Exploring the Molecular and Immune Landscape of Lung Cancer Associated with Cystic Airspaces: Implications for Prognosis and Therapeutic Strategies

Exploring the Molecular and Immune Landscape of Lung Cancer Associated with Cystic Airspaces: Implications for Prognosis and Therapeutic Strategies | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Research Article Exploring the Molecular and Immune Landscape of Lung Cancer Associated with Cystic Airspaces: Implications for Prognosis and Therapeutic Strategies Xiang Zheng, Li Qiu, Ying Huang, Ran Cheng, Si Huang, Ke Xu, Wei Cai, and 6 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-3448810/v1 This work is licensed under a CC BY 4.0 License Status: Posted Version 1 posted You are reading this latest preprint version Abstract Objective: To explore the molecular biological characteristics of lung cancer associated with cystic airspaces (LCCA) and its potential roles on prognosis. Methods: A total of 165 LCCAs and 201 non-LCCAs were enrolled in this study. Bulk RNA sequencing was implemented in eight LCCAs and nine non-LCCAs to explore the differentially expressed genes. TCGA data were used to analyze LCCA-specific genes that associated with overall survival. Results: The median age was 60 (IQR 53 to 65) years in LCCA cohort. We found LCCA were predominant in men and had less visceral pleura invasion (VPI) or lympho-vascular invasion (LVI). Moreover, LCCA presented with higher histological heterogeneity. Kaplan-Meier analysis showed that patients of age more than 60 and positive VPI had significantly less PFS in LCCA. Cox regression suggested that LCCA, micropapillary subtype proportion and VPI were the independent risk factors for PFS. LCCA had up-regulated pathways associated with EMT, angiogenesis and cell migration. In addition, LCCA displayed higher levels of immunosuppressor infiltration (M2 macrophages, CAFs and MDSCs) and distinct cell death and metabolic patterns. BCR/TCR repertoire analysis revealed less BCR richness, clonality and high-abundance shared clonotypes in LCCA. Finally, Cox regression analysis identified that four cystic-specific genes, KCNK3 , NRN1 , PARVB and TRHDE-AS1 , were associated with OS of LUAD. And cystic-specific risk scores (CSRSs) were calculated to construct a nomogram, which performance well. Conclusions: Our study for the first time indicated significantly distinct molecular biological and immune characteristics between LCCA and non-LCCA, which provide complementary prognostic values in early-stage NSCLC. Lung cancer Lung cancer associated with cystic airspaces Prognosis Gene expression BCR/TCR repertoire Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Figure 6 Introduction Although clinical rarity, there has been increasingly reports recently of a unique type of lung cancer characterized by the presence of round or irregular air-containing cavities with cystic walls radiographically [ 1 – 3 ] . Terminologically, some entities were used loosely in association with this type of lung cancer [ 4 ] . In this study, we define the term “lung cancer associated with cystic airspaces” (i.e., LCCA) to refer to lung cancer that manifests as solitary cystic airspaces and has thin walls of 4 mm or less along the lesion circumference [ 5 , 6 ] . The comprehensive nature of LCCA is still limited, though most previous studies have described the clinicopathological and radiographical features of LCCA. Accumulating evidence suggests that LCCA occurs in approximately 1% ~ 4% lung cancer [ 3 , 7 , 8 ] , and they are predominantly (about 90%) adenocarcinomas and more frequently found in men and smokers [ 4 ] . A four-type classification system was proposed to summarize the morphologic patterns of CT observations in 2015 [ 9 ] . Long-term outcomes are not well-defined as regarded to classification or stage. Still, it is reported that LCCA had a worse prognosis than the non-LCCA [ 3 , 10 ] and the wall thickness is an independent prognostic factor [ 11 ] . Moreover, the nodule pattern is associated with poor differentiation, shorter survival and higher PD-L1 expression than other subtypes [ 2 , 3 , 12 ] . However, little is known about the genetic landscape of LCCA. The purpose of this study is to investigate the transcriptional profiles of LCCA as compare to non-LCCA or normal lung tissue. Materials and Methods Patients and samples Patients with cystic lesions in the lungs were retrospectively reviewed in our institute from January 1st 2017 to July 1st 2023. All the patients were screened according to the following inclusion criteria: (1) age ≥ 18 years; (2) the target lesion histologically confirmed as lung carcinoma; and (3) available medical records for clinicopathological features and follow-ups (Figure S1 A). Another cohort of patients with solid lesions in the lungs (i.e., non-LCCA) were enrolled according to the same criteria and matched for TNM stage and histology. All the slides were examined by the same pathologist to confirm the diagnosis. Clinicopathological characteristics including age, gender, smoking status, surgical procedure, histological subtype, visceral pleura invasion (VPI), cancer cells spread through air spaces (STAS), lymphovascular thrombus, TNM stage according 8th edition staging system for lung cancer and gene mutations. In this study, tumor samples from 8 patients with LCCA and 9 patients with non-LCCA, as well as 10 samples of adjacent normal tissue from above patients were collected. All the samples were stored in liquid nitrogen immediately after surgical resection (Figure S1 ). CT image and morphology-based classifications All the chest computed tomographic (CT) scans were performed on the multi-detector CT scanners (SIEMENS SOMATOM Definition AS 128, SIEMENS SOMATOM Perspective 128, GE Revolution CT 256), abiding by the following protocols: CT scans were performed without contrast administration and obtained at the end of inspiration during a single deep breath-hold, using the automatic tube current modulation with 120 kV, 150–300 mA, matrix size of 512 × 512, field of view (FOV) of 300–350 mm, and 1.0 mm reconstructed section thickness. The lung window was set as a width of 1200 Hounsfield units (HU) and a level of -600 HU. Two independent thoracic radiologists with 10 and 12 years of working experience reviewed the CT images and were blinded to pathology. The radiographical features including lesion location, entire lesion size, nodule density, cyst size, wall component, nodule size/wall thickness, number of cystic airspaces, shape of cystic airspaces, and pleural retraction, were evaluated. A four-type classification system proposed by Mascalchi [ 9 ] was applied and assigned to each LCCA. Type I refers to a nodule extruding from the wall of the cystic airspace; type II refers to a nodule projecting into the cystic airspace from the wall; type III is that of the cyst wall thickening along the lesion circumference; and type IV refers to multiple cystic lesions intermixed with soft tissue septum (Figure S1 C-F). The inter-reviewer consistency of the classifications between the two radiologists was assessed. In case of discrepancy, a consensus was reached by discussion and examined by a third senior radiologist with 20 years of experience. RNA Extraction and Quality Control Total RNA was isolated from tissues using TRIzol reagent (Life Technologies, CA, USA). The quantity and quality of total RNA samples were measured using a NanoDrop ND-1000 (Wilmington, DE, USA). RNA integrity was assessed using an Agilent 2100 Bioanalyzer (Agilent Technologies, Santa Clara, CA, USA). Subsequent analyses were performed on samples with RNA Integrity Number (RIN) greater than 7. cDNA Library Construction and High-throughput Sequencing After extracting the total RNA of the sample, the rRNA was removed by the Ribosome Kit (NEB, USA). The circular RNA library construction also required RNase R (Biosearch Technologies , UK) to remove the linear RNA, and then the RNA was fragmented (the average fragment length was about 200nt), using M-MuLV Reverse Transcriptase (Thermo Fisher Scientific, USA) reverse transcription to synthesize single-stranded cDNA, then double-stranded cDNA is synthesized and purified, followed by end repair and adapter primers. PCR amplification was performed using LongAmp™ Taq 2X Master Mix, SR Primer for Illumina and index primers (NEB, USA). PCR products were purified on 8% polyacrylamide gels. DNA fragments corresponding to 140–160 bp were recovered and dissolved in 8 mL of elution buffer. Finally, library quality was assessed on the Agilent Bioanalyzer 2100 system using the DNA High Sensitivity Chip. Sequencing was finally performed using an Illumina HiSeq2500 sequencer (RiboBio, Guangzhou, China) by performing paired-end runs with 100 bp read length. Pre-processing of sequencing reads Sequencing read quality was inspected using the FastQC [ 13 ] . Adapter removal and read trimming were performed using Trimmomatic [ 14 ] to removing trailing sequences below a phred quality score of 20 and drop reads below length of 50 bp. Sequencing data were aligned to the human reference genome hg19 using HISAT2 [ 15 ] . HTSeq was subsequently employed to count the reads numbers mapped to each gene [ 16 ] . Read counts were further normalized to transcripts per kilobase of exon model per million mapped reads (TPM) to quantify the expression levels of mRNAs or ncRNAs. Processing of LUAD (Lung carcinoma) data from TCGA (The Cancer Genome Altas) database The raw count and HTseq-FPKM gene expression data and corresponding LUAD clinical and survival data were downloaded from Xena database ( http://xena.ucsc.edu/ ). The FPKM (Fragments per kilobase of transcript per million mapped reads) data were transformed to TPM following the formula: \({TPM}_{i}=\frac{{FPKM}_{i}\times {10}^{6}}{\sum _{i=1}^{N}{FPKM}_{i}}\) A total of 412 cases were enrolled (Figure S1 B), and all the data had undergone process of normalization, log 2 -transformation and removal of batch effect before analyses. The cases were randomly assigned into training set and validation set at a ratio of 7:3. Differential expression analysis Differential expression was normalized and analyzed by DESeq2 [ 17 ] in R using read counts as input. The p-values were attained by the Wald test and corrected for multiple testing using the Benjamini-Hochberg method. To screen the significant differentially expressed genes, log 2 fold-change (absolute value) greater than 0.5 and p-value less than 0.05 was used as threshold for miRNAs, and log 2 fold-change (absolute value) greater than 1 and p-value less than 0.05 was used as threshold for mRNAs and lncRNAs. We used “ggplot2” package in R to draw the volcano plots and heatmaps. Functional enrichment analysis In order to identify the Gene Ontology (GO) annotations and pathways in which differentially expressed genes were enriched, GO term and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway enrichment analyses were performed using the “culsterProfiler” [ 18 ] package and visualized by bubble charts using the “ggplot2” package. The Benjamini-Hochberg method was used to correct the p-values. Pathway activity analysis Gene set enrichment analysis (GSEA) was used to identify the significantly regulated gene sets in LCCAs, as compared to non-LCCAs, using the “GSEA” function in the “clusterProfiler” package. Further, the GSVA (Gene Set Variation Analysis) [ 19 ] algorithm was used to comprehensively score the enrichment of related gene sets and assess the potential biological alterations between the two groups. The gene sets used for GSEA and GSVA were downloaded from the Molecular Signatures Database (MSigDB) website ( https://www.gsea-msigdb.org/gsea/msigdb ), and the gene set collections of “Hallmarks”, “C2.CP”, “C3”, “C4” and “C5.GO” were applied. A series of gene sets from literature (Supplementary) was also analyzed. ceRNA regulatory network To construct the ceRNA regulatory network, we first predicted and screened the matched differentially expressed lncRNAs and miRNAs using miRcode [ 20 ] database. The candidate target genes of differentially expressed miRNAs were predicted using the following database: miRDB [ 21 ] , miRTarBase [ 22 ] , and TargetScan [ 23 ] . Then, we integrated the interactions between miRNAs and mRNAs or lncRNAs and visualized the ceRNA regulatory network using Cytoscape (version 3.10.0). Defining diversity score to quantify the histological heterogeneity and BCR/TCR clonality We define a diversity score to characterize the histological heterogeneity of the samples, which was calculated using the Shannon-Wiener Index: $${H}^{{\prime }}= -\sum _{i=1}^{R}{p}_{i}\times {log}_{2}{p}_{i}$$ Here, \(R\) is the number of histological subtypes (lepidic, acinar, papillary, micropapillary, solid and complex glandular), i.e., 6, and \({p}_{i}\) refers to the proportion of corresponding subtype in the tumor ( \(0\le {p}_{i}\le 1\) ). The diversity score ranges from 0 to 2.585, with 0 indicating that a specific subtype accounts for 100% of the tumor and the maximum value indicating that all the six subtypes are present evenly. The evenness and clonality score were calculated using the following formula: $$Evenness=\frac{Diversity}{\text{l}\text{o}\text{g}\left(Richness\right)}$$ $$Clonality=1-Evenness$$ Calculating of the intratumor heterogeneity, stemness, and IC50 The “DEPTH” [ 24 ] package is an algorithm to quantify intratumor heterogeneity using the normalized RNA-Seq data as input. The stemness score, denoted as mRNAsi index, was calculated as Zhang [ 25 ] described. Higher mRNAsi scores refer to more aggressive biological processes in cancer stem cells (CSCs) and more tumor dedefferention. The “oncoPredict” package was used for drug response prediction. The reference GDSC2 pre-processed training expression data and CTRP response data were download from https://osf.io/c6tfx/ . The input RNA-Seq data was TPM-normalized and log2-tranformed. Deconvolution of tumor microenvironment (TME) from gene expression profiles ImmunoPhenoScore was calculated using the method proposed by Charoentong [ 26 ] , which was an aggregation of the expression of representative immune-related genes and presence of immune effector cells or suppressor cells. Instead of the proposed scoring system, we adopted the sums of the four components. From the RNA-seq profiles of the samples, we first applied the xCell [ 27 ] algorithm to calculate the panoramic cellular infiltration score of the TME. Then the TIDE [ 28 ] module under Python framework was used to decrypt abnormal T cell function and infiltration related to tumor immune evasion. The normalized and log2-transoformed gene expression data were used as input. Exploring the repertoire profiles of BCR and TCR MiXCR [ 29 ] is an integrated platform for interactive analysis of B cell and T cell receptor repertoire, starting from sequencing data to biological perspective. MiXCR implements a series of highly efficient algorithms aligning the sequencing data to reference V-, D-, J- and C- gene segment database, and yields CDR3 regions of various gene recombination (IGH, IGK and IGL for BCR, TRAD, TRB and TRG for TCR) and the frequency for each clonotype. The downstream processes and analyses of repertoire profiles were performed by “immunarch” [ 30 ] package, which compatibly works fine with MiXCR’s output format. Iterative feature selection To screen the candidate gene associated for prognosis, we applied machine learning methods to iteratively select the important features (candidate genes). Random Forest is commonly-used machine learning algorithm consisting of many decision trees. In this study, we set the number of trees as 100. Least absolute shrinkage and selection operator (LASSO) is a regression method that applied regularization to enhance the accuracy and interpretability of the model. After shrinkage, it chose the non-zero-coefficient variables as the important features. Forward and backward stepwise regression is a linear regression mainly aim to solve multivariate collinearity. It was used to screen variables in multivariate analyses, achieving the minimum AIC (Akaike Information Criterion). Statistical analysis All statistical analyses were conducted using R software (version 4.1.0). Between-group comparisons were performed using Wilcoxon rank-sum test for continuous, Fisher’s exact test or Chi-squared test for discrete variables, using the R function “wilcox.test”, “fisher.test” and “chisq.test”, respectively. Distribution of continuous variables was checked by Kolmogorov-Smirnov test. Survival analyses including log-rank tests, Kaplan-Meier curve, univariate and multivariate Cox regression analyses were performed using the R package “survival”. A two-sided p-value of less than 0.05 was considered statistically significant. Results Baseline of the study cohorts The clinicopathological characteristics of patients are presented in Table 1 . A total of 165 patients with LCCA and 201 patients with non-LCCA that matched for TNM stage and histology were enrolled in this study. The median follow-up was 17 (IQR 6 to 43) months. The median age was 60 (IQR 53 to 65) years in LCCA cohort. In consistence with literature, LCCA was found more frequently in men (76%). The predominant histological type was invasive non-mucinous adenocarcinoma (INMA) in 75% of the LCCAs, followed by 13% of minimally invasive adenocarcinoma (MIA), 5% of squamous cell carcinoma (SCC) and 4% of invasive mucinous adenocarcinoma (IMA). There were 5 cases mixed with INMA and IMA, and one case of large cell carcinoma (LCC). The clinicopathological features between LCCA and non-LCCA cohort were similar, except that LCCAs were prone to occur more in men (76% vs 50%, p < 0.001) and associated with less visceral pleura invasion (VPI, 6% vs 18%, p = 0.00118) or lympho-vascular invasion (LVI, 8% vs 23%, p < 0.001) in our study. EGFR and KRAS mutation were found positive in 39% and 9% of the cases in LCCA cohort, respectively. Table 1 Baseline characteristics of all patients. Variable LCCA (n = 165, 45.1%) non-LCCA (n = 201, 54.9%) p value Age, median (IQR) 60 (53–65) 60 (53–65) 0.278 Gender, (%) < 0.001 Male 125 (76%) 100 (50%) Female 40 (24%) 101 (50%) Smoking status, (%) 0.333 Never 107 (65%) 141 (70%) Current or ever 58 (35%) 60 (30%) Histological subtype, (%) 0.597 IMA 7 (4%) 8 (4%) INMA 123 (75%) 163 (82%) LCC 1 (1%) 0 (0%) MIA 21 (13%) 19 (9%) Mixed 5 (3%) 3 (1%) SCC 8 (5%) 8 (4%) Grade, (%) 0.135 1 20 (12%) 12 (6%) 2 87 (53%) 122 (61%) 3 28 (17%) 37 (18%) Unknown 30 (18%) 30 (15%) VPI, (%) 0.00118 Positive 10 (6%) 36 (18%) Negative 155 (94%) 165 (82%) STAS, (%) 0.253 Positive 18 (11%) 14 (7%) Negative 147 (89%) 187 (93%) LVI, (%) < 0.001 Positive 13 (8%) 46 (23%) Negative 152 (92%) 155 (77%) pT, median (IQR) 2 (1.4–2.5) 2.1 (1.6–3) 0.0176 pN, (%) 0.0785 N0 162 (98%) 189 (94%) N1 0 (0%) 5 (2%) N2 3 (2%) 7 (3%) Stage, (%) 0.245 IA 131 (79%) 141 (70%) IB 23 (14%) 37 (18%) II 7 (4%) 15 (7%) III 4 (2%) 8 (4%) EGFR, (%) 0.359 Positive 64 (39%) 86 (43%) Negative 43 (26%) 40 (20%) Unknown 58 (35%) 75 (37%) KRAS, (%) 0.135 Positive 15 (9%) 8 (4%) Negative 92 (56%) 118 (59%) Unknown 58 (35%) 75 (37%) Data are presented as interquartile range (IQR) or numbers (frequency). LCCA: lung cancer associated with cystic airspaces; IMA: invasive mucinous adenocarcinoma; INMA: invasive non-mucinous adenocarcinoma; MIA: minimally invasive adenocarcinoma; Mixed: contained IMA and INMA; SCC: lung squamous cell carcinoma; LCC: large cell carcinoma; STAS: spread through air space; VPI: visceral pleura invasion; LVI: lympho-vascular invasion; EGFR: epidermal growth factor receptor; KRAS: Kirsten rats arcomaviral oncogene homolog. p value < 0.05 marked statistical significance. LCCAs were associated with histological heterogeneity and poor PFS As regarding to the survival stratification, combinations of histological subtypes (lepidic, acinar, papillary, micropapillary, solid and complex glandular) and predominant proportion were applied in the newly proposed grading system for invasive adenocarcinoma of LUAD in 2020 [ 31 ] . For both the LCCA and non-LCCA cohorts, we determined the proportion of each histological subtype component for cases with INMA (Fig. 1A). Next, we defined a metric, known as the diversity score, to measure the histological heterogeneity of LNMA samples (see the Methods section). The diversity scores in LCCAs were significantly higher ( p = 0.022) than in non-LCCAs (Fig. 1B). We further found patients of age less than 60 ( p < 0.001), never smoking ( p < 0.001), negative VPI ( p = 0.0084) and negative LVI ( p = 0.017) in LCCAs presented more significant histological heterogeneity (Fig. 1C-F), while no differences were found among groups of gender and tumor size (Figure S2 A-B). To investigate whether the presenting of cystic airspaces was associated with prognosis of early-stage lung cancer, Kaplan-Meier curves were plotted to compare between LCCA cohort and non-LCCA cohort. Although not significant, it was suggested that LCCAs and a high diversity score were associated with poorer progression-free survival (PFS) than that of non-LCCAs. Yet in the stratification analysis, patients of age more than 60 (HR 5.36 [1.03–27.83], p = 0.046) and positive VPI (HR 17.46 [1.81–168.68], p = 0.013) were indicated to have significantly less PFS in LCCAs. Furthermore, LCCA (HR 6.13 [1.26–29.80], p = 0.0025), micropapillary subtype fraction higher than 0.15 (HR 7.25 [1.48–36.37], p = 0.016) and VPI positive (HR 8.94 [1.90–42.02], p = 0.006) were proven to be the independent risk factors for PFS in the univariate and multivatiate Cox regression analyses (Table 2 ). As for the LCCA cohort alone, VPI positive and STAS positive were the significant predictors for PFS (Table S1 ). Table 2 Univariate and multivariate Cox regression analyses of LCCA and non-LCCA Variable Total (N) Univariate analysis Multivariate analysis Coefficient HR (95% CI) P value HR (95% CI) P value Group 294 LCCA 166 Reference Reference non-LCCA 128 3.3(0.78–14.17) 0.102 6.13 (1.3–29.8) 0.025 1.81 MPP 294 Low 263 Reference Reference High 31 8.56(2.14–34.23) 0.002 7.25 (1.5–36.4) 0.016 1.98 VPI 294 Negative 252 Reference Reference Positive 42 4.44 (1.10–17.9) 0.036 8.94 (1.9–42.0) 0.006 2.19 STAS 294 Negative 263 Reference Reference Positive 31 8.7 (2.04–37.16) 0.003 3.6 (0.65–19.8) 0.139 1.28 LCCA: lung cancer associated with cystic airspaces; MPP: Micropapillary; STAS: spread through air space; VPI: visceral pleura invasion; LVI: lympho-vascular invasion; HR: hazard ratio. p value < 0.05 marked statistical significance. Differentially expressed mRNA, lncRNA and miRNA profiles by bulk RNA-seq To further characterize the transcriptome alterations underlying patients with LCCA and non-LCCA, we collected 8 tumor samples from LCCA and 9 from non-LCCA, as well as 10 adjacent non-neoplastic tissues (ANT) from the above patients (Fig. 2A). Bulk RNA-sequencing (RNA-seq) was performed for each group. As expected, the samples were separated into three clusters in the principal component analysis (PCA) (Fig. 2B), though the differences between LCCA and non-LCCA on PC1 was not very obvious, as compared to ANTs. To profile the genomic features of LCCAs, we first performed differential gene expression analysis between LCCAs and ANTs. A total of 2383 mRNAs (1581 up-regulated and 802 down-regulated), 1151 lncRNAs (706 up-regulated and 445 down-regulated), and 68 miRNAs (42 up-regulated and 26 down-regulated) were identified (Fig. 2C-E). A total of 27 differentially expressed lncRNAs and three miRNAs (miR-142, miR-200b and miR-490) were matched into 38 miRNA-lncRNA interactions, whereas the three miRNAs targeted to 200 candidate mRNAs to form 205 miRNA-lncRNA pairs. Of the candidate mRNAs, 16 were fund significantly differentially expressed (Fig. 2F). Finally, a network of 2357 competitive lncRNA-miRNA-mRNA links was constructed (Fig. 2G). Next, we compared gene expression between LCCAs and non-LCCAs. A total of 970 up-regulated and 275 down-regulated genes were identified (Fig. 3A, Figure S3 A-C). GO and KEGG pathway enrichment analyses revealed that the differentially up-regulated genes were remarkably associated with cell proliferation, migration, communication, ion channel or receptor activities and angiogenesis, et al (Fig. 3B). The related GO terms or pathways were further confirmed significantly up-regulated by GSEA framework (Fig. 3C-D, Figure S3 D-F). We also found that molecular signatures associated with angiogenesis mediated by macrophage [ 32 ] , regulation of stromal components [ 33 ] , infiltration of cancer-related fibroblasts (CAFs) [ 34 ] and myeloid-derived suppressor cells (MDSCs) [ 28 ] were up-regulated in LCCAs (Fig. 3E-F, Figure S3 G-I). LCCA displayed distinct biochemical characteristics GSVA was used to comprehensively score the enrichment of related GO terms and pathways of each sample, and differentially regulated gene sets were assessed between the two cohorts (Fig. 4A, Supplementary Data 1). Several GO terms or pathways associate with “regulation of lymphangiogenesis”, “woulding healing involved in inflammatory response”, “actin filament reorganization”, and “leukotriene metabolic pathway”, were significantly up-regulated in LCCAs. We also compared the gene sets from literatures. The samples of LCCA displayed higher levels of gene sets associated with epithelial-to-mesenchymal transition (EMT) [ 35 ] , angiogenesis [ 32 , 36 ] and distal metastasis [ 37 ] , which may explain their poor prognosis (Fig. 4B). Moreover, the pathways related to ferroptosis, oxeiptosis and intrinsic apoptosis, were significantly up-regulated, while that of extrinsic apoptosis were down-regulated, indicating a distinct death episode was orchestrated in LCCAs. In addition, biochemical processes associated with galactose metabolism, prostaglandin biosynthesis and glycosaminoglycan degradation were enhanced, as compared to non-LCCAs. To elucidate the tumor biological behavior of LCCA, we calculate the scores of intratumor genetic heterogeneity and stemness of each sample. Interestingly, the intratumor heterogeneity (ITH) scores were significantly lower ( p = 0.016) in LCCAs (Fig. 4C), inconsistent with their higher histological heterogeneity. The stemness score between two groups were similar (Fig. 4D). We further used “oncoPredict” to predict the IC50 of each sample to a series of agents. It was suggested that samples in the LCCA group displayed similar sensitivity to the chemical agents that were the most commonly used in adjuvant chemotherapy settings for NSCLC (non-small cell lung cancer), including docetaxel, gemcitabine and cisplatin (Fig. 4E-F, Figure S4 A). As the most common oncogenic drivers in NSCLC, EGFR has become an important therapeutic target for the treatment of these tumors. Samples in the LCCA group seemed to be more resistant to gefitinib and afatinib (Fig. 4G, Figure S4 B), though not significant. However, they showed comparable IC50s to Osimertinib, as compared to that of non-LCCA (Fig. 4H). LCCAs exhibited dysfunctional immune response and distinct BCR clonality To investigate the differences in the immune environment between LCCA and non-LCCA group, we first quantified the immune-related gene sets by GSVA. We found that gene sets associated with T cell exhaustion [ 28 ] , M2 macrophage polarization [ 38 ] , MDSCs [ 28 ] and CAF [ 34 ] infiltration were significantly up-regulated in LCCAs (Fig. 5A), suggesting a more immunosuppressive environment. Next, we calculate the ImmunoPhenoScore (IPS), which showed a trend of decline in the LCCA group (Fig. 5B). To further deconvolute the TME, we used the xCell algorithm to quantify the various cellular components (Fig. 5C). It was showed that macrophages in LCCAs manifested an anti-inflammatory M2 phenotype, not M1 phenotype, as indicated by the dramatically higher scores in M2 macrophages ( p < 0.001). Next, we explored the dysfunctional and exclusive status of TME using the TIDE algorithm. We found that cancer-associated fibroblast (CAF) scores were higher in LCCAs ( p = 0.046), resulting a trend of increase in dysfunction scores and comprehensive TIDE scores (Fig. 5D). The scores related to cytotoxic T cells and exclusion scores were comparable. These data suggested LCCA displayed a more immunosuppressive profiles in the TME, which may be attributed to polarization of M2-macrophage and infiltration of suppressors. To investigate the expansion of B cell and T cell populations in LCCA and non-LCCA, we performed clonality analyses using the MiXCR software. We used the clean fastq data as input to identify CDR3 sequencing of BCRs/TCRs. A total of 147,752 BCR and 1,092 TCR clonotypes were identified in our study. A metric, denoted as richness, was applied to measure the clonal abundance, which referred to the total number of unique clonotypes in the B/T cell population. We compared the BCR and TCR richness between samples of LCCA and non-LCCA (Fig. 5E-F). It showed that LCCA had lower ( p = 0.012) BCR richness, while TCR in the two groups displayed similar clonal abundance. Next, we evaluated the BCR/TCR repertoire based on the fraction of the unique clonotypes, calculating the Shannon-Wiener Index mentioned above. Three metrics were adopted. Diversity score, i.e., Shannon index, represents the heterogeneity of BCR/TCR clonotype. Clonality score, in opposed to evenness score, indicates the potential expansion or predominance of certain B/T cell clones. There were no significant differences between LCCA and non-LCCA, though a trend of enhancement of BCR diversity in non-LCCA (Fig. 5G-H). We then explore the overlap clonotypes of BCR/TCR repertoire. A total of 42,639 clonotypes among 348 shared pairs were identified. Figure 5I revealed the shared BCR/TCR clonotypes among all the samples. Almost all the shared clones came from CDR3 sequences of BCR, except for only one TCR clonotype of TRAD shared between groups. And it seemed that LCCA tended to have less shared clonotypes of high abundance (Fig. 5J). Moreover, the between-group and within-group shared clonotypes exhibited comparable number and distribution (Fig. 5K). These results suggested that repertoire diversity and clonal expansion of B cells, instead of T cells, were impaired in LCCA. Cystic-specific genes predicted overall survival (OS) in LUAD Whether the differentially expressed in LCCA, as compared to non-LCCA, were associated with the prognosis of LUAD. First, we screened the LCCA-related genes at a threshold of absolute log 2 fold-change more than 1 and adjusted p-value less than 0.05. Then, the raw count expression data, corresponding clinical and survival data of LUAD were downloaded from the TCGA data portal. Finally, a list of 214 LCCA-related genes and 412 LUAD cases were identified, and the cases was randomly split into training set and validation set (see the Methods section). The clinical characteristics of training set and validation set were listed in Table 3. To screen for candidate prognosis-related genes, Random Forest (RF) algorithm, LASSO regression (Figure S5 A-B) and stepwise regression were performed, and a union of the above results enrolled 23 candidate genes (Fig. 6A). In the univariate and multivariate Cox regression analyses, four genes ( KCNK3 , NRN1 , PARVB and TRHDE-AS1 ) were finally identified to be the independently predictive factor for overall survival of LUAD (Fig. 6B, Figure S5 C-F). Then, we constructed a cystic-specific risk score (CSRS, \(Risk Score= -0.249\times KCNK3-0.321\times NRN1+0.64\times PARVB+0.571\times TRHDE-AS1\) ). Kaplan-Meier analysis revealed that CSRS was dramatically associated with poor OS (Fig. 6C, Figure S5 G, p < 0.0001). Multivariate Cox regression analysis suggested that CSRS (HR 3.33 [2.06-5.4], p < 0.001) was an independent predictor for OS (Figure S6 A-B). Further, we built a CSRS-based nomogram to predict the OS of LUAD (Fig. 6D). The nomogram model showed good performance in both the training set and validation set (Fig. 6E, Figure S5 H-I) with a c-index of 0.707 (95% CI 0.682–0.732) in the whole population. The 1-year, 3-year and 5-year AUCs of ROC was 0.79 (0.70–0.87), 0.75 (0.67–0.83) and 0.78 (0.69–0.88), respectively. Moreover, DCA showed that the nomogram, other than CSRS alone yielded better net benefit (Fig. 6F, Figure S5 J-K). Discussion Lung cancer associated with cystic airspaces (LCCA) is a clinically rare type of lung cancer. Although many previous reports have delineated the clinicopathological and radiographic features, there is still a gap of bedside-to-bench recognition of LCCAs, from morphological and molecular pathological mechanisms to biological and genetic traits. To best of our knowledge, this is the first study to investigate the profiles of molecule regulatory network, as well as tumor and immune microenvironment at gene level. Our results indicate that LCCA and non-LCCA had substantial differences in tumor cell biology and immune microenvironment. Specifically, LCCAs are more inclined to exhibit migratory and immunosuppressive nature, and distinct BCR clonality and cell death orchestration. Moreover, a list of cystic-specific genes could predict the overall survival of LUAD. In line with previous study [3, 39] , we found that LCCA and micropapillary component were the risk factors for recurrence in early-stage lung cancer. It is reported that micropapillary component is associated with higher tumor mutational burden, genome alteration and copy number amplifications, whole-genome doubling rate [40] . In our study, we comprehensively score all the components of histological subtypes, and we found that LCCAs displayed higher histological heterogeneity and a trend for BCR diversity, though lower ITH scores inferred by “DEPTH” algorithm. The histological heterogeneity was more prominent in younger patients and never smokers. It is well-established that lymphovascular invasion (VPI) and tumor spread through air spaces (STAS) are unfavorable prognostic factor in early-stage lung cancer [41, 42] . LCCAs seemed to be more aggressive than non-LCCAs, in the settings of positive VPI or STAS in our study. Indeed, GSEA and GSVA analyses suggest that pathways associated with EMT, angiogenesis and distal metastasis were significantly enriched in LCCA. As suggested in our study, it might be related to the significant degradation of glycosaminoglycan, a component of extracellular matrix, which play an important role in cancer metastasis [43] . These functional traits are consistent with the above clinical manifestations. We found LCCA exhibited distinct cell death pattern and metabolic profiles. The ferroptosis pathway was significantly up-regulated in LCCA. Ferroptosis is a manifestation of the imbalance between oxidative stress and antioxidant damage [44] . Besides a unique cell death pattern, it plays a dual role in anti-tumor immunity, which is related to the composition and function of immune cells in the tumor microenvironment [45] . During ferroptosis, the lipid peroxidation products can serve as an important "find me" signal to recruit antigen-presenting cells and other immune cells to the tumor microenvironment. However, some immune reactive cells are also sensitive to ferroptosis. Ma found that CD36-mediated fatty acid uptake induced ferroptosis of CD8+ T cells and weakened their anti-tumor activity, leading to tumor growth [46] . Moreover, there might also be a preference to activation of intrinsic apoptosis and suppression of extrinsic apoptosis in LCCA according our results. It is well-accepted that deregulation of apoptosis can lead to the development of different diseases and it can also influence the treatment of some of these diseases. It is necessary to understand the mechanisms by which way of apoptosis occurs and interacts with other biological systems. Therefore, further work is warranted to investigate the potential mechanisms that orchestrate the polarity of apoptotic pattern. Immune evasion is a hallmark of malignancy. In fact, tumor cells have developed different mechanisms to dysfunction and circumvent the anti-tumor immune response [47, 48] . Immunosuppressive components in the TME, including myeloid-derived suppressive cells (MDSCs), tumor-associated macrophages (TAMs), regulatory T cells (Tregs), cancer-associated fibroblasts (CAFs), tumor-associated neutrophils, and tumor-associated dendritic cells, created a disordered niche where T cell reside or be excluded, and are critical factors correlated with immune resistance [49] . We found that gene sets associated TAMs, CAFs and MDSCs were up-regulated in LCCAs. Consistent with suppressor infiltration from inferred gene sets, the TME deconvolution analysis show high level of M2 macrophage polarization and CAFs activity. Importantly, our study also highlights the impaired B cell roles in LCCA. The BCR/TCR repertoire identified significantly more unique BCR clonotypes the TCR clonotypes. However, a lowered BCR richness, clonality and high-abundance shared clonotypes were found in LCCA, which may further deteriorate the hormonal anti-tumor immunity. It is proposed that a “check-valve” mechanism plays a role in the formation of airspaces. Under this circumstance, as the gas continues increase, the alveolar wall was damaged by both tumor cells and high pressure, and resulting the destruction of elastic fibers and induction of stromal hyperplasia, which may pave the way for infiltration and activation of suppressors and hamper B cell clonal expansion. It should also be noted that the infer of BCR/TCR repertoire was based on alignment of RNA-seq data to a known VDJ sequence database. A more profound such as BCR/TCR sequencing at single cell level will provide a comprehensive perspective of BCR/TCR repertoire with huge V-, D-, J- section combinations. The limitation of this study came from the small sample size of sequenced, due to its clinical rarity, and the non-LCCA cohort was matched for TNM stage and histology, which may comprise selection bias. Yet, we evaluated the molecule regulatory network and immune profiles of LCCAs at gene level for the first time, which show a distinct biological trait as compared to non-LCCAs. Deeper sequencing of large sample size and inherent molecular mechanisms were warranted further investigation. Declarations Ethics Statement The studies involving human participants were reviewed and approved by the Ethics Committee of the First Affiliated Hospital of Guangzhou Medical University and were carried out in accordance with the World Medical Association’s Declaration of Helsinki. The patients provided their written informed consent to participate in this study. Authorship contribution statement Q.L, Z.X., H.Y., and L.J. contributed study design, data analysis and paper writing. C.R., X.K. and C.W.P. collected samples and generated data. W.W. and Z.X. reviewed the CT images and D.Y. examined the results. C.F, W.W, H.Z.X and H S H provided study materials or patients. All authors contributed to the article and approved the submitted version. Competing Interests The authors declare no competing interests. Funding This work was supported by the Basic Research Program of Guangzhou Municipal Science and Technology Bureau (No: 2060206). References ZHU H, ZHANG L, HUANG Z, et al. Lung adenocarcinoma associated with cystic airspaces: Predictive value of CT features in assessing pathologic invasiveness [J]. Eur J Radiol, 2023, 165: 110947. MA Z, WANG S, ZHU H, et al. Comprehensive investigation of lung cancer associated with cystic airspaces: predictive value of morphology [J]. Eur J Cardiothorac Surg, 2022, 62(5). SHEN Y, ZHANG Y, GUO Y, et al. Prognosis of lung cancer associated with cystic airspaces: A propensity score matching analysis [J]. Lung Cancer, 2021, 159: 111-6. DETTERBECK F C, KUMBASAR U, LI A X, et al. Lung cancer with air lucency: a systematic review and clinical management guide [J]. J Thorac Dis, 2023, 15(2): 731-46. TAN Y, GAO J, WU C, et al. CT Characteristics and Pathologic Basis of Solitary Cystic Lung Cancer [J]. Radiology, 2019, 291(2): 495-501. HANSELL D M, BANKIER A A, MACMAHON H, et al. Fleischner Society: glossary of terms for thoracic imaging [J]. Radiology, 2008, 246(3): 697-722. FAROOQI A O, CHAM M, ZHANG L, et al. Lung cancer associated with cystic airspaces [J]. AJR Am J Roentgenol, 2012, 199(4): 781-6. FINTELMANN F J, BRINKMANN J K, JECK W R, et al. Lung Cancers Associated With Cystic Airspaces: Natural History, Pathologic Correlation, and Mutational Analysis [J]. J Thorac Imaging, 2017, 32(3): 176-88. MASCALCHI M, ATTINA D, BERTELLI E, et al. Lung cancer associated with cystic airspaces [J]. J Comput Assist Tomogr, 2015, 39(1): 102-8. TOYOKAWA G, SHIMOKAWA M, KOZUMA Y, et al. Invasive features of small-sized lung adenocarcinoma adjoining emphysematous bullae [J]. Eur J Cardiothorac Surg, 2018, 53(2): 372-8. WATANABE Y, KUSUMOTO M, YOSHIDA A, et al. Cavity Wall Thickness in Solitary Cavitary Lung Adenocarcinomas Is a Prognostic Indicator [J]. Ann Thorac Surg, 2016, 102(6): 1863-71. SHEN Y, XU X, ZHANG Y, et al. Lung cancers associated with cystic airspaces: CT features and pathologic correlation [J]. Lung Cancer, 2019, 135: 110-5. ANDREWS S. FastQC: A Quality Control Tool for High Throughput Sequence Data [Z]. 2010 BOLGER A M, LOHSE M, USADEL B. Trimmomatic: a flexible trimmer for Illumina sequence data [J]. Bioinformatics, 2014, 30(15): 2114-20. KIM D, PAGGI J M, PARK C, et al. Graph-based genome alignment and genotyping with HISAT2 and HISAT-genotype [J]. Nat Biotechnol, 2019, 37(8): 907-15. ANDERS S, PYL P T, HUBER W. HTSeq--a Python framework to work with high-throughput sequencing data [J]. Bioinformatics, 2015, 31(2): 166-9. LOVE M I, HUBER W, ANDERS S. Moderated estimation of fold change and dispersion for RNA-seq data with DESeq2 [J]. Genome Biol, 2014, 15(12): 550. WU T, HU E, XU S, et al. clusterProfiler 4.0: A universal enrichment tool for interpreting omics data [J]. Innovation (Camb), 2021, 2(3): 100141. HANZELMANN S, CASTELO R, GUINNEY J. GSVA: gene set variation analysis for microarray and RNA-seq data [J]. BMC Bioinformatics, 2013, 14: 7. JEGGARI A, MARKS D S, LARSSON E. miRcode: a map of putative microRNA target sites in the long non-coding transcriptome [J]. Bioinformatics, 2012, 28(15): 2062-3. WONG N, WANG X. miRDB: an online resource for microRNA target prediction and functional annotations [J]. Nucleic Acids Res, 2015, 43(Database issue): D146-52. HUANG H Y, LIN Y C, CUI S, et al. miRTarBase update 2022: an informative resource for experimentally validated miRNA-target interactions [J]. Nucleic Acids Res, 2022, 50(D1): D222-d30. AGARWAL V, BELL G W, NAM J W, et al. Predicting effective microRNA target sites in mammalian mRNAs [J]. Elife, 2015, 4. LI M, ZHANG Z, LI L, et al. An algorithm to quantify intratumor heterogeneity based on alterations of gene expression profiles [J]. Commun Biol, 2020, 3(1): 505. ZHANG Y, TSENG J T, LIEN I C, et al. mRNAsi Index: Machine Learning in Mining Lung Adenocarcinoma Stem Cell Biomarkers [J]. Genes (Basel), 2020, 11(3). CHAROENTONG P, FINOTELLO F, ANGELOVA M, et al. Pan-cancer Immunogenomic Analyses Reveal Genotype-Immunophenotype Relationships and Predictors of Response to Checkpoint Blockade [J]. Cell Rep, 2017, 18(1): 248-62. ARAN D. Cell-Type Enrichment Analysis of Bulk Transcriptomes Using xCell [J]. Methods Mol Biol, 2020, 2120: 263-76. JIANG P, GU S, PAN D, et al. Signatures of T cell dysfunction and exclusion predict cancer immunotherapy response [J]. Nat Med, 2018, 24(10): 1550-8. BOLOTIN D A, POSLAVSKY S, DAVYDOV A N, et al. Antigen receptor repertoire profiling from RNA-seq data [J]. Nat Biotechnol, 2017, 35(10): 908-11. TEAM I. immunarch: An R Package for Painless Analysis of Large-Scale Immune Repertoire Data [Z]. 2019 MOREIRA A L, OCAMPO P S S, XIA Y, et al. A Grading System for Invasive Pulmonary Adenocarcinoma: A Proposal From the International Association for the Study of Lung Cancer Pathology Committee [J]. J Thorac Oncol, 2020, 15(10): 1599-610. MASIERO M, SIMOES F C, HAN H D, et al. A core human primary tumor angiogenesis signature identifies the endothelial orphan receptor ELTD1 as a key regulator of angiogenesis [J]. Cancer Cell, 2013, 24(2): 229-41. ZENG D, LI M, ZHOU R, et al. Tumor Microenvironment Characterization in Gastric Cancer Identifies Prognostic and Immunotherapeutically Relevant Gene Signatures [J]. Cancer Immunol Res, 2019, 7(5): 737-50. TIROSH I, IZAR B, PRAKADAN S M, et al. Dissecting the multicellular ecosystem of metastatic melanoma by single-cell RNA-seq [J]. Science, 2016, 352(6282): 189-96. DAMRAUER J S, HOADLEY K A, CHISM D D, et al. Intrinsic subtypes of high-grade bladder cancer reflect the hallmarks of breast cancer biology [J]. Proc Natl Acad Sci U S A, 2014, 111(8): 3110-5. SJODAHL G, LAUSS M, LOVGREN K, et al. A molecular taxonomy for urothelial carcinoma [J]. Clin Cancer Res, 2012, 18(12): 3377-86. KANG Y, SIEGEL P M, SHU W, et al. A multigenic program mediating breast cancer metastasis to bone [J]. Cancer Cell, 2003, 3(6): 537-49. AZIZI E, CARR A J, PLITAS G, et al. Single-Cell Map of Diverse Immune Phenotypes in the Breast Tumor Microenvironment [J]. Cell, 2018, 174(5): 1293-308 e36. LI Y, BYUN A J, CHOE J K, et al. Micropapillary and Solid Histologic Patterns in N1 and N2 Lymph Node Metastases Are Independent Factors of Poor Prognosis in Patients With Stages II to III Lung Adenocarcinoma [J]. J Thorac Oncol, 2023, 18(5): 608-19. CASO R, SANCHEZ-VEGA F, TAN K S, et al. The Underlying Tumor Genomics of Predominant Histologic Subtypes in Lung Adenocarcinoma [J]. J Thorac Oncol, 2020, 15(12): 1844-56. URUGA H, FUJII T, FUJIMORI S, et al. Semiquantitative Assessment of Tumor Spread through Air Spaces (STAS) in Early-Stage Lung Adenocarcinomas [J]. J Thorac Oncol, 2017, 12(7): 1046-51. VAAHTOMERI K, ALITALO K. Lymphatic Vessels in Tumor Dissemination versus Immunotherapy [J]. Cancer Res, 2020, 80(17): 3463-5. OLIVEIRA-FERRER L, LEGLER K, MILDE-LANGOSCH K. Role of protein glycosylation in cancer metastasis [J]. Semin Cancer Biol, 2017, 44: 141-52. SNYDER A G, HUBBARD N W, MESSMER M N, et al. Intratumoral activation of the necroptotic pathway components RIPK1 and RIPK3 potentiates antitumor immunity [J]. Sci Immunol, 2019, 4(36). LEI G, ZHUANG L, GAN B. Targeting ferroptosis as a vulnerability in cancer [J]. Nat Rev Cancer, 2022, 22(7): 381-96. KIM R, HASHIMOTO A, MARKOSYAN N, et al. Ferroptosis of tumour neutrophils causes immune suppression in cancer [J]. Nature, 2022, 612(7939): 338-46. VINAY D S, RYAN E P, PAWELEC G, et al. Immune evasion in cancer: Mechanistic basis and therapeutic strategies [J]. Semin Cancer Biol, 2015, 35 Suppl: S185-S98. SHARMA P, HU-LIESKOVAN S, WARGO J A, et al. Primary, Adaptive, and Acquired Resistance to Cancer Immunotherapy [J]. Cell, 2017, 168(4): 707-23. TIE Y, TANG F, WEI Y Q, et al. Immunosuppressive cells in cancer: mechanisms and potential therapeutic targets [J]. J Hematol Oncol, 2022, 15(1): 61. Additional Declarations No competing interests reported. Supplementary Files FigureS1.png FigureS2.png FigureS3.png FigureS4.png FigureS5.png FigureS6.png SupplementaryData1msigdbgsvadegsresults.csv SupplementaryData2diygenesetdegsresults.csv TableS1.docx Cite Share Download PDF Status: Posted Version 1 posted You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-3448810","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":240349468,"identity":"6a8074cf-485f-41c7-8ded-5cb18efde649","order_by":0,"name":"Xiang Zheng","email":"","orcid":"","institution":"The First Affiliated Hospital of Guangzhou Medical University","correspondingAuthor":false,"submittingAuthor":false,"prefix":"","firstName":"Xiang","middleName":"","lastName":"Zheng","suffix":""},{"id":240349469,"identity":"c9ee89e0-89bc-4f27-a39a-11b85bdcc402","order_by":1,"name":"Li Qiu","email":"","orcid":"","institution":"The First Affiliated Hospital of Guangzhou Medical University","correspondingAuthor":false,"submittingAuthor":false,"prefix":"","firstName":"Li","middleName":"","lastName":"Qiu","suffix":""},{"id":240349470,"identity":"7eed095f-2e3a-4d3c-927c-d91c31cc7668","order_by":2,"name":"Ying Huang","email":"","orcid":"","institution":"The First Affiliated Hospital of Guangzhou Medical University","correspondingAuthor":false,"submittingAuthor":false,"prefix":"","firstName":"Ying","middleName":"","lastName":"Huang","suffix":""},{"id":240349471,"identity":"d23d97ba-3881-49f2-b355-d6a37200adc8","order_by":3,"name":"Ran Cheng","email":"","orcid":"","institution":"The First Affiliated Hospital of Guangzhou Medical University","correspondingAuthor":false,"submittingAuthor":false,"prefix":"","firstName":"Ran","middleName":"","lastName":"Cheng","suffix":""},{"id":240349472,"identity":"b433da16-34cd-41c6-aa8a-448e2617d1e6","order_by":4,"name":"Si Huang","email":"","orcid":"","institution":"The First Affiliated Hospital of Guangzhou Medical University","correspondingAuthor":false,"submittingAuthor":false,"prefix":"","firstName":"Si","middleName":"","lastName":"Huang","suffix":""},{"id":240349473,"identity":"9b8b6a37-6695-4513-8b7c-1103f0ad0c57","order_by":5,"name":"Ke Xu","email":"","orcid":"","institution":"Tongji University","correspondingAuthor":false,"submittingAuthor":false,"prefix":"","firstName":"Ke","middleName":"","lastName":"Xu","suffix":""},{"id":240349474,"identity":"561889af-1beb-426f-bf48-0c211dfc1f25","order_by":6,"name":"Wei Cai","email":"","orcid":"","institution":"Shantou Central Hospital","correspondingAuthor":false,"submittingAuthor":false,"prefix":"","firstName":"Wei","middleName":"","lastName":"Cai","suffix":""},{"id":240349475,"identity":"39a3b40c-ed5b-4afa-8eee-4c13695b1ba5","order_by":7,"name":"Yu Deng","email":"","orcid":"","institution":"The First Affiliated Hospital of Guangzhou Medical University","correspondingAuthor":false,"submittingAuthor":false,"prefix":"","firstName":"Yu","middleName":"","lastName":"Deng","suffix":""},{"id":240349476,"identity":"3a7a81b2-798c-42da-ae94-38599c2bed7e","order_by":8,"name":"Wei Wang","email":"","orcid":"","institution":"The First Affiliated Hospital of Guangzhou Medical University","correspondingAuthor":false,"submittingAuthor":false,"prefix":"","firstName":"Wei","middleName":"","lastName":"Wang","suffix":""},{"id":240349477,"identity":"57d2351c-7abf-4328-a29f-4345b3efc417","order_by":9,"name":"Xi Zhong","email":"","orcid":"","institution":"Affiliated Cancer Hospital \u0026 Institute of Guangzhou Medical University","correspondingAuthor":false,"submittingAuthor":false,"prefix":"","firstName":"Xi","middleName":"","lastName":"Zhong","suffix":""},{"id":240349478,"identity":"c1f18ebe-04bb-400c-a482-03fa63ec4466","order_by":10,"name":"Fei Cui","email":"","orcid":"","institution":"The First Affiliated Hospital of Guangzhou Medical University","correspondingAuthor":false,"submittingAuthor":false,"prefix":"","firstName":"Fei","middleName":"","lastName":"Cui","suffix":""},{"id":240349479,"identity":"3c5f200e-785a-4ea6-8894-9c0811512df6","order_by":11,"name":"Zhexue Hao","email":"","orcid":"","institution":"The First Affiliated Hospital of Guangzhou Medical University","correspondingAuthor":false,"submittingAuthor":false,"prefix":"","firstName":"Zhexue","middleName":"","lastName":"Hao","suffix":""},{"id":240349480,"identity":"ad9ba049-18c9-4e38-8eb8-33aeb1c77813","order_by":12,"name":"Jun Liu","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAAwUlEQVRIiWNgGAWjYBACPoYDDIwNDAxybOzNB4jTwgbVYszHcyyBWC0MYC2J8yRyFIjUwnjGTHLGr8PpbQw5DAw/KrYRY8uxNMmNfYdz2xjOHmDsOXObGC2Hj9182APUwtiXwMzYRpSWg20gLelszDwGxGoB2rLhx+EENjbitRxL/zmzId2wjYct4SBRfuGXOGNs2PPHWl5+/uODD35UEKGFQeIAMGLamsHsA0SoB1nTACT+1BGneBSMglEwCkYmAACNP0D5qKA4xAAAAABJRU5ErkJggg==","orcid":"","institution":"The First Clinical Medical College of Henan University","correspondingAuthor":true,"submittingAuthor":false,"prefix":"","firstName":"Jun","middleName":"","lastName":"Liu","suffix":""}],"badges":[],"createdAt":"2023-10-15 13:59:12","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-3448810/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-3448810/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":44803874,"identity":"3924dbb6-90ce-4cd4-9762-64144dd7675b","added_by":"auto","created_at":"2023-10-17 17:19:34","extension":"jpg","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":1705588,"visible":true,"origin":"","legend":"\u003cp\u003eLegend not included with this version\u003c/p\u003e","description":"","filename":"Figure1.jpg","url":"https://assets-eu.researchsquare.com/files/rs-3448810/v1/4e509a7edafc2876c546ab24.jpg"},{"id":44803873,"identity":"28463ee3-4c3c-4e82-9ccd-fc98c5de5f89","added_by":"auto","created_at":"2023-10-17 17:19:33","extension":"jpg","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":1562942,"visible":true,"origin":"","legend":"\u003cp\u003eLegend not included with this version\u003c/p\u003e","description":"","filename":"Figure2.jpg","url":"https://assets-eu.researchsquare.com/files/rs-3448810/v1/14c9e9906319b6bfa4fbf45e.jpg"},{"id":44805541,"identity":"449fc63c-f327-427e-85cf-8c713cccbeef","added_by":"auto","created_at":"2023-10-17 17:35:34","extension":"jpg","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":1813739,"visible":true,"origin":"","legend":"\u003cp\u003eLegend not included with this version\u003c/p\u003e","description":"","filename":"Figure3.jpg","url":"https://assets-eu.researchsquare.com/files/rs-3448810/v1/f4b820e52612625f31651046.jpg"},{"id":44805540,"identity":"8838aedd-b033-4684-8cdf-67fc4e778d9b","added_by":"auto","created_at":"2023-10-17 17:35:34","extension":"jpg","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":1746745,"visible":true,"origin":"","legend":"\u003cp\u003eLegend not included with this version\u003c/p\u003e","description":"","filename":"Figure4.jpg","url":"https://assets-eu.researchsquare.com/files/rs-3448810/v1/3050ff930dad29339e87db5f.jpg"},{"id":44804870,"identity":"638f2931-6761-4cb6-83ac-25cc83013b27","added_by":"auto","created_at":"2023-10-17 17:27:34","extension":"jpg","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":1685234,"visible":true,"origin":"","legend":"\u003cp\u003eLegend not included with this version\u003c/p\u003e","description":"","filename":"Figure5.jpg","url":"https://assets-eu.researchsquare.com/files/rs-3448810/v1/7c8da377e9fc3f9b3745f5b5.jpg"},{"id":44805802,"identity":"02b6c18c-8a42-466b-a186-35d9f920d92c","added_by":"auto","created_at":"2023-10-17 17:43:34","extension":"jpg","order_by":6,"title":"Figure 6","display":"","copyAsset":false,"role":"figure","size":1295196,"visible":true,"origin":"","legend":"\u003cp\u003eLegend not included with this version\u003c/p\u003e","description":"","filename":"Figure6.jpg","url":"https://assets-eu.researchsquare.com/files/rs-3448810/v1/4c7e5904d751c5f32293360e.jpg"},{"id":45496990,"identity":"f355b170-546a-4cba-8d9e-b60c39fa79c6","added_by":"auto","created_at":"2023-10-31 01:59:25","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":1406410,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-3448810/v1/5592b3fe-0a4f-4280-bfd8-d75cd79d0e6a.pdf"},{"id":44803878,"identity":"3e4d0573-15e3-494c-8382-930de8ccf8c7","added_by":"auto","created_at":"2023-10-17 17:19:34","extension":"png","order_by":2,"title":"","display":"","copyAsset":false,"role":"supplement","size":2216656,"visible":true,"origin":"","legend":"","description":"","filename":"FigureS1.png","url":"https://assets-eu.researchsquare.com/files/rs-3448810/v1/a70f9395add035d68c2a8b23.png"},{"id":44803875,"identity":"71aacd7b-c845-48b3-b5fb-5c36f4824820","added_by":"auto","created_at":"2023-10-17 17:19:34","extension":"png","order_by":3,"title":"","display":"","copyAsset":false,"role":"supplement","size":244656,"visible":true,"origin":"","legend":"","description":"","filename":"FigureS2.png","url":"https://assets-eu.researchsquare.com/files/rs-3448810/v1/65ac434c4fc0a246972808d0.png"},{"id":44804873,"identity":"1df1974a-822a-4bef-a543-d34a87f034f8","added_by":"auto","created_at":"2023-10-17 17:27:34","extension":"png","order_by":4,"title":"","display":"","copyAsset":false,"role":"supplement","size":451991,"visible":true,"origin":"","legend":"","description":"","filename":"FigureS3.png","url":"https://assets-eu.researchsquare.com/files/rs-3448810/v1/cd8c4e569c023c575ad649a1.png"},{"id":44803876,"identity":"e2d4c4c8-64ec-4f40-86da-16f51e064e4b","added_by":"auto","created_at":"2023-10-17 17:19:34","extension":"png","order_by":5,"title":"","display":"","copyAsset":false,"role":"supplement","size":75718,"visible":true,"origin":"","legend":"","description":"","filename":"FigureS4.png","url":"https://assets-eu.researchsquare.com/files/rs-3448810/v1/cbaa8e77e93bf6e2e301039c.png"},{"id":44803886,"identity":"9f405511-7044-4327-83a4-1178512258ca","added_by":"auto","created_at":"2023-10-17 17:19:34","extension":"png","order_by":6,"title":"","display":"","copyAsset":false,"role":"supplement","size":476202,"visible":true,"origin":"","legend":"","description":"","filename":"FigureS5.png","url":"https://assets-eu.researchsquare.com/files/rs-3448810/v1/a8309305d2202c243a3281b1.png"},{"id":44804872,"identity":"7e7effb4-44cf-4bf1-813d-c6bc262a0bd5","added_by":"auto","created_at":"2023-10-17 17:27:34","extension":"png","order_by":7,"title":"","display":"","copyAsset":false,"role":"supplement","size":397671,"visible":true,"origin":"","legend":"","description":"","filename":"FigureS6.png","url":"https://assets-eu.researchsquare.com/files/rs-3448810/v1/560098f0b71b3ed96ab99ad8.png"},{"id":44803887,"identity":"926254a2-2cf4-400e-8528-2bbc78cd931d","added_by":"auto","created_at":"2023-10-17 17:19:34","extension":"csv","order_by":8,"title":"","display":"","copyAsset":false,"role":"supplement","size":2778446,"visible":true,"origin":"","legend":"","description":"","filename":"SupplementaryData1msigdbgsvadegsresults.csv","url":"https://assets-eu.researchsquare.com/files/rs-3448810/v1/6516303662c7135b465c0634.csv"},{"id":44803880,"identity":"7b223075-3584-4c8f-9e04-37a133d45cae","added_by":"auto","created_at":"2023-10-17 17:19:34","extension":"csv","order_by":9,"title":"","display":"","copyAsset":false,"role":"supplement","size":78555,"visible":true,"origin":"","legend":"","description":"","filename":"SupplementaryData2diygenesetdegsresults.csv","url":"https://assets-eu.researchsquare.com/files/rs-3448810/v1/003975326bc4b4101c52da52.csv"},{"id":44803885,"identity":"1a757cc1-1801-4349-a358-08ac77325381","added_by":"auto","created_at":"2023-10-17 17:19:34","extension":"docx","order_by":10,"title":"","display":"","copyAsset":false,"role":"supplement","size":15395,"visible":true,"origin":"","legend":"","description":"","filename":"TableS1.docx","url":"https://assets-eu.researchsquare.com/files/rs-3448810/v1/498b373c78303b8e32d91d02.docx"}],"financialInterests":"No competing interests reported.","formattedTitle":"Exploring the Molecular and Immune Landscape of Lung Cancer Associated with Cystic Airspaces: Implications for Prognosis and Therapeutic Strategies","fulltext":[{"header":"Introduction","content":"\u003cp\u003eAlthough clinical rarity, there has been increasingly reports recently of a unique type of lung cancer characterized by the presence of round or irregular air-containing cavities with cystic walls radiographically\u003csup\u003e[\u003cspan additionalcitationids=\"CR2\" citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e]\u003c/sup\u003e. Terminologically, some entities were used loosely in association with this type of lung cancer\u003csup\u003e[\u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e]\u003c/sup\u003e. In this study, we define the term \u0026ldquo;lung cancer associated with cystic airspaces\u0026rdquo; (i.e., LCCA) to refer to lung cancer that manifests as solitary cystic airspaces and has thin walls of 4 mm or less along the lesion circumference\u003csup\u003e[\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e, \u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e]\u003c/sup\u003e.\u003c/p\u003e \u003cp\u003eThe comprehensive nature of LCCA is still limited, though most previous studies have described the clinicopathological and radiographical features of LCCA. Accumulating evidence suggests that LCCA occurs in approximately 1% ~ 4% lung cancer\u003csup\u003e[\u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e, \u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e, \u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e]\u003c/sup\u003e, and they are predominantly (about 90%) adenocarcinomas and more frequently found in men and smokers\u003csup\u003e[\u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e]\u003c/sup\u003e. A four-type classification system was proposed to summarize the morphologic patterns of CT observations in 2015\u003csup\u003e[\u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e]\u003c/sup\u003e. Long-term outcomes are not well-defined as regarded to classification or stage. Still, it is reported that LCCA had a worse prognosis than the non-LCCA\u003csup\u003e[\u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e, \u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e]\u003c/sup\u003e and the wall thickness is an independent prognostic factor\u003csup\u003e[\u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e]\u003c/sup\u003e. Moreover, the nodule pattern is associated with poor differentiation, shorter survival and higher PD-L1 expression than other subtypes\u003csup\u003e[\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e, \u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e, \u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e]\u003c/sup\u003e.\u003c/p\u003e \u003cp\u003eHowever, little is known about the genetic landscape of LCCA. The purpose of this study is to investigate the transcriptional profiles of LCCA as compare to non-LCCA or normal lung tissue.\u003c/p\u003e"},{"header":"Materials and Methods","content":"\u003cdiv id=\"Sec3\" class=\"Section2\"\u003e \u003ch2\u003ePatients and samples\u003c/h2\u003e \u003cp\u003ePatients with cystic lesions in the lungs were retrospectively reviewed in our institute from January 1st 2017 to July 1st 2023. All the patients were screened according to the following inclusion criteria: (1) age\u0026thinsp;\u0026ge;\u0026thinsp;18 years; (2) the target lesion histologically confirmed as lung carcinoma; and (3) available medical records for clinicopathological features and follow-ups (Figure \u003cspan refid=\"MOESM1\" class=\"InternalRef\"\u003eS1\u003c/span\u003eA). Another cohort of patients with solid lesions in the lungs (i.e., non-LCCA) were enrolled according to the same criteria and matched for TNM stage and histology. All the slides were examined by the same pathologist to confirm the diagnosis. Clinicopathological characteristics including age, gender, smoking status, surgical procedure, histological subtype, visceral pleura invasion (VPI), cancer cells spread through air spaces (STAS), lymphovascular thrombus, TNM stage according 8th edition staging system for lung cancer and gene mutations. In this study, tumor samples from 8 patients with LCCA and 9 patients with non-LCCA, as well as 10 samples of adjacent normal tissue from above patients were collected. All the samples were stored in liquid nitrogen immediately after surgical resection (Figure \u003cspan refid=\"MOESM1\" class=\"InternalRef\"\u003eS1\u003c/span\u003e).\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec4\" class=\"Section2\"\u003e \u003ch2\u003eCT image and morphology-based classifications\u003c/h2\u003e \u003cp\u003eAll the chest computed tomographic (CT) scans were performed on the multi-detector CT scanners (SIEMENS SOMATOM Definition AS 128, SIEMENS SOMATOM Perspective 128, GE Revolution CT 256), abiding by the following protocols: CT scans were performed without contrast administration and obtained at the end of inspiration during a single deep breath-hold, using the automatic tube current modulation with 120 kV, 150\u0026ndash;300 mA, matrix size of 512 \u0026times; 512, field of view (FOV) of 300\u0026ndash;350 mm, and 1.0 mm reconstructed section thickness. The lung window was set as a width of 1200 Hounsfield units (HU) and a level of -600 HU. Two independent thoracic radiologists with 10 and 12 years of working experience reviewed the CT images and were blinded to pathology. The radiographical features including lesion location, entire lesion size, nodule density, cyst size, wall component, nodule size/wall thickness, number of cystic airspaces, shape of cystic airspaces, and pleural retraction, were evaluated. A four-type classification system proposed by Mascalchi\u003csup\u003e[\u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e]\u003c/sup\u003e was applied and assigned to each LCCA. Type I refers to a nodule extruding from the wall of the cystic airspace; type II refers to a nodule projecting into the cystic airspace from the wall; type III is that of the cyst wall thickening along the lesion circumference; and type IV refers to multiple cystic lesions intermixed with soft tissue septum (Figure \u003cspan refid=\"MOESM1\" class=\"InternalRef\"\u003eS1\u003c/span\u003eC-F). The inter-reviewer consistency of the classifications between the two radiologists was assessed. In case of discrepancy, a consensus was reached by discussion and examined by a third senior radiologist with 20 years of experience.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec5\" class=\"Section2\"\u003e \u003ch2\u003eRNA Extraction and Quality Control\u003c/h2\u003e \u003cp\u003eTotal RNA was isolated from tissues using TRIzol reagent (Life Technologies, CA, USA). The quantity and quality of total RNA samples were measured using a NanoDrop ND-1000 (Wilmington, DE, USA). RNA integrity was assessed using an Agilent 2100 Bioanalyzer (Agilent Technologies, Santa Clara, CA, USA). Subsequent analyses were performed on samples with RNA Integrity Number (RIN) greater than 7.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec6\" class=\"Section2\"\u003e \u003ch2\u003ecDNA Library Construction and High-throughput Sequencing\u003c/h2\u003e \u003cp\u003eAfter extracting the total RNA of the sample, the rRNA was removed by the Ribosome Kit (NEB, USA). The circular RNA library construction also required RNase R (Biosearch Technologies\u003c/p\u003e \u003cp\u003e, UK) to remove the linear RNA, and then the RNA was fragmented (the average fragment length was about 200nt), using M-MuLV Reverse Transcriptase (Thermo Fisher Scientific, USA) reverse transcription to synthesize single-stranded cDNA, then double-stranded cDNA is synthesized and purified, followed by end repair and adapter primers. PCR amplification was performed using LongAmp\u0026trade; Taq 2X Master Mix, SR Primer for Illumina and index primers (NEB, USA). PCR products were purified on 8% polyacrylamide gels. DNA fragments corresponding to 140\u0026ndash;160 bp were recovered and dissolved in 8 mL of elution buffer. Finally, library quality was assessed on the Agilent Bioanalyzer 2100 system using the DNA High Sensitivity Chip. Sequencing was finally performed using an Illumina HiSeq2500 sequencer (RiboBio, Guangzhou, China) by performing paired-end runs with 100 bp read length.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec7\" class=\"Section2\"\u003e \u003ch2\u003ePre-processing of sequencing reads\u003c/h2\u003e \u003cp\u003eSequencing read quality was inspected using the FastQC\u003csup\u003e[\u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e13\u003c/span\u003e]\u003c/sup\u003e. Adapter removal and read trimming were performed using Trimmomatic\u003csup\u003e[\u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e]\u003c/sup\u003e to removing trailing sequences below a phred quality score of 20 and drop reads below length of 50 bp. Sequencing data were aligned to the human reference genome hg19 using HISAT2\u003csup\u003e[\u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e]\u003c/sup\u003e. HTSeq was subsequently employed to count the reads numbers mapped to each gene\u003csup\u003e[\u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e]\u003c/sup\u003e. Read counts were further normalized to transcripts per kilobase of exon model per million mapped reads (TPM) to quantify the expression levels of mRNAs or ncRNAs.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec8\" class=\"Section2\"\u003e \u003ch2\u003eProcessing of LUAD (Lung carcinoma) data from TCGA (The Cancer Genome Altas) database\u003c/h2\u003e \u003cp\u003eThe raw count and HTseq-FPKM gene expression data and corresponding LUAD clinical and survival data were downloaded from Xena database (\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttp://xena.ucsc.edu/\u003c/span\u003e\u003cspan address=\"http://xena.ucsc.edu/\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e). The FPKM (Fragments per kilobase of transcript per million mapped reads) data were transformed to TPM following the formula:\u003c/p\u003e \u003cp\u003e \u003cspan class=\"InlineEquation\"\u003e \u003cspan class=\"mathinline\"\u003e\\({TPM}_{i}=\\frac{{FPKM}_{i}\\times {10}^{6}}{\\sum _{i=1}^{N}{FPKM}_{i}}\\)\u003c/span\u003e \u003c/span\u003e \u003c/p\u003e \u003cp\u003eA total of 412 cases were enrolled (Figure \u003cspan refid=\"MOESM1\" class=\"InternalRef\"\u003eS1\u003c/span\u003eB), and all the data had undergone process of normalization, log\u003csub\u003e2\u003c/sub\u003e-transformation and removal of batch effect before analyses. The cases were randomly assigned into training set and validation set at a ratio of 7:3.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec9\" class=\"Section2\"\u003e \u003ch2\u003eDifferential expression analysis\u003c/h2\u003e \u003cp\u003eDifferential expression was normalized and analyzed by DESeq2\u003csup\u003e[\u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e]\u003c/sup\u003e in R using read counts as input. The p-values were attained by the Wald test and corrected for multiple testing using the Benjamini-Hochberg method. To screen the significant differentially expressed genes, log\u003csub\u003e2\u003c/sub\u003e fold-change (absolute value) greater than 0.5 and p-value less than 0.05 was used as threshold for miRNAs, and log\u003csub\u003e2\u003c/sub\u003e fold-change (absolute value) greater than 1 and p-value less than 0.05 was used as threshold for mRNAs and lncRNAs. We used \u0026ldquo;ggplot2\u0026rdquo; package in R to draw the volcano plots and heatmaps.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec10\" class=\"Section2\"\u003e \u003ch2\u003eFunctional enrichment analysis\u003c/h2\u003e \u003cp\u003eIn order to identify the Gene Ontology (GO) annotations and pathways in which differentially expressed genes were enriched, GO term and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway enrichment analyses were performed using the \u0026ldquo;culsterProfiler\u0026rdquo;\u003csup\u003e[\u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e]\u003c/sup\u003e package and visualized by bubble charts using the \u0026ldquo;ggplot2\u0026rdquo; package. The Benjamini-Hochberg method was used to correct the p-values.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec11\" class=\"Section2\"\u003e \u003ch2\u003ePathway activity analysis\u003c/h2\u003e \u003cp\u003eGene set enrichment analysis (GSEA) was used to identify the significantly regulated gene sets in LCCAs, as compared to non-LCCAs, using the \u0026ldquo;GSEA\u0026rdquo; function in the \u0026ldquo;clusterProfiler\u0026rdquo; package. Further, the GSVA (Gene Set Variation Analysis)\u003csup\u003e[\u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e19\u003c/span\u003e]\u003c/sup\u003e algorithm was used to comprehensively score the enrichment of related gene sets and assess the potential biological alterations between the two groups. The gene sets used for GSEA and GSVA were downloaded from the Molecular Signatures Database (MSigDB) website (\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://www.gsea-msigdb.org/gsea/msigdb\u003c/span\u003e\u003cspan address=\"https://www.gsea-msigdb.org/gsea/msigdb\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e), and the gene set collections of \u0026ldquo;Hallmarks\u0026rdquo;, \u0026ldquo;C2.CP\u0026rdquo;, \u0026ldquo;C3\u0026rdquo;, \u0026ldquo;C4\u0026rdquo; and \u0026ldquo;C5.GO\u0026rdquo; were applied. A series of gene sets from literature (Supplementary) was also analyzed.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec12\" class=\"Section2\"\u003e \u003ch2\u003eceRNA regulatory network\u003c/h2\u003e \u003cp\u003eTo construct the ceRNA regulatory network, we first predicted and screened the matched differentially expressed lncRNAs and miRNAs using miRcode\u003csup\u003e[\u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e20\u003c/span\u003e]\u003c/sup\u003e database. The candidate target genes of differentially expressed miRNAs were predicted using the following database: miRDB\u003csup\u003e[\u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e21\u003c/span\u003e]\u003c/sup\u003e, miRTarBase\u003csup\u003e[\u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e22\u003c/span\u003e]\u003c/sup\u003e, and TargetScan\u003csup\u003e[\u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e23\u003c/span\u003e]\u003c/sup\u003e. Then, we integrated the interactions between miRNAs and mRNAs or lncRNAs and visualized the ceRNA regulatory network using Cytoscape (version 3.10.0).\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec13\" class=\"Section2\"\u003e \u003ch2\u003eDefining diversity score to quantify the histological heterogeneity and BCR/TCR clonality\u003c/h2\u003e \u003cp\u003eWe define a diversity score to characterize the histological heterogeneity of the samples, which was calculated using the Shannon-Wiener Index:\u003cdiv id=\"Equa\" class=\"Equation\"\u003e\u003cdiv format=\"TEX\" class=\"mathdisplay\" id=\"FileID_Equa\" name=\"EquationSource\"\u003e\n$${H}^{{\\prime }}= -\\sum _{i=1}^{R}{p}_{i}\\times {log}_{2}{p}_{i}$$\u003c/div\u003e\u003c/div\u003e\u003c/p\u003e \u003cp\u003eHere, \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(R\\)\u003c/span\u003e\u003c/span\u003e is the number of histological subtypes (lepidic, acinar, papillary, micropapillary, solid and complex glandular), i.e., 6, and \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\({p}_{i}\\)\u003c/span\u003e\u003c/span\u003e refers to the proportion of corresponding subtype in the tumor (\u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(0\\le {p}_{i}\\le 1\\)\u003c/span\u003e\u003c/span\u003e). The diversity score ranges from 0 to 2.585, with 0 indicating that a specific subtype accounts for 100% of the tumor and the maximum value indicating that all the six subtypes are present evenly. The evenness and clonality score were calculated using the following formula:\u003cdiv id=\"Equb\" class=\"Equation\"\u003e\u003cdiv format=\"TEX\" class=\"mathdisplay\" id=\"FileID_Equb\" name=\"EquationSource\"\u003e\n$$Evenness=\\frac{Diversity}{\\text{l}\\text{o}\\text{g}\\left(Richness\\right)}$$\u003c/div\u003e\u003c/div\u003e\u003cdiv id=\"Equc\" class=\"Equation\"\u003e\u003cdiv format=\"TEX\" class=\"mathdisplay\" id=\"FileID_Equc\" name=\"EquationSource\"\u003e\n$$Clonality=1-Evenness$$\u003c/div\u003e\u003c/div\u003e\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec14\" class=\"Section2\"\u003e \u003ch2\u003eCalculating of the intratumor heterogeneity, stemness, and IC50\u003c/h2\u003e \u003cp\u003eThe \u0026ldquo;DEPTH\u0026rdquo;\u003csup\u003e[\u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e24\u003c/span\u003e]\u003c/sup\u003e package is an algorithm to quantify intratumor heterogeneity using the normalized RNA-Seq data as input. The stemness score, denoted as mRNAsi index, was calculated as Zhang\u003csup\u003e[\u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e25\u003c/span\u003e]\u003c/sup\u003e described. Higher mRNAsi scores refer to more aggressive biological processes in cancer stem cells (CSCs) and more tumor dedefferention. The \u0026ldquo;oncoPredict\u0026rdquo; package was used for drug response prediction. The reference GDSC2 pre-processed training expression data and CTRP response data were download from \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://osf.io/c6tfx/\u003c/span\u003e\u003cspan address=\"https://osf.io/c6tfx/\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e. The input RNA-Seq data was TPM-normalized and log2-tranformed.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec15\" class=\"Section2\"\u003e \u003ch2\u003eDeconvolution of tumor microenvironment (TME) from gene expression profiles\u003c/h2\u003e \u003cp\u003eImmunoPhenoScore was calculated using the method proposed by Charoentong\u003csup\u003e[\u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e26\u003c/span\u003e]\u003c/sup\u003e, which was an aggregation of the expression of representative immune-related genes and presence of immune effector cells or suppressor cells. Instead of the proposed scoring system, we adopted the sums of the four components. From the RNA-seq profiles of the samples, we first applied the xCell\u003csup\u003e[\u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e27\u003c/span\u003e]\u003c/sup\u003e algorithm to calculate the panoramic cellular infiltration score of the TME. Then the TIDE\u003csup\u003e[\u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e28\u003c/span\u003e]\u003c/sup\u003e module under Python framework was used to decrypt abnormal T cell function and infiltration related to tumor immune evasion. The normalized and log2-transoformed gene expression data were used as input.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec16\" class=\"Section2\"\u003e \u003ch2\u003eExploring the repertoire profiles of BCR and TCR\u003c/h2\u003e \u003cp\u003eMiXCR\u003csup\u003e[\u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e29\u003c/span\u003e]\u003c/sup\u003e is an integrated platform for interactive analysis of B cell and T cell receptor repertoire, starting from sequencing data to biological perspective. MiXCR implements a series of highly efficient algorithms aligning the sequencing data to reference V-, D-, J- and C- gene segment database, and yields CDR3 regions of various gene recombination (IGH, IGK and IGL for BCR, TRAD, TRB and TRG for TCR) and the frequency for each clonotype. The downstream processes and analyses of repertoire profiles were performed by \u0026ldquo;immunarch\u0026rdquo;\u003csup\u003e[\u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e30\u003c/span\u003e]\u003c/sup\u003e package, which compatibly works fine with MiXCR\u0026rsquo;s output format.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec17\" class=\"Section2\"\u003e \u003ch2\u003eIterative feature selection\u003c/h2\u003e \u003cp\u003eTo screen the candidate gene associated for prognosis, we applied machine learning methods to iteratively select the important features (candidate genes). Random Forest is commonly-used machine learning algorithm consisting of many decision trees. In this study, we set the number of trees as 100. Least absolute shrinkage and selection operator (LASSO) is a regression method that applied regularization to enhance the accuracy and interpretability of the model. After shrinkage, it chose the non-zero-coefficient variables as the important features. Forward and backward stepwise regression is a linear regression mainly aim to solve multivariate collinearity. It was used to screen variables in multivariate analyses, achieving the minimum AIC (Akaike Information Criterion).\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec18\" class=\"Section2\"\u003e \u003ch2\u003eStatistical analysis\u003c/h2\u003e \u003cp\u003eAll statistical analyses were conducted using R software (version 4.1.0). Between-group comparisons were performed using Wilcoxon rank-sum test for continuous, Fisher\u0026rsquo;s exact test or Chi-squared test for discrete variables, using the R function \u0026ldquo;wilcox.test\u0026rdquo;, \u0026ldquo;fisher.test\u0026rdquo; and \u0026ldquo;chisq.test\u0026rdquo;, respectively. Distribution of continuous variables was checked by Kolmogorov-Smirnov test. Survival analyses including log-rank tests, Kaplan-Meier curve, univariate and multivariate Cox regression analyses were performed using the R package \u0026ldquo;survival\u0026rdquo;. A two-sided p-value of less than 0.05 was considered statistically significant.\u003c/p\u003e \u003c/div\u003e"},{"header":"Results","content":"\u003cdiv id=\"Sec20\" class=\"Section2\"\u003e \u003ch2\u003eBaseline of the study cohorts\u003c/h2\u003e \u003cp\u003eThe clinicopathological characteristics of patients are presented in Table\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e. A total of 165 patients with LCCA and 201 patients with non-LCCA that matched for TNM stage and histology were enrolled in this study. The median follow-up was 17 (IQR 6 to 43) months. The median age was 60 (IQR 53 to 65) years in LCCA cohort. In consistence with literature, LCCA was found more frequently in men (76%). The predominant histological type was invasive non-mucinous adenocarcinoma (INMA) in 75% of the LCCAs, followed by 13% of minimally invasive adenocarcinoma (MIA), 5% of squamous cell carcinoma (SCC) and 4% of invasive mucinous adenocarcinoma (IMA). There were 5 cases mixed with INMA and IMA, and one case of large cell carcinoma (LCC). The clinicopathological features between LCCA and non-LCCA cohort were similar, except that LCCAs were prone to occur more in men (76% vs 50%, \u003cem\u003ep\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.001) and associated with less visceral pleura invasion (VPI, 6% vs 18%, \u003cem\u003ep\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.00118) or lympho-vascular invasion (LVI, 8% vs 23%, \u003cem\u003ep\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.001) in our study. EGFR and KRAS mutation were found positive in 39% and 9% of the cases in LCCA cohort, respectively.\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab1\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 1\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eBaseline characteristics of all patients.\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"4\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eVariable\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eLCCA\u003c/p\u003e \u003cp\u003e(n\u0026thinsp;=\u0026thinsp;165, 45.1%)\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003enon-LCCA\u003c/p\u003e \u003cp\u003e(n\u0026thinsp;=\u0026thinsp;201, 54.9%)\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003e\u003cem\u003ep\u003c/em\u003e value\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cb\u003eAge, median (IQR)\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e60 (53\u0026ndash;65)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e60 (53\u0026ndash;65)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.278\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cb\u003eGender, (%)\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e\u003cb\u003e\u0026lt;\u0026thinsp;0.001\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eMale\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e125 (76%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e100 (50%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eFemale\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e40 (24%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e101 (50%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cb\u003eSmoking status, (%)\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.333\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eNever\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e107 (65%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e141 (70%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eCurrent or ever\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e58 (35%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e60 (30%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cb\u003eHistological subtype, (%)\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.597\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eIMA\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e7 (4%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e8 (4%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eINMA\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e123 (75%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e163 (82%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eLCC\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e1 (1%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0 (0%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eMIA\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e21 (13%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e19 (9%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eMixed\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e5 (3%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e3 (1%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eSCC\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e8 (5%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e8 (4%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cb\u003eGrade, (%)\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.135\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e20 (12%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e12 (6%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e2\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e87 (53%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e122 (61%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e3\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e28 (17%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e37 (18%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eUnknown\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e30 (18%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e30 (15%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cb\u003eVPI, (%)\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e\u003cb\u003e0.00118\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ePositive\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e10 (6%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e36 (18%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eNegative\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e155 (94%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e165 (82%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cb\u003eSTAS, (%)\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.253\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ePositive\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e18 (11%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e14 (7%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eNegative\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e147 (89%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e187 (93%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cb\u003eLVI, (%)\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e\u003cb\u003e\u0026lt;\u0026thinsp;0.001\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ePositive\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e13 (8%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e46 (23%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eNegative\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e152 (92%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e155 (77%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cb\u003epT, median (IQR)\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e2 (1.4\u0026ndash;2.5)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e2.1 (1.6\u0026ndash;3)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.0176\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cb\u003epN, (%)\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.0785\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eN0\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e162 (98%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e189 (94%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eN1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0 (0%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e5 (2%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eN2\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e3 (2%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e7 (3%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cb\u003eStage, (%)\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.245\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eIA\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e131 (79%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e141 (70%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eIB\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e23 (14%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e37 (18%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eII\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e7 (4%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e15 (7%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eIII\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e4 (2%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e8 (4%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cb\u003eEGFR, (%)\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.359\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ePositive\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e64 (39%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e86 (43%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eNegative\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e43 (26%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e40 (20%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eUnknown\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e58 (35%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e75 (37%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cb\u003eKRAS, (%)\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.135\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ePositive\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e15 (9%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e8 (4%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eNegative\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e92 (56%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e118 (59%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eUnknown\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e58 (35%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e75 (37%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003ctfoot\u003e \u003ctr\u003e\u003ctd colspan=\"4\"\u003eData are presented as interquartile range (IQR) or numbers (frequency). LCCA: lung cancer associated with cystic airspaces; IMA: invasive mucinous adenocarcinoma; INMA: invasive non-mucinous adenocarcinoma; MIA: minimally invasive adenocarcinoma; Mixed: contained IMA and INMA; SCC: lung squamous cell carcinoma; LCC: large cell carcinoma; STAS: spread through air space; VPI: visceral pleura invasion; LVI: lympho-vascular invasion; EGFR: epidermal growth factor receptor; KRAS: Kirsten rats arcomaviral oncogene homolog. \u003cem\u003ep\u003c/em\u003e value\u0026thinsp;\u0026lt;\u0026thinsp;0.05 marked statistical significance.\u003c/td\u003e\u003c/tr\u003e \u003c/tfoot\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec21\" class=\"Section2\"\u003e \u003ch2\u003eLCCAs were associated with histological heterogeneity and poor PFS\u003c/h2\u003e \u003cp\u003eAs regarding to the survival stratification, combinations of histological subtypes (lepidic, acinar, papillary, micropapillary, solid and complex glandular) and predominant proportion were applied in the newly proposed grading system for invasive adenocarcinoma of LUAD in 2020\u003csup\u003e[\u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e31\u003c/span\u003e]\u003c/sup\u003e. For both the LCCA and non-LCCA cohorts, we determined the proportion of each histological subtype component for cases with INMA (Fig.\u0026nbsp;1A). Next, we defined a metric, known as the diversity score, to measure the histological heterogeneity of LNMA samples (see the Methods section). The diversity scores in LCCAs were significantly higher (\u003cem\u003ep\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.022) than in non-LCCAs (Fig.\u0026nbsp;1B). We further found patients of age less than 60 (\u003cem\u003ep\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.001), never smoking (\u003cem\u003ep\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.001), negative VPI (\u003cem\u003ep\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.0084) and negative LVI (\u003cem\u003ep\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.017) in LCCAs presented more significant histological heterogeneity (Fig.\u0026nbsp;1C-F), while no differences were found among groups of gender and tumor size (Figure \u003cspan refid=\"MOESM2\" class=\"InternalRef\"\u003eS2\u003c/span\u003eA-B).\u003c/p\u003e \u003cp\u003eTo investigate whether the presenting of cystic airspaces was associated with prognosis of early-stage lung cancer, Kaplan-Meier curves were plotted to compare between LCCA cohort and non-LCCA cohort. Although not significant, it was suggested that LCCAs and a high diversity score were associated with poorer progression-free survival (PFS) than that of non-LCCAs. Yet in the stratification analysis, patients of age more than 60 (HR 5.36 [1.03\u0026ndash;27.83], \u003cem\u003ep\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.046) and positive VPI (HR 17.46 [1.81\u0026ndash;168.68], \u003cem\u003ep\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.013) were indicated to have significantly less PFS in LCCAs. Furthermore, LCCA (HR 6.13 [1.26\u0026ndash;29.80], \u003cem\u003ep\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.0025), micropapillary subtype fraction higher than 0.15 (HR 7.25 [1.48\u0026ndash;36.37], \u003cem\u003ep\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.016) and VPI positive (HR 8.94 [1.90\u0026ndash;42.02], \u003cem\u003ep\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.006) were proven to be the independent risk factors for PFS in the univariate and multivatiate Cox regression analyses (Table\u0026nbsp;\u003cspan refid=\"Tab2\" class=\"InternalRef\"\u003e2\u003c/span\u003e). As for the LCCA cohort alone, VPI positive and STAS positive were the significant predictors for PFS (Table \u003cspan refid=\"MOESM1\" class=\"InternalRef\"\u003eS1\u003c/span\u003e).\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab2\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 2\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eUnivariate and multivariate Cox regression analyses of LCCA and non-LCCA\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"7\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c6\" colnum=\"6\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c7\" colnum=\"7\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\" morerows=\"1\" rowspan=\"2\"\u003e \u003cp\u003eVariable\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\" morerows=\"1\" rowspan=\"2\"\u003e \u003cp\u003eTotal (N)\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colspan=\"2\" nameend=\"c4\" namest=\"c3\"\u003e \u003cp\u003eUnivariate analysis\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colspan=\"2\" nameend=\"c6\" namest=\"c5\"\u003e \u003cp\u003eMultivariate analysis\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c7\"\u003e \u003cp\u003eCoefficient\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eHR (95% CI)\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003eP value\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c5\"\u003e \u003cp\u003eHR (95% CI)\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c6\"\u003e \u003cp\u003eP value\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cb\u003eGroup\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e294\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eLCCA\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e166\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eReference\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eReference\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003enon-LCCA\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e128\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e3.3(0.78\u0026ndash;14.17)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.102\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e6.13 (1.3\u0026ndash;29.8)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.025\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e1.81\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cb\u003eMPP\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e294\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eLow\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e263\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eReference\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eReference\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eHigh\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e31\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e8.56(2.14\u0026ndash;34.23)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.002\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e7.25 (1.5\u0026ndash;36.4)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.016\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e1.98\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cb\u003eVPI\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e294\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eNegative\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e252\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eReference\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eReference\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ePositive\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e42\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e4.44 (1.10\u0026ndash;17.9)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.036\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e8.94 (1.9\u0026ndash;42.0)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.006\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e2.19\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cb\u003eSTAS\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e294\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eNegative\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e263\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eReference\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eReference\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ePositive\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e31\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e8.7 (2.04\u0026ndash;37.16)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.003\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e3.6 (0.65\u0026ndash;19.8)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.139\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e1.28\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003ctfoot\u003e \u003ctr\u003e\u003ctd colspan=\"7\"\u003eLCCA: lung cancer associated with cystic airspaces; MPP: Micropapillary; STAS: spread through air space; VPI: visceral pleura invasion; LVI: lympho-vascular invasion; HR: hazard ratio. \u003cem\u003ep\u003c/em\u003e value\u0026thinsp;\u0026lt;\u0026thinsp;0.05 marked statistical significance.\u003c/td\u003e\u003c/tr\u003e \u003c/tfoot\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec22\" class=\"Section2\"\u003e \u003ch2\u003eDifferentially expressed mRNA, lncRNA and miRNA profiles by bulk RNA-seq\u003c/h2\u003e \u003cp\u003eTo further characterize the transcriptome alterations underlying patients with LCCA and non-LCCA, we collected 8 tumor samples from LCCA and 9 from non-LCCA, as well as 10 adjacent non-neoplastic tissues (ANT) from the above patients (Fig.\u0026nbsp;2A). Bulk RNA-sequencing (RNA-seq) was performed for each group. As expected, the samples were separated into three clusters in the principal component analysis (PCA) (Fig.\u0026nbsp;2B), though the differences between LCCA and non-LCCA on PC1 was not very obvious, as compared to ANTs.\u003c/p\u003e \u003cp\u003eTo profile the genomic features of LCCAs, we first performed differential gene expression analysis between LCCAs and ANTs. A total of 2383 mRNAs (1581 up-regulated and 802 down-regulated), 1151 lncRNAs (706 up-regulated and 445 down-regulated), and 68 miRNAs (42 up-regulated and 26 down-regulated) were identified (Fig.\u0026nbsp;2C-E). A total of 27 differentially expressed lncRNAs and three miRNAs (miR-142, miR-200b and miR-490) were matched into 38 miRNA-lncRNA interactions, whereas the three miRNAs targeted to 200 candidate mRNAs to form 205 miRNA-lncRNA pairs. Of the candidate mRNAs, 16 were fund significantly differentially expressed (Fig.\u0026nbsp;2F). Finally, a network of 2357 competitive lncRNA-miRNA-mRNA links was constructed (Fig.\u0026nbsp;2G).\u003c/p\u003e \u003cp\u003eNext, we compared gene expression between LCCAs and non-LCCAs. A total of 970 up-regulated and 275 down-regulated genes were identified (Fig.\u0026nbsp;3A, Figure \u003cspan refid=\"MOESM3\" class=\"InternalRef\"\u003eS3\u003c/span\u003eA-C). GO and KEGG pathway enrichment analyses revealed that the differentially up-regulated genes were remarkably associated with cell proliferation, migration, communication, ion channel or receptor activities and angiogenesis, et al (Fig.\u0026nbsp;3B). The related GO terms or pathways were further confirmed significantly up-regulated by GSEA framework (Fig.\u0026nbsp;3C-D, Figure \u003cspan refid=\"MOESM3\" class=\"InternalRef\"\u003eS3\u003c/span\u003eD-F). We also found that molecular signatures associated with angiogenesis mediated by macrophage\u003csup\u003e[\u003cspan citationid=\"CR32\" class=\"CitationRef\"\u003e32\u003c/span\u003e]\u003c/sup\u003e, regulation of stromal components\u003csup\u003e[\u003cspan citationid=\"CR33\" class=\"CitationRef\"\u003e33\u003c/span\u003e]\u003c/sup\u003e, infiltration of cancer-related fibroblasts (CAFs)\u003csup\u003e[\u003cspan citationid=\"CR34\" class=\"CitationRef\"\u003e34\u003c/span\u003e]\u003c/sup\u003e and myeloid-derived suppressor cells (MDSCs)\u003csup\u003e[\u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e28\u003c/span\u003e]\u003c/sup\u003e were up-regulated in LCCAs (Fig.\u0026nbsp;3E-F, Figure \u003cspan refid=\"MOESM3\" class=\"InternalRef\"\u003eS3\u003c/span\u003eG-I).\u003c/p\u003e \u003cdiv id=\"Sec23\" class=\"Section3\"\u003e \u003ch2\u003eLCCA displayed distinct biochemical characteristics\u003c/h2\u003e \u003cp\u003eGSVA was used to comprehensively score the enrichment of related GO terms and pathways of each sample, and differentially regulated gene sets were assessed between the two cohorts (Fig.\u0026nbsp;4A, Supplementary Data 1). Several GO terms or pathways associate with \u0026ldquo;regulation of lymphangiogenesis\u0026rdquo;, \u0026ldquo;woulding healing involved in inflammatory response\u0026rdquo;, \u0026ldquo;actin filament reorganization\u0026rdquo;, and \u0026ldquo;leukotriene metabolic pathway\u0026rdquo;, were significantly up-regulated in LCCAs. We also compared the gene sets from literatures. The samples of LCCA displayed higher levels of gene sets associated with epithelial-to-mesenchymal transition (EMT)\u003csup\u003e[\u003cspan citationid=\"CR35\" class=\"CitationRef\"\u003e35\u003c/span\u003e]\u003c/sup\u003e, angiogenesis\u003csup\u003e[\u003cspan citationid=\"CR32\" class=\"CitationRef\"\u003e32\u003c/span\u003e, \u003cspan citationid=\"CR36\" class=\"CitationRef\"\u003e36\u003c/span\u003e]\u003c/sup\u003e and distal metastasis\u003csup\u003e[\u003cspan citationid=\"CR37\" class=\"CitationRef\"\u003e37\u003c/span\u003e]\u003c/sup\u003e, which may explain their poor prognosis (Fig.\u0026nbsp;4B). Moreover, the pathways related to ferroptosis, oxeiptosis and intrinsic apoptosis, were significantly up-regulated, while that of extrinsic apoptosis were down-regulated, indicating a distinct death episode was orchestrated in LCCAs. In addition, biochemical processes associated with galactose metabolism, prostaglandin biosynthesis and glycosaminoglycan degradation were enhanced, as compared to non-LCCAs.\u003c/p\u003e \u003cp\u003eTo elucidate the tumor biological behavior of LCCA, we calculate the scores of intratumor genetic heterogeneity and stemness of each sample. Interestingly, the intratumor heterogeneity (ITH) scores were significantly lower (\u003cem\u003ep\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.016) in LCCAs (Fig.\u0026nbsp;4C), inconsistent with their higher histological heterogeneity. The stemness score between two groups were similar (Fig.\u0026nbsp;4D). We further used \u0026ldquo;oncoPredict\u0026rdquo; to predict the IC50 of each sample to a series of agents. It was suggested that samples in the LCCA group displayed similar sensitivity to the chemical agents that were the most commonly used in adjuvant chemotherapy settings for NSCLC (non-small cell lung cancer), including docetaxel, gemcitabine and cisplatin (Fig.\u0026nbsp;4E-F, Figure \u003cspan refid=\"MOESM4\" class=\"InternalRef\"\u003eS4\u003c/span\u003eA). As the most common oncogenic drivers in NSCLC, EGFR has become an important therapeutic target for the treatment of these tumors. Samples in the LCCA group seemed to be more resistant to gefitinib and afatinib (Fig.\u0026nbsp;4G, Figure \u003cspan refid=\"MOESM4\" class=\"InternalRef\"\u003eS4\u003c/span\u003eB), though not significant. However, they showed comparable IC50s to Osimertinib, as compared to that of non-LCCA (Fig.\u0026nbsp;4H).\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv id=\"Sec24\" class=\"Section2\"\u003e \u003ch2\u003eLCCAs exhibited dysfunctional immune response and distinct BCR clonality\u003c/h2\u003e \u003cp\u003eTo investigate the differences in the immune environment between LCCA and non-LCCA group, we first quantified the immune-related gene sets by GSVA. We found that gene sets associated with T cell exhaustion\u003csup\u003e[\u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e28\u003c/span\u003e]\u003c/sup\u003e, M2 macrophage polarization\u003csup\u003e[\u003cspan citationid=\"CR38\" class=\"CitationRef\"\u003e38\u003c/span\u003e]\u003c/sup\u003e, MDSCs\u003csup\u003e[\u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e28\u003c/span\u003e]\u003c/sup\u003e and CAF\u003csup\u003e[\u003cspan citationid=\"CR34\" class=\"CitationRef\"\u003e34\u003c/span\u003e]\u003c/sup\u003e infiltration were significantly up-regulated in LCCAs (Fig.\u0026nbsp;5A), suggesting a more immunosuppressive environment. Next, we calculate the ImmunoPhenoScore (IPS), which showed a trend of decline in the LCCA group (Fig.\u0026nbsp;5B). To further deconvolute the TME, we used the xCell algorithm to quantify the various cellular components (Fig.\u0026nbsp;5C). It was showed that macrophages in LCCAs manifested an anti-inflammatory M2 phenotype, not M1 phenotype, as indicated by the dramatically higher scores in M2 macrophages (\u003cem\u003ep\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.001). Next, we explored the dysfunctional and exclusive status of TME using the TIDE algorithm. We found that cancer-associated fibroblast (CAF) scores were higher in LCCAs (\u003cem\u003ep\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.046), resulting a trend of increase in dysfunction scores and comprehensive TIDE scores (Fig.\u0026nbsp;5D). The scores related to cytotoxic T cells and exclusion scores were comparable. These data suggested LCCA displayed a more immunosuppressive profiles in the TME, which may be attributed to polarization of M2-macrophage and infiltration of suppressors.\u003c/p\u003e \u003cp\u003eTo investigate the expansion of B cell and T cell populations in LCCA and non-LCCA, we performed clonality analyses using the MiXCR software. We used the clean fastq data as input to identify CDR3 sequencing of BCRs/TCRs. A total of 147,752 BCR and 1,092 TCR clonotypes were identified in our study. A metric, denoted as richness, was applied to measure the clonal abundance, which referred to the total number of unique clonotypes in the B/T cell population. We compared the BCR and TCR richness between samples of LCCA and non-LCCA (Fig.\u0026nbsp;5E-F). It showed that LCCA had lower (\u003cem\u003ep\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.012) BCR richness, while TCR in the two groups displayed similar clonal abundance. Next, we evaluated the BCR/TCR repertoire based on the fraction of the unique clonotypes, calculating the Shannon-Wiener Index mentioned above. Three metrics were adopted. Diversity score, i.e., Shannon index, represents the heterogeneity of BCR/TCR clonotype. Clonality score, in opposed to evenness score, indicates the potential expansion or predominance of certain B/T cell clones. There were no significant differences between LCCA and non-LCCA, though a trend of enhancement of BCR diversity in non-LCCA (Fig.\u0026nbsp;5G-H). We then explore the overlap clonotypes of BCR/TCR repertoire. A total of 42,639 clonotypes among 348 shared pairs were identified. Figure\u0026nbsp;5I revealed the shared BCR/TCR clonotypes among all the samples. Almost all the shared clones came from CDR3 sequences of BCR, except for only one TCR clonotype of TRAD shared between groups. And it seemed that LCCA tended to have less shared clonotypes of high abundance (Fig.\u0026nbsp;5J). Moreover, the between-group and within-group shared clonotypes exhibited comparable number and distribution (Fig.\u0026nbsp;5K). These results suggested that repertoire diversity and clonal expansion of B cells, instead of T cells, were impaired in LCCA.\u003c/p\u003e \u003cdiv id=\"Sec25\" class=\"Section3\"\u003e \u003ch2\u003eCystic-specific genes predicted overall survival (OS) in LUAD\u003c/h2\u003e \u003cp\u003eWhether the differentially expressed in LCCA, as compared to non-LCCA, were associated with the prognosis of LUAD. First, we screened the LCCA-related genes at a threshold of absolute log\u003csub\u003e2\u003c/sub\u003e fold-change more than 1 and adjusted p-value less than 0.05. Then, the raw count expression data, corresponding clinical and survival data of LUAD were downloaded from the TCGA data portal. Finally, a list of 214 LCCA-related genes and 412 LUAD cases were identified, and the cases was randomly split into training set and validation set (see the Methods section). The clinical characteristics of training set and validation set were listed in Table\u0026nbsp;3. To screen for candidate prognosis-related genes, Random Forest (RF) algorithm, LASSO regression (Figure \u003cspan refid=\"MOESM5\" class=\"InternalRef\"\u003eS5\u003c/span\u003eA-B) and stepwise regression were performed, and a union of the above results enrolled 23 candidate genes (Fig.\u0026nbsp;6A). In the univariate and multivariate Cox regression analyses, four genes (\u003cem\u003eKCNK3\u003c/em\u003e, \u003cem\u003eNRN1\u003c/em\u003e, \u003cem\u003ePARVB\u003c/em\u003e and \u003cem\u003eTRHDE-AS1\u003c/em\u003e) were finally identified to be the independently predictive factor for overall survival of LUAD (Fig.\u0026nbsp;6B, Figure \u003cspan refid=\"MOESM5\" class=\"InternalRef\"\u003eS5\u003c/span\u003eC-F). Then, we constructed a cystic-specific risk score (CSRS, \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(Risk Score= -0.249\\times KCNK3-0.321\\times NRN1+0.64\\times PARVB+0.571\\times TRHDE-AS1\\)\u003c/span\u003e\u003c/span\u003e). Kaplan-Meier analysis revealed that CSRS was dramatically associated with poor OS (Fig.\u0026nbsp;6C, Figure \u003cspan refid=\"MOESM5\" class=\"InternalRef\"\u003eS5\u003c/span\u003eG, p\u0026thinsp;\u0026lt;\u0026thinsp;0.0001). Multivariate Cox regression analysis suggested that CSRS (HR 3.33 [2.06-5.4], \u003cem\u003ep\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.001) was an independent predictor for OS (Figure \u003cspan refid=\"MOESM6\" class=\"InternalRef\"\u003eS6\u003c/span\u003eA-B).\u003c/p\u003e \u003cp\u003eFurther, we built a CSRS-based nomogram to predict the OS of LUAD (Fig.\u0026nbsp;6D). The nomogram model showed good performance in both the training set and validation set (Fig.\u0026nbsp;6E, Figure \u003cspan refid=\"MOESM5\" class=\"InternalRef\"\u003eS5\u003c/span\u003eH-I) with a c-index of 0.707 (95% CI 0.682\u0026ndash;0.732) in the whole population. The 1-year, 3-year and 5-year AUCs of ROC was 0.79 (0.70\u0026ndash;0.87), 0.75 (0.67\u0026ndash;0.83) and 0.78 (0.69\u0026ndash;0.88), respectively. Moreover, DCA showed that the nomogram, other than CSRS alone yielded better net benefit (Fig.\u0026nbsp;6F, Figure \u003cspan refid=\"MOESM5\" class=\"InternalRef\"\u003eS5\u003c/span\u003eJ-K).\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e"},{"header":"Discussion","content":"\u003cp\u003eLung cancer associated with cystic airspaces (LCCA) is a clinically rare type of lung cancer. Although many previous reports have delineated the clinicopathological and radiographic features, there is still a gap of bedside-to-bench recognition of LCCAs, from morphological and molecular pathological mechanisms to biological and genetic traits. To best of our knowledge, this is the first study to investigate the profiles of molecule regulatory network, as well as tumor and immune microenvironment at gene level. Our results indicate that LCCA and non-LCCA had substantial differences in tumor cell biology and immune microenvironment. Specifically, LCCAs are more inclined to exhibit migratory and immunosuppressive nature, and distinct BCR clonality and cell death orchestration. Moreover, a list of cystic-specific genes could predict the overall survival of LUAD.\u003c/p\u003e\n\u003cp\u003eIn line with previous study\u003csup\u003e[3, 39]\u003c/sup\u003e, we found that LCCA and micropapillary component were the risk factors for recurrence in early-stage lung cancer. It is reported that micropapillary component is associated with higher tumor mutational burden, genome alteration and copy number amplifications, whole-genome doubling rate\u003csup\u003e[40]\u003c/sup\u003e. In our study, we comprehensively score all the components of histological subtypes, and we found that LCCAs displayed higher histological heterogeneity and a trend for BCR diversity, though lower ITH scores inferred by \u0026ldquo;DEPTH\u0026rdquo; algorithm. The histological heterogeneity was more prominent in younger patients and never smokers. It is well-established that lymphovascular invasion (VPI) and tumor spread through air spaces (STAS) are unfavorable prognostic factor in early-stage lung cancer\u003csup\u003e[41, 42]\u003c/sup\u003e. LCCAs seemed to be more aggressive than non-LCCAs, in the settings of positive VPI or STAS in our study. Indeed, GSEA and GSVA analyses suggest that pathways associated with EMT, angiogenesis and distal metastasis were significantly enriched in LCCA. As suggested in our study, it might be related to the significant degradation of glycosaminoglycan, a component of extracellular matrix, which play an important role in cancer metastasis\u003csup\u003e[43]\u003c/sup\u003e. These functional traits are consistent with the above clinical manifestations.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eWe found LCCA exhibited distinct cell death pattern and metabolic profiles. The ferroptosis pathway was significantly up-regulated in LCCA. Ferroptosis is a manifestation of the imbalance between oxidative stress and antioxidant damage\u003csup\u003e[44]\u003c/sup\u003e. Besides a unique cell death pattern, it plays a dual role in anti-tumor immunity, which is related to the composition and function of immune cells in the tumor microenvironment\u003csup\u003e[45]\u003c/sup\u003e. During ferroptosis, the lipid peroxidation products can serve as an important \u0026quot;find me\u0026quot; signal to recruit antigen-presenting cells and other immune cells to the tumor microenvironment. However, some immune reactive cells are also sensitive to ferroptosis. Ma found that CD36-mediated fatty acid uptake induced ferroptosis of CD8+ T cells and weakened their anti-tumor activity, leading to tumor growth\u003csup\u003e[46]\u003c/sup\u003e. Moreover, there might also be a preference to activation of intrinsic apoptosis and suppression of extrinsic apoptosis in LCCA according our results. It is well-accepted that deregulation of apoptosis can lead to the development of different diseases and it can also influence the treatment of some of these diseases. It is necessary to understand the mechanisms by which way of apoptosis occurs and interacts with other biological systems. Therefore, further work is warranted to investigate the potential mechanisms that orchestrate the polarity of apoptotic pattern.\u003c/p\u003e\n\u003cp\u003eImmune evasion is a hallmark of malignancy. In fact, tumor cells have developed different mechanisms to dysfunction and circumvent the anti-tumor immune response\u003csup\u003e[47, 48]\u003c/sup\u003e. Immunosuppressive components in the TME, including myeloid-derived suppressive cells (MDSCs), tumor-associated macrophages (TAMs), regulatory T cells (Tregs), cancer-associated fibroblasts (CAFs), tumor-associated neutrophils, and tumor-associated dendritic cells, created a disordered niche where T cell reside or be excluded, and are critical factors correlated with immune resistance\u003csup\u003e[49]\u003c/sup\u003e. We found that gene sets associated TAMs, CAFs and MDSCs were up-regulated in LCCAs. Consistent with suppressor infiltration from inferred gene sets, the TME deconvolution analysis show high level of M2 macrophage polarization and CAFs activity.\u003c/p\u003e\n\u003cp\u003eImportantly, our study also highlights the impaired B cell roles in LCCA. The BCR/TCR repertoire identified significantly more unique BCR clonotypes the TCR clonotypes. However, a lowered BCR richness, clonality and high-abundance shared clonotypes were found in LCCA, which may further deteriorate the hormonal anti-tumor immunity. It is proposed that a \u0026ldquo;check-valve\u0026rdquo; mechanism plays a role in the formation of airspaces. Under this circumstance, as the gas continues increase, the alveolar wall was damaged by both tumor cells and high pressure, and resulting the destruction of elastic fibers and induction of stromal hyperplasia, which may pave the way for infiltration and activation of suppressors and hamper B cell clonal expansion.\u003c/p\u003e\n\u003cp\u003eIt should also be noted that the infer of BCR/TCR repertoire was based on alignment of RNA-seq data to a known VDJ sequence database. A more profound such as BCR/TCR sequencing at single cell level will provide a comprehensive perspective of BCR/TCR repertoire with huge V-, D-, J- section combinations. The limitation of this study came from the small sample size of sequenced, due to its clinical rarity, and the non-LCCA cohort was matched for TNM stage and histology, which may comprise selection bias. Yet, we evaluated the molecule regulatory network and immune profiles of LCCAs at gene level for the first time, which show a distinct biological trait as compared to non-LCCAs. Deeper sequencing of large sample size and inherent molecular mechanisms were warranted further investigation.\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003eEthics Statement\u003c/p\u003e\n\u003cp\u003eThe studies involving human participants were reviewed and approved by the Ethics Committee of the First Affiliated Hospital of Guangzhou Medical University and were carried out in accordance with the World Medical Association\u0026rsquo;s Declaration of Helsinki. The patients provided their written informed consent to participate in this study.\u003c/p\u003e\n\u003cp\u003eAuthorship contribution statement\u003c/p\u003e\n\u003cp\u003eQ.L, Z.X., H.Y., and L.J. contributed study design, data analysis and paper writing. C.R., X.K. and C.W.P. collected samples and generated data. W.W. and Z.X. reviewed the CT images and D.Y. examined the results. C.F, W.W, H.Z.X and H S H provided study materials or patients. All authors contributed to the article and approved the submitted version.\u003c/p\u003e\n\u003cp\u003eCompeting Interests\u003c/p\u003e\n\u003cp\u003eThe authors declare no competing interests.\u003c/p\u003e\n\u003cp\u003eFunding\u003c/p\u003e\n\u003cp\u003eThis work was supported by the Basic Research Program of Guangzhou Municipal Science and Technology Bureau (No: 2060206).\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\n\u003cli\u003eZHU H, ZHANG L, HUANG Z, et al. Lung adenocarcinoma associated with cystic airspaces: Predictive value of CT features in assessing pathologic invasiveness [J]. Eur J Radiol, 2023, 165: 110947.\u003c/li\u003e\n\u003cli\u003eMA Z, WANG S, ZHU H, et al. Comprehensive investigation of lung cancer associated with cystic airspaces: predictive value of morphology [J]. Eur J Cardiothorac Surg, 2022, 62(5).\u003c/li\u003e\n\u003cli\u003eSHEN Y, ZHANG Y, GUO Y, et al. Prognosis of lung cancer associated with cystic airspaces: A propensity score matching analysis [J]. Lung Cancer, 2021, 159: 111-6.\u003c/li\u003e\n\u003cli\u003eDETTERBECK F C, KUMBASAR U, LI A X, et al. Lung cancer with air lucency: a systematic review and clinical management guide [J]. J Thorac Dis, 2023, 15(2): 731-46.\u003c/li\u003e\n\u003cli\u003eTAN Y, GAO J, WU C, et al. CT Characteristics and Pathologic Basis of Solitary Cystic Lung Cancer [J]. Radiology, 2019, 291(2): 495-501.\u003c/li\u003e\n\u003cli\u003eHANSELL D M, BANKIER A A, MACMAHON H, et al. Fleischner Society: glossary of terms for thoracic imaging [J]. Radiology, 2008, 246(3): 697-722.\u003c/li\u003e\n\u003cli\u003eFAROOQI A O, CHAM M, ZHANG L, et al. Lung cancer associated with cystic airspaces [J]. AJR Am J Roentgenol, 2012, 199(4): 781-6.\u003c/li\u003e\n\u003cli\u003eFINTELMANN F J, BRINKMANN J K, JECK W R, et al. Lung Cancers Associated With Cystic Airspaces: Natural History, Pathologic Correlation, and Mutational Analysis [J]. J Thorac Imaging, 2017, 32(3): 176-88.\u003c/li\u003e\n\u003cli\u003eMASCALCHI M, ATTINA D, BERTELLI E, et al. Lung cancer associated with cystic airspaces [J]. J Comput Assist Tomogr, 2015, 39(1): 102-8.\u003c/li\u003e\n\u003cli\u003eTOYOKAWA G, SHIMOKAWA M, KOZUMA Y, et al. Invasive features of small-sized lung adenocarcinoma adjoining emphysematous bullae [J]. Eur J Cardiothorac Surg, 2018, 53(2): 372-8.\u003c/li\u003e\n\u003cli\u003eWATANABE Y, KUSUMOTO M, YOSHIDA A, et al. Cavity Wall Thickness in Solitary Cavitary Lung Adenocarcinomas Is a Prognostic Indicator [J]. Ann Thorac Surg, 2016, 102(6): 1863-71.\u003c/li\u003e\n\u003cli\u003eSHEN Y, XU X, ZHANG Y, et al. Lung cancers associated with cystic airspaces: CT features and pathologic correlation [J]. Lung Cancer, 2019, 135: 110-5.\u003c/li\u003e\n\u003cli\u003eANDREWS S. FastQC: A Quality Control Tool for High Throughput Sequence Data [Z]. 2010\u003c/li\u003e\n\u003cli\u003eBOLGER A M, LOHSE M, USADEL B. Trimmomatic: a flexible trimmer for Illumina sequence data [J]. Bioinformatics, 2014, 30(15): 2114-20.\u003c/li\u003e\n\u003cli\u003eKIM D, PAGGI J M, PARK C, et al. Graph-based genome alignment and genotyping with HISAT2 and HISAT-genotype [J]. Nat Biotechnol, 2019, 37(8): 907-15.\u003c/li\u003e\n\u003cli\u003eANDERS S, PYL P T, HUBER W. HTSeq--a Python framework to work with high-throughput sequencing data [J]. Bioinformatics, 2015, 31(2): 166-9.\u003c/li\u003e\n\u003cli\u003eLOVE M I, HUBER W, ANDERS S. Moderated estimation of fold change and dispersion for RNA-seq data with DESeq2 [J]. Genome Biol, 2014, 15(12): 550.\u003c/li\u003e\n\u003cli\u003eWU T, HU E, XU S, et al. clusterProfiler 4.0: A universal enrichment tool for interpreting omics data [J]. Innovation (Camb), 2021, 2(3): 100141.\u003c/li\u003e\n\u003cli\u003eHANZELMANN S, CASTELO R, GUINNEY J. GSVA: gene set variation analysis for microarray and RNA-seq data [J]. BMC Bioinformatics, 2013, 14: 7.\u003c/li\u003e\n\u003cli\u003eJEGGARI A, MARKS D S, LARSSON E. miRcode: a map of putative microRNA target sites in the long non-coding transcriptome [J]. Bioinformatics, 2012, 28(15): 2062-3.\u003c/li\u003e\n\u003cli\u003eWONG N, WANG X. miRDB: an online resource for microRNA target prediction and functional annotations [J]. Nucleic Acids Res, 2015, 43(Database issue): D146-52.\u003c/li\u003e\n\u003cli\u003eHUANG H Y, LIN Y C, CUI S, et al. miRTarBase update 2022: an informative resource for experimentally validated miRNA-target interactions [J]. Nucleic Acids Res, 2022, 50(D1): D222-d30.\u003c/li\u003e\n\u003cli\u003eAGARWAL V, BELL G W, NAM J W, et al. Predicting effective microRNA target sites in mammalian mRNAs [J]. Elife, 2015, 4.\u003c/li\u003e\n\u003cli\u003eLI M, ZHANG Z, LI L, et al. An algorithm to quantify intratumor heterogeneity based on alterations of gene expression profiles [J]. Commun Biol, 2020, 3(1): 505.\u003c/li\u003e\n\u003cli\u003eZHANG Y, TSENG J T, LIEN I C, et al. mRNAsi Index: Machine Learning in Mining Lung Adenocarcinoma Stem Cell Biomarkers [J]. Genes (Basel), 2020, 11(3).\u003c/li\u003e\n\u003cli\u003eCHAROENTONG P, FINOTELLO F, ANGELOVA M, et al. Pan-cancer Immunogenomic Analyses Reveal Genotype-Immunophenotype Relationships and Predictors of Response to Checkpoint Blockade [J]. Cell Rep, 2017, 18(1): 248-62.\u003c/li\u003e\n\u003cli\u003eARAN D. Cell-Type Enrichment Analysis of Bulk Transcriptomes Using xCell [J]. Methods Mol Biol, 2020, 2120: 263-76.\u003c/li\u003e\n\u003cli\u003eJIANG P, GU S, PAN D, et al. Signatures of T cell dysfunction and exclusion predict cancer immunotherapy response [J]. Nat Med, 2018, 24(10): 1550-8.\u003c/li\u003e\n\u003cli\u003eBOLOTIN D A, POSLAVSKY S, DAVYDOV A N, et al. Antigen receptor repertoire profiling from RNA-seq data [J]. Nat Biotechnol, 2017, 35(10): 908-11.\u003c/li\u003e\n\u003cli\u003eTEAM I. immunarch: An R Package for Painless Analysis of Large-Scale Immune Repertoire Data [Z]. 2019\u003c/li\u003e\n\u003cli\u003eMOREIRA A L, OCAMPO P S S, XIA Y, et al. A Grading System for Invasive Pulmonary Adenocarcinoma: A Proposal From the International Association for the Study of Lung Cancer Pathology Committee [J]. J Thorac Oncol, 2020, 15(10): 1599-610.\u003c/li\u003e\n\u003cli\u003eMASIERO M, SIMOES F C, HAN H D, et al. A core human primary tumor angiogenesis signature identifies the endothelial orphan receptor ELTD1 as a key regulator of angiogenesis [J]. Cancer Cell, 2013, 24(2): 229-41.\u003c/li\u003e\n\u003cli\u003eZENG D, LI M, ZHOU R, et al. Tumor Microenvironment Characterization in Gastric Cancer Identifies Prognostic and Immunotherapeutically Relevant Gene Signatures [J]. Cancer Immunol Res, 2019, 7(5): 737-50.\u003c/li\u003e\n\u003cli\u003eTIROSH I, IZAR B, PRAKADAN S M, et al. Dissecting the multicellular ecosystem of metastatic melanoma by single-cell RNA-seq [J]. Science, 2016, 352(6282): 189-96.\u003c/li\u003e\n\u003cli\u003eDAMRAUER J S, HOADLEY K A, CHISM D D, et al. Intrinsic subtypes of high-grade bladder cancer reflect the hallmarks of breast cancer biology [J]. Proc Natl Acad Sci U S A, 2014, 111(8): 3110-5.\u003c/li\u003e\n\u003cli\u003eSJODAHL G, LAUSS M, LOVGREN K, et al. A molecular taxonomy for urothelial carcinoma [J]. Clin Cancer Res, 2012, 18(12): 3377-86.\u003c/li\u003e\n\u003cli\u003eKANG Y, SIEGEL P M, SHU W, et al. A multigenic program mediating breast cancer metastasis to bone [J]. Cancer Cell, 2003, 3(6): 537-49.\u003c/li\u003e\n\u003cli\u003eAZIZI E, CARR A J, PLITAS G, et al. Single-Cell Map of Diverse Immune Phenotypes in the Breast Tumor Microenvironment [J]. Cell, 2018, 174(5): 1293-308 e36.\u003c/li\u003e\n\u003cli\u003eLI Y, BYUN A J, CHOE J K, et al. Micropapillary and Solid Histologic Patterns in N1 and N2 Lymph Node Metastases Are Independent Factors of Poor Prognosis in Patients With Stages II to III Lung Adenocarcinoma [J]. J Thorac Oncol, 2023, 18(5): 608-19.\u003c/li\u003e\n\u003cli\u003eCASO R, SANCHEZ-VEGA F, TAN K S, et al. The Underlying Tumor Genomics of Predominant Histologic Subtypes in Lung Adenocarcinoma [J]. J Thorac Oncol, 2020, 15(12): 1844-56.\u003c/li\u003e\n\u003cli\u003eURUGA H, FUJII T, FUJIMORI S, et al. Semiquantitative Assessment of Tumor Spread through Air Spaces (STAS) in Early-Stage Lung Adenocarcinomas [J]. J Thorac Oncol, 2017, 12(7): 1046-51.\u003c/li\u003e\n\u003cli\u003eVAAHTOMERI K, ALITALO K. Lymphatic Vessels in Tumor Dissemination versus Immunotherapy [J]. Cancer Res, 2020, 80(17): 3463-5.\u003c/li\u003e\n\u003cli\u003eOLIVEIRA-FERRER L, LEGLER K, MILDE-LANGOSCH K. Role of protein glycosylation in cancer metastasis [J]. Semin Cancer Biol, 2017, 44: 141-52.\u003c/li\u003e\n\u003cli\u003eSNYDER A G, HUBBARD N W, MESSMER M N, et al. Intratumoral activation of the necroptotic pathway components RIPK1 and RIPK3 potentiates antitumor immunity [J]. Sci Immunol, 2019, 4(36).\u003c/li\u003e\n\u003cli\u003eLEI G, ZHUANG L, GAN B. Targeting ferroptosis as a vulnerability in cancer [J]. Nat Rev Cancer, 2022, 22(7): 381-96.\u003c/li\u003e\n\u003cli\u003eKIM R, HASHIMOTO A, MARKOSYAN N, et al. Ferroptosis of tumour neutrophils causes immune suppression in cancer [J]. Nature, 2022, 612(7939): 338-46.\u003c/li\u003e\n\u003cli\u003eVINAY D S, RYAN E P, PAWELEC G, et al. Immune evasion in cancer: Mechanistic basis and therapeutic strategies [J]. Semin Cancer Biol, 2015, 35 Suppl: S185-S98.\u003c/li\u003e\n\u003cli\u003eSHARMA P, HU-LIESKOVAN S, WARGO J A, et al. Primary, Adaptive, and Acquired Resistance to Cancer Immunotherapy [J]. Cell, 2017, 168(4): 707-23.\u003c/li\u003e\n\u003cli\u003eTIE Y, TANG F, WEI Y Q, et al. Immunosuppressive cells in cancer: mechanisms and potential therapeutic targets [J]. J Hematol Oncol, 2022, 15(1): 61.\u003c/li\u003e\n\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":true,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true},"keywords":"Lung cancer, Lung cancer associated with cystic airspaces, Prognosis, Gene expression, BCR/TCR repertoire","lastPublishedDoi":"10.21203/rs.3.rs-3448810/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-3448810/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003e\u003cstrong\u003eObjective:\u003c/strong\u003e To explore the molecular biological characteristics of lung cancer associated with cystic airspaces (LCCA) and its potential roles on prognosis.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eMethods:\u003c/strong\u003e A total of 165 LCCAs and 201 non-LCCAs were enrolled in this study. Bulk RNA sequencing was implemented in eight LCCAs and nine non-LCCAs to explore the differentially expressed genes. TCGA data were used to analyze LCCA-specific genes that associated with overall survival.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eResults:\u003c/strong\u003e The median age was 60 (IQR 53 to 65) years in LCCA cohort. We found LCCA were predominant in men and had less visceral pleura invasion (VPI) or lympho-vascular invasion (LVI). Moreover, LCCA presented with higher histological heterogeneity. Kaplan-Meier analysis showed that patients of age more than 60 and positive VPI had significantly less PFS in LCCA. Cox regression suggested that LCCA, micropapillary subtype proportion and VPI were the independent risk factors for PFS. LCCA had up-regulated pathways associated with EMT, angiogenesis and cell migration. In addition, LCCA displayed higher levels of immunosuppressor infiltration (M2 macrophages, CAFs and MDSCs) and distinct cell death and metabolic patterns. BCR/TCR repertoire analysis revealed less BCR richness, clonality and high-abundance shared clonotypes in LCCA. Finally, Cox regression analysis identified that four cystic-specific genes, \u003cem\u003eKCNK3\u003c/em\u003e, \u003cem\u003eNRN1\u003c/em\u003e, \u003cem\u003ePARVB\u003c/em\u003e and \u003cem\u003eTRHDE-AS1\u003c/em\u003e,\u003cem\u003e \u003c/em\u003ewere associated with OS of LUAD. And cystic-specific risk scores (CSRSs) were calculated to construct a nomogram, which performance well.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eConclusions:\u003c/strong\u003e Our study for the first time indicated significantly distinct molecular biological and immune characteristics between LCCA and non-LCCA, which provide complementary prognostic values in early-stage NSCLC.\u003c/p\u003e","manuscriptTitle":"Exploring the Molecular and Immune Landscape of Lung Cancer Associated with Cystic Airspaces: Implications for Prognosis and Therapeutic Strategies","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2023-10-17 17:19:28","doi":"10.21203/rs.3.rs-3448810/v1","editorialEvents":[{"type":"communityComments","content":0}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"6ab9a339-8a71-4ad3-a438-905097e416a8","owner":[],"postedDate":"October 17th, 2023","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[],"tags":[],"updatedAt":"2023-10-31T01:59:09+00:00","versionOfRecord":[],"versionCreatedAt":"2023-10-17 17:19:28","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-3448810","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-3448810","identity":"rs-3448810","version":["v1"]},"buildId":"WrCJVZZCHTDjtuVLN7oU0","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}