Abstract
Background
With the rapid development of immunotherapy for solid tumors, the exploration of immune characteristics is becoming more and more important. Although single biomarkers such as programmed cell death-ligand 1 (PD-L1) expression and tumor mutational burden (TMB) are clinically used, their predictive value is imperfect across molecular subgroups and clinical settings, more exploration of immune characteristics will bring more opportunities for clinical diagnosis and treatment. Given the high morbidity and mortality of lung adenocarcinoma (LUAD) in the Chinese population, the aim of this study is to explore the immune characteristics of LUAD patients and lay the foundation for better prognosis and immunotherapy.
Methods
Eight Gene Expression Omnibus (GEO) cohorts were used to identify immune and prognostically relevant genes. An immune-related gene set (IRGS) score predictive model was constructed using the single-sample gene set enrichment analysis (ssGSEA) algorithm and internally validated. The performance of the model was further verified in five external validation cohorts. To evaluate immune cell infiltration, Tumor Immune Estimation Resource (TIMER), XCELL, which digitally portrays the tissue cellular heterogeneity landscape, and Cell-type Identification by Estimating Relative Subsets of RNA Transcripts (CIBERSORT) were applied to quantify the relative proportions of the infiltrating immune cells.
Results
The patients with high IRGS scores had significantly better overall survival (OS) than those with low IRGS scores [hazard ratio (HR) =0.56; 95% confidence interval (CI): 0.46–0.68; P<0.001] in the training set. Similar results were obtained in the validation set (HR =0.45; 95% CI: 0.33–0.6; P<0.001). Further validation in five external cohorts yielded consistent results [GSE31210: P<0.001; GSE68465: P=0.04; Chen_2019: P=0.03; The Cancer Genome Atlas (TCGA)_LUAD: P=0.002; Clinical Proteomic Tumor Analysis Consortium (CPTAC)_LUAD: P=0.04]. In the tumor microenvironment analysis, the patients with high IRGS scores had higher levels of T cells, B cells, dendritic cells (DCs), and neutrophils. The immunotherapy analysis of a public cohort showed that the patients with high IRGS scores had better progression-free survival (PFS) after immunotherapy than those with low IRGS scores (P=0.01).
Conclusions
Patients with high IRGS scores had a better prognosis and improved immune efficacy. The IRGS score model appears to have a good predictive performance, indicating its potential value in clinical applications.
Keywords: Immune-related gene set (IRGS), predictive model, tumor microenvironment (TME), lung adenocarcinoma (LUAD)
Highlight box.
Key findings
• A novel immune-related gene signature (IRGS) score model was constructed based on eight Gene Expression Omnibus cohorts using a single-sample gene set enrichment analysis, and its predictive power was validated in five independent external lung adenocarcinoma (LUAD) cohorts.
• High IRGS scores were significantly associated with improved overall survival [hazard ratio (HR) =0.56 in the training set; HR =0.45 in the validation set] and better progression-free survival following immunotherapy (P=0.01).
• The patients with high IRGS scores exhibited increased immune cell infiltration, indicating a more active tumor immune microenvironment.
What is known and what is new?
• The immune characteristics of the tumor microenvironment are crucial in determining patient prognosis and the immunotherapy response in LUAD. Various immune-related biomarkers exist, but few have demonstrated consistent performance across multiple datasets in predicting immunotherapy outcomes.
• This study introduced a robust IRGS score model that showed consistent prognostic and predictive value across multiple LUAD datasets. The IRGS score model also demonstrated superior performance compared to existing immune and genomic instability markers in stratifying patient prognosis and the immunotherapy response.
What is the implication, and what should change now?
• The IRGS score model demonstrated superior performance in predicting the immunotherapy response.
• The IRGS score model may serve as a reliable biomarker that can be used to guide prognostication and therapeutic decision making in LUAD, and has strong potential for clinical application.
Introduction
Lung cancer is a malignant tumor with high incidence and mortality rates worldwide. According to Global Cancer Statistics (GLOBOCAN) 2020, lung cancer ranks first in terms of both incidence and mortality in China, and has a 5-year survival rate of less than 20%, and thus poses a significant threat to public health (1,2). Non-small cell lung cancer (NSCLC) is the most common type of lung cancer, accounting for 85% of all lung cancer cases (2). Lung adenocarcinoma (LUAD) is the most common histological subtype of NSCLC in China, accounting for 35% of all cases (3). In recent years, the advent of immune checkpoint inhibitors (ICIs) has brought about a paradigm shift in the treatment of NSCLC (4,5), consequently propelling tumor immunology characteristics to the forefront of research.
In recent years, ICIs targeting programmed cell death 1 (PD-1) or its ligand, programmed cell death-ligand 1 (PD-L1), have been widely used in the treatment of tumors, and significant progress has been made in the treatment and overall management of locally advanced and advanced NSCLC (6). However, most clinical studies have reported that the objective response rate (ORR) of NSCLC patients who receive first-line immunotherapy combined with chemotherapy is still not more than 70% (7,8), while the ORR of those who receive PD-L1 is more than 50%, and the ORR of those who receive cemiplimab monotherapy is only 46.5% (9). LUAD is often associated with driver gene mutations, so the expression level of PD-L1 is often affected by it and is relatively low, EGFR-driven or low TMB types are more likely to have immune escape phenotypes (10-12). In addition, unlike the European and American populations, the PD-L1 level of LUAD in the Chinese population is not significantly different from that in the normal population (12). Therefore, more immune characteristics in LUAD patients of China are worth exploring.
A previous study has shown that stromal cells in the tumor immune microenvironment are genetically stable and attractive therapeutic targets for reducing the risk of drug resistance and tumor recurrence (13). However, due to the metabolic activity disorder of tumor cells, tumor infiltrating immune cells usually experience metabolic stress, resulting in an impaired anti-tumor immune response and immune genome abnormalities (14,15). Therefore, the role of abnormally expressed tumor immune-related genes in tumor immunity has become a new direction in tumor research, and the multi-target method of inhibiting immune-related genomes may provide a more effective way to treat cancer.
This study used bioinformatics methods to analyze and evaluate the immune-related gene set (IRGS) scores of LUAD patients. Immune system genes associated with prognosis were studied and a prognosis prediction model was built to aid in patient diagnosis and therapy in clinical settings. A better understanding of the immune characteristics of LUAD and its related immune regulatory gene expression is of great significance for the future diagnosis and treatment of such patients. We present this article in accordance with the TRIPOD reporting checklist (available at https://jtd.amegroups.com/article/view/10.21037/jtd-2025-1616/rc).
Methods
Data acquisition
LUAD RNA-sequencing, phenotype, and survival data were downloaded from The Cancer Genome Atlas (TCGA) via the Genomic Data Commons (GDC) Data Portal (https://portal.gdc.cancer.gov) and the Clinical Proteomic Tumor Analysis Consortium (CPTAC) (https://kb.linkedomics.org/download#LUAD). The GSE11969, GSE30213, GSE31210, GSE26939, GSE30219, GSE37745, GSE42127, GSE50081, GSE68465, and GSE72094 cohorts were downloaded from the Gene Expression Omnibus (GEO) database. Among them, GSE31210 and GSE68465 were used as the independent external validation sets, and the others were used as training sets. The LUAD cohort (Chen_2019) was downloaded from the literature. The RNA-sequencing and phenotype data for LUAD were downloaded from CPTAC. Immune-related genes were collected from the ImmPort database (https://www.immport.org/shared/genelists). The study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments.
Gene selection
Using the eight GEO-LUAD cohorts, a single-factor Cox analysis was conducted to explore the correlation between the expression of each gene and overall survival (OS). A meta-analysis was performed for each gene, and the genes were filtered based on a meta P value <0.001. Samples were divided into two groups based on the enrichment level of immune-related genes: immune-related gene enrichment signature (immsign) and immune-related gene enrichment poor (immpoor) based on a meta hazard ratio (HR) <1 and an HR ≥1.
IRGS model construction and validation
After batch correction using the combat algorithm, the total number of patients in the eight cohorts (n=1,350) was divided into a training group (n=945) and a validation group (n=405) at a ratio of 7:3. The enrichment score for the immsign and immpoor gene sets was calculated using the single-sample gene set enrichment analysis (ssGSEA) algorithm. The score of the two gene sets were estimated using the “ssGSEA” algorithm with default parameters in the “GSVA” R package. The IRGS score was calculated by subtracting the enrichment score of immpoor from immsign, and the calculation formula is as follows:
| [1] |
The median value of the IRGS score in the training set was used as a cut-off value to group patients and compare differences in OS between the two groups. The same cut-off value was applied to the validation set. Additionally, the independent external validation was performed in two other GEO cohorts (GSE31210 and GSE68465) and three RNA-sequencing cohorts (i.e., the TCGA, CPTAC, and Chen_2019 cohorts).
IRGS score independence analysis
Independent factors, including the IRGS score, genetic variation, indicators of genomic instability, and clinical factors, were analyzed by univariate and multivariate Cox regression analyses. The clinical factors included age, sex, and stage
Functional enrichment analysis
The Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathways that were significantly enriched in each IRGS score group were identified using the R package “clusterProfiler”.
Subtype and tumor microenvironment analysis
The correlation between the IRGS score and genomic instability markers and immune therapy-related markers was explored to gain an understanding of the characteristics of the IRGS scores in different subgroups. Through tumor Immune Estimation Resource (TIMER; http://timer.cistrome.org/) to evaluate immune cell infiltration, which digitally portrays the tissue cellular heterogeneity landscape.
ICI cohort analysis
A survival curve analysis was conducted of the publicly available LUAD cohort (Chen_2019) of 21 patients who received immunotherapy, and differences in the efficacy of immunotherapy in different IRGS score groups were compared.
Statistical analysis
A Kaplan-Meier curve analysis was conducted with log-rank tests to compare the statistical differences in OS. The effect of the model on patient survival was assessed via single-factor and multi-factor Cox regression analyses. Mann-Whitney U tests were used to compare differences between two groups of continuous variables, and Chi-squared tests were used to compare differences between categorical variables. Spearman’s method was used to calculate the correlation coefficients and P values of the IRGS score and other factors. R language (version 4.3.0) was used for the data analysis and visualization. A P value <0.05 was considered statistically significant.
Results
Gene selection
A Cox analysis of the genes in the eight cohorts was conducted, and a meta-analysis was performed, and 15 candidate genes with a meta P value <0.001, including 14 with a HR <1 and one with a HR ≥1 were identified (Figure 1).
Figure 1.
Gene screening profiles: (A) Cox analysis results; and (B) meta-analysis results. CI, confidence interval; HR, hazard ratio; IV. Fixed + Random, the inverse variance method combines fixed-effect and random-effect models; SE, standard error.
IRGS model construction and validation
The IRGS score was calculated by subtracting the immpoor from immsign. Using the median in the training set as the cut-off value, the 945 samples were divided into two groups: the high IRGS score group (n=473) and the low IRGS score group (n=472). The patients in the high IRGS score group had significantly better OS than those in the low IRGS score group [HR =0.56; 95% confidence interval (CI): 0.46–0.68; P<0.001] (Figure 2A). Similarly, in the validation set, the 405 samples were stratified into two groups using the same cut-off value. The patients in the high IRGS score group (n=210) had significantly better OS than those in the low IRGS score group (n=195) (HR =0.45; 95% CI: 0.33–0.6; P<0.001) (Figure 2B).
Figure 2.
Kaplan-Meier survival curves showing the relationship between different levels of the IRGS scores and OS in the (A) training set, (B) validation set, (C) GSE31210 set, (D) CPTAC_LUAD set, (E) GSE68465 set, (F) Chen_2019 set, (G) TCGA_LUAD set. CI, confidence interval; CPTAC, Clinical Proteomic Tumor Analysis Consortium; HR, hazard ratio; IRGS, immune-related gene set; LUAD, lung adenocarcinoma; OS, overall survival; TCGA, The Cancer Genome Atlas.
To further confirm the prognostic performance of the IRGS score model, validation was conducted in five external cohorts. The results from the five cohorts consistently showed that the patients in the high IRGS score group had significantly better OS than those in the low IRGS score group (GSE31210: HR =0.31, 95% CI: 0.15–0.64, P<0.001; GSE68465: HR =0.76, 95% CI: 0.59–0.99, P=0.04; Chen_2019: HR =0.52, 95% CI: 0.28–0.95, P=0.03; TCGA_LUAD: HR =0.62, 95% CI: 0.46–0.68, P=0.002; CPTAC_LUAD: HR =0.41, 95% CI: 0.18–0.97, P=0.04) (Figure 2C-2G).
IRGS score independence analysis
Single-factor and multi-factor analyses were conducted of the risk associated with the IRGS scores, and clinical features. The results revealed significant correlations between age, gender, stage, IRGS risk, and IRGS scores, and prognosis in the training, internal validation, and multiple external validation sets. In some cohorts, the single-factor analysis showed a significant association between the epidermal growth factor receptor mutation and prognosis; however, no such significant correlations were found in the multi-factor analysis (tables available at https://cdn.amegroups.cn/static/public/jtd-2025-1616-1.xlsx).
Functional enrichment analysis
To explore the mechanistic correlation between the IRGS score and prognosis, a differential expression analysis was conducted by grouping samples based on high and low IRGS scores. Differential genes (high versus low) were identified, and GO and KEGG annotations were performed separately for the upregulated and downregulated genes. The results revealed that the upregulated genes in the high IRGS score group were mainly associated with B-cell and T-cell differentiation signal pathways. While the upregulated genes in the low IRGS score group were mainly associated with the DNA damage response (DDR) pathway (Figure S1, tables available at https://cdn.amegroups.cn/static/public/jtd-2025-1616-2.xlsx).
Subtype analysis
In the analysis of clinical features, the IRGS was found to be significantly correlated with age (P=0.02), gender (P=0.002), and stage (P<0.001). Among the genomic instability-related biomarkers, the patients with a low tumor mutation burden (TMB), homologous recombination deficiency (HRD), number of telomeric allelic imbalance (NTAI), and loss of heterozygosity (LOH) levels had significantly higher IRGS scores than those with high levels (P<0.001), whereas the patients with low translesion synthesis (TLS) had significantly lower IRGS scores than those with high TLS (P<0.001). The subgroup analysis of the immune checkpoint blockade (ICB) gene markers indicated significantly higher levels of PDCD1 (PD-1), cytotoxic T lymphocyte-associated antigen-4 (CTLA4), CD274 (PD-L1), LAG3, TIGIT, and HAVCR2 in the high IRGS score group (all P<0.001) (Figure 3, Figure S2).
Figure 3.
Association between the IRGS score and clinical and genomic influencing factors of (A) age, (B) sex, (C) stage, (D) TP53, (E) EGFR, (F) TLS, (G) TMB, (H) HRD, (I) NTAI, and (J) LOH. EGFR, epidermal growth factor receptor; HRD, homologous recombination deficiency; IRGS, immune-related gene set; LOH, loss of heterozygosity; Mut, mutation; NTAI, number of telomeric allelic imbalance; TLS, tertiary lymphoid structures; TMB, tumor mutation burden; Wt, wild type.
Tumor microenvironment analysis
To explore the correlation between the IRGS score and the tumor immune microenvironment, the TIMER algorithm was used to calculate the correlation between the IRGS score and immune cells. B cells, T cells, granulocytes, and dendritic cells (DCs) all exhibited positive correlations with the IRGS score (Figure 4A). A further analysis of the differences in immune cell levels between the high and low IRGS score groups revealed that the patients in the low IRGS score group had significantly lower levels of B cells, T cells, granulocytes, and DCs than those in the high IRGS score group (P<0.001) (Figure 4B).
Figure 4.
Correlation between the IRGS score and immune cells. (A) The association between the IRGS score and the immune cells. (B) The differences in immune cell levels between the different IRGS score groups. DC, dendritic cell; IRGS, immune-related gene set.
ICI cohort analysis
A cohort of 21 patients from a public LUAD population receiving immunotherapy was enrolled, their IRGS was calculated, and an analysis of the correlation between prognosis and IRGS scores was performed. The results indicated that patients in the high IRGS score group had significantly improved progression-free survival (PFS) following immunotherapy than those in the low IRGS score group (P=0.01) (Figure 5A). There was no significant difference between the two groups in terms of the ORR; however, the proportion of patients achieving a complete response (CR) or partial response (PR) was higher in the high IRGS score group than the low IRGS score group (P=0.63) (Figure 5B).
Figure 5.
Kaplan-Meier survival curves of the relationship between different levels of the IRGS score and prognosis in the immunotherapy cohort in terms of: (A) PFS, and (B) the ORR. CI, confidence interval; CR, complete response; HR, hazard ratio; IRGS, immune-related gene set; ORR, objective response rate; PD, progressive disease; PFS, progression-free survival; PR, partial response; SD, stable disease.
Discussion
In recent years, breakthroughs related to ICIs have led to advancements in the treatment of lung cancer. An increasing number of immunotherapy regimens for lung cancer have been approved. Compared to traditional chemotherapy, ICIs have lower toxicity and superior efficacy (16-18). However, even with the exploration of biomarkers such as PD-L1 and TMB, the ORR of immunotherapy remains limited, particularly for monotherapy (19-21). Therefore, exploring the immune characteristics and associated genes of tumor patients holds clinical significance. This study conducted a ssGSEA of immune genes from the ImmPort database, constructed an IRGS score model using eight GEO cohorts, and explored IRGS score model mechanisms and features.
Given the clinical significance of predicting the prognosis of lung cancer patients, a considerable amount of research has been conducted to explore prognosis risks in recent years (22-24). However, currently no guidelines or consensus-recommended predictive models exist, indicating the limitations of existing research. Immune-related models have attracted more attention in recent years, among which Estimate Score is the most widely used model (22,23). Unlike in European and American populations where a high stable score is associated with a better prognosis, it is associated with poorer immunity in Asian populations (12). In addition, most models are often constructed by LASSO regression model for different gene pathways or functional pathways, while ssGSEA is based on rank calculation, which can better quantify different factors and avoid the problem of overfitting of LASSO (25,26). This study used a ssGSEA to analyze immune-related genes, and then constructed an IRGS score model that not only showed good prognostic prediction for patients in the training and internal validation sets but also performed well in five external validation cohorts. Thus, the predictive model showed strong performance across different LUAD cohorts and is widely applicable in clinical practice.
The activation of B cell-related pathways plays a crucial role in immunity. B cells are considered the main effector cells of humoral immunity, inhibiting tumor progression by secreting immunoglobulins, promoting T cell responses, and directly killing cancer cells (27). A study has shown that B cell-related pathways are enriched in ribosomal scanning, protein synthesis, and metabolism processes in NSCLC cell lines (28). Further, the high expression of B cell-related genes has been shown to be correlated with better immunotherapy efficacy in multiple cancers, including lung cancer, melanoma, and colorectal cancer (29). These findings suggest that the upregulation of B cell-related pathways is associated with prognosis and immunotherapy response in NSCLC. Moreover, the upregulation of both T and B cell-related pathways has been shown to be correlated with immune checkpoint markers (30). Several clinical studies have shown that the high expression of markers such as PD-L1 and CTLA4 is associated with better immunotherapy efficacy, and guidelines recommend immunotherapy monotherapy for patients with high PD-L1 expression (31-33). This aligns with our findings that a high IRGS score is correlated with elevated levels of ICB markers and better immunotherapy efficacy.
As a dynamic system, the tumor microenvironment includes immune cells, stromal cells, cancer cells, complex cytokines, and chemokines (34). Among them, immune-related cells and factors participate in the whole process of tumorigenesis, proliferation and development is particularly important (35,36). Previous research has shown that patients with higher levels of CD8+ T-cell, CD4+ T-cell, and B-cell infiltration have a better prognosis and better immune efficacy (19,37-39). A retrospective study reported significantly longer PFS in patients with a high infiltration of CD8+ T cells undergoing monotherapy (40). In addition, in solid tumors such as head and neck malignancies, a higher level of B-cell infiltration has been associated with the better efficacy of immunosuppressive agents (41). In our study, patients with high IRGS scores had higher levels of T cells, B cells, DCs, and neutrophils. This not only aligns with the better prognosis observed in patients with high IRGS scores in the aforementioned model, but also corresponds with the upregulated signaling pathways associated with IRGS score model. Further, our research also showed that the patients with high IRGS scores had a better prognosis and immune efficacy than those with low IRGS scores. In the immunotherapy cohort analysis, the patients with high IRGS scores had significantly better PFS, and a higher proportion of these patients achieved a CR and PR. These results are consistent with previous study concerning ovarian cancer patients (42), indicating the reliability of the IRGS score in predicting patient prognosis and immune efficacy.
Additionally, we conducted an analysis of the correlation between genomic instability markers and the IRGS. The results showed that only TLS was higher in patients with high IRGS scores, while the TMB, HRD, LOH, NTAI, and other markers were significantly lower in patients with high IRGS scores. This indicates that the IRGS model can identify patients who cannot be distinguished by other biomarkers and can better predict patient prognosis and response to immunotherapy. In a study on prostate cancer, patients with better immune-related model scores had significantly lower TMB and DDR scores (43). Similarly, a recent study on cervical cancer has also confirmed that immune-related genes are closely associated with the immune microenvironment and can be used to predict prognosis (44). Immune models can therefore help to identify individuals who are more likely to benefit from immunotherapy.
This study had a number of limitations. First, this study only conducted the analyses and model validation using public cohorts. The clinical immune efficacy analysis also used public cohorts; additionally, real clinical patients were not included to verify the prediction of immune efficacy by the IRGS score. Second, in terms of functional evaluation, only GO and KEGG analyses were conducted, and no cell or animal experiments were performed to confirm the mechanism. These factors require further validation, to provide new insights and to increase the applicability of our results in clinical settings.
Conclusions
In conclusion, this study used the ssGSEA algorithm to identify 15 immune-related and prognostically significant genes in public cohorts of LUAD patients. An IRGS score model was established and validated. Patients with high IRGS scores had a better prognosis and enhanced immune efficacy. The IRGS score model exhibited superior predictive performance, and thus has potential value in clinical applications.
Supplementary
The article’s supplementary files as
Acknowledgments
None.
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.
Reporting Checklist: The authors have completed the TRIPOD reporting checklist. Available at https://jtd.amegroups.com/article/view/10.21037/jtd-2025-1616/rc
Funding: None.
Conflicts of Interest: All authors have completed the ICMJE uniform disclosure form (available at https://jtd.amegroups.com/article/view/10.21037/jtd-2025-1616/coif). The authors have no conflicts of interest to declare.
(English Language Editor: L. Huleatt)
References
- 1.Sung H, Ferlay J, Siegel RL, et al. Global Cancer Statistics 2020: GLOBOCAN Estimates of Incidence and Mortality Worldwide for 36 Cancers in 185 Countries. CA Cancer J Clin 2021;71:209-49. 10.3322/caac.21660 [DOI] [PubMed] [Google Scholar]
- 2.Li W, Wu M, Wang Q, et al. A comparative genomics analysis of lung adenocarcinoma for Chinese population by using panel of recurrent mutations. J Biomed Res 2020;35:11-20. 10.7555/JBR.34.20200068 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Li D, Shi J, Dong X, et al. Epidemiological characteristics and risk factors of lung adenocarcinoma: A retrospective observational study from North China. Front Oncol 2022;12:892571. 10.3389/fonc.2022.892571 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Pakkala S, Owonikoko TK. Immune checkpoint inhibitors in small cell lung cancer. J Thorac Dis 2018;10:S460-7. 10.21037/jtd.2017.12.51 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Xiong A, Wang J, Zhou C. Immunotherapy in the First-Line Treatment of NSCLC: Current Status and Future Directions in China. Front Oncol 2021;11:757993. 10.3389/fonc.2021.757993 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.NCCN. The NCCN NSCLC clinical practice guidelines in oncology (version 1.2024). Fort Washington: NCCN; 2024 [2023-12-21]. Available online: http://www.nccn.org/professionals/physician_gls/f_guidelines.asp
- 7.Zhou C, Chen G, Huang Y, et al. Camrelizumab plus carboplatin and pemetrexed as first-line therapy for advanced non-squamous non-small-cell lung cancer: 5-year outcomes of the CameL randomized phase 3 study. J Immunother Cancer 2024;12:e009240. 10.1136/jitc-2024-009240 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Wang Z, Wu L, Li B, et al. Toripalimab Plus Chemotherapy for Patients With Treatment-Naive Advanced Non-Small-Cell Lung Cancer: A Multicenter Randomized Phase III Trial (CHOICE-01). J Clin Oncol 2023;41:651-63. 10.1200/JCO.22.00727 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Kilickap S, Baramidze A, Sezer A, et al. Cemiplimab Monotherapy for First-Line Treatment of Patients with Advanced NSCLC With PD-L1 Expression of 50% or Higher: Five-Year Outcomes of EMPOWER-Lung 1. J Thorac Oncol 2025;20:941-54. 10.1016/j.jtho.2025.03.033 [DOI] [PubMed] [Google Scholar]
- 10.Zhang XC, Wang J, Shao GG, et al. Comprehensive genomic and immunological characterization of Chinese non-small cell lung cancer patients. Nat Commun 2019;10:1772. 10.1038/s41467-019-09762-1 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Jiao XD, Zhang XC, Qin BD, et al. Tumor mutation burden in Chinese cancer patients and the underlying driving pathways of high tumor mutation burden across different cancer types. Ann Transl Med 2020;8:860. 10.21037/atm-20-3807 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Bie F, Tian H, Sun N, et al. Comprehensive analysis of PD-L1 expression, tumor-infiltrating lymphocytes, and tumor microenvironment in LUAD: differences between Asians and Caucasians. Clin Epigenetics 2021;13:229. 10.1186/s13148-021-01221-3 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Quail DF, Joyce JA. Microenvironmental regulation of tumor progression and metastasis. Nat Med 2013;19:1423-37. 10.1038/nm.3394 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Zhong Y, She Y, Deng J, et al. Deep Learning for Prediction of N2 Metastasis and Survival for Clinical Stage I Non-Small Cell Lung Cancer. Radiology 2022;302:200-11. 10.1148/radiol.2021210902 [DOI] [PubMed] [Google Scholar]
- 15.Li X, Wenes M, Romero P, et al. Navigating metabolic pathways to enhance antitumour immunity and immunotherapy. Nat Rev Clin Oncol 2019;16:425-41. 10.1038/s41571-019-0203-7 [DOI] [PubMed] [Google Scholar]
- 16.Mok TSK, Wu YL, Kudaba I, et al. Pembrolizumab versus chemotherapy for previously untreated, PD-L1-expressing, locally advanced or metastatic non-small-cell lung cancer (KEYNOTE-042): a randomised, open-label, controlled, phase 3 trial. Lancet 2019;393:1819-30. 10.1016/S0140-6736(18)32409-7 [DOI] [PubMed] [Google Scholar]
- 17.Zhou C, Chen G, Huang Y, et al. Camrelizumab plus carboplatin and pemetrexed versus chemotherapy alone in chemotherapy-naive patients with advanced non-squamous non-small-cell lung cancer (CameL): a randomised, open-label, multicentre, phase 3 trial. Lancet Respir Med 2021;9:305-14. 10.1016/S2213-2600(20)30365-9 [DOI] [PubMed] [Google Scholar]
- 18.Yang Y, Wang Z, Fang J, et al. Efficacy and Safety of Sintilimab Plus Pemetrexed and Platinum as First-Line Treatment for Locally Advanced or Metastatic Nonsquamous NSCLC: a Randomized, Double-Blind, Phase 3 Study (Oncology pRogram by InnovENT anti-PD-1-11). J Thorac Oncol 2020;15:1636-46. 10.1016/j.jtho.2020.07.014 [DOI] [PubMed] [Google Scholar]
- 19.Sezer A, Kilickap S, Gümüş M, et al. Cemiplimab monotherapy for first-line treatment of advanced non-small-cell lung cancer with PD-L1 of at least 50%: a multicentre, open-label, global, phase 3, randomised, controlled trial. Lancet 2021;397:592-604. 10.1016/S0140-6736(21)00228-2 [DOI] [PubMed] [Google Scholar]
- 20.Reck M, Rodríguez-Abreu D, Robinson AG, et al. Pembrolizumab versus Chemotherapy for PD-L1-Positive Non-Small-Cell Lung Cancer. N Engl J Med 2016;375:1823-33. 10.1056/NEJMoa1606774 [DOI] [PubMed] [Google Scholar]
- 21.Marabelle A, Fakih M, Lopez J, et al. Association of tumour mutational burden with outcomes in patients with advanced solid tumours treated with pembrolizumab: prospective biomarker analysis of the multicohort, open-label, phase 2 KEYNOTE-158 study. Lancet Oncol 2020;21:1353-65. 10.1016/S1470-2045(20)30445-9 [DOI] [PubMed] [Google Scholar]
- 22.Yang D, Ma X, Song P. A prognostic model of non small cell lung cancer based on TCGA and ImmPort databases. Sci Rep 2022;12:437. 10.1038/s41598-021-04268-7 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Zhang F, Wang J, Xu Y, et al. Co-occurring genomic alterations and immunotherapy efficacy in NSCLC. NPJ Precis Oncol 2022;6:4. 10.1038/s41698-021-00243-7 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Lai YH, Chen WN, Hsu TC, et al. Overall survival prediction of non-small cell lung cancer by integrating microarray and clinical data with deep learning. Sci Rep 2020;10:4679. 10.1038/s41598-020-61588-w [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Hänzelmann S, Castelo R, Guinney J. GSVA: gene set variation analysis for microarray and RNA-seq data. BMC Bioinformatics 2013;14:7. 10.1186/1471-2105-14-7 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Simon N, Friedman J, Hastie T, et al. Regularization Paths for Cox's Proportional Hazards Model via Coordinate Descent. J Stat Softw 2011;39:1-13. 10.18637/jss.v039.i05 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Tokunaga R, Naseem M, Lo JH, et al. B cell and B cell-related pathways for novel cancer treatments. Cancer Treat Rev 2019;73:10-9. 10.1016/j.ctrv.2018.12.001 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Ye Q, Guo NL, Single B. Cell Gene Co-Expression Networks Implicated in Prognosis, Proliferation, and Therapeutic Responses in Non-Small Cell Lung Cancer Bulk Tumors. Cancers (Basel) 2022;14:3123. 10.3390/cancers14133123 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Lundberg A, Li B, Li R. B cell-related gene signature and cancer immunotherapy response. Br J Cancer 2022;126:899-906. 10.1038/s41416-021-01674-6 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.Baldanzi G. Immune Checkpoint Receptors Signaling in T Cells. Int J Mol Sci 2022;23:3529. 10.3390/ijms23073529 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.Gandhi L, Rodríguez-Abreu D, Gadgeel S, et al. Pembrolizumab plus Chemotherapy in Metastatic Non-Small-Cell Lung Cancer. N Engl J Med 2018;378:2078-92. 10.1056/NEJMoa1801005 [DOI] [PubMed] [Google Scholar]
- 32.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. 10.3390/cells11030320 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33.Graves EE, Maity A, Le QT. The tumor microenvironment in non-small-cell lung cancer. Semin Radiat Oncol 2010;20:156-63. 10.1016/j.semradonc.2010.01.003 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34.Giatromanolaki A, Koukourakis IM, Balaska K, et al. Programmed death-1 receptor (PD-1) and PD-ligand-1 (PD-L1) expression in non-small cell lung cancer and the immune-suppressive effect of anaerobic glycolysis. Med Oncol 2019;36:76. 10.1007/s12032-019-1299-4 [DOI] [PubMed] [Google Scholar]
- 35.Roma-Rodrigues C, Mendes R, Baptista PV, et al. Targeting Tumor Microenvironment for Cancer Therapy. Int J Mol Sci 2019;20:840. 10.3390/ijms20040840 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 36.Thommen DS, Koelzer VH, Herzig P, et al. A transcriptionally and functionally distinct PD-1(+) CD8(+) T cell pool with predictive potential in non-small-cell lung cancer treated with PD-1 blockade. Nat Med 2018;24:994-1004. 10.1038/s41591-018-0057-z [DOI] [PMC free article] [PubMed] [Google Scholar]
- 37.Fassan M, Cavallin F, Guzzardo V, et al. PD-L1 expression, CD8+ and CD4+ lymphocyte rate are predictive of pathological complete response after neoadjuvant chemoradiotherapy for squamous cell cancer of the thoracic esophagus. Cancer Med 2019;8:6036-48. 10.1002/cam4.2359 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 38.Leong TL, Bryant VL. B cells in lung cancer-not just a bystander cell: a literature review. Transl Lung Cancer Res 2021;10:2830-41. 10.21037/tlcr-20-788 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 39.Guo J, Shu T, Zhang H, et al. A combined model of serum neutrophil extracellular traps, CD8(+) T cells, and tumor proportion score provides better prediction of PD-1 inhibitor efficacy in patients with NSCLC. FEBS J 2024;291:3403-16. 10.1111/febs.17144 [DOI] [PubMed] [Google Scholar]
- 40.Wang J, Zhong Y, Wu L. Immunotherapy for Advanced Non-small Cell Lung Cancer: Research Progress and Perspectives. Cancer Research on Prevention and Treatment 2024;51:409-18. [Google Scholar]
- 41.Li K, Zhang C, Zhou R, et al. Single cell analysis unveils B cell-dominated immune subtypes in HNSCC for enhanced prognostic and therapeutic stratification. Int J Oral Sci 2024;16:29. 10.1038/s41368-024-00292-1 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 42.Wolde T, Bhardwaj V, Reyad-Ul-Ferdous M, et al. The Integrated Bioinformatic Approach Reveals the Prognostic Significance of LRP1 Expression in Ovarian Cancer. Int J Mol Sci 2024;25:7996. 10.3390/ijms25147996 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 43.Han S, Shi T, Liao Y, et al. Tumor immune contexture predicts recurrence after prostatectomy and efficacy of androgen deprivation and immunotherapy in prostate cancer. J Transl Med 2023;21:194. 10.1186/s12967-022-03827-4 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 44.Bao Q, Tang W, Tang W, et al. Identification of immune-related cervical cancer prognostic biomarkers and construction of prognostic model based on tumor microenvironment. Eur J Med Res 2025;30:261. 10.1186/s40001-025-02515-5 [DOI] [PMC free article] [PubMed] [Google Scholar]