diagnosis, prognosis and immune infiltrates of colorectal cancer by bioinformatics analysis and experimental validation

Introduction

Colorectal cancer (CRC) is a major public health concern worldwide. It ranks as the third most frequently diagnosed cancer and is a leading cause of cancer-related deaths (1,2). This malignancy not only impacts affected individuals but also creates considerable financial strain on healthcare systems, due to the complex treatments required and high costs associated with managing advanced-stage disease (3). Currently, the primary clinical interventions for CRC comprise surgical resection and cytotoxic chemotherapy. Nevertheless, the efficacy of these modalities remains suboptimal, particularly in advanced stages, where 5-year survival rates remain persistently low (4). In the realm of clinical diagnostics, while certain molecular biomarkers are currently utilized to aid in the identification of CRC, the majority are subject to significant limitations, notably inadequate sensitivity and specificity for early-stage detection. Furthermore, these markers are often unable to concurrently provide a robust assessment of both tumor progression and long-term prognosis. Additionally, the key biological mechanisms underlying tumor progression and therapy resistance are not yet fully understood (5). Due to the increasing incidence and worse clinical prognosis of CRC, it is particularly significant to understand the development process of CRC and identify relevant prognostic factors.

Emerging evidence shows that neurotransmitters and their receptors play key roles in regulating the tumor microenvironment (6,7). These signaling molecules mediate diverse cellular processes including proliferation, apoptosis, and motility through receptor-ligand interactions (8-10). Recent investigations have highlighted the substantial impact of neurotransmitter receptor-associated genes (NRGs) on CRC and disease progression. These effects involve complex signaling networks and intercellular communication. For example, serotonin (5-hydroxytryptamine, 5-HT) promotes migration and triggers epithelial-mesenchymal transition (EMT) in CRC cells by activating the 5-HT2B receptor (HTR2B) receptor, processes closely associated with tumor invasion and metastasis (11). Moreover, additional neurotransmitters such as dopamine and gamma-aminobutyric acid (GABA) contribute to modulating CRC biology by influencing both tumor growth and immune microenvironment remodeling (12,13). Therefore, thoroughly studying the roles of NRGs in CRC pathogenesis and progression is crucial to understanding tumor biology and developing novel therapies.

This research examines the expression patterns of gamma-aminobutyric acid type A receptor subunit delta (GABRD), a gene encoding a neurotransmitter receptor subunit. We also explore the gene’s associations with clinicopathological characteristics, prognostic significance, and immune cell infiltration in CRC. GABA, the primary inhibitory neurotransmitter in the central nervous system, is widely distributed in the brain and spinal cord. Existing literature indicates that GABA receptors are present in multiple cancer types and modulate neoplastic cell proliferation and migration (13,14). Specifically, the delta subunit of gamma-aminobutyric acid type A receptor (δ-GABA_AR), encoded by GABRD, plays a key role in tumor development and progression (15). Analyses across multiple cancer types (pan-cancer analyses) reveal dysregulated GABRD expression in various malignancies. Studies document markedly elevated GABRD levels in hepatocellular carcinoma specimens (16). In low-grade glioma cases, patients exhibiting high GABRD expression show more favorable clinical outcomes compared to those with reduced expression, who frequently experience an unfavorable prognosis (17). Furthermore, Liu and colleagues have verified upregulated GABRD messenger RNA (mRNA) expression in CRC tissues; this indicates its potential utility as a prognostic indicator for CRC patients (18). These differential expression patterns underscore GABRD’s potential therapeutic relevance. However, the specific biological roles and mechanisms of GABRD remain unclear, representing a significant knowledge gap in current research. Our investigation seeks to bridge this gap in understanding of GABRD’s specific biological roles and mechanisms by evaluating its dual potential as both a prognostic indicator and therapeutic target in CRC.

This research integrates bioinformatics analysis with experimental verification to study the function of NRGs, focusing specifically on GABRD in CRC. The analytical framework incorporates multi-omics datasets from The Cancer Genome Atlas (TCGA) and Gene Expression Omnibus (GEO) repositories and employs sophisticated computational approaches such as differential gene expression profiling, least absolute shrinkage and selection operator (LASSO) Cox regression modeling, protein-protein interaction (PPI) network construction and immune cell infiltration analysis. This study is expected to identify prognostic NRGs markers, clarify the functional pathways of GABRD, and evaluate its potential as both a diagnostic indicator and a therapeutic target in CRC. By focusing on GABRD, this work aims to develop novel diagnostic and therapeutic strategies to improve clinical outcomes in CRC management. We present this article in accordance with the TRIPOD and MDAR reporting checklists (available at https://tcr.amegroups.com/article/view/10.21037/tcr-2026-1-0016/rc).

Methods

Processing and normalization of gene expression data

Transcriptomic profiles for colon adenocarcinoma (COAD) and rectal adenocarcinoma (READ) were retrieved from TCGA database (http://cancergenome.nih.gov). The dataset included 644 cancerous tissues and 51 normal tissues. Clinical information for these samples was also obtained from the same source. We normalized raw sequencing counts and associated clinical metadata from the TCGA-COADREAD dataset using the limma R package. To ensure data quality, genes with expression values below 1 in more than 50% of samples were removed before downstream analysis. For validation, we acquired independent datasets (GSE28000, GSE41258, and GSE44076) containing genome-wide expression profiles. These datasets were obtained from GEO repository (https://www.ncbi.nlm.nih.gov/gds). The demographic characteristics of the GSE28000 and GSE44076 datasets are presented in Table S1 and Table S2, respectively. Furthermore, we curated a list of 1,113 NRGs (available online: https://cdn.amegroups.cn/static/public/tcr-2026-1-0016-1.xls) from the GeneCards database (https://www.genecards.org/).

Identification of differentially expressed genes (DEGs) in CRC

To study the molecular mechanisms of CRC and identify key biological pathways, we first normalized the TCGA-COADREAD dataset using the limma package. Subsequently, we used the DESeq2 package to perform differential expression analysis on the count data. Genes meeting the stringent criteria of |log₂FC| >2 and adjusted P value (P.adj) <0.05 were classified as DEGs. Specifically, genes with log₂FC >2 and P.adj <0.05 were classified as upregulated DEGs, whereas those with log₂FC <−2 and P.adj <0.05 were classified as downregulated DEGs. To identify NRGs involved in CRC, we filtered DEGs with |log₂FC| >2 and P.adj <0.05 from the TCGA-COADREAD dataset. To further narrow down candidate genes, we performed intersection analysis between these filtered DEGs and known NRGs, with the results graphically represented using a Venn diagram. The differential expression patterns were visualized through a volcano plot generated using the ggplot2 package in R.

Machine learning approach for identifying critical genes

We constructed our prognostic prediction model using the LASSO regression method. We carefully optimized the regularization parameter through rigorous 10-fold cross-validation. Through comprehensive LASSO Cox regression analyses, we initially identified 154 differentially expressed NRGs (DNRGs). After applying stringent selection criteria, we identified 9 core genes to build a prognostic risk model. This model shows strong potential to predict overall survival (OS) in CRC patients. It provides valuable clinical insights for outcome evaluation and the development of therapeutic strategies.

Construction and analysis of PPI networks with identification of key hub genes

We constructed the PPI network using DEGs from the STRING database (http://string-db.org). We visualized the network and performed subsequent analyses using Cytoscape software (https://www.cytoscape.org). To identify the most significant hub genes, we used the “Cytohubba” plugin in Cytoscape to select the top 20 genes with the highest maximal clique centrality (MCC) scores.

Survival analysis

A comprehensive survival analysis was conducted on the training cohort dataset. The Kaplan-Meier method and univariate Cox proportional hazards regression models were used, implemented via the “survival” R package (v4.4.1). We systematically conducted both univariate and multivariate Cox regression analyses to evaluate the association between clinical variables and patient survival outcomes. In the initial univariate screening, variables with marginal significance (P<0.1) were identified as potential prognostic factors and included in the multivariate Cox regression model for further validation. To better visualize and interpret the results, we generated forest plots of the multivariate regression outcomes using the ggplot2 package in R. The main goal of these analyses was to identify and validate prognostic biomarkers that provide meaningful insights into patient survival patterns.

Functional enrichment analysis

Gene Ontology (GO) analysis (19) is a widely used bioinformatics method for studying functional enrichment in large datasets. It covers three main domains: molecular functions (MF), cellular components (CC), and biological processes (BP). The Kyoto Encyclopedia of Genes and Genomes (KEGG) (20) is a comprehensive repository that includes genomic data, disease associations, pharmacological information, as well as metabolic pathways. In this study, we used the “clusterProfiler” package (v4.4.4) for systematic enrichment analysis of GO and KEGG pathways to explore the biological significance of our dataset (21). Furthermore, we performed Gene Set Enrichment Analysis (GSEA, v4.2.1) to explore gene function related to different GABRD expression levels, focusing on comparisons between high and low expression groups. We applied rigorous statistical thresholds in all analyses, considering only results with an adjusted P value <0.05 and a false discovery rate (FDR) <0.25 as biologically significant, ensuring the validity and reliability of our findings.

Correlation analysis between GABRD expression levels and immune cell infiltration

We analyzed 24 distinct immune cell types to evaluate immune cell infiltration. We quantified the relative abundance of these immune cells in COADREAD samples using single-sample Gene Set Enrichment Analysis (ssGSEA) with the GSVA R package (22). We further investigate potential associations between GABRD expression levels and immune cell infiltration patterns by performing Spearman’s rank correlation analysis. Additionally, we compared immune infiltration profiles between GABRD high- and low-expression groups using the Wilcoxon rank-sum test.

Correlation analysis between GABRD expression levels and clinicopathological characteristics

Comprehensive clinicopathological data, including OS, disease-specific survival (DSS), and progression-free interval (PFI), were retrieved from the TCGA-COADREAD cohort for CRC patients. Predictive nomograms incorporating both clinical parameters and GABRD expression levels were developed using the “rms” package (v6.3) in R to estimate OS probabilities in CRC cases. We systematically evaluated the potential correlations between GABRD expression levels and various clinicopathological features using logistic regression modeling.

Analysis of immune checkpoint expression profiles

We used the limma R package to identify differential expression patterns among multiple immune checkpoint molecules, and statistical associations between GABRD expression levels and immune checkpoint markers were assessed using Spearman’s rank correlation test.

Database resources for investigating the association of GABRD expression with tumor mutational burden (TMB), microsatellite instability (MSI), and mutant-allele tumor heterogeneity (MATH) in CRC

We examined potential correlations between GABRD expression levels and key tumor genomic features. These features included TMB, MSI, and MATH. For this analysis, we used the SangerBox web-based analytical platform (http://sangerbox.com/Tool). This publicly accessible bioinformatics tool provides comprehensive analysis capabilities for the TCGA dataset (23). We obtained somatic mutation data for analysis from TCGA database. We performed subsequent genomic alteration assessments using the R package “Maftools”, which systematically evaluates mutation patterns in genes with differential GABRD expression.

Nomogram validation

Using multivariate Cox regression analysis, we developed prognostic nomograms that incorporate independent risk factors to predict 1-, 3-, and 5-year survival probabilities. We evaluated the predictive accuracy of the nomograms using calibration curves and statistical analyses.

Drug sensitivity prediction

We used the pRRophetic package in R software to estimate the half-maximal inhibitory concentration (IC₅₀) of anticancer drugs. The IC₅₀ quantifies the potency of a compound in inhibiting specific cellular or molecular targets.

Docking prediction of GABRD with drug molecules

The amino acid sequence of the GABRD protein was obtained from the UniProt database. The Simplified Molecular Input Line Entry System (SMILES) representations for the pharmacological compounds were extracted from the ChEMBL repository. These compounds include afatinib (CHEMBL1173655), BI-2536 (CHEMBL513909), crizotinib (CHEMBL601719), cytarabine (CHEMBL803), dasatinib (CHEMBL1421), docetaxel (CHEMBL3545252), gefitinib (CHEMBL939), MK-2206 (CHEMBL1079175), obatoclax mesylate (CHEMBL2107358), paclitaxel (CHEMBL428647), and PD-0325901 (CHEMBL507361). Molecular interaction analyses were conducted using AlphaFold 3, a computational tool developed by DeepMind, with computational resources featuring an RTX A6000 GPU (NVIDIA Corporation, Santa Clara, CA, USA) via the OpenBayes platform (Hangzhou, China). The generated crystallographic information files (CIF) were subsequently processed and visualized using PyMOL (open-source edition), with particular emphasis on characterizing hydrogen bonding interactions between the protein target and ligand molecules.

Ethics statement

CRC specimens and matched adjacent non-tumor tissues were collected from the Pathology Department of the Affiliated Hospital of Jiaxing University in Zhejiang Province, China. A total of 139 patients who underwent surgical resection at this institution were randomly selected for tissue procurement. This study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments. The study was approved by the Institutional Review Board of the Affiliated Hospital of Jiaxing University (No. 2025-LP-795). Individual consent for this retrospective analysis was waived.

Immunohistochemical (IHC) staining

After dewaxing in aqueous solution, tissue sections underwent antigen retrieval by microwave irradiation for 18 minutes. After cooling to room temperature, the sections were treated with 3% hydrogen peroxide solution for 7 minutes to block endogenous peroxidase activity. Primary antibody incubation was performed overnight at 4 °C using GABRD antibody (No. abs148357, Absin Bioscience Inc., Shanghai, China) at a dilution of 1:50. Subsequently, sections were incubated with appropriate secondary antibodies for 1 hour at room temperature. Immunoreactivity was visualized as characteristic brown staining using 3,3'-diaminobenzidine (DAB), followed by a 1-minute hematoxylin nuclear counterstain. The processed sections were then permanently mounted using a neutral resin-based mounting medium. Finally, microscopic examination and digital image acquisition were performed. Two independent pathologists blinded to clinical data performed IHC evaluation using a standardized semi-quantitative scoring system. Staining intensity was graded as 0 (absent), 1 (mild), 2 (moderate), or 3 (intense). The percentage of positive tumor cells was categorized as 0 (0–10%), 1 (11–25%), 2 (26–50%), 3 (51–75%), or 4 (76–100%). The final immunoreactivity score was calculated by multiplying the intensity and percentage scores, producing a composite score from 0 to 12.

Cell culture and transfection methodology

We obtained the human CRC cell line HCT116 from the National Collection of Authenticated Cell Cultures (Shanghai, China). Cells were cultured in McCoy’s 5A medium (Thermo Fisher Scientific, Waltham, MA, USA) supplemented with 10% fetal bovine serum (FBS; Gibco, Carlsbad, CA, USA) and 1% penicillin-streptomycin solution. To investigate the functional role of GABRD, gene silencing was performed using small interfering RNA (siRNA). Three distinct siRNA sequences targeting GABRD (siGABRD-1, siGABRD-2, and siGABRD-3), along with a scrambled negative control siRNA (siNC), were synthesized by RiboBio (Guangzhou, China). The siRNA sequences are listed in Table 1. Transfection was performed using Lipofectamine™ 3000 (Thermo Fisher Scientific) according to the manufacturer’s instructions. After transfection, HCT116 cells were incubated for 48 hours to ensure efficient knockdown of GABRD expression.

Real-time quantitative polymerase chain reaction (RT-qPCR) analysis

Total RNA was isolated using TRIzol reagent (Thermo Fisher Scientific). Reverse transcription was performed using the SuperScript™ III First-Strand Synthesis SuperMix kit (Thermo Fisher Scientific) to generate complementary DNA (cDNA). Gene expression was quantified by RT-qPCR. The primer sequences, designed by Sangon Biotech (Shanghai, China), were as follows: GABRD forward primer 5'-CATGCTGGACCTGGAGAGCTA-3' and reverse primer 5'-CGGTAGCTGGTGATGGTGAACT-3'.

Western blot analysis

Protein extraction from HCT116 cells was performed using RIPA lysis buffer (Thermo Fisher Scientific) that was supplemented with a protease inhibitor cocktail (Thermo Fisher Scientific). Protein samples were separated by sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) and then transferred onto polyvinylidene difluoride (PVDF) membranes (Merck Millipore, Burlington, MA, USA). Membranes were blocked with 5% bovine serum albumin (BSA) at room temperature for 1 hour. Immunoblotting was performed using primary antibodies: rabbit polyclonal anti-GABRD (Proteintech, Wuhan, China; dilution 1:1,000) and rabbit monoclonal anti-β-actin (Abcam, Waltham, MA, USA; dilution 1:2,000). This was followed by incubation with appropriate secondary antibodies. Protein bands were visualized on X-ray film (Huadong Medicine, Hangzhou, China) and quantified using Image J imaging software.

Cell proliferation analysis

After centrifugation, harvested cells were resuspended in culture medium at a density of 1×10⁴ cells/mL. Subsequently, 100 µL aliquots of the cell suspension were seeded into individual wells of a 96-well plate. Cellular proliferation was assessed using the CCK-8 assay (Dojindo Laboratories, Kumamoto, Japan) at 24, 48, and 72 hours, according to the manufacturer’s protocol. Absorbance measurements at 450 nm (OD₄₅₀) were recorded and analyzed using GraphPad Prism software, to evaluate proliferative activity. All experiments were performed in triplicate to ensure reproducibility.

Colony formation assay

For the colony formation assay, transfected HCT116 cells were plated at 1,000 cells per well in a 6-well plate and cultured for one week. After incubation, we fixed the colonies with 4% paraformaldehyde (Sangon Biotech) for 30 minutes, followed by staining with 0.1% crystal violet (Sangon Biotech. Finally, the colonies were counted for further analysis.

Wound healing assay

Cells were seeded in 6-well plates for the wound healing assay. A 100 µL pipette tip was used to create uniform scratches across the cell monolayer, which were oriented perpendicular to a pre-marked reference line. After creating the scratch, images of the wound area were acquired using a light microscope (Olympus, Tokyo, Japan), and cell migration was evaluated at specified time points (0 and 48 h). Quantitative analysis of wound closure was performed using ImageJ software. The wound healing rate was calculated as follows: [(initial scratch width at 0 h – scratch width at 48 h) / initial scratch width at 0 h] × 100%. Each experiment was performed in triplicate to ensure reproducibility of the data.

Statistical analysis

Bioinformatics analyses were conducted using R software (version 4.2.1). Differences among multiple groups were evaluated using one-way analysis of variance (ANOVA). For comparisons between two groups, the Wilcoxon rank-sum test (also known as the Mann-Whitney U test) was applied. In all statistical tests, P<0.05 was considered statistically significant.

Results

Identification of differentially expressed overlapping NRGs

To investigate NRGs implicated in CRC, we performed a comparative analysis of DEGs between COADREAD tumor samples and adjacent normal tissues. The limma package was used to identify DEGs (|log₂FC| >2, adjusted P value <0.05), which yielded 2,667 upregulated and 1,809 downregulated genes. From the GeneCards database, we extracted 1,113 NRGs, of which 154 were identified as DNRGs in CRC. The volcano plot illustrates the global expression patterns of DEGs (Figure 1A), and the Venn diagram demonstrates the overlap between NRGs and DEGs (Figure 1B), which emphasizes their potential relevance in CRC pathogenesis.

Figure 1 Construction of a prognostic model based on DNRGs. (A) Differential gene analysis volcano plot of CRC tissues (group: tumor) and adjacent normal tissues (group: normal) of the TCGA-COADREAD dataset. (B) Venn diagram showed 154 neurotransmitter receptor-related DEGs in COADREAD. (C) LASSO regression analysis was conducted on the genes associated with the neurotransmitter receptor-related DEGs. (D) The forest plot illustrates the prognostic significance of genes associated with the neurotransmitter receptor-related DEGs. (E) The comparison of expression levels of neurotransmitter receptor-related DEGs between the normal group and the CRC group. *, P<0.05; ***, P<0.001. CI, confidence interval; COADREAD, colon adenocarcinoma and rectum adenocarcinoma; CRC, colorectal cancer; DEGs, differentially expressed genes; DNRGs, differentially expressed NRGs; HR, hazard ratio; LASSO, least absolute shrinkage and selection operator; TCGA, The Cancer Genome Atlas; TPM, transcripts per million.

For prognostic model construction, we implemented LASSO Cox regression analysis to select optimal candidate genes from the DNRGs. This approach generated a nine-gene signature with non-zero regression coefficients, including GABRD, MMP3, CDKN2A, GPX3, H19, CLU, ALPP, LRP2, and ITLN1 (Figure 1C). Subsequent univariate Cox regression revealed significant associations between the expression levels of GABRD, CDKN2A, GPX3, H19, CLU, ALPP, and LRP2 and OS in CRC patients (Figure 1D), with some genes showing positive and others negative correlations with prognosis. Notably, elevated expression of GABRD, CDKN2A, H19, ALPP, and LRP2 was observed in tumor tissues compared to normal controls, whereas reduced expression of GPX3, CLU, and ITLN1 was detected (Figure 1E). These distinct expression profiles not only reflect tumor-specific molecular features but also serve as valuable indicators for clinical prognosis assessment. Further investigations are warranted to elucidate the functional significance and mechanistic contributions of these genes in CRC progression and therapeutic response.

Expression correlation and prognosis of DNRGs

We further analyzed the functional relationships among these nine pivotal genes in CRC. Spearman’s correlation analysis revealed significant associations between GABRD and several genes: CDKN2A, GPX3, H19, CLU, LRP2, and ITLN1 (Figure 2A-2C). Kaplan-Meier survival analysis showed that higher expression of GABRD, CDKN2A, GPX3, CLU, ALPP, and LRP2 strongly correlated with poor clinical outcomes in CRC patients. Specifically, overexpression of these genes was associated with shorter survival and faster disease progression. Conversely, higher expression of MMP3 and ITLN1 was associated with better patient outcomes, indicating a protective role (Figure 2D). These differential expression patterns suggest distinct biological roles for these genes in CRC pathogenesis, with some potentially acting as oncogenic drivers, while others function as tumor suppressors.

Figure 2 The correlation and prognostic analysis of DNRGs. (A) The chord diagram illustrates the relationships among significant genes. (B) The heatmap illustrates the relationships among significant genes. (C) The grid chart illustrates the relationship among significant genes. (D) The Kaplan-Meier survival curves demonstrate notable variations in OS linked to nine essential genes. DNRGs, differentially expressed neurotransmitter receptor-associated genes; HR, hazard ratio; OS, overall survival.

High expression of GABRD in CRC

Receiver operating characteristic (ROC) analysis showed that GABRD had the highest diagnostic accuracy (Figure S1); therefore, we focused our subsequent investigations on GABRD. Our initial analysis revealed differential expression patterns of GABRD across multiple cancer types. Figure S2A shows that GABRD expression levels were elevated in 17 different malignancies. This increase was statistically significant (P<0.05), especially in COAD and READ. Comparative analysis of matched tumor-normal pairs has demonstrated consistent upregulation of GABRD expression in various cancers, particularly in COAD and READ (Figure S2B). We further examined the TCGA-COADREAD dataset, which includes both paired and unpaired samples. This analysis confirmed that GABRD expression is significantly higher in tumor tissues compared to normal tissues (P<0.05, Figure 3A,3B). To validate these findings, we integrated data from three independent GEO datasets (GSE28000, GSE41258, and GSE44076). These datasets collectively reinforced the observation of GABRD overexpression in CRC relative to normal tissues (Figure 3C-3E). Furthermore, single-cell analysis of two distinct sample sources from the Tumor Immune Single-cell Hub (TISCH) database showed GABRD expression in specific cell populations (Figure S3). ROC curve analysis showed that GABRD had strong diagnostic value, with an impressive area under the curve (AUC) of 0.981 (Figure 3F), indicating high accuracy in detecting CRC. Finally, IHC staining on clinical specimens was performed to further validate the expression level of GABRD in adenomatous polyps and CRC. Representative IHC staining patterns are illustrated in Figure 3G. Quantitative assessment revealed significantly elevated GABRD expression levels in adenomatous polyps and CRC specimens compared to matched adjacent normal tissues (P<0.001, Figure 3H,3I). These findings position GABRD as a promising diagnostic biomarker, potentially facilitating early disease identification and improved clinical management of CRC patients.

Figure 3 High expression level of GABRD in CRC tissues. GABRD expression in (A) unpaired and (B) paired samples from the TCGA-COADREAD database. Expression of GABRD in the (C) GSE28000, (D) GSE41258 and (E) GSE44076 databases. (F) The ROC curve area was 0.981 indicating GABRD was a biomarker of diagnostic of CRC. (G) The representative images of IHC staining of GABRD in adenomatous polyp tissues, CRC tissues and normal tissues. (H) The intensity IHC score of adenomatous polyp tissues, CRC tissues and normal tissues. (I) The intensity IHC score of 59 CRC tissues and normal tissues. ***, P<0.001. AUC, area under the curve; CI, confidence interval; COADREAD, colon adenocarcinoma and rectum adenocarcinoma; CRC, colorectal cancer; FPR, false positive rate; IHC, immunohistochemical; ROC, receiver operating characteristic; TCGA, The Cancer Genome Atlas; TPM, transcripts per million; TPR, true positive rate.

Correlation between GABRD expression levels and clinicopathological characteristics of CRC patients

We investigated the relationship between GABRD expression and clinical characteristics in CRC. For this purpose, we analyzed data from 644 COADREAD samples obtained from the TCGA database (Table 2). The study evaluated the correlation between GABRD expression levels and 15 clinicopathological variables. These variables included tumor staging (T, N, M stages), pathological stage, demographic factors (gender, age, race), anthropometric measures (body mass index), serum markers (CEA level), histopathological features (perineural invasion and lymphatic invasion), clinical outcomes—OS, DSS, PFI—and neoplasm type (Figure 4). Statistical analysis revealed significant associations between elevated GABRD expression and several clinical parameters: T stage (P<0.05, Figure 4A), N stage (P<0.001, Figure 4B), M stage (P<0.001, Figure 4C), pathological stage (P<0.001, Figure 4D), patient age (P<0.01, Figure 4F), perineural invasion (P<0.05, Figure 4I), lymphatic invasion (P<0.001, Figure 4J), OS events (P<0.001, Figure 4K), DSS (P<0.001, Figure 4L), PFI (P<0.001, Figure 4M), and neoplasm type (P<0.01, Figure 4N). These findings show that GABRD is overexpressed in COADREAD samples and strongly correlates with key clinical features, indicating its potential role in CRC progression and prognosis.

Figure 4 In the TCGA-COADREAD database, the relationship between the expression of GABRD and clinicopathological factors. (A-O) The relationship between the expression of GABRD and fifteen clinicopathological variables, including T stage, N stage, M stage, pathologic stage, gender, age, BMI, CEA level, perineural invasion, lymphatic invasion, OS event, DSS event, PFI event, neoplasm type and race. ns, P≥0.05; *, P<0.05; **, P<0.01; ***, P<0.001. BMI, body mass index; CEA, carcinoembryonic antigen; COADREAD, colon adenocarcinoma and rectum adenocarcinoma; DSS, disease-specific survival; OS, overall survival; PFI, progression-free interval; TCGA, The Cancer Genome Atlas; TPM, transcripts per million.

Prognostic value of GABRD in CRC patients

To investigate the prognostic significance of GABRD expression in CRC, we stratified TCGA patients into two groups based on median expression: high GABRD expression (upper 50th percentile) and low GABRD expression (lower 50th percentile). Subsequent survival analysis revealed that elevated GABRD expression was correlated with adverse clinical outcomes in COADREAD patients across multiple survival metrics, including OS (HR =1.91, P<0.001), DSS (HR =2.89, P<0.001), and PFI (HR =1.83, P<0.001) (Figure 5A-5C). Further subgroup analyses showed that high GABRD expression was consistently associated with worse prognosis in patients with COAD and its histological subtypes. This association remained significant across clinical subgroups, including patients with T3 stage, N2 lymph node metastasis, M0 status, age ≤65 or >65 years, and those who underwent R0 resection (Figure S4). For a more robust assessment of prognostic factors, we conducted comprehensive univariate and multivariate Cox regression analyses. These analyses identified GABRD expression, patient age, advanced disease stage (III/IV), distant metastasis (M1), and regional lymph node involvement (N1) as independent predictors of OS in COADREAD patients (Figure 5D,5E). Importantly, the high GABRD expression group consistently demonstrated inferior survival outcomes compared to their low-expression counterparts.

Figure 5 The impact of GABRD level on prognosis in CRC patients was evaluated utilizing Kaplan-Meier. (A) OS, (B) DSS and (C) PFI for CRC patients with high- vs. low-GABRD. Forest map of OS with CRC patients based on univariate (D) and multivariate (E) Cox analysis. CI, confidence interval; CRC, colorectal cancer; DSS, disease-specific survival; OS, overall survival; PFI, progression-free interval.

Functional enrichment analysis of DEGs related to GABRD

A total of 646 DEGs were identified between groups with high and low GABRD expression levels. Of these, 232 genes (35.9%) were upregulated, and 414 genes (64.1%) were downregulated, meeting the criteria of an adjusted P value <0.05 and |log₂FC| >1 (Figure 6A). We used the String online database to analyze these DEGs and construct a PPI network. The network visualization from String was imported into Cytoscape software to identify key modules and hub genes (Figure 6B). Then, using the cytoHubba plugin in Cytoscape, we identified the top 20 hub genes: COL1A1, TAGLN, ACTA2, MYH11, CNN1, ACTG2, MYL9, ACTC1, BGN, ELN, COL6A2, FBLN1, PRELP, THBS2, MFAP4, LMOD1, MGP, COMP, CILP, and THBS4 (Figure 6C). Focusing on these 20 hub genes, their correlations with GABRD were analyzed, and the correlation network is shown in Figure 6D. To further understand the functional relevance of these genes, we performed GO and KEGG pathway enrichment analyses on all DEGs. The GO analysis showed that BP were mainly enriched in nucleosome assembly, protein-DNA complex assembly, DNA replication-independent chromatin assembly, and epidermal cell differentiation. For CC, enrichment occurred in protein-DNA complexes and DNA packaging complexes. MF were enriched in receptor ligand activity, signaling receptor activator activity, and serine-type endopeptidase inhibitor activity. KEGG pathway analysis identified significant enrichment in neutrophil extracellular trap formation and neuroactive ligand-receptor interaction (Figure 6E and Table 3). Subsequently, GSEA performed on all genes demonstrated that Hedgehog signaling, Wnt/β-catenin signaling, angiogenesis, and TGF-β signaling pathways were significantly enriched in the group with high GABRD expression levels (Figure 6F).

Figure 6 DEGs related to GABRD, the PPI network of the DEGs and its functional enrichment analysis utilizing GSEA, GO and KEGG. (A) Blue and red dots indicated the vitally down- and up-regulated DEGs in the Volcano plot, respectively. (B) The PPI network of the DEGs was constructed using the Cytoscape software. (C) Identification of 20 hub DEGs. (D) The top 20 hub DEGs positively correlated with GABRD level. (E) KEGG, GO, and (F) GSEA analysis of DEGs. DEG, differentially expressed gene; FDR, false discovery rate; GO, Gene Ontology; GSEA, Gene Set Enrichment Analysis; KEGG, Kyoto Encyclopedia of Genes and Genomes; NES, normalized enrichment score; PPI, protein-protein interaction; TPM, transcripts per million.

Association between GABRD expression and immune cell infiltration

Our analysis revealed significant correlations between GABRD expression levels and various immune cell types (Figure 7). Positive associations were observed with natural killer (NK) cells, regulatory T cells (Treg), plasmacytoid dendritic cells (pDC), and immature dendritic cells (iDC); while negative correlations were found with Th2 cells, central memory T cells (Tcm), and T helper cells (all P<0.05; Figure 7B). Comparative evaluation of immune infiltration scores showed statistically significant differences between high and low GABRD expression groups across multiple immune cell types, including dendritic cells (DC), eosinophils, macrophages, mast cells, NK cells, and various T cell populations (P<0.05, Figure 7A). To better understand these relationships, we conducted detailed correlation analyses. We examined GABRD expression levels and immune infiltration scores for specific cell types, including NK cells, Treg, pDC, iDC, Th2 cells, and Tcm. These analyses yielded highly significant associations (P<0.001) and are illustrated in Figure 7D-7I. We visualized the comprehensive immune landscape using a correlation chord diagram that integrates GABRD expression with all immune cell infiltration scores (Figure 7C). Collectively, based on the observed correlations, these findings strongly suggest that elevated GABRD expression is associated with an immunosuppressive tumor microenvironment.

Figure 7 Correlation between GABRD level and immune cells infiltration in COADREAD. (A) The enrichment scores of 24 immune cells were compared between GABRD-high and GABRD-low groups, and the results were shown using box plots. (B) The lollipop plot represented the correlation between 24 immune cells and GABRD. (C) The chord diagram served to illustrate the correlation, where blue lines indicated negative correlation and red lines indicated positive correlation. (D-I) Correlation between GABRD expression and immune cells., including NK cells, Treg, pDC, iDC, Th2 cells and Tcm. ns, P≥0.05; *, P<0.05; **, P<0.01; ***, P<0.001. COADREAD, colon adenocarcinoma and rectum adenocarcinoma; iDC, immature dendritic cell; MDSC, myeloid-derived suppressor cell; NK, natural killer; pDC, plasmacytoid dendritic cell; Tcm, central memory T cell; TFH, follicular helper T cell; TPM, transcripts per million; Treg, regulatory T cell.

Construction and validation of a nomogram based on the independent factors

We systematically analyzed multiple independent variables to construct an integrated prognostic nomogram. This approach was used to evaluate the clinical outcomes of COADREAD patients. Our findings demonstrated a significant inverse correlation between the cumulative risk score and patient survival; specifically, elevated total scores corresponded to unfavorable clinical outcomes (Figure 8A). Calibration curve analysis rigorously evaluated the nomogram’s predictive accuracy and confirmed its robust performance (Figure 8B). Time-dependent ROC curve analysis showed that GABRD had discrimination ability with AUC values exceeding 0.5 for 1-, 3-, and 5-year survival predictions (Figure 8C). Importantly, our study identified GABRD as a significant independent prognostic determinant in COADREAD patients. GABRD is a genetic marker linked to COADREAD predisposition and influences disease progression.

Figure 8 Construction of prognostic nomogram model. (A) A nomogram for predicting 1-, 3- and 5-year OS survival of COADREAD. (B) The calibration curve for the 1-, 3- and 5-year OS survival nomogram. (C) Time-dependent ROC curves. AUC, area under the curve; COADREAD, colon adenocarcinoma and rectum adenocarcinoma; FPR, false positive rate; OS, overall survival; ROC, receiver operating characteristic; TPR, true positive rate.

Association between GABRD and immune checkpoint molecules in CRC

Analysis revealed significant downregulation of TNFSF14, LAIR1, NRP1, CD27, and HHLA2 immune checkpoint components in CRC specimens (Figure 9A). In contrast, the immune checkpoint markers TNFRSF4, TNFRSF25, CD44, and CD276 were significantly upregulated in tumor tissues (Figure 9A). We further examined the correlations between the transcript levels of these checkpoint molecules and the expression of GABRD (Figure 9B,9C). The results showed that GABRD expression showed a significant positive correlation with CD276, NRP1, TNFRSF25, and TNFRSF4 (Figure 9D-9G), whereas it was significantly negatively correlated with HHLA2 and CD44 (Figure 9H,9I). These observations suggest that GABRD participates in the regulatory network governing specific immune checkpoint pathways in CRC pathogenesis.

Figure 9 Correlation of GABRD expression and immune checkpoints in COADREAD. (A) Expression of immune checkpoints in COADREAD. (B,C) Heatmap of co-expression and correlation between GABRD and immune checkpoints. (D-I) Expression correlation between CD276, NRP1, TNFRSF25, TNFRSF4, HHLA2, CD44 and GABRD. ***, P<0.001. COADREAD, colon adenocarcinoma and rectum adenocarcinoma; TPM, transcripts per million.

Association of GABRD expression with TMB, MSI, and MATH

To elucidate the immunological characteristics associated with differential GABRD expression patterns, we conducted comprehensive genomic mutation profiling. Analysis identified 15 genes with significant differences in mutation frequencies between high and low GABRD expression groups (Figure 10A), with missense mutations being the main variant type. Our investigation demonstrated significant inverse correlations between GABRD expression levels and both TMB (R=−0.136, P=0.008, Figure 10B) and MSI (R=−0.211, P<0.001, Figure 10C) in COADREAD. Conversely, a positive correlation was observed between GABRD expression and MATH (R=0.139, P=0.007, Figure 10D), a measure of intratumor genetic heterogeneity. Further analysis confirmed that TMB levels were significantly elevated in low GABRD expression groups compared to high expression groups across both COADREAD and COAD datasets (P<0.05, Figure 10E-10G). These findings suggest that GABRD expression status could serve as a biomarker for tumor immunogenicity in CRC.

Figure 10 Relationship of GABRD expression with TMB, MSI and MATH. (A) Significantly mutated genes in the mutated CRC samples of the high GABRD expression and the low GABRD expression groups, respectively. (B) Relationships between GABRD expression and TMB. (C) Relationships between GABRD expression and MSI. (D) Relationships between GABRD expression and MATH. (E-G) The boxplot shows the differences in TMB between high- and low-GABRD expression groups. ns, P≥0.05; *, P<0.05; **, P<0.01. COAD, colon adenocarcinoma; CRC, colorectal cancer; MATH, mutant-allele tumor heterogeneity; MSI, microsatellite instability; READ, rectum adenocarcinoma; TCGA, The Cancer Genome Atlas; TMB, tumor mutational burden.

IC₅₀ score

When evaluating patient responsiveness to targeted pharmacological interventions, IC₅₀ serves as a critical parameter. Using data provided by the Genomics of Drug Sensitivity in Cancer (GDSC), we analyzed potential variations in IC₅₀ values of chemotherapeutic agents across different GABRD expression subgroups. Our investigation revealed elevated IC₅₀ values for cytarabine (Drug ID: 1006), docetaxel (Drug ID: 1007), gefitinib (Drug ID: 1010), afatinib (Drug ID: 1032), PD-0325901 (Drug ID: 1060), obatoclax mesylate (Drug ID: 1068), paclitaxel (Drug ID: 1080), crizotinib (Drug ID: 1083), and BI-2536 (Drug ID: 1086) in the GABRD high-expression group (Figure 11A-11I). Significantly reduced IC₅₀ values were observed for ZM447439 (Drug ID: 1050), MK.2206 (Drug ID: 1053), and dasatinib (Drug ID: 1079) in the GABRD high-expression group compared to the low-expression group (Figure 11J-11L).

Figure 11 A breakdown of the IC₅₀ values for 12 targeted chemotherapeutic and targeted agents across the various GABRD expression groups. IC₅₀, half-maximal inhibitory concentration.

Molecular docking validation

AlphaFold 3 successfully predicted the binding interfaces between multiple pharmaceutical compounds and the GABRD protein. The analysis revealed hydrogen bond formation between the β-sheet domain of GABRD and several drugs. These drugs include afatinib (Figure 12A), ZM-447439 (Figure 12B), paclitaxel (Figure 12C), gefitinib (Figure 12D), cytarabine (Figure 12E), crizotinib (Figure 12F), and BI-2536 (Figure 12G). A conserved interaction pattern was observed: afatinib, crizotinib, cytarabine, gefitinib, and ZM-447439 all contact glutamate residue 71 (Glu71) of GABRD. Furthermore, specific drug-residue interactions were identified: afatinib formed an additional hydrogen bond with serine 155 (Ser155), BI-2536 interacted via serine 197 (Ser197), and paclitaxel formed a hydrogen bond with arginine 145 (Arg145). These structural observations suggest that these compounds may bind to distinct sites and follow different mechanistic pathways when interacting with GABRD.

Figure 12 Molecular docking interfaces between various drug-like molecules and GABRD (UniProt ID: O14764). (A) Afatinib, (B) ZM-447439, (C) paclitaxel, (D) gefitinib, (E) cytarabine, (F) crizotinib, and (G) BI-2536 with GABRD, as predicted by AlphaFold 3 and rendered with PyMOL (open-source version). The protein models are rendered in cartoon representation, and the ligand models are shown in sticks representation. Hydrogen bonds are depicted as yellow dashed lines. Protein residues and ligand atoms participating in these hydrogen bond interactions are highlighted in red.

Upregulation of GABRD enhanced the proliferation and migration capabilities of CRC cells

To investigate the functional significance of GABRD in CRC pathogenesis, we performed GABRD knockdown in HCT116 cells. The silencing efficiency was confirmed by Western blot analysis and RT-qPCR (Figure 13A,13B). Subsequent CCK-8 viability assays demonstrated a marked decrease in cellular viability following GABRD depletion (Figure 13C). Colony formation experiments revealed a significant reduction in proliferation after GABRD suppression (Figure 13D,13E). Wound healing assays, which evaluated migratory potential, showed significant inhibition of HCT116 cell motility after GABRD knockdown (Figure 13F,13G). These collective findings, demonstrating suppressed CRC cell proliferation and migration following GABRD depletion, strongly indicate that GABRD functions as a positive regulator in CRC progression.

Figure 13 Knockdown of GABRD inhibited CRC progression. (A,B) Down expressed GABRD in HCT116 cells was confirmed by Western blotting (A) and RT-qPCR (B). (C) CCK-8 assay was used to detect the proliferation of HCT116 cells. (D,E) Colony formation assay was used to test the proliferative capacity of HCT116 cells (0.1% crystal violet). (F,G) Wound-healing assay was used to test the migratory capacity of HCT116 cells (200×). *, P<0.05; **, P<0.01; ***, P<0.001; ****, P<0.0001. CCK-8, Cell Counting Kit-8; CRC, colorectal cancer; OD, optical density; RT-qPCR, reverse transcription quantitative polymerase chain reaction.

Discussion

CRC is a major global health burden and one of the most common cancers causing high mortality worldwide. The development of CRC results from complex interactions among genetic factors, environmental exposures, and behavioral risks that promote tumor initiation and progression (24,25). Diagnostic techniques and therapeutic options have improved significantly, including surgical resection, cytotoxic agents, and targeted drugs. However, early detection remains limited, often resulting in advanced-stage diagnoses and poor clinical outcomes (26). The molecular diversity of CRC poses therapeutic challenges, emphasizing the urgent need for novel molecular markers to improve early diagnosis and guide precision medicine. This study aims to address these challenges by investigating the role of NRGs, especially GABRD, in CRC pathogenesis, ultimately advancing precise diagnostic methods and personalized treatments.

In this investigation, we aimed to clarify the role of NRGs, especially GABRD, in CRC development. To achieve this, we combined extensive bioinformatics approaches with experimental verification. We used multi-omics datasets from TCGA and GEO repositories and applied sophisticated computational methods, such as differential expression profiling and LASSO Cox regression, to establish a 9-gene prognostic model comprising GABRD, MMP3, CDKN2A, GPX3, H19, CLU, ALPP, LRP2, and ITLN1. Our results revealed a significant correlation between the transcriptional activity of these genes and OS in CRC patients, highlighting the value of NRGs as prognostic markers. Importantly, elevated expression of multiple genes in the signature—particularly GABRD, CDKN2A, ALPP, and LRP2—correlated with unfavorable prognosis, reinforcing the concept that these genes contribute critically to tumor development and patient survival. Building on this, our data identify GABRD as a key player, showing significant overexpression in CRC specimens and linkage to poor clinical outcomes. The novel analytical framework used in this study highlights the potential of GABRD as both a diagnostic biomarker and therapeutic target in CRC. Additionally, it opens avenues for deeper investigation into the molecular mechanisms of GABRD and its relevance to personalized treatment strategies.

Functional enrichment analysis of DEGs associated with GABRD demonstrated significant enrichment in several key signaling pathways, including Hedgehog, Wnt/β-catenin, and TGF-β signaling cascades. The Hedgehog pathway, essential for cellular differentiation, proliferation, and tissue morphogenesis, is dysregulated in various malignancies, especially CRC (27). Similarly, TGF-β signaling participates in both tumor-suppressive and tumor-promoting activities, reflecting its complex involvement in oncogenesis (28). The association of GABRD with these pathways suggests its potential as a molecular target, and pharmacological modulation of GABRD could yield novel therapeutic approaches for CRC management. Future studies should focus on experimentally validating these pathway associations and assessing the clinical relevance of targeting GABRD within these signaling pathways. Additionally, our investigation showed GABRD’s significant involvement in angiogenesis and extracellular matrix (ECM) remodeling, supported by PPI network analyses that identified key nodes like COL1A1 and TAGLN. Angiogenesis represents a fundamental BP supporting neoplastic expansion and metastatic dissemination (29), while ECM components critically influence tumor microenvironment dynamics and malignant cell behavior (30,31). These findings imply that GABRD may function not merely as a diagnostic marker, but as an active contributor to CRC pathogenesis. Delineating the mechanistic relationships between GABRD and these BPs could reveal innovative treatment modalities and deepen our comprehension of CRC pathogenesis. Notably, our pathway analysis identified significant enrichment in neuroactive ligand-receptor interactions, suggesting neuromodulatory effects on CRC progression. Emerging evidence implicates neurotransmitter receptors, including GABRD, in tumor microenvironment regulation and malignant cell modulation (32). The intersection between neurotransmitter signaling, immune cell recruitment, and tumor evolution represents a promising research frontier. Clarifying GABRD’s role in shaping the tumor-immune interface may facilitate the development of novel immunotherapeutic interventions for CRC. Future studies should elucidate the precise mechanisms through which GABRD influences these pathways and assess its suitability as a therapeutic target in combinatorial approaches designed to augment anti-tumor immune responses.

The immune system plays a crucial role in the development and progression of CRC. Our research identifies significant correlations between GABRD expression levels and distinct immune cell populations. Interestingly, higher GABRD expression was positively associated with NK cells, Tregs, pDCs, and iDCs. In contrast, it showed negative correlations with Th2 cells, Tcms, and T helper cells. These findings suggest that GABRD may foster an immunosuppressive environment, facilitating tumor immune evasion. Existing literature has established the critical role of NK cells in tumor surveillance and Tregs in maintaining immune tolerance within tumors (33). Furthermore, the identified relationship between GABRD and immune checkpoint proteins like TNFSF14 and CD276 supports its possible involvement in immune regulation, indicating that GABRD could be a key player in CRC immune checkpoint pathways. Additionally, tumors with high versus low GABRD expression show substantial differences in immune cell infiltration across 14 categories. This highlights GABRD’s potential influence on the immune profile of CRC. This observation corroborates growing evidence that tumor-associated immune cells significantly influence treatment efficacy and clinical outcomes (34). In conclusion, our investigation demonstrates the complex interplay between GABRD expression and immune cell infiltration in CRC, suggesting GABRD’s dual potential as both a prognostic indicator and a possible target for immune-based therapies. Additional studies are needed to confirm these results. Furthermore, treatment approaches that exploit GABRD’s function in modifying the immune microenvironment of CRC should be developed.

To assess the prognostic significance of GABRD, we analyzed its relationship with clinicopathological features using data from TCGA database. Our findings demonstrated significant correlations between GABRD expression and several key tumor characteristics, including TNM staging, pathological stage, perineural invasion, and lymphatic invasion. Elevated GABRD expression was particularly associated with increased risks of lymphatic metastasis and distant metastasis, suggesting poorer clinical outcomes for patients with COADREAD. Survival analysis further confirmed that individuals exhibiting high GABRD expression experienced significantly reduced OS, PFI, and DSS. These observations remained consistent across various subgroup analyses. Additionally, we developed a nomogram as a clinical prognostic prediction model based on multivariate Cox regression results. We subsequently validated its predictive accuracy. The calibration curves showed excellent agreement between predicted and observed OS rates at 1, 3, and 5 years. These results indicate that the nomogram developed in our study represents a novel and clinically valuable tool for prognostic assessment. In summary, there is a significant correlation between GABRD expression level and key clinical prognostic indicators of patients, indicating that GABRD is a potential prognostic biomarker that can supplement the existing prognostic evaluation system and provide a more accurate stratification basis for the prognostic assessment of CRC patients. However, prior to its clinical application, further validation in larger-scale independent cohorts is still required to confirm its stability and reliability as a prognostic biomarker.

The discovery of molecular markers for CRC plays a pivotal role in improving early detection capabilities and developing individualized therapeutic approaches. Our research identifies GABRD as a promising biomarker. Its elevated expression is significantly associated with advanced disease progression and unfavorable clinical outcomes. These observations suggest GABRD has potential as a diagnostic indicator and a target for therapeutic intervention. Through ROC curve assessment, we established that GABRD has exceptional discriminative capacity, with an area under the curve (AUC) of 0.981 for distinguishing CRC cases from healthy controls. This outstanding diagnostic performance suggests that incorporating GABRD evaluation could significantly improve current CRC screening, which often misses early, more treatable malignancies. Building on its diagnostic utility, we further investigated GABRD’s role in chemotherapy responsiveness by analyzing IC₅₀ measurements. This analysis underscores GABRD’s potential utility in chemotherapy regimen selection. The correlation between GABRD expression patterns and drug sensitivity profiles supports the implementation of precision medicine strategies in CRC management, potentially increasing treatment success rates while reducing unnecessary toxicity. Molecular docking simulations revealed detailed structural interactions between GABRD and various anticancer compounds, providing valuable information to guide rational drug design. These computational findings lay the groundwork for future studies focused on developing optimized treatment protocols for CRC patients with elevated GABRD expression, potentially leading to more effective and targeted therapeutic approaches.

Several limitations inherent in this study must be considered when interpreting the results and their clinical relevance. Primarily, our analysis depends heavily on data retrieved from the TCGA and GEO repositories. Consequently, the demographic profiles of these cohorts may not adequately reflect global population diversity, as they are skewed toward specific ethnicities and geographic regions while underrepresenting other groups. Such demographic constraints could compromise the external validity of our findings, given that GABRD expression patterns and prognostic significance might differ across various populations. Although we validated our bioinformatic results using 139 clinical samples, this validation set is limited in size and lacks demographic heterogeneity, thereby restricting the broader applicability of our conclusions. Secondly, this research examines GABRD as a single molecular marker without evaluating its potential synergy with established CRC biomarkers. Relying on single markers often yields suboptimal diagnostic accuracy and prognostic precision; conversely, integrating multiple biomarkers could offer a more robust framework for CRC diagnosis, prognosis, and clinical decision-making. Thirdly, our study did not include patient-reported outcomes (PROs) and quality of life (QoL) metrics. As GABRD-directed therapeutic strategies move toward clinical implementation, assessing their influence on patient well-being—encompassing physical comfort, psychological status, and overall QoL—is essential for gauging clinical utility and patient acceptance. Prospectively, the assessment of GABRD expression holds significant promise for refining the differential diagnosis and risk stratification of individuals with CRC, while also serving as a viable surveillance marker. When integrated with standard imaging modalities and CEA assays, this approach could effectively monitor disease progression or detect early recurrence. Furthermore, compared to the established gold standard of colonoscopy and emerging non-invasive modalities such as ctDNA-based assays, the cost-effectiveness and practical viability of GABRD detection remain insufficiently characterized for implementation in large-scale clinical settings and population-based screening programs. Finally, despite confirming the prospective value of GABRD for prognosis and early detection, the absence of standardized assay protocols and insufficient validation across diverse, multi-center independent cohorts currently hinders its immediate clinical translation. Future investigations will address these gaps by expanding sample sizes to include demographically and geographically diverse cohorts, examining the combinatorial effects of GABRD alongside other biomarkers, incorporating PROs and QoL evaluations, and developing standardized detection methodologies to strengthen the generalizability, reliability, and clinical translatability of our work.

In this study, we used siRNA to silence GABRD expression in HCT116 CRC cells. Our results showed that GABRD knockdown significantly reduced cell proliferation and metastasis potential. GSEA revealed that gene sets associated with EMT, TGF-β signaling, and angiogenesis were generally upregulated in the GABRD high-expression group and downregulated in the GABRD low-expression group. This suggests that GABRD may influence the progression of CRC by regulating EMT and the TGF-β signaling pathway, which represents a direction for our future research. These findings strongly suggest that GABRD is crucial for CRC tumorigenesis, progression, and cancer cell dissemination. Importantly, our study identifies GABRD as a potential target for new CRC therapies.

Conclusions

In summary, our research demonstrates the critical involvement of GABRD in CRC, establishing it as both a valuable prognostic indicator and a potential therapeutic target. There is a strong association between elevated GABRD expression and unfavorable clinical outcomes. Moreover, its link with specific immune cell infiltration characteristics suggests that GABRD may modulate tumor microenvironment interactions and facilitate immune escape processes. Furthermore, the identification of distinct drug responsiveness patterns based on GABRD expression provides new opportunities for personalized therapeutic interventions. Nevertheless, limitations such as the need for mechanistic validation and sample variability highlight the necessity for further studies. Future efforts should focus on translating these findings into clinical practice by utilizing independent validation datasets and conducting functional experiments to confirm GABRD’s mechanistic role in CRC progression and treatment response. Collectively, these findings provide a foundation that supports the development of novel strategies to improve CRC treatment and prognosis.

Acknowledgments

We sincerely thank Junshuang Guo and Jingwen Lin for their valuable contributions, continued support, encouragement and guidance. During the preparation of this manuscript, the authors used ChatGPT-4o in order to refine the structure of the text and enhance its academic clarity while strictly preserving the original content without AI-generated writing. The authors have reviewed and edited the output and take full responsibility for the content of this publication.

Footnote

Reporting Checklist: The authors have completed the TRIPOD and MDAR reporting checklists. Available at https://tcr.amegroups.com/article/view/10.21037/tcr-2026-1-0016/rc

Data Sharing Statement: Available at https://tcr.amegroups.com/article/view/10.21037/tcr-2026-1-0016/dss

Peer Review File: Available at https://tcr.amegroups.com/article/view/10.21037/tcr-2026-1-0016/prf

Funding: This study was supported by the Key Medical Discipline of Jiaxing City (Pathology, 2023-zc-017), the Zhejiang Medical and Health Science and Technology Program (2025KY1584), and the “Morning Star” Talent Cultivation Program of Jiaxing First Hospital (2024-QMX-033).

Conflicts of Interest: All authors have completed the ICMJE uniform disclosure form (available at https://tcr.amegroups.com/article/view/10.21037/tcr-2026-1-0016/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 was conducted in accordance with the Declaration of Helsinki and its subsequent amendments. The study was approved by the Institutional Review Board of the Affiliated Hospital of Jiaxing University (No. 2025-LP-795). Individual consent for this retrospective analysis was waived.

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. McCabe MA, Mauro AJ, Schoen RE. Novel colorectal cancer screening methods - opportunities and challenges. Nat Rev Clin Oncol 2025;22:581-91. [Crossref] [PubMed]
2. Bray F, Laversanne M, Sung H, et al. Global cancer statistics 2022: GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185 countries. CA Cancer J Clin 2024;74:229-63. [Crossref] [PubMed]
3. Cromo M, Randazzo A, Miller DP Jr, et al. Facilitators of and Barriers to Implementation of a Tablet-Based Digital Health Program for Colorectal Cancer Screening in Primary Care: Qualitative Pragmatic Implementation Study. JMIR Mhealth Uhealth 2026;14:e74206. [Crossref] [PubMed]
4. Housini M, Dariya B, Ahmed N, et al. Colorectal cancer: Genetic alterations, novel biomarkers, current therapeutic strategies and clinical trials. Gene 2024;892:147857. [Crossref] [PubMed]
5. Yanova M, Maltseva D, Tonevitsky A. Sidedness matters: single-cell perspectives on left- and right-sided colorectal cancer. Front Cell Dev Biol 2025;13:1720996. [Crossref] [PubMed]
6. Zhang L, Jing W, Li Y, et al. Neurotransmitters: Key regulators of the tumor immune microenvironment. Semin Immunol 2026;81:102015. [Crossref] [PubMed]
7. Jiang H, Zhang G, Sun W, et al. The role of neurotransmitter signaling in colorectal cancer progression: Opportunities for drug repurposing. Cytokine Growth Factor Rev 2026;88:88-104. [Crossref] [PubMed]
8. Jobling P, Pundavela J, Oliveira SM, et al. Nerve-Cancer Cell Cross-talk: A Novel Promoter of Tumor Progression. Cancer Res 2015;75:1777-81. [Crossref] [PubMed]
9. Wang X, Yan Y, Li L, et al. The influence of neuro-tumor interactions on tumorigenesis and therapeutic response. Exp Hematol Oncol 2026;15:18. [Crossref] [PubMed]
10. Zhang Y, Wang N, Zhang L, et al. Serotonin (5-Hydroxytryptamine): Metabolism, Signaling, Biological Functions, Diseases, and Emerging Therapeutic Opportunities. MedComm (2020) 2025;6:e70383.
11. Li T, Wei L, Zhang X, et al. Serotonin Receptor HTR2B Facilitates Colorectal Cancer Metastasis via CREB1-ZEB1 Axis-Mediated Epithelial-Mesenchymal Transition. Mol Cancer Res 2024;22:538-54. [Crossref] [PubMed]
12. Huang Q, Hu B, Zhang P, et al. Neuroscience of cancer: unraveling the complex interplay between the nervous system, the tumor and the tumor immune microenvironment. Mol Cancer 2025;24:24. [Crossref] [PubMed]
13. Tang D, Orlandi P, Li Q, et al. GABAergic signaling in colorectal cancer: Mechanistic insights, tumor microenvironment crosstalk, and therapeutic opportunities. Biochim Biophys Acta Rev Cancer 2025;1880:189414. [Crossref] [PubMed]
14. Buffon VA, Filho JMR, Malafaia O, et al. GABAergic Influences on Medulloblastoma. Brain Sci 2025;15:746. [Crossref] [PubMed]
15. Wu M, Kim KY, Park WC, et al. Enhanced expression of GABRD predicts poor prognosis in patients with colon adenocarcinoma. Transl Oncol 2020;13:100861. [Crossref] [PubMed]
16. Sarathi A, Palaniappan A. Novel significant stage-specific differentially expressed genes in hepatocellular carcinoma. BMC Cancer 2019;19:663. [Crossref] [PubMed]
17. Zhang H, Zhang L, Tang Y, et al. Systemic screening identifies GABRD, a subunit gene of GABAA receptor as a prognostic marker in adult IDH wild-type diffuse low-grade glioma. Biomed Pharmacother 2019;118:109215. [Crossref] [PubMed]
18. Liu T, Fang Y. Research for Expression and Prognostic Value of GABRD in Colon Cancer and Coexpressed Gene Network Construction Based on Data Mining. Comput Math Methods Med 2021;2021:5544182. [Crossref] [PubMed]
19. Gang X, Sun Y, Li F, et al. Identification of key genes associated with rheumatoid arthritis with bioinformatics approach. Medicine (Baltimore) 2017;96:e7673. [Crossref] [PubMed]
20. Droste C, De Las Rivas J. Path2enet: generation of human pathway-derived networks in an expression specific context. BMC Genomics 2016;17:731. [Crossref] [PubMed]
21. 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]
22. Hänzelmann S, Castelo R, Guinney J. GSVA: gene set variation analysis for microarray and RNA-seq data. BMC Bioinformatics 2013;14:7. [Crossref] [PubMed]
23. Wang J, Song Z, Ren L, et al. Pan-cancer analysis supports MAPK12 as a potential prognostic and immunotherapeutic target in multiple tumor types, including in THCA. Oncol Lett 2022;24:445. [Crossref] [PubMed]
24. Davoody S, Tayebi Z, Sharif-Zak M, et al. Assessing the role of Escherichia coli and Klebsiella pneumoniae in colorectal cancer oncogene expression: insights from microbial colonization phenotypes. Mol Biol Rep 2025;52:828. [Crossref] [PubMed]
25. Modeel S, Siwach S, Dolkar P, et al. Emerging risk factors and the role of gut microbiota in immunomodulation and therapeutic implications in colorectal cancer. Cancer Pathog Ther 2026;4:14-30. [Crossref] [PubMed]
26. İlhan B, Kuraş S, Kılıç B, et al. Exploratory Analysis of Circulating Serum miR-197-3p, miR-1236, and miR-1271 Expression in Early Breast Cancer. Int J Mol Sci 2025;26:8944. [Crossref] [PubMed]
27. Liu Y, Zhong Y, Sang Y, et al. Molecular characteristics and cancer immunity of LRP1B and its relationship with the Hedgehog signaling pathway in colorectal cancer. Front Immunol 2025;16:1567102. [Crossref] [PubMed]
28. Aftabi S, Barzegar Behrooz A, Cordani M, et al. Therapeutic targeting of TGF-β in lung cancer. FEBS J 2025;292:1520-57. [Crossref] [PubMed]
29. Chen Z, Liang Z, Cai T, et al. ANGPTL3 promotes colorectal carcinoma progression and metastasis through regulating COL1A2 transcription via interacting with integrin αVβ3. Cancer Lett 2026;638:218150. [Crossref] [PubMed]
30. Karamanos NK, Piperigkou Z, Gourdoupi C, et al. Extracellular matrix matters: matrix-based bioscaffolds in advancing translational cancer research and targeted therapy. Am J Physiol Cell Physiol 2025;328:C1957-63. [Crossref] [PubMed]
31. Nonnast E, Mira E, Mañes S. The role of laminins in cancer pathobiology: a comprehensive review. J Transl Med 2025;23:83. [Crossref] [PubMed]
32. Li RQ, Zhao XH, Zhu Q, et al. Exploring neurotransmitters and their receptors for breast cancer prevention and treatment. Theranostics 2023;13:1109-29. [Crossref] [PubMed]
33. Ahn M, Mun JG, Han Y, et al. Cancer cell-derived extracellular vesicles: a potential target for overcoming tumor immunotherapy resistance and immune evasion strategies. Front Immunol 2025;16:1601266. [Crossref] [PubMed]
34. Franzè MS, Saffioti F, Mavroeidis VK. Interactions between tumor microenvironment and resistance to transarterial and systemic treatments for HCC. Cancer Drug Resist 2025;8:33. [Crossref] [PubMed]