Prognostic chromatin remodeling signature stratifies survival outcomes in lung adenocarcinoma patients

Introduction

Lung adenocarcinoma (LUAD), constituting over 80% of non-small cell lung cancer cases, persists as a major global health challenge with a dismal 5-year survival rate below 20% for advanced stages (1,2). While advancements in targeted therapies and immune checkpoint inhibitors have improved clinical management, therapeutic resistance driven by genomic instability, epigenetic dysregulation, and tumor microenvironment reprogramming continues to hinder treatment efficacy (3,4). Emerging evidence highlights chromatin remodeling, an epigenetic mechanism regulating transcriptional accessibility through nucleosome repositioning, as a critical player in oncogenesis and therapy resistance, though its prognostic implications in LUAD remain poorly characterized (5,6).

Chromatin remodeling refers to a dynamic process that regulates gene expression by altering nucleosome positioning and chromatin accessibility, thereby influencing transcription factor binding to promoter regions (7). The core machinery of chromatin remodeling primarily involves four major protein complexes: SWI/SNF complex, ISWI complex, INO80 complex, and CHD complex, which participate in DNA damage repair, transcriptional activation or silencing, genomic stability, and cell differentiation (8-11). Increasing evidence suggests that genetic alterations in chromatin remodeling complexes can promote tumorigenesis by activating oncogenes or inhibiting tumor suppressor genes. For example, mutations in the core subunit of the SWI/SNF complex, ARID1A, have been reported to drive tumor proliferation, invasion, and drug resistance in LUAD (12); overexpression of CHD4, a key component of the CHD complex, can inhibit DNA damage repair and promote lung cancer cell proliferation and chemoresistance (13); the INO80 complex can enhance cancer cell proliferation and invasion by increasing chromatin accessibility and promoting oncogene expression (14). These findings suggest that dysregulated chromatin remodeling may serve as a key driver in LUAD progression. While several molecular biomarkers and prediction models exist for LUAD, their predictive accuracy and clinical applicability are often limited. This highlights an unmet need for novel, robust prognostic biomarkers. Chromatin remodeling genes (CRGs) represent a compelling new class of candidates, as they regulate broad transcriptional programs underlying tumor heterogeneity. However, for a prediction model to be clinically viable, it cannot rely on molecular data alone.

Moreover, emerging studies indicate that chromatin remodeling plays an essential role in regulating the tumor immune microenvironment (TIME) (15). The TIME refers to a complex ecosystem consisting of immune cells, stromal cells, endothelial cells, and tumor-associated fibroblasts, which plays a pivotal role in cancer progression, immune evasion, and therapeutic response (16). Tumor cells can manipulate chromatin remodeling to alter the expression of immune checkpoint molecules (such as PD-L1) and reprogram the functionality of tumor-infiltrating immune cells (such as CD8⁺ T cells and Treg cells), thereby promoting immune evasion (17). For instance, ARID1A deficiency has been shown to enhance PD-L1 expression, thereby enabling tumor cells to evade immune surveillance (18); similarly, overexpression of CHD4 can suppress antigen presentation, reduce T-cell activity, and ultimately lead to poor immunotherapy response (19). Additionally, aberrant chromatin remodeling can influence chemotherapeutic drug sensitivity, contributing to acquired resistance (20). For example, the inactivation of the SWI/SNF complex has been linked to increased resistance to paclitaxel and platinum-based chemotherapy in LUAD (21). Hence, systematically exploring the expression characteristics of CRGs and their impact on TIME and therapeutic response in LUAD is crucial for identifying novel predictive biomarkers and developing effective therapeutic strategies.

Given the aforementioned findings, the present study aimed to systematically investigate the expression patterns, prognostic significance, and biological functions of chromatin remodeling-associated genes in LUAD. We constructed a prognostic risk model based on CRGs and evaluated its predictive value for overall survival (OS) and therapeutic response. Furthermore, we explored the correlation between the risk model and the TIME, providing potential therapeutic targets for LUAD immunotherapy and precision medicine. Our findings offer novel insights into the role of chromatin remodeling in LUAD progression and highlight promising strategies for individualized treatment. We present this article in accordance with the TRIPOD and MDAR reporting checklists (available at https://tcr.amegroups.com/article/view/10.21037/tcr-2025-1699/rc).

Methods

Data collection and processing

Transcriptomic profiles, somatic mutation data, and clinical records of LUAD patients were curated from The Cancer Genome Atlas (TCGA) and Gene Expression Omnibus (GEO) cohorts (GSE31210, GSE50081), with inclusion criteria requiring complete survival information. The study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments. To integrate TCGA (RNA-seq) with two GEO datasets (microarray), we first performed uniform processing: raw counts from TCGA were converted to log₂(FPKM+1), and the already normalized expression matrices of the GEO datasets were log₂-transformed. Subsequently, we extracted the common set of genes shared across the three platforms and applied the “ComBat” algorithm to correct for batch effects arising from different data sources. Through a search of the GeneCards database (https://www.genecards.org), 117 CRGs (Table S1) were identified and designated as the final cohort for subsequent research.

Consensus clustering analysis of CRGs

Through “ConsensusClusterPlus” package in R, unsupervised clustering analysis was applied to classify patients into different molecular subtypes based on the mRNA expression profiles of CRGs. Consensus clustering is a common research method for cancer subtype classification. Samples can be divided into several subtypes according to different sets of omics data, so as to find new disease subtypes or compare and analyze different subtypes. The distribution of the subtypes was confirmed as per the expression profiles of the genes using principal component analysis (PCA).

Identification of differentially expressed genes (DEGs)

DEGs identified from comparisons among the three chromatin remodeling-based subtypes (using limma with |log₂FC| >1 and adjusted P<0.05) served as the foundation for prognostic model building.

Development of CRG-based prognostic Model

The DEGs was then evaluated using univariate Cox regression (P<0.05) to filter for those with individual survival prediction power. This approach ensured that our final model genes are both key effectors of the subtype biology and strong predictors of patient outcome. Subsequently, candidate genes underwent least absolute shrinkage and selection operator (LASSO) Cox regression analysis via the glmnet R package to optimize feature selection and control model overfitting. The established risk scoring system stratified patients into high- and low-risk subgroups using median risk scores as the cutoff. Survival disparities between subgroups were evaluated through Kaplan-Meier curves with log-rank testing. Model discrimination capacity was quantified using time-dependent receiver operating characteristic (ROC) curve analysis.

Immune analysis

This study employed integrated bioinformatics approaches to evaluate the tumor immune microenvironment. The CIBERSORT algorithm was applied to quantify the infiltration proportions of 22 immune cell subtypes based on gene expression data, with 1,000 permutations set for calculation and results retaining P<0.05 considered reliable. Single-sample Gene Set Enrichment Analysis (ssGSEA) was performed to assess the activation status of immune-related pathways. Immune gene sets were obtained from MSigDB, and enrichment analysis was conducted using the clusterProfiler package. The Tumor Immune Dysfunction and Exclusion (TIDE) algorithm was utilized to predict immunotherapy response. Standardized expression matrices were input via the TIDE online platform (http://tide.dfci.harvard.edu) to obtain TIDE scores, as well as dysfunction and exclusion indices.

Tumor mutational burden (TMB) analysis

TMB data of LUAD patients were obtained from the TCGA database. TMB was defined as the total number of somatic mutations per megabase of the genome.

Drug susceptibility analysis

To explore the potential association between the CRG risk score and sensitivity to chemotherapeutic or targeted drugs, the pRRophetic package in R was used to calculate the half-maximal inhibitory concentration (IC₅₀) of commonly used anti-cancer drugs in LUAD patients.

Establishment of a nomogram scoring system

A nomogram integrating risk score, age, gender, and TNM stage was developed to predict 1-, 3-, 5-year survival. Calibration curves quantified prediction accuracy.

Cell culture and transfection

LUAD cell lines A549 and H1299 (Cell Bank of Chinese Academy of Sciences) were maintained in RPMI-1640 medium with 10% fetal bovine serum at 37 ℃/5% CO₂. For GJB3 silencing, siRNA (GenePharma) was transfected using Lipofectamine 3000 (Thermo Fisher), with knockdown efficiency validated by subsequent real-time quantitative polymerase chain reaction (RT-qPCR).

Western blot analysis

Total protein was extracted using RIPA lysis buffer and quantified by the BCA method. GJB3 protein expression levels were detected by Western Blot. Equal amounts of protein samples were separated by SDS-PAGE and transferred onto PVDF membranes. The membranes were blocked with 5% skim milk and then incubated overnight at 4 ℃ with rabbit anti-human GJB3 primary antibody (1:1,000, Proteintech, 12880-1-AP) and mouse anti-human GAPDH primary antibody (1:5,000, Proteintech, 60004-1-Ig). The following day, after incubation with HRP-conjugated secondary antibodies at room temperature for 1 hour, the blots were visualized using an ECL chemiluminescent substrate. Band intensity was analyzed with ImageJ software, and the relative expression of GJB3 was calculated after normalization to GAPDH. Experiments were independently repeated three times.

RT-qPCR

Total RNA extracted with TRIzol (Invitrogen) was reverse-transcribed into cDNA using PrimeScript RT Kit (Takara). SYBR Green Master Mix (Takara) was employed for amplification on a QuantStudio 6 system (Applied Biosystems). GJB3 expression was normalized to GAPDH via the 2^−ΔΔCt method. Primer sequences were presented in Table S2.

Cell proliferation assay

Cells (5×10³/well) seeded in 96-well plates were assessed using Cell Counting Kit-8 (CCK-8) (Beyotime). Absorbance at 450 nm was measured at 24/48/72 h post-seeding using a SpectraMax microplate reader.

Transwell migration and invasion assays

For migration assays, 1×10⁵ cells in serum-free medium were added to the upper chamber of Transwell inserts (8-µm pore, Corning). For invasion assays, inserts were pre-coated with Matrigel (1:8 dilution, Corning). Complete medium (600 µL) served as chemoattractant in the lower chamber. After 24 h incubation, cells traversing the membrane were fixed with 4% paraformaldehyde, stained with 0.1% crystal violet, and quantified under an IX73 microscope. All experiments were performed with n=3.

Statistical analysis

All statistical analyses were performed using R software (v4.2.1). A P value <0.05 was considered statistically significant. Specific tests included the Wilcoxon rank-sum test for group comparisons, the log-rank test for Kaplan-Meier survival analysis, and Cox regression for univariate and multivariate prognostic analysis. Correlation analyses were conducted using Spearman’s method. The LASSO Cox regression was implemented with the ‘glmnet’ package.

Results

Molecular patterns of CRGs with distinct survival in LUAD

To investigate the expression patterns and prognostic significance of CRGs in LUAD, we conducted consensus clustering analysis based on the expression profiles of CRGs. As shown in Figure 1A, LUAD patients were classified into three distinct molecular subtypes (k=3) with relatively high intra-cluster correlations. PCA revealed significant differences in gene expression profiles among the three subtypes (Figure 1B). Kaplan-Meier survival analysis indicated that these subtypes were closely related to OS, with Cluster A showing the poorest prognosis and Cluster C showing the most favorable survival outcomes (Figure 1C). Moreover, we explored the immune landscape among the three subtypes and observed significant differences in immune cell infiltration patterns (Figure 1D). This strongly suggests that chromatin remodeling exerts a profound influence on the immune microenvironment in LUAD, potentially by regulating the expression of immune-related genes.

Figure 1 Molecular subtyping of LUAD based on chromatin remodeling genes. (A) Consensus matrix heatmap with k=3 clustering. (B) Principal component analysis plot displaying the distribution of the three clusters. (C) Survival analysis by Kaplan-Meier method. (D) Box plots illustrating the relative fractions of 23 immune cell types across the three subtypes, as estimated by the CIBERSORT algorithm. Statistical significance was assessed using the Kruskal-Wallis test among the three groups; **, P<0.01, ***, P<0.001. LUAD, lung adenocarcinoma; MDSC, myeloid-derived suppressor cells.

Development and validation of CRGs signature

To identify subtype-specific prognostic genes, we performed differential expression analysis among the three subtypes, yielding 1,844 significant genes (online supplementary table: https://cdn.amegroups.cn/static/public/tcr-2025-1699-1.xlsx). These genes were subsequently subjected to univariable Cox proportional hazards regression analysis to assess their individual prognostic value. Finally, the resulting genes from the Cox analysis were incorporated into a LASSO regression model for further refinement. For this analysis, LUAD patients were randomly allocated into training and validation cohorts. LASSO regression with 10-fold cross-validation was applied to refine the prognostic gene signature, and the optimal penalty parameter (λ) was determined by the value corresponding to the minimum cross-validation error (Figure S1A,S1B). As a result, 19 CRGs with significant prognostic value were selected to construct the risk prediction model. The heatmaps from both the training and test cohorts demonstrated distinct expression patterns of these 19 CRGs across different risk groups, further supporting the effectiveness of the model (Figure 2A,2B). The distribution of survival status and risk scores clearly distinguished high-risk patients from low-risk patients in both cohorts (Figure 2C-2F). Kaplan-Meier curves demonstrated significantly reduced OS in high-risk patients versus low-risk counterparts across both training (Figure 2G) and validation cohorts (Figure 2H). Time-dependent ROC analysis achieved area under the curve (AUC) values of 0.754, 0.755, and 0.756 for 1-, 2-, and 3-year OS predictions in the training set (Figure 2I), with corresponding validation cohort AUCs of 0.701, 0.670, and 0.633 (Figure 2J), confirming robust predictive capacity. Consistent prognostic performance was further verified in the combined cohort through survival trends and risk score patterns (Figure S2). Stratified survival analysis revealed significantly worse OS in high-risk subgroups across age categories (Figure 3A,3B), gender divisions, and early T stage (T1–2) patients (Figure 3C-3E). While T3–4 stage high-risk patients displayed a survival disadvantage, statistical significance was not attained (P=0.06, Figure 3F). Risk score distributions highlighted elevated scores in males and advanced T stage (T3–4) subgroups compared to females and early T stage (T1–2) counterparts (Figure 3G-3I), with no age-dependent variations. These observations underscore the model’s discriminative power, particularly in male and advanced T stage populations.

Figure 2 Construction and validation of the chromatin remodeling-based prognostic signature. (A,B) Heatmaps of 19-gene signature and clinical characteristics in training (A) and test (B) cohorts. (C-F) Survival status and risk score distributions across datasets. (G,H) Stratified survival analysis in training (G) and test (H) sets. (I,J) ROC curves for 1-, 2-, and 3-year survival prediction. AUC, area under the curve; ROC, receiver operating characteristic.

Figure 3 Clinical correlation and broad applicability of the prognostic signature. Prognostic stratification by (A,B) age, (C,D) gender, and (E,F) tumor stage. Boxplots comparing risk scores between (G) age groups, (H) genders, and (I) T stage subgroups.

Building a prognostic nomogram

To improve the clinical applicability of the CRGs-based risk model, we integrated the risk score with clinicopathological features to construct a prognostic nomogram. Univariate and multivariate Cox regression analyses revealed that the risk score was an independent predictor of OS in both the training (Figure 4A,4B) and test cohorts (Figure 4C,4D). A nomogram was then developed to predict the 1-, 3-, and 5-year OS of LUAD patients (Figure 4E). The calibration curves demonstrated a high degree of concordance between the predicted and actual survival rates, supporting the reliability of the nomogram (Figure 4F).

Figure 4 Independent prognostic value and clinical utility of the risk model. (A-D) Univariate/multivariate Cox regression in training (A,B) and test (C,D) cohorts. (E) Nomogram integrating clinical parameters for 1-, 3-, and 5-year survival prediction. (F) Calibration curves demonstrating model accuracy. ***, P<0.001. CI, confidence interval; OS, overall survival.

Assessing immune infiltration and checkpoints

Given the significant impact of chromatin remodeling on the TIME, we further investigated the correlation between the risk score and immune infiltration. As shown in Figure 5A, patients in the high-risk group exhibited higher infiltration levels of anti-tumor immune cells such as CD8⁺ T cells and macrophage cells, whereas the infiltration of immune cells like neutrophils cells was significantly decreased. Moreover, the high-risk group showed lower expression levels of immune function-related pathways (Figure 5B) and upregulation of immune checkpoint molecules such as CD276, PDCD1, and CD274 (Figure 5C). Notably, the result showed that the TIDE score of the low-risk group was lower (Figure 5D), which means that the low-risk group may be more sensitive to immunotherapy.

Figure 5 Immune landscape analysis. (A) Enrichment of 16 immune cell types analyzed by ssGSEA between the high- and low-risk groups. (B) Enrichment of 13 immune functions analyzed by ssGSEA between the high- and low-risk groups. (C) Differential expression of key immune checkpoint molecules between the two risk groups. (D) Comparison of TIDE scores between the high- and low-risk groups. ns, P>0.05; *, P<0.05; **, P<0.01; ***, P<0.001; ****, P<0.0001. ssGSEA, Single-sample Gene Set Enrichment Analysis; TIDE, Tumor Immune Dysfunction and Exclusion.

TMB analysis

Given the potential role of TMB in predicting immunotherapy response, we next analyzed the relationship between TMB and the risk score. The results demonstrated that TMB was significantly higher in the high-risk group compared to the low-risk group (Figure 6A). Kaplan-Meier survival analysis showed that patients with low TMB exhibited poorer OS (Figure 6B). The waterfall plot further illustrated the top 20 most frequently mutated genes in both groups, with TP53, TTN, MUC16, RYR2, and CSMD3 being the most common mutations in high-risk patients (Figure 6C,6D).

Figure 6 Analysis of TMB and its prognostic value. (A) Comparison of TMB scores between the high- and low-risk groups. (B) Survival analysis of patients stratified by high versus low TMB. (C,D) Waterfall plots depicting the top 20 mutated genes in the high-risk (C) and low-risk (D) groups. H, high; L, low; TMB, tumor mutation burden.

Analyzing drug susceptibility

To explore the potential therapeutic value of the CRGs-based risk model, we evaluated the association between risk scores and drug sensitivity. Correlation analysis revealed that several model genes were closely associated with drug sensitivity, suggesting their potential role as therapeutic targets (Figure 7A). The IC₅₀ values of various chemotherapeutic agents and targeted drugs were compared between the two risk groups. The results indicated that high-risk patients were significantly less sensitive to conventional chemotherapy drugs such as MIRA-1, PFI3, BIBR-1532, nilotinib, temozolomide, and CZC24832 (Figure 7B).

Figure 7 Assessment of therapeutic relevance based on the risk model. (A) Correlation heatmap depicting the associations between drug sensitivity (IC₅₀), the expression of key genes, and the risk score. (B) Comparison of the estimated IC₅₀ values for candidate drugs between the high- and low-risk groups. *, P<0.05; **, P<0.01; ***, P<0.001. IC₅₀, half maximal inhibitory concentration.

GJB3 promotes proliferation and metastasis in LUAD cells

To further validate the functional relevance of the risk model, we selected GJB3, one of the most upregulated genes in the high-risk group, for in vitro experiments. The knockdown efficiency of GJB3 in A549 and H1299 cells was confirmed by Western blot and qRT-PCR (Figure 8A,8B). Functional assays revealed that GJB3 knockdown significantly suppressed the proliferation of both A549 and H1299 cells (Figure 8C). Moreover, Transwell assays demonstrated that the invasive and migratory capacities of LUAD cells were significantly reduced upon GJB3 silencing (Figure 8D,8E). These findings suggested that GJB3 plays a crucial role in promoting LUAD cell proliferation and metastasis and may serve as a potential therapeutic target for LUAD patients.

Figure 8 GJB3 knockdown impairs the oncogenic behavior of LUAD cells. (A,B) Efficient knockdown of GJB3 was confirmed by Western blot and qRT-PCR in A549 and H1299 cells. (C) The proliferative capacity of cells was significantly reduced after GJB3 silencing, as measured by CCK‑8 assay. (D,E) GJB3 knockdown markedly attenuated the migration and invasion capabilities of LUAD cells, shown by representative Transwell images (stained with crystal violet) at 200× magnification (D) and their quantification (E). All data are shown as mean ± SD (n=3). ****, P<0.0001. CCK-8, Cell Counting Kit-8; LUAD, lung adenocarcinoma; OD, optical density; qRT-PCR, quantitative reverse transcription polymerase chain reaction; SD, standard deviation.

Discussion

This study systematically explored the expression patterns, prognostic significance, and biological functions of CRGs in LUAD. By performing consensus clustering analysis, we identified three distinct molecular subtypes of LUAD based on the expression profiles of CRGs, which exhibited significant differences in gene expression patterns, immune microenvironment characteristics, and OS outcomes. Furthermore, we successfully constructed and validated a robust prognostic risk model based on CRGs and subsequently confirmed the oncogenic role of the high-risk gene GJB3 in LUAD through in vitro experiments. These findings provide critical theoretical and practical insights for elucidating the regulatory role of chromatin remodeling in LUAD progression and offer promising opportunities for developing CRG-based precision therapeutic strategies.

One of the most significant findings in this study was the identification of three distinct molecular subtypes of LUAD based on the expression profiles of CRGs. The results of consensus clustering and PCA revealed that these three molecular subtypes exhibited significant differences in gene expression patterns and clinical outcomes. Among them, Cluster A had the worst prognosis, while Cluster C showed the most favorable survival outcomes. This observation is consistent with previous studies in other malignancies, which suggest that aberrant chromatin remodeling may profoundly influence tumorigenesis, immune microenvironment composition, and therapeutic response (22-24). In recent years, accumulating evidence has indicated that chromatin remodeling plays a crucial role in regulating the TIME and is closely associated with immune evasion in cancer (25). In this study, we further explored the relationship between CRGs and immune cell infiltration in LUAD. The results showed that the high-risk group exhibited a significantly enhanced immune activation state, with a marked increase in the infiltration of antitumor immune cells, such as CD8⁺ T cells. This finding implies that aberrant chromatin remodeling may promote immune evasion by altering the immune microenvironment, thereby accelerating LUAD progression. In addition, we observed significantly elevated expression levels of immune checkpoint molecules, including PD-L1, PD-1, and IDO1, in the high-risk group, suggesting that these patients might benefit more from immune checkpoint inhibitors. This finding is consistent with previous studies demonstrating that abnormal chromatin remodeling can enhance immune evasion by upregulating immune checkpoint expression (26). Notably, the TMB was significantly higher in the high-risk group, further indicating that these patients may respond favorably to immunotherapy (27).

In terms of prognostic prediction, we successfully constructed and validated a CRG-based risk score model composed of 19 CRGs significantly associated with OS. LASSO regression and Cox regression analyses showed that this model demonstrated excellent predictive performance in the training cohorts, with the time-dependent ROC curve achieving an AUC value greater than 0.7. This finding indicates that the risk score model could effectively stratify LUAD patients based on their survival outcomes and aid in making personalized treatment decisions. In a previous study, a model based on two immune-related genes was constructed to predict the prognosis of LUAD patients, with AUCs of 0.7061, 0.6816, and 0.6747 at 1, 2, and 3 years, respectively (28). In this study, the model achieved AUCs of 0.754, 0.755, and 0.756 at 1, 2, and 3 years, respectively, indicating that the model established in this study has slightly better predictability. Furthermore, the model also demonstrates better predictability in long-term follow-up. Furthermore, univariate and multivariate Cox regression analyses revealed that the risk score was an independent prognostic factor, irrespective of age, gender, and TNM stage. Thus, the CRG-based prognostic model may serve as a promising tool for clinicians to identify high-risk LUAD patients and implement personalized therapeutic interventions accordingly.

Another key finding in our study was the experimental validation of the oncogenic role of GJB3 in LUAD progression. GJB3 encodes the gap junction beta-3 protein, which has been widely recognized for its pro-oncogenic functions in various cancers (29). Our analysis revealed that GJB3 was significantly upregulated in the high-risk LUAD patients and was strongly correlated with poor prognosis. To confirm the functional role of GJB3, we conducted in vitro experiments and demonstrated that knockdown of GJB3 significantly inhibited the proliferation, migration, and invasion of LUAD cells. These results suggest that GJB3 may promote LUAD progression through multiple mechanisms, including enhancing intercellular communication, promoting epithelial-mesenchymal transition, and activating critical oncogenic signaling pathways such as Wnt/β-catenin and Notch signaling. Therefore, targeting GJB3 or its downstream signaling pathways may represent a promising therapeutic strategy for improving clinical outcomes in high-risk LUAD patients.

Despite the important findings of our study, several limitations should be noted. First, the study relied primarily on public datasets from TCGA and GEO, lacking large-scale prospective cohort validation. Future studies should collect clinical tissue samples to prospectively validate the expression profiles of CRGs and the prognostic performance of the risk model. Second, the functional validation of GJB3 was limited to in vitro experiments. Further in vivo animal models and patient-derived organoid models are needed to explore the underlying molecular mechanisms of GJB3 in LUAD progression and metastasis. Specifically, future work should prioritize uncovering molecular mechanisms of GJB3, such as its interacting partners and downstream signaling pathways. Additionally, although we observed a significant association between CRGs and the immune microenvironment, the specific regulatory mechanisms through which CRGs reshape the TIME and influence immunotherapeutic response remain to be elucidated. Future investigations should focus on delineating the mechanistic basis of CRG-mediated immune regulation and therapy resistance to provide new therapeutic targets for LUAD precision medicine.

Conclusions

This study comprehensively investigated the expression profiles, prognostic significance, and biological functions of CRGs in LUAD. We successfully constructed and validated a CRG-based risk score model, which demonstrated excellent prognostic predictive ability and clinical applicability. Moreover, we identified GJB3 as a critical oncogene that promotes LUAD progression, suggesting it as a promising therapeutic target. Our findings provide novel insights into the role of chromatin remodeling in LUAD and offer valuable evidence to support the development of CRG-targeted therapy strategies for LUAD patients.

Acknowledgments

The authors would like to thank Nanjing Integrated Traditional Chinese and Western Medicine Hospital Affiliated with Nanjing University of Chinese Medicine for their support and assistance with this study.

Footnote

Reporting Checklist: The authors have completed the TRIPOD and MDAR reporting checklists. Available at https://tcr.amegroups.com/article/view/10.21037/tcr-2025-1699/rc

Data Sharing Statement: Available at https://tcr.amegroups.com/article/view/10.21037/tcr-2025-1699/dss

Peer Review File: Available at https://tcr.amegroups.com/article/view/10.21037/tcr-2025-1699/prf

Funding: None.

Conflicts of Interest: All authors have completed the ICMJE uniform disclosure form (available at https://tcr.amegroups.com/article/view/10.21037/tcr-2025-1699/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. The study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments.

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. Siegel RL, Miller KD, Wagle NS, et al. Cancer statistics, 2023. CA Cancer J Clin 2023;73:17-48. [Crossref] [PubMed]
2. Herbst RS, Morgensztern D, Boshoff C. The biology and management of non-small cell lung cancer. Nature 2018;553:446-54. [Crossref] [PubMed]
3. Tang S, Qin C, Hu H, et al. Immune Checkpoint Inhibitors in Non-Small Cell Lung Cancer: Progress, Challenges, and Prospects. Cells 2022;11:320. [Crossref] [PubMed]
4. Reck M, Remon J, Hellmann MD. First-Line Immunotherapy for Non-Small-Cell Lung Cancer. J Clin Oncol 2022;40:586-97. [Crossref] [PubMed]
5. Sarkar D, Leung EY, Baguley BC, et al. Epigenetic regulation in human melanoma: past and future. Epigenetics 2015;10:103-21. [Crossref] [PubMed]
6. Sun D, Teng F, Xing P, et al. ARID1A serves as a receivable biomarker for the resistance to EGFR-TKIs in non-small cell lung cancer. Mol Med 2021;27:138. [Crossref] [PubMed]
7. Lorch Y, Kornberg RD. Chromatin-remodeling for transcription. Q Rev Biophys 2017;50:e5. [Crossref] [PubMed]
8. Centore RC, Sandoval GJ, Soares LMM, et al. Mammalian SWI/SNF Chromatin Remodeling Complexes: Emerging Mechanisms and Therapeutic Strategies. Trends Genet 2020;36:936-50. [Crossref] [PubMed]
9. Li Y, Gong H, Wang P, et al. The emerging role of ISWI chromatin remodeling complexes in cancer. J Exp Clin Cancer Res 2021;40:346. [Crossref] [PubMed]
10. Conaway RC, Conaway JW. The INO80 chromatin remodeling complex in transcription, replication and repair. Trends Biochem Sci 2009;34:71-7. [Crossref] [PubMed]
11. Muhammad T, Pastore SF, Good K, et al. Chromatin gatekeeper and modifier CHD proteins in development, and in autism and other neurological disorders. Psychiatr Genet 2023;33:213-32. [Crossref] [PubMed]
12. Mullen J, Kato S, Sicklick JK, et al. Targeting ARID1A mutations in cancer. Cancer Treat Rev 2021;100:102287. [Crossref] [PubMed]
13. Zhang J, Shih DJH, Lin SY. The Tale of CHD4 in DNA Damage Response and Chemotherapeutic Response. J Cancer Res Cell Ther 2019;3:052.
14. Sarkar S, Kiely R, McHugh PJ. The Ino80 chromatin-remodeling complex restores chromatin structure during UV DNA damage repair. J Cell Biol 2010;191:1061-8. [Crossref] [PubMed]
15. Yang J, Xu J, Wang W, et al. Epigenetic regulation in the tumor microenvironment: molecular mechanisms and therapeutic targets. Signal Transduct Target Ther 2023;8:210. [Crossref] [PubMed]
16. Chu X, Li X, Zhang Y, et al. Integrative single-cell analysis of human colorectal cancer reveals patient stratification with distinct immune evasion mechanisms. Nat Cancer 2024;5:1409-26. [Crossref] [PubMed]
17. Chen C, Wang Z, Ding Y, et al. Tumor microenvironment-mediated immune evasion in hepatocellular carcinoma. Front Immunol 2023;14:1133308. [Crossref] [PubMed]
18. Fukumoto T, Fatkhutdinov N, Zundell JA, et al. HDAC6 Inhibition Synergizes with Anti-PD-L1 Therapy in ARID1A-Inactivated Ovarian Cancer. Cancer Res 2019;79:5482-9. [Crossref] [PubMed]
19. Tomofuji Y, Takaba H, Suzuki HI, et al. Chd4 choreographs self-antigen expression for central immune tolerance. Nat Immunol 2020;21:892-901. [Crossref] [PubMed]
20. Bagheri M, Mohamed GA, Mohamed Saleem MA, et al. Pharmacological induction of chromatin remodeling drives chemosensitization in triple-negative breast cancer. Cell Rep Med 2024;5:101504. [Crossref] [PubMed]
21. Emmings E, Mullany S, Chang Z, et al. Targeting Mitochondria for Treatment of Chemoresistant Ovarian Cancer. Int J Mol Sci 2019;20:229. [Crossref] [PubMed]
22. Means AL. ATRX Links Chromatin Remodeling to Inflammation and Tumorigenesis in the Pancreas. Cell Mol Gastroenterol Hepatol 2019;7:233-4. [Crossref] [PubMed]
23. Giammona A, De Vellis C, Crivaro E, et al. Tumor-derived GLI1 promotes remodeling of the immune tumor microenvironment in melanoma. J Exp Clin Cancer Res 2024;43:214. [Crossref] [PubMed]
24. Xu G, Chhangawala S, Cocco E, et al. ARID1A determines luminal identity and therapeutic response in estrogen-receptor-positive breast cancer. Nat Genet 2020;52:198-207. [Crossref] [PubMed]
25. Kumar K, Pareek A, Kaur R. SWI/SNF complex-mediated chromatin remodeling in Candida glabrata promotes immune evasion. iScience 2024;27:109607. [Crossref] [PubMed]
26. Miao D, Margolis CA, Gao W, et al. Genomic correlates of response to immune checkpoint therapies in clear cell renal cell carcinoma. Science 2018;359:801-6. [Crossref] [PubMed]
27. Dvir K, Giordano S, Leone JP. Immunotherapy in Breast Cancer. Int J Mol Sci 2024;25:7517. [Crossref] [PubMed]
28. Luo C, Lei M, Zhang Y, et al. Systematic construction and validation of an immune prognostic model for lung adenocarcinoma. J Cell Mol Med 2020;24:1233-44. [Crossref] [PubMed]
29. Zeng J, Li X, Zhang Y, et al. GJB3: a comprehensive biomarker in pan-cancer prognosis and immunotherapy prediction. Aging (Albany NY) 2024;16:7647-67. [Crossref] [PubMed]