
==== Front
Oncol Lett
Oncol Lett
OL
Oncology Letters
1792-1074
1792-1082
D.A. Spandidos

10.3892/ol.2024.14682
OL-28-5-14682
Articles
Analysis and identification of mRNAsi‑related expression signatures via RNA sequencing in lung cancer
Yan Bo 1
Chen Yong 2
Wang Zhouyu 3
Li Jing 3
Wang Ruiru 3
Pan Xufeng 2
Li Boyi 4
Li Rong 1
1 Clinical Research Unit, Shanghai Chest Hospital, School of Medicine, Shanghai Jiao Tong University, Shanghai 200050, P.R. China
2 Department of Thoracic Surgery, Shanghai Chest Hospital, School of Medicine, Shanghai Jiao Tong University, Shanghai 200050, P.R. China
3 Berry Oncology Corporation, Beijing 100102, P.R. China
4 Kanghui Biotechnology Corporation, Shenyang, Liaoning 110042, P.R. China
Correspondence to: Dr Rong Li, Clinical Research Unit, Shanghai Chest Hospital, School of Medicine, Shanghai Jiao Tong University, 241 West Huaihai Road, Changning, Shanghai 200050, P.R. China, E-mail: xkyylirong@163.com
11 2024
13 9 2024
13 9 2024
28 5 54902 4 2024
15 8 2024
Copyright: © 2024 Yan et al.
2024
https://creativecommons.org/licenses/by-nc-nd/4.0/ This is an open access article distributed under the terms of the Creative Commons Attribution-NonCommercial-NoDerivs License, which permits use and distribution in any medium, provided the original work is properly cited, the use is non-commercial and no modifications or adaptations are made.
High stemness index scores are associated with poor survival in patients with lung cancer. Studies on the mRNA expression-based stemness index (mRNAsi) are typically conducted using tumor tissues; however, mRNAsi-related expression signatures based on cell-free RNA (cfRNA) are yet to be comprehensively investigated. The present study aimed to elucidate the gene expression profiles of tumor stemness in lung cancer tissues and corresponding cfRNAs in blood, and to assess their links with immune infiltration. Tumor tissue, paracancerous tissue, peripheral blood and lymph node samples were collected from patients with stage I–III non-small cell lung cancer and RNA sequencing was performed. The TCGAbiolinks package was used to calculate the mRNAsi for each of these four types of sample. Weighted gene co-expression network analysis and differentially expressed gene analyses were performed to investigate mRNAsi-related genes, and pathway enrichment analysis was performed using the Kyoto Encyclopedia of Genes and Genomes (KEGG) orthology-based annotation system. In addition, the STAR-Fusion tool was used to detect fusion variants, and CIBERSORT was used to analyze the correlations of stemness signatures in tissues and blood with immune cell infiltration. The mRNAsi values in peripheral blood and lymph nodes were found to be higher than those in cancer tissues. ‘Hematopoietic cell lineage’ was the only KEGG pathway enriched in mRNAsi-related genes in both lung cancer tissues and peripheral blood. In addition, the protein tyrosine phosphatase receptor type C associated protein gene was the only gene commonly associated with the mRNAsi in these two types of sample. The expression of mRNAsi-related genes was increased in the dendritic and Treg cells in tumor tissues, but was elevated in Treg and CD8 cells in the blood. In conclusion, cfRNAs in the blood exhibit unique stemness signatures that have potential for use in the diagnosis of lung cancer.

lung cancer
mRNAsi
cell-free RNA
immune infiltration
diagnosis
National Natural Science Foundation of China81773273 The Special Project of Clinical Research in Health Field of Shanghai Municipal Health Commission202240019 This study was funded by the National Natural Science Foundation of China (grant no. 81773273) and The Special Project of Clinical Research in Health Field of Shanghai Municipal Health Commission (grant no. 202240019).
==== Body
pmcIntroduction

Approximately 2.2 million new lung cancer cases and 1.8 million deaths of patients with lung cancer were reported worldwide in 2020 (1). Although considerable progress has been achieved in lung cancer diagnosis and treatment, the five-year survival rate for this disease remains low (2). The survival rate for lung cancer is closely associated with the disease stage, and the five-year survival rate of patients with stage I lung cancer can reach 85.5–90.2% (3). Therefore, the identification of novel and reliable biomarkers for the early diagnosis of lung cancer is critical.

Cancer stem cells (CSCs) are cells that exhibit stem cell properties, have the ability for self-renewal, and promote the invasion and growth of tumor cells in cancer (4). The presence of CSCs leads to cellular heterogeneity in malignancies as well as innate antibiotic resistance and enhanced invasive capacity, which contribute to cancer progression and metastasis (5). Several conventional CSC markers, including CD44, CD90 and SRY-box transcription factor 2, have been identified in lung cancer; however, these markers are not used universally in clinical practice (6). The mRNA expression-based stemness index (mRNAsi) is a quantitative measure that indicates the degree of similarity between tumors and stem cells (7,8). This index is based on mRNA expression profiles and reflects the stemness characteristics of the transcriptome (9). Previous studies have revealed that lung cancer tissues exhibit higher mRNAsi values than normal tissues and suggest that high mRNAsi values are associated with poor survival in patients with lung adenocarcinoma (LUAD) (10,11).

Although circulating tumor DNA (ctDNA) exhibits potential for the detection of lung cancer, its sensitivity and specificity in the diagnosis of early lung cancer is not optimal (12). Circulating cell-free RNA (cfRNA) is actively secreted by cells or released by apoptotic and necrotic cells into the blood (13). Plasma cfRNAs have been shown to reflect the systemic response to a growing tumor and to indicate the tissue origin of the tumor (14). The release of cfRNA into the blood circulation by early tumors allows upregulated, tumor-specific and tumor-derived RNAs to be identified in the blood at an early stage, which may facilitate the early diagnosis of tumors (15). Studies have indicated that the combined analysis of cfRNA and ctDNA is complementary, improving the early diagnosis of lung cancer (16,17).

As most studies on mRNAsi are typically conducted using tumor tissues, whether cfRNA-based mRNAsi can be used for the early diagnosis of lung cancer remains unclear. In the present study, primary lung cancer tissues, adjacent non-cancerous tissues and lymph nodes were collected for RNA sequencing and analysis to identify gene expression signatures stratified by mRNAsi. Furthermore, paired blood samples were collected for cfRNA sequencing and the determination of a stemness gene expression signature of cfRNAs based on tissue stemness genes. Subsequently, the genes and pathways associated with tissue and blood stemness were analyzed. Finally, correlations of tissue mRNAsi and the cfRNA stemness index with immune infiltration were examined.

Materials and methods

Tissue and blood samples

Tumor tissue, paracancerous tissue (>5 cm away from the tumor edge), peripheral blood and lymph node samples were obtained from 22 patients with stage I–III non-small cell lung cancer (NSCLC) between October 2021 and March 2022 at the Shanghai Chest Hospital, Shanghai Jiao Tong University School of Medicine (Shanghai, China). None of the patients had received any chemotherapy, targeted therapy or immunotherapy before surgery. The clinicopathological data of the patients, including their sex, age, pathological and molecular pathological characteristics (based on TNM staging) (18), and recurrence, were collected.

The freshly collected tissues were snap-frozen and stored in liquid nitrogen. Blood samples were temporarily stored at 4°C. Plasmapheresis was completed within 4 h of the blood being drawn from the patient, and the plasma was stored immediately in a refrigerator at −80°C.

RNA sequencing of samples

Tissue total RNA was extracted from frozen tissues using a MagMAX FFPE DNA/RNA Ultra Kit (Thermo Fisher Scientific, Inc.) following the standard protocol provided by the manufacturer but skipping the deparaffinization process. The concentration of the RNA was measured by NanoDrop™ 2000 spectrophotometer (Thermo Fisher Scientific, Inc.) and the quality of the RNA was evaluated using TapeStation 4200 system (Agilent Technologies, Inc.), and RNA with DV200 ≥40% was considered qualified. Qualified RNA was reverse-transcribed with random hexamers for first strand synthesize and underwent second strand synthesizing, end modification and polyadenylation, sequencing adapter ligation and library amplification using a Hieff NGS® Ultima Dual-mode RNA Library Prep Kit [Yeasen Biotechnology (Shanghai) Co., Ltd.]. Purified libraries were pooled and sequenced using a NovaSeq 6000 system (Illumina, Inc.) with PE150 mode. Low-quality sequences were filtered out prior to downstream analysis.

Cell free RNA was extracted from plasma samples using SpCap-cfRNA extraction and library preparation technology, which is a modified from the method by Jacobsen et al (19) specifically for low abundance plasma cell free RNA extraction and library preparation. In brief, the SpCap-cfRNA technology utilizes poly-T oligo-coated magnatic beads to capture cell free RNA from the plasma, and after extensive wash, captured RNA is reverse-transcribed on beads to make RNA libraries for sequencing. The cfRNA libraries were pooled and sequenced using NovaSeq 6000 system (Illumina, Inc.) with PE150 mode. The sequencing data were subjected to bioinformatics analysis.

Calculation of mRNAsi

The R package TCGAbiolinks (2.18.0) was used to calculate the mRNAsi of different sample types (20). Stemness signatures were analyzed using Progenitor Cell Biology Consortium stemSig (version 2.18.0) (21). The mRNAsi, a continuous variable, was used to categorise the patients into high- and low-mRNAsi groups for different sample types, based on the median.

Weighted gene co-expression network analysis (WGCNA) and identification of differentially expressed genes (DEGs)

The WGCNA package (version 1.71) (22) was used construct coexpression networks. In this analysis, the Pearson's correlation coefficients of genes were calculated to obtain correlation matrices, and appropriate thresholds calculated by the function pickSoftThreshold in WGCNA were selected to measure the connectivity between genes. Adjacency matrices were converted to topological overlap matrices (TOM), and the corresponding differences (1-TOM) were calculated. Genes with similar expression patterns were then grouped into co-expression modules. First, WGCNA of the tumor tissues was performed. In addition, the DEGs of tumor and paracancerous tissues were screened using the EdgeR package (version 3.34.0) (23) with the criteria of fold change >2 and adjusted P-value <0.05. Correlation analyses were then performed to determine the WGCNA modules most relevant to the mRNAsi of tumor tissues and blood. The tissue and peripheral blood samples were categorized into high and low mRNAsi score groups using the median value as a cutoff. DEGs between the high and low mRNAsi groups were screened using the aforementioned EdgeR package. Pathway enrichment analysis of the identified genes was performed using the Kyoto Encyclopedia of Genes and Genomes (KEGG) orthology-based annotation system (version 3.0) (https://www.genome.jp/kegg/) with default parameters.

Identification of fusion transcripts

The STAR-Fusion tool (version 1.8.1) was used to detect fusion transcripts based on the plasma cfRNA sequencing data (24). In the gene expression analysis, STAR was used to map the sequencing data to the reference genome (hg38) and define the transcript coordinates based on the gene annotation format file from GENCODE (version 27; GRCh38) (https://www.gencodegenes.org/human/release_27.html). Gene expression abundance was quantified as reads per kilobase per million mapped reads by using the Cufflinks package (version 2.2.1) (25). The EdgeR package was used to identify DEGs between the fusion-positive and normal samples.

Correlation between stemness index and immune cell infiltration

The gene expression profiles were uploaded to the CIBERSORT website (https://cibersortx.stanford.edu/) to analyze the immune cells infiltrating the tumor tissue. Using LM22 as a reference, 1,000 random permutation operations were performed, and the results were screened based on a significance threshold of P<0.05. The ggplot2 package (version 3.4.4) (https://github.com/tidyverse/ggplot2) was used to analyze differences in immune cell populations between normal paracancerous tissues and tumors. The packages limma (version 3.54.2) (https://bioconductor.org/packages/limma), reshape2 (version 1.4.4) (https://github.com/hadley/reshape) and ggExtra (version 0.10.1) (https://github.com/daattali/ggExtra) were used to evaluate the relationships between the mRNAsi-related genes and immune infiltrating cells. Spearman's rank correlation tests were used for analysis, and the ggplot2 package (version 3.4.4) (https://github.com/tidyverse/ggplot2) was used to generate correlation heat maps.

Statistical analysis

R software (version 4.1.2) (https://www.r-project.org/) was used to statistically analyze all data. Differences in mRNAsi values between different types of matched samples were assessed using the Friedman test followed by Wilcoxon signed-rank tests with Bonferroni correction. An unpaired t-test was used to analyze the statistical significance of differences in normally distributed variables between two groups, and the Wilcoxon rank-sum test was used to compare non-normally distributed variables between two groups. Kaplan-Meier (K-M) analysis and log-rank tests were used to compare survival between two groups. P<0.05 was considered to indicate a statistically significant result.

Results

Clinical data

A total of 22 patients were included in the present study (Table I). The age of the patients ranged from 46 to 74 years, with an average of 63.8 years. Among all patients, 63.6% were women, 90.9% had adenocarcinoma, 86.3% had stage I or II cancer, and 40.9% had poorly differentiated tumors.

mRNAsi correlates with the prognosis of patients with lung cancer

By evaluating the median value of mRNAsi in the peripheral blood and cancer tissue samples of patients with lung cancer, the patients were categorized into high and low mRNAsi groups for each sample type. K-M survival analysis revealed no significant difference in overall survival between the high and low mRNAsi groups for peripheral blood or cancer tissue samples (Fig. 1). This may be attributed to the short duration of follow-up, as the median follow-up time was only 820 days. To investigate the relationship between mRNAsi and disease recurrence, the recurrence rates of patients in the groups over two years were calculated. In both the peripheral blood and cancer tissue samples, the two-year recurrence rate of patients in the high mRNAsi group was 27.3%, whereas that of patients in the low mRNAsi group was 18.2%. Notably, when the peripheral blood and cancer tissue samples both had high mRNAsi values, the two-year recurrence rate of the patients was 18.2%, which was twice that of patients with low mRNAsi values.

Comparison of mRNAsi values in peripheral blood and tissues

To investigate the association between tumor stemness and metastasis, the mRNAsi values of paracancerous tissues, cancer tissues, lymph nodes and peripheral blood were compared. The results revealed that the mRNAsi value of tumor tissues was significantly higher than that of paracancerous normal tissues, and the mRNAsi value of lymph nodes was significantly higher than that of tumor tissues. Also, the mRNAsi value of cfRNA in peripheral blood was higher than that in cancer tissues, albeit not significantly, and was significantly higher than that in paracancerous tissues (Fig. 2).

Genes and pathways associated with lung cancer development

WGCNA of the tumor tissues revealed that several genes, including ALKBH3, MED19, protein tyrosine phosphatase receptor type C associated protein (PTPRN), ERAL1 and PPME1 are associated with lung cancer (Table SI). KEGG analysis indicated that these genes are enriched in the pathways ‘steroid biosynthesis’ and ‘other types of O-glycan biosynthesis’ (Fig. 3A). The DEGs associated with tumorigenesis, identified by the comparison of gene expression in tumor tissue with that in paracancerous tissue, are presented as a volcano plot (Fig. 3B; Table SII). Upregulated genes are enriched in the ‘cell cycle’, ‘protein digestion and absorption’ and ‘metabolism of xenobiotics by cytochrome P450’ pathways (Fig. 3C). By contrast, downregulated genes are enriched in the ‘neuroactive ligand-receptor interaction’, ‘calcium signaling pathway’ and ‘vascular smooth muscle contraction’ pathways (Fig. 3D).

Genes and pathways associated with the mRNAsi of tumor tissues and peripheral blood

WGCNA and DEG analyses were conducted to identify genes associated with the mRNAsi of tumor tissues and peripheral blood. First, WGCNA was used to perform the cluster analysis of genes associated with the mRNAsi of tumor tissues (Fig. 4A). The results indicated that CYBC1, DEF6, LIME1, CD3D and CD7 are among the genes associated with the mRNAsi of tumor tissues (Table SIII). The pathways found to be associated with this mRNAsi include ‘hematopoietic cell lineage’ and ‘T-cell receptor signaling pathway’ (Fig. 4B). The stemness index of the tissues was used to categorise the tumors into high and low mRNAsi groups, and the DEGs between the two groups were determined (Table SIV). The pathways enriched in the upregulated genes are ‘cell cycle’, ‘oocyte meiosis’ and ‘metabolism of xenobiotics by cytochrome P450’ (Fig. 4C). The pathways enriched with the downregulated genes are ‘cytokine-cytokine receptor interaction’, ‘TNF signaling pathway’ and ‘PI3K-Akt signaling pathway’ (Fig. 4D).

The WGCNA results revealed that AEN, BCL7A, BLNK, CCT3 and CD22 are among the genes associated with the mRNAsi of peripheral blood (Fig. 4E, Table SV). These genes are enriched in ‘hematopoietic cell lineage’, ‘B-cell receptor signaling pathway’ and ‘NF-kappa B signaling pathway’ (Fig. 4F). PTPRCAP was the only gene found to be associated with the stemness index in both tissues and peripheral blood (Fig. S1). The DEGs associated with the stemness index in the peripheral blood are presented in Table SVI. The upregulated genes are enriched in ‘PI3K-Akt signaling pathway’ (Fig. 4G), whereas the downregulated genes are enriched in ‘PPAR signaling pathway’, ‘ECM receptor interaction’ and ‘protein digestion and absorption’ (Fig. 4H).

Identification of rare fusion variants through cfRNA sequencing

Seven rare fusion variants, specifically IGSF9-SCAMP3, SEC62-ALCAM, CHD1L-ARHGAP26, EML4-ALK, ZFAND3-FKBP5, RASSF3-RHPN2 and PPOX-KHDC4, were identified by cfRNA sequencing (Table II). These results demonstrate that rare fusion variants can be identified using cfRNA sequencing.

Correlation between mRNAsi and immune cell infiltration

Correlations between immune cell levels and the mRNAsi in tumor tissues or peripheral blood were analyzed. The results indicate that M2 macrophages and CD4+ memory T cells are activated in the cancer tissue microenvironment (Fig. 5A). In addition, TMEM177, MPRS15, PSMB3, NOP58 and TONSL genes correlate with mRNAsi in the immune microenvironment of the cancer tissues (Fig. 5B). By contrast, the genes that correlate with cfRNA stemness in immune cells include RPS4X, CD52, BCL7A, HVCN1, AEN and SMAD3 (Fig. 5C). The mRNAsi values of dendritic cells, regulatory T cells (Tregs), and M2 macrophages in the tumor tissues are higher than those in other types of immune cells (Fig. 5D). The cfRNA stemness index values of Tregs and CD8+ T cells are higher than those of other types of immune cells (Fig. 5E). In addition, the cfRNA stemness of dendritic cells and M2 macrophages is lower than the stemness of the cancer tissue. Finally, the associations between the degree of stemness and various immune cells were analyzed. The results revealed that Tregs are closely associated with tissue and blood stemness (Figs. S2 and S3). Notably, the mRNAsi-associated immune cells and genes differ between the tumor tissues and peripheral blood.

Discussion

The treatment and prognosis of lung cancer vary according to the extent of the disease at diagnosis (26). Early detection and diagnosis methods, particularly non-invasive approaches, are crucial for improving the overall survival rate and prognosis of patients with lung cancer (27). Numerous studies have focused on the identification of biomarkers for the early detection of lung cancer based on cfDNA analysis (28–30). In the present study, cfRNA sequencing was used to explore the genes, pathways and immune cells associated with lung cancer cell stemness.

The mRNAsi value of metastatic lymph nodes was found to be significantly higher than that of tumor tissues. In addition, a previous study has reported that the mRNAsi value of metastatic tumors is considerably higher than that of primary tumors, particularly in prostate and pancreatic cancers (7). Furthermore, a comparison of two-year recurrence rates in the current study revealed that high tumor and blood mRNAsi values are associated with a poorer prognosis, which is consistent with the results of previous studies (31,32). Studies of cfRNAs in lung cancer have typically focused on miRNAs or a few known cancer-associated mRNAs (33). Plasma-based cfRNA transcriptome analysis has been demonstrated to be useful for the early detection, diagnosis and monitoring of NSCLC (17). In the present study, the stemness index of cfRNA was higher than that of tumor tissues. Additional studies are necessary to elucidate the relationship between cfRNAs and tissue stemness in lung cancer.

WGCNA revealed that the ‘hematopoietic cell lineage’ pathway is enriched in mRNAsi-related genes in lung cancer tissue and blood. Hematopoietic cell lineages have previously been indicated to be involved in the proliferation, differentiation and adhesion of lung cancer cells (34). The present study also revealed that PTPRCAP is the only gene associated with the stemness index in peripheral blood as well as in tumor tissues. PTPRCAP is an immunological marker of triple-negative breast cancer, the expression of which positively correlates with disease-free survival (35). Additionally, PTPRCAP is a natural killer cell-associated gene that is associated with the response of patients with chronic myeloid leukemia to tyrosine kinase inhibitor (TKI) treatment (36). However, the role of PTPRCAP in lung cancer, particularly its effect on lung cancer cell stemness, is unknown. Therefore, additional studies are necessary to investigate the mechanism of action of PTPRCAP in lung cancer progression.

Seven rare fusion variants, namely IGSF9-SCAMP3, SEC62-ALCAM, CHD1L-ARHGAP26, EML4-ALK, ZFAND3-FKBP5, RASSF3-RHPN2 and PPOX-KHDC4, were identified in the present study using cfRNA sequencing. Anaplastic lymphoma kinase (ALK) fusion has been found in 3–7% of patients with NSCLC (37). Considerable progress has been achieved in the treatment of ALK+ NSCLC using TKIs, as demonstrated by clinical trials (38,39). EML4-ALK is the most frequently observed fusion variant resulting from ALK rearrangement in NSCLC (40). For patients with EML4-ALK+ NSCLC, the National Comprehensive Cancer Network guidelines recommend crizotinib, ceritinib or alectinib as first-line therapies (41). However, to the best of our knowledge, the other fusion variants have not been reported in lung cancer and require validation.

Immune cells are a crucial part of the tumor microenvironment (TME) and play a crucial role in tumor immunotherapy (42). The present study indicated that M2 macrophages and CD4+ memory T cells are activated in the microenvironment of lung cancer tissues. Tumor-associated macrophages (TAMs) are derived from peripheral blood monocytes that infiltrate solid tumor tissues (43). TAMs acquire an M2-type macrophage phenotype in response to signals from the TME and contribute to tumor metastasis (44). In addition, the CSC-mediated accumulation of M2 macrophages is reported to contribute to the generation of an immunosuppressive, pro-tumorigenic TME (42). A high invasive abundance of CD4+ memory-activated T cells has been demonstrated to be associated with poor survival in patients with LUAD (45). Furthermore, in the present study, high mRNAsi values, indicating the upregulated expression of stemness-associated genes, were observed in the dendritic and Treg cells in tissues and in Treg and CD8+ T cells in the blood, which may be attributed to the environments in which the cells are located. CSCs are able to alter the composition and functional characteristics of tumor-specific effector T cells and promote the expansion of immunosuppressive tumor-promoting Tregs (46). In addition to regulating the immune environment, CSCs exhibit various properties that allow them to directly evade the effector mechanisms of immune cells (47). Since a high cfRNA stemness index was associated with increased Treg enrichment, an elevated cfRNA stemness index is indicative of an immunosuppressive state.

In summary, the present study reveals the unique stemness gene expression characteristics of blood cfRNAs, offering novel insights for the development of non-invasive early lung cancer detection methods. However, the study has some limitations. First, only 22 lung cancer samples were analyzed. The results of such a small sample could be biased; therefore, future studies with more samples or patients from multiple centers are necessary to support the findings. Second, as the study is based on bioinformatics analysis, it has certain shortcomings in terms of experimental validation. Therefore, well-designed basic experiments and clinical trials are critical for confirmation of the results.

Supplementary Material

Supporting Data

Supporting Data

Supporting Data

Supporting Data

Supporting Data

Supporting Data

Supporting Data

Acknowledgements

Not applicable.

Availability of data and materials

The data generated in the present study may be found in the CNGB Sequence Archive of China National GeneBank DataBase (https://db.cngb.org/) under accession number CNP0005912.

Authors' contributions

RL, BY and YC conceived and designed the study. BY drafted the manuscript. RL and ZW reviewed and revised the manuscript. YC and XP performed the experiments. ZW, JL, RW and BL conducted the data analysis and prepared the figures. RL, BY and BL confirm the authenticity of all the raw data. All authors read and approved the final version of the manuscript.

Ethics approval and consent to participate

The study was approved by the Ethics Committee of Shanghai Chest Hospital (Shanghai, China; approval no. KS1740), and complied with the principles of the Declaration of Helsinki. All patients who participated in the study signed informed consent forms.

Patient consent for publication

Not applicable.

Competing interests

ZW, JL, and RW are employees of Berry Oncology Corporation, and some of the materials used in the study were provided by Berry Oncology Corporation. BL is an employee of Liaoning Kanghui Biotechnology Corporation, and assays and analyses in this study were provided free of charge by Liaoning Kanghui Biotechnology Corporation. The other authors declare that they have no competing interests.

Figure 1. Relationships between mRNAsi values and overall survival in patients with non-small cell lung cancer assessed using K-M analysis. K-M comparison of high and low mRNAsi groups in (A) peripheral blood and (B) tumor tissues. mRNAsi, mRNA expression-based stemness index; K-M, Kaplan-Meier; P-high, high peripheral blood mRNAsi; P-low, low peripheral blood mRNAsi; T-high, high tumor tissue mRNAsi; T-low, low tumor tissue mRNAsi.

Figure 2. Comparison of mRNAsi among different matched patient samples. All groups were compared by the Friedman test, and comparisons between two groups were performed using Wilcoxon signed-rank tests with Bonferroni correction. **P<0.01, ***P<0.001. N, normal paracancerous tissues; T, tumor tissues; P, peripheral blood; L, lymph node.

Figure 3. Genes and pathways associated with lung cancer. (A) Pathway enrichment analysis of the genes found to be associated with lung cancer by weighted gene co-expression network analysis. (B) DEGs associated with lung tumorigenesis. Pathways enriched by (C) upregulated DEGs and (D) downregulated DEGs. DEGs, differentially expressed genes; TvsN, tumor tissues vs. normal paracancerous tissues; Up, upregulated genes; Down, downregulated genes; NS, not significant; P-adj, adjusted P-value.

Figure 4. Genes and pathways associated with the mRNAsi of lung cancer tissues and peripheral blood. (A) WGCNA was used to analyze the genes associated with the mRNAsi in tissues and (B) enriched KEGG pathways were identified. Pathways enriched in (C) upregulated and (D) downregulated DEGs associated with the tissue mRNAsi. (E) WGCNA analysis of the genes associated with the mRNAsi in peripheral blood and (F) enriched KEGG pathways. Pathways enriched in (G) upregulated and (H) downregulated DEGs associated with the mRNAsi of peripheral blood. mRNAsi, mRNA expression-based stemness index; WGCNA, weighted gene co-expression network analysis; KEGG, Kyoto Encyclopedia of Genes and Genomes; DEGs, differentially expressed genes.

Figure 5. Immune cells associated with the stemness of lung cancer tissue and cfRNA. (A) Analysis of the infiltration of various immune cells in tissues. (B) Correlation between the mRNAsi-related genes of tissues and immune cell infiltration. (C) Correlation between cfRNA stemness index-related genes and immune cell infiltration. Correlation between immune cells and the (D) mRNAsi of tumor tissues and (E) cfRNA stemness index. cfRNA, cell-free DNA; mRNAsi, mRNA expression-based stemness index.

Table I. Clinical characteristics of the enrolled patients.

Characteristics	Values	
Mean age, years	63.8	
Sex, n (%)		
  Female	14 (63.6)	
  Male	8 (36.4)	
Histology, n (%)		
  Adenocarcinoma	20 (90.9)	
  Squamous cell carcinoma	2 (9.1)	
Differentiation degree, n (%)		
  Poor	9 (40.9)	
  Moderate	13 (59.1)	
T stage, n (%)		
  T1	11 (50.0)	
  T2	7 (31.8)	
  T3	3 (13.6)	
  T4	1 (4.5)	
N stage, n (%)		
  N0	20 (90.9)	
  N1	0 (0.0)	
  N2	2 (9.1)	
  N3	0 (0.0)	
M stage		
  M0	22 (100.0)	
  M1	0 (0.0)	
pTMN stage		
  I	16 (72.7)	
  II	3 (13.6)	
  III	3 (13.6)	

Table II. Rare fusion variants in tissues identified by RNA sequencing.

Fusion variant	Junction Read Count	Spanning Frag Count	
IGSF9-SCAMP3	37	8	
SEC62-ALCAM	27	13	
CHD1L-ARHGAP26	17	11	
EML4-ALK	11	13	
ZFAND3-FKBP5	39	19	
RASSF3-RHPN2	68	21	
PPOX-KHDC4	9	17	
Junction Read Count indicates the overlapping reads at fusion breakpoints. Spanning Frag Count indicates that reads do not overlap the breakpoint position, but have read pairs on either side of the breakpoint (spanning the breakpoint).
==== Refs
References

1 Sung H Ferlay J Siegel RL Laversanne M Soerjomataram I Jemal A Bray F Global cancer statistics 2020: Globocan estimates of incidence and mortality worldwide for 36 cancers in 185 countries CA Cancer J Clin 71 209 249 2021 10.3322/caac.21660 33538338
2 Samarth N Gulhane P Singh S Immunoregulatory framework and the role of miRNA in the pathogenesis of NSCLC-A systematic review Front Oncol 12 1089320 2022 10.3389/fonc.2022.1089320 36620544
3 Shi JF Wang L Wu N Li JL Hui ZG Liu SM Yang BY Gao SG Ren JS Huang HY Clinical characteristics and medical service utilization of lung cancer in China, 2005–2014: Overall design and results from a multicenter retrospective epidemiologic survey Lung Cancer 128 91 100 2019 10.1016/j.lungcan.2018.11.031 30642458
4 Shi X Liu Y Cheng S Hu H Zhang J Wei M Zhao L Xin S Cancer stemness associated with prognosis and the efficacy of immunotherapy in adrenocortical carcinoma Front Oncol 11 651622 2021 10.3389/fonc.2021.651622 34367952
5 Bhuvaneswari MS Priyadharsini S Balaganesh N Theenathayalan R Hailu TA Investigating the lung adenocarcinoma stem cell biomarker expressions using machine learning approaches Biomed Res Int 2022 3518190 2022 10.1155/2022/3518190 36193299
6 Wan R Liao H Liu J Zhou L Yin Y Mu T Wei J Development of a 5-gene signature to evaluate lung adenocarcinoma prognosis based on the features of cancer stem cells Biomed Res Int 2022 4404406 2022 10.1155/2022/4404406 35480140
7 Su C Zheng J Chen S Tuo J Su J Ou X Chen S Wang C Identification of key genes associated with cancer stem cell characteristics in Wilms' tumor based on bioinformatics analysis Ann Transl Med 10 1204 2022 10.21037/atm-22-4477 36544656
8 Malta TM Sokolov A Gentles AJ Burzykowski T Poisson L Weinstein JN Kamińska B Huelsken J Omberg L Gevaert O Machine learning identifies stemness features associated with oncogenic dedifferentiation Cell 173 338 354.e15 2018 10.1016/j.cell.2018.03.034 29625051
9 Liao Y Xiao H Cheng M Fan X Bioinformatics analysis reveals biomarkers with cancer stem cell characteristics in lung squamous cell carcinoma Front Genet 11 427 2020 10.3389/fgene.2020.00427 32528520
10 Hou S Xu H Liu S Yang B Li L Zhao H Jiang C Integrated bioinformatics analysis identifies a new stemness index-related survival model for prognostic prediction in lung adenocarcinoma Front Genet 13 860268 2022 10.3389/fgene.2022.860268 35464867
11 Li N Li Y Zheng P Zhan X Cancer stemness-based prognostic immune-related gene signatures in lung adenocarcinoma and lung squamous cell carcinoma Front Endocrinol (Lausanne) 12 755805 2021 10.3389/fendo.2021.755805 34745015
12 Li RY Liang ZY Circulating tumor DNA in lung cancer: Real-time monitoring of disease evolution and treatment response Chin Med J (Engl) 133 2476 2485 2020 10.1097/CM9.0000000000001097 32960843
13 Heitzer E Haque IS Roberts CES Speicher MR Current and future perspectives of liquid biopsies in genomics-driven oncology Nat Rev Genet 20 71 88 2019 10.1038/s41576-018-0071-5 30410101
14 Roskams-Hieter B Kim HJ Anur P Wagner JT Callahan R Spiliotopoulos E Kirschbaum CW Civitci F Spellman PT Thompson RF Plasma cell-free RNA profiling distinguishes cancers from pre-malignant conditions in solid and hematologic malignancies NPJ Precis Oncol 6 28 2022 10.1038/s41698-022-00270-y 35468987
15 Larson MH Pan W Kim HJ Mauntz RE Stuart SM Pimentel M Zhou Y Knudsgaard P Demas V Aravanis AM Jamshidi A A comprehensive characterization of the cell-free transcriptome reveals tissue- and subtype-specific biomarkers for cancer detection Nat Commun 12 2357 2021 10.1038/s41467-021-22444-1 33883548
16 Sorber L Zwaenepoel K Jacobs J De Winne K Goethals S Reclusa P Van Casteren K Augustus E Lardon F Roeyen G Circulating cell-free DNA and RNA analysis as liquid biopsy: Optimal centrifugation protocol Cancers (Basel) 11 458 2019 10.3390/cancers11040458 30935089
17 Seneviratne C Shetty AC Geng X McCracken C Cornell J Mullins K Jiang F Stass S A pilot analysis of circulating cfRNA transcripts for the detection of lung cancer Diagnostics (Basel) 12 2897 2022 10.3390/diagnostics12122897 36552904
18 Goldstraw P Chansky K Crowley J Rami-Porta R Asamura H Eberhardt WE Nicholson AG Groome P Mitchell A Bolejack V The IASLC lung cancer staging project: Proposals for revision of the TNM stage groupings in the forthcoming (Eighth) edition of the TNM classification for lung cancer J Thorac Oncol 11 39 51 2016 10.1016/j.jtho.2015.09.009 26762738
19 Jacobsen N Nielsen PS Jeffares DC Eriksen J Ohlsson H Arctander P Kauppinen S Direct isolation of poly(A)+ RNA from 4 M guanidine thiocyanate-lysed cell extracts using locked nucleic acid-oligo(T) capture Nucleic Acids Res 32 e64 2004 10.1093/nar/gnh056 15096560
20 Colaprico A Silva TC Olsen C Garofano L Cava C Garolini D Sabedot TS Malta TM Pagnotta SM Castiglioni I TCGAbiolinks: An R/Bioconductor package for integrative analysis of TCGA data Nucleic Acids Res 44 e71 2016 10.1093/nar/gkv1507 26704973
21 Chen D Liu J Zang L Xiao T Zhang X Li Z Zhu H Gao W Yu X Integrated machine learning and bioinformatic analyses constructed a novel stemness-related classifier to predict prognosis and immunotherapy responses for hepatocellular carcinoma patients Int J Biol Sci 18 360 373 2022 10.7150/ijbs.66913 34975338
22 Langfelder P Horvath S WGCNA: An R package for weighted correlation network analysis BMC Bioinformatics 9 559 2008 10.1186/1471-2105-9-559 19114008
23 Robinson MD McCarthy DJ Smyth GK edgeR: A Bioconductor package for differential expression analysis of digital gene expression data Bioinformatics 26 139 140 2010 10.1093/bioinformatics/btp616 19910308
24 Zhang X Wang F Yan F Huang D Wang H Gao B Gao Y Hou Z Lou J Li W Yan J Identification of a novel HOOK3-FGFR1 fusion gene involved in activation of the NF-kappaB pathway Cancer Cell Int 22 40 2022 10.1186/s12935-022-02451-y 35081975
25 Trapnell C Roberts A Goff L Pertea G Kim D Kelley DR Pimentel H Salzberg SL Rinn JL Pachter L Differential gene and transcript expression analysis of RNA-seq experiments with TopHat and Cufflinks Nat Protoc 7 562 578 2012 10.1038/nprot.2012.016 22383036
26 Zhao Y Jia S Zhang K Zhang L Serum cytokine levels and other associated factors as possible immunotherapeutic targets and prognostic indicators for lung cancer Front Oncol 13 1064616 2023 10.3389/fonc.2023.1064616 36874133
27 Shai S Patolsky F Drori H Scheinman EJ Davidovits E Davidovits G Tirman S Arber N Katz A Adir Y A novel, accurate, and non-invasive liquid biopsy test to measure cellular immune responses as a tool to diagnose early-stage lung cancer: A clinical trials study Respir Res 24 52 2023 10.1186/s12931-023-02358-w 36788537
28 Mathios D Johansen JS Cristiano S Medina JE Phallen J Larsen KR Bruhm DC Niknafs N Ferreira L Adleff V Detection and characterization of lung cancer using cell-free DNA fragmentomes Nat Commun 12 5060 2021 10.1038/s41467-021-24994-w 34417454
29 Sugimoto A Matsumoto S Udagawa H Itotani R Usui Y Umemura S Nishino K Nakachi I Kuyama S Daga H A large-scale prospective concordance study of plasma- and tissue-based next-generation targeted sequencing for advanced non-small cell lung cancer (LC-SCRUM-Liquid) Clin Cancer Res 29 1506 1514 2023 10.1158/1078-0432.CCR-22-1749 36201167
30 Wang S Meng F Li M Bao H Chen X Zhu M Liu R Xu X Yang S Wu X Multidimensional cell-free DNA fragmentomic assay for detection of early-stage lung cancer Am J Respir Crit Care Med 207 1203 1213 2023 10.1164/rccm.202109-2019OC 36346614
31 Chen M Wang X Wang W Gui X Li Z Immune- and stemness-related genes revealed by comprehensive analysis and validation for cancer immunity and prognosis and its nomogram in lung adenocarcinoma Front Immunol 13 829057 2022 10.3389/fimmu.2022.829057 35833114
32 Wang H Wang Y Luo W Zhang X Cao R Yang Z Duan J Wang K Integrative stemness characteristics associated with prognosis and the immune microenvironment in lung adenocarcinoma BMC Pulm Med 22 463 2022 10.1186/s12890-022-02184-8 36471379
33 Müller S Janke F Dietz S Sültmann H Circulating MicroRNAs as potential biomarkers for lung cancer Recent Results Cancer Res 215 299 318 2020 10.1007/978-3-030-26439-0_16 31605236
34 Narayanan S Wu ZX Wang JQ Ma H Acharekar N Koya J Yoganathan S Fang S Chen ZS Pan Y The spleen tyrosine kinase inhibitor, entospletinib (GS-9973) restores chemosensitivity in lung cancer cells by modulating ABCG2-mediated multidrug resistance Int J Biol Sci 17 2652 2665 2021 10.7150/ijbs.61229 34326700
35 Marchetti P Antonov A Anemona L Vangapandou C Montanaro M Botticelli A Mauriello A Melino G Catani MV New immunological potential markers for triple negative breast cancer: IL18R1, CD53, TRIM, Jaw1, LTB, PTPRCAP Discov Oncol 12 6 2021 10.1007/s12672-021-00401-0 35201443
36 Park H Kang S Kim I Kim S Kim HJ Shin DY Kim DY Lee KH Ahn JS Sohn SK The association of genetic alterations with response rate in newly diagnosed chronic myeloid leukemia patients Leuk Res 114 106791 2022 10.1016/j.leukres.2022.106791 35101736
37 Wu R Liu S Lv G Deng C Wang R Zhang S Zhu D Wang L Lei Y Luo Z First-line crizotinib therapy is effective for a novel SEC31A-anaplastic lymphoma kinase fusion in a patient with stage IV lung adenocarcinoma: A case report and literature reviews Anticancer Drugs 34 294 301 2023 10.1097/CAD.0000000000001408 36730620
38 Zou Z Gu Y Liang L Hao X Fan C Xin T Zhao S Liu Z Guo Y Ma K Alectinib as first-line treatment for advanced ALK-positive non-small cell lung cancer in the real-world setting: Preliminary analysis in a Chinese cohort Transl Lung Cancer Res 11 2495 2506 2022 10.21037/tlcr-22-803 36636411
39 Shaw AT Solomon BJ Chiari R Riely GJ Besse B Soo RA Kao S Lin CC Bauer TM Clancy JS Lorlatinib in advanced ROS1-positive non-small-cell lung cancer: A multicentre, open-label, single-arm, phase 1–2 trial Lancet Oncol 20 1691 1701 2019 10.1016/S1470-2045(19)30655-2 31669155
40 Guo W Liang J Zhang D Huang X Lv Y Lung adenocarcinoma harboring complex EML4-ALK fusion and BRAF V600E co-mutation responded to alectinib Medicine (Baltimore) 101 e30913 2022 10.1097/MD.0000000000030913 36221356
41 Ettinger DS Wood DE Aisner DL Akerley W Bauman JR Bharat A Bruno DS Chang JY Chirieac LR D'Amico TA NCCN guidelines insights: Non-small cell lung cancer, version 2.2021 J Natl Compr Canc Netw 19 254 266 2021 10.6004/jnccn.2021.0013 33668021
42 Steven A Fisher SA Robinson BW Immunotherapy for lung cancer Respirology 21 821 833 2016 10.1111/resp.12789 27101251
43 Liu P Liu Y Chen L Fan Z Luo Y Cui Y Anemoside A3 inhibits macrophage M2-like polarization to prevent triple-negative breast cancer metastasis Molecules 28 1611 2023 10.3390/molecules28041611 36838599
44 Najafi M Goradel NH Farhood B Salehi E Nashtaei MS Khanlarkhani N Khezri Z Majidpoor J Abouzaripour M Habibi M Macrophage polarity in cancer: A review J Cell Biochem 120 2756 2765 2019 10.1002/jcb.27646 30270458
45 Chen H Lin R Lin W Chen Q Ye D Li J Feng J Cheng W Zhang M Qi Y An immune gene signature to predict prognosis and immunotherapeutic response in lung adenocarcinoma Sci Rep 12 8230 2022 10.1038/s41598-022-12301-6 35581376
46 Fridman WH Zitvogel L Sautès-Fridman C Kroemer G The immune contexture in cancer prognosis and treatment Nat Rev Clin Oncol 14 717 734 2017 10.1038/nrclinonc.2017.101 28741618
47 Müller L Tunger A Plesca I Wehner R Temme A Westphal D Meier F Bachmann M Schmitz M Bidirectional crosstalk between cancer stem cells and immune cell subsets Front Immunol 11 140 2020 10.3389/fimmu.2020.00140 32117287
