Single-cell transcriptome analysis and multi-omics-based biomarker for predicting responsiveness to immune checkpoint inhibitor therapy, and information provision method using same

Abstract

The present invention relates to a biomarker and an information provision method, for predicting whether or not a mutation is present in an epidermal growth factor receptor (EGFR) by utilizing single-cell multi-omics analysis in a non-small cell lung cancer (NSCLC) patient, and predicting responsiveness to immune checkpoint inhibitor (ICI) therapy. The present invention provides, as an information provision method for predicting responsiveness to ICI therapy in an NSCLC patient, the information provision method comprising the steps of: analyzing a sub-population composition ratio of CD8-positive T cells by performing single-cell multi-omics analysis of a lung sample of a patient; and measuring the frequency of dysfunctional CD8 T cells (CD8_Td) and the frequency of naive-like cells (CD8_Tn) among the CD8-positive T cells, wherein, when the frequency of CD8_Td decreases and the frequency of CD8_Tn increases in EGFR mutation-positive patients, it is predicted that responsiveness to ICI therapy is low.

Description

Single-cell transcriptome analysis and multi-omics-based biomarkers for predicting immune checkpoint inhibitor therapeutic responsiveness and a method for providing information using the same

[0001] The present invention relates to a biomarker and a method for providing information to predict the presence of epidermal growth factor receptor (EGFR) mutations and the responsiveness to immune checkpoint inhibitor (ICI) treatment using single-cell multi-omics analysis in patients with non-small cell lung cancer (NSCLC).

[0002] Tumor cells act on host immunity in several ways to evade immune defenses in the tumor microenvironment. This phenomenon is generally referred to as "cancer immune escape." One of the most important components of this system is the immunosuppressive co-signal (immune checkpoint) mediated by the PD1 (Programmed cell death protein 1) receptor and its ligand, PD-L1 (Programmed death-ligand 1).

[0003] PD-1 receptors and PD-L1 ligands play essential roles in immune modulation. PD-1, expressed on activated T cells, is activated by PD-L1 and PD-L2, expressed by stromal cells, tumor cells, or both, thereby initiating T-cell apoptosis and localized immunosuppression, potentially providing an immunogenic environment for tumor development and growth. Conversely, inhibition of these interactions can enhance localized T-cell responses and mediate anti-tumor activity.

[0004] Therefore, recently, immunotherapy, specifically immune checkpoint inhibitors (ICIs), has been developed and is being used in cancer treatment as a therapeutic method that activates the human immune system to kill cancer cells. The most widely used immune checkpoint inhibitors are those using monoclonal antibodies against PD-1 or PD-L1. These inhibitors eliminate cancer cells by inhibiting the binding of PD-1 on the surface of T cells to PD-L1 on the surface of cancer cells, thereby activating immune cells such as T cells. This has dramatically increased the survival rate of cancer patients, and it has been reported that in some patients, there has been no recurrence of cancer for several years.

[0005] However, it has been reported that even when these inhibitors are used, patients do not respond, resulting in low therapeutic efficacy, although the specific cause has not been identified. In particular, while there are clinical observations that non-small cell lung cancer patients with EGFR mutations show low responsiveness to immunotherapies, a comprehensive single-cell level immunological elucidation of the molecular mechanisms has not been sufficiently conducted.

[0006] Tertiary lymphoid structures (TLS) are ectopic lymphoid tissues formed in chronically inflammatory tissues, particularly in the tumor microenvironment, and play an important role in tumor immune responses. Follicular helper T cells (Tfh) and the chemokine CXCL13 secreted by them play a key role in the formation of TLS. CXCL13, also known as CXC motif chemokine ligand 13, is a small cytokine with chemotactic activity against B cells.

[0007] The composition and functional status of CD4-positive and CD8-positive T cell subpopulations within a tumor are important factors determining the quality and direction of the tumor immune response. However, no studies have been reported to date that simultaneously analyzed the transcriptome, surface proteome, and T cell receptor repertoire at the single-cell level to analyze the effects of EGFR mutations on the composition and function of T cell subpopulations within a tumor.

[0008] Therefore, it is necessary to precisely elucidate the tumor immune microenvironment of EGFR-mutated non-small cell lung cancer patients through single-cell multi-omics analysis, and based on this, to develop reliable biomarkers for predicting EGFR mutations and responsiveness to immune checkpoint inhibitor treatment.

[0009] The present invention aims to provide a method for providing information to predict the presence of EGFR mutations using single-cell multi-omics analysis in patients with non-small cell lung cancer.

[0010] The present invention aims to provide a method for providing information to predict therapeutic responsiveness to immune checkpoint inhibitors in patients with non-small cell lung cancer.

[0011] The present invention aims to provide a method for evaluating the inhibition of tumor-associated tertiary lymphoid structure formation by EGFR mutations in non-small cell lung cancer.

[0012] The present invention aims to provide a biomarker panel for determining the EGFR mutation status of patients with non-small cell lung cancer.

[0013] The present invention aims to provide a diagnostic kit for determining the EGFR mutation status of patients with non-small cell lung cancer.

[0014] The purposes of the present disclosure are not limited to those mentioned above, and other purposes and advantages of the present disclosure not mentioned may be understood from the following description and will be more clearly understood by the embodiments of the present disclosure. Furthermore, it will be readily apparent that the purposes and advantages of the present disclosure can be realized by the means and combinations thereof set forth in the claims.

[0015] The present invention provides a method for providing information to predict whether an epidermal growth factor receptor (EGFR) mutation exists in a patient with non-small cell lung cancer (NSCLC), comprising the steps of: performing single-cell multi-omics analysis on a lung sample of the patient to analyze the composition ratio of CD4-positive T cells in a subgroup; and measuring the frequency of follicular helper T cells (CD4_Tfh) among the CD4-positive T cells and the level of CXCL13 expression, wherein if the frequency of follicular helper T cells and CXCL13 expression are significantly reduced compared to a normal control group or an EGFR wild-type patient, the patient is predicted to have an EGFR mutation.

[0016] In addition, the present invention provides a method for providing information for predicting the EGFR mutation status and responsiveness to immune checkpoint inhibitor treatment in patients with non-small cell lung cancer using a biomarker panel that essentially includes CXCL13 and additionally includes two or more genes selected from the group consisting of MS4A1, BANK1, CXCR4, IL7R, HPGD, AQP1, RAB11FIP1, SPOCK2, POU2F, CXCL8, TNFAIP3, and IGFBP3.

[0017] The above EGFR mutations may be one or more selected from the group consisting of exon 19 deletion (E19del) mutations and L858R point mutations.

[0018] The above subgroup of CD4-positive T cells may include naive-like T cells (CD4_Tn), regulatory T cells (CD4_Treg), effector / memory T cells (CD4_Tem), and follicular helper T cells (CD4_Tfh).

[0019] The present invention also provides a method for providing information for predicting the therapeutic responsiveness to an immune checkpoint inhibitor (ICI) in a patient with non-small cell lung cancer, comprising the steps of: performing single-cell multi-omics analysis on a lung sample of the patient to analyze the subpopulation composition ratio of CD8-positive T cells; and measuring the frequency of dysfunctional CD8 T cells (CD8_Td) and the frequency of naive-like cells (CD8_Tn) among the CD8-positive T cells, wherein if the frequency of CD8_Td decreases and the frequency of CD8_Tn increases in a patient with EGFR mutation-positive cancer, the therapeutic responsiveness to the immune checkpoint inhibitor is predicted to be low.

[0020] The above subgroup of CD8-positive T cells may include uncontacted T-like cells (CD8_Tn), predysfunctional T cells (CD8_Tpd), dysfunctional T cells (CD8_Td), and cytotoxic T cells (CD8_Tc).

[0021] The above single-cell multi-omics analysis may involve simultaneously performing whole transcriptome analysis (WTA), surface proteome analysis (AbSeq / CITE-seq), and T-cell receptor / B-cell receptor (TCR / BCR) repertoire analysis at the single-cell level.

[0022] The above single-cell multi-omics analysis may involve capturing a single cell using a microwell cartridge system and constructing a library using the BD Rhapsody WTA Amplification reagent kit and the BD Rhapsody TCR / BCR Assays reagent kit.

[0023] The present invention also provides a method for evaluating inhibition of tumor-associated tertiary lymphoid structure (TLS) formation by EGFR mutations in non-small cell lung cancer, comprising the steps of: measuring follicular helper T cell (CD4_Tfh) frequency and CXCL13 expression in a patient's lung tumor sample through single-cell multi-omics analysis; and evaluating CD4_Tfh frequency in a selected tumor-responsive T cell population through TCR clonality analysis, wherein the reduction in CD4_Tfh frequency and CXCL13 expression is utilized as an indicator of inhibition of TLS formation by EGFR mutations.

[0024] The present invention also provides a biomarker panel for determining the EGFR mutation status of a patient with non-small cell lung cancer (NSCLC), comprising: (a) cell frequency (%) of follicular helper T cells (CD4_Tfh) within a CD4-positive T cell population; (b) CXCL13 mRNA expression levels in said CD4_Tfh cells; (c) cell frequency (%) of dysfunctional T cells (CD8_Td) within a CD8-positive T cell population; and (d) cell frequency (%) of uncontacted T-like cells (CD8_Tn) within a CD8-positive T cell population, wherein, in the case of EGFR mutation positivity, the values ​​of said (a) and (b) are decreased, said (c) is decreased, and said (d) is increased compared to an EGFR wild-type or normal control.

[0025] The present invention also provides a diagnostic kit for determining the EGFR mutation status of a patient with non-small cell lung cancer (NSCLC), comprising: one or more gene expression measurement reagents selected from the group consisting of PDCD1, ICOS, FOXP3, IL2RA, IFNG, SELL, and TCF7 for classifying CD4-positive T cells into follicular helper T cells (CD4_Tfh), regulatory T cells (CD4_Treg), effector memory T cells (CD4_Tem), and uncontacted T-like cells (CD4_Tn); and one or more gene expression measurement reagents selected from the group consisting of CXCL13 and IL21 for measuring the functional activity of follicular helper T cells in the classified CD4_Tfh, wherein the measurement reagents are primers, probes, or antibodies that specifically bind to the mRNA of the gene or to the protein encoded by the gene.

[0026] The present invention also provides a multi-gene signature composition for NSCLC companion diagnosis comprising a preparation for measuring mRNA or protein levels of 10 or more genes selected from CXCL13 and Group 2 functional markers.

[0027] The present invention also provides a liquid biopsy-based information-providing method for non-invasively providing information for predicting immunotherapeutic responsiveness or EGFR mutation status in a patient with non-small cell lung cancer (NSCLC), comprising the step of measuring the expression levels of 10 or more combinations selected from CXCL13 and Group 2 functional markers from the patient's blood, plasma, serum, or circulating immune cells.

[0028] The present invention can provide information for non-invasively predicting the presence of EGFR mutations by measuring the frequency of follicular helper T cells (CD4_Tfh) among CD4-positive T cell subpopulations and the CXCL13 expression level using single-cell multi-omics analysis in patients with non-small cell lung cancer.

[0029] The present invention can prevent unnecessary immunotherapy and contribute to establishing effective treatment strategies by providing information to predict low therapeutic responsiveness to immune checkpoint inhibitors in patients with EGFR mutations through compositional analysis of CD8-positive T cell subgroups.

[0030] The present invention provides a novel biomarker panel and diagnostic kit that elucidates the mechanism of inhibition of tertiary lymphoid structure (TLS) formation within tumors by EGFR mutations at the single-cell level, which can be utilized for precise immunological classification and companion diagnosis of non-small cell lung cancer patients.

[0031] The present invention can provide a method for providing information for predicting therapeutic responsiveness or prognosis to immune checkpoint inhibitors using biomarkers based on single-cell multi-omics analysis.

[0032] The present invention can dramatically improve diagnostic accuracy and reliability through a biomarker panel combining two or more of MS4A1, BANK1, CXCR4, IL7R, HPGD, AQP1, RAB11FIP1, SPOCK2, POU2F, CXCL8, TNFAIP3, and IGFBP3 compared to the CXCL13 single marker.

[0033] Figure 1 shows an overall overview of a single-cell multi-omics analysis method using non-small cell lung cancer patients and normal controls.

[0034] Figure 2 shows the results of the cellular heterogeneity analysis of CD4-positive T cells. (a) Single-cell UMAP visualization and comparison of CD4 T cell subgroup frequencies by normal / tumor group, and (b) density distribution of characteristic marker genes (SELL, TCF7, CCR7, IFNG, CCL3, CCL4, FOXP3, IL2RA, CTLA4, PDCD1, IL21, CXCL13) of each subgroup.

[0035] Figure 3 shows the results of the analysis of cell heterogeneity by EGFR mutation type in CD4-positive T cells. (c) frequency of CD4 T cell subgroups by EGFR type (Wild, E19del, L858R) in tumor and clonal T cell populations, (d) CD4_Tfh cell density UMAP by EGFR type, (e) frequency distribution by EGFR type in each subgroup, (f) volcano plot of differentially expressed genes in CD4_Tfh, (g) CXCL13 expression UMAP by EGFR type, and (h) violin plot of CXCL13 expression levels for each CD4 subgroup.

[0036] Figure 4 shows the results of the analysis of the status of CD8-positive T cells. (a) Single-cell UMAP visualization and comparison of CD8 T cell subgroup frequencies by normal / tumor group, (b) density distribution of characteristic marker genes (IL7R, CCR7, TCF7, GZMK, PDCD1, CTLA4, FCGR3A, PRF1, CX3CR1) and surface proteomes (TIGIT, CD279 / PD-1, CD39, CD278 / ICOS, CD103, Tim-3, CD226, CD11b, CD56, CD45RA) of each subgroup.

[0037] Figure 5 shows the results of the status analysis of CD8-positive T cells by EGFR mutation type. (c) frequency of CD8 T cell subgroups by EGFR type in tumor and clonal T cell populations, (d) CD8_Td cell density UMAP by EGFR type, (e) frequency distribution by EGFR type for each subgroup, and (f) violin plots of IL7R, GZMK, CTLA4, and CXCL13 expression levels for each CD8 subgroup.

[0038] Figure 6 shows a biomarker panel for diagnosing non-small cell lung cancer selected according to the present invention and expression levels according to cell type.

[0039] The inventors confirmed that non-small cell lung cancer patients with EGFR mutations have low responsiveness to immunotherapies and precisely analyzed the intratumoral T cell immune environment of these patients through single-cell multi-omics analysis. As a result, it was confirmed that in non-small cell lung cancer patients with EGFR mutations, the frequency of follicular helper T cells (CD4_Tfh) among CD4-positive T cells is significantly reduced and the expression level of CXCL13 is lowered, thereby inhibiting the formation of tertiary lymphoid structures (TLS); along with this, the frequency of dysfunctional T cells (CD8_Td) among CD8-positive T cells is reduced and the frequency of uncontacted T-like cells (CD8_Tn) is increased, thereby significantly reducing the population of tumor-responsive T cells that can be reactivated by immune checkpoint inhibitors. The present invention is based on these findings.

[0040] In this specification, "non-small cell lung cancer (NSCLC)" is a major type of lung cancer, including adenocarcinoma, squamous cell carcinoma, large cell carcinoma, etc., and accounts for more than 85% of all lung cancers.

[0041] In this specification, "EGFR (epidermal growth factor receptor) mutation" means an activating mutation occurring in the EGFR gene, including, but not limited to, exon 19 deletion (E19del) mutations and L858R point mutations in exon 21. E19del mutations and L858R point mutations are the most common EGFR activating mutations observed in non-small cell lung cancer, accounting for about 85–90% of all EGFR mutations.

[0042] In this specification, "single cell multi-omics analysis" refers to a technology that simultaneously analyzes multiple omics information, such as genomics, transcriptomics, and proteomics, at the single-cell level. The present invention is characterized by simultaneously performing whole transcriptome analysis (WTA), surface proteomics (AbSeq / CITE-seq), and T-cell receptor / B-cell receptor (TCR / BCR) repertoire analysis at the single-cell level.

[0043] In this specification, "biomarker panel" refers to a marker combination comprising CXCL13 as an essential marker and additionally including two or more genes selected from the group consisting of MS4A1, BANK1, CXCR4, IL7R, HPGD, AQP1, RAB11FIP1, SPOCK2, POU2F, CXCL8, TNFAIP3, and IGFBP3. Preferably, it includes all of CXCL13, MS4A1, BANK1, CXCR4, IL7R, HPGD, AQP1, RAB11FIP1, SPOCK2, POU2F, CXCL8, TNFAIP3, and IGFBP3. The markers are derived by analyzing single-cell transcriptome and multi-omics analysis (scRNA-seq) data using the Seurat algorithm and consist of B-cell specific markers, chemokine receptors, immune regulatory factors, and functional markers related to TLS formation. By combining two or more of the above genes compared to the CXCL13 single marker, the reliability of determining EGFR mutation status and predicting ICI treatment responsiveness can be increased.

[0044] In this specification, "subgroups of CD4-positive T cells" refer to a population of CD4-expressing T cells classified according to their molecular function and expression markers, and include naive-like T cells (CD4_Tn), regulatory T cells (CD4_Treg), effector / memory T cells (CD4_Tem), and follicular helper T cells (CD4_Tfh). Each subgroup is distinguished by the following characteristic marker genes.

[0045] CD4_Tn (non-contact T-like cells) are characterized by high expression of SELL (CD62L), TCF7, and CCR7. CD4_Treg (regulatory T cells) are characterized by high expression of FOXP3, IL2RA (CD25), and CTLA4. CD4_Tem (effect memory T cells) are characterized by high expression of IFNG, CCL3, and CCL4. CD4_Tfh (follicular helper T cells) are characterized by high expression of PDCD1 (PD-1), ICOS, CXCL13, and IL21, among which CXCL13 is a key chemokine that promotes TLS formation.

[0046] In this specification, "subgroups of CD8-positive T cells" are T cell populations expressing CD8 classified according to their functional status, and include uncontacted T cells (CD8_Tn), predysfunctional T cells (CD8_Tpd), dysfunctional T cells (CD8_Td), and cytotoxic T cells (CD8_Tc). CD8_Tn has the characteristic of highly expressing IL7R, CCR7, and TCF7, CD8_Tpd has the characteristic of highly expressing GZMK, CD8_Td has the characteristic of highly expressing multiple immune checkpoint molecules such as PDCD1 (PD-1), CTLA4, and TIGIT, and CD8_Tc has the characteristic of highly expressing FCGR3A, PRF1, and CX3CR1.

[0047] In this specification, "follicular helper T cell (CD4_Tfh)" is a subgroup of CD4-positive T cells that supports germinal center responses and assists in the production of antibodies by B cells. In the tumor microenvironment, CD4_Tfh plays a key role in the formation and maintenance of TLS and secretes CXCL13 to attract B cells to the tumor site.

[0048] In this specification, "tertiary lymphoid structure (TLS)" refers to ectopic lymphoid tissue formed within the tumor microenvironment, meaning a structured aggregate of immune cells containing both T cell regions and B cell regions. The presence of TLS within a tumor is known to have a positive correlation with responsiveness to immunotherapy and prognosis in various cancer types.

[0049] In this specification, "CXCL13" is also referred to as CXC motif chemokine ligand 13 and is a small circulating cytokine that is chemotactic toward B cells. CXCL13 is primarily secreted by CD4_Tfh cells and serves to attract B cells and T cells expressing CXCR5 to the TLS.

[0050] In this specification, "level (or expression level) of the gene or protein of CXCL13" is used to mean both the mRNA expression level of the CXCL13 gene and the expression level of the CXCL13 protein expressed therefrom.

[0051] In this specification, "dysfunctional CD8 T cell (CD8_Td)" refers to CD8-positive T cells whose function is impaired by chronic antigen stimulation in the tumor microenvironment, and is characterized by high expression of immune checkpoint molecules such as PD-1, CTLA4, and TIGIT. These cells are the most important target population that can be reactivated by immune checkpoint inhibitors (ICIs), and it is expected that the higher the frequency of CD8_Td within the tumor, the higher the therapeutic responsiveness to ICIs.

[0052] In this specification, "immune checkpoint inhibitor (ICI)" means a substance that inhibits, interferes with, or modulates one or more immune checkpoint proteins wholly or partially, and includes, but is not limited to, drugs that specifically bind to PD-1 (Programmed cell death protein 1) or PDL1 (Programmed death-ligand 1). In one specific example, the drug that specifically binds to PD-1 may be an anti-PD-1 antibody, and the drug that specifically binds to PD-L1 may be an anti-PD-L1 antibody.

[0053] In this specification, "liquid biopsy" refers to a diagnostic technique that obtains tumor-related information non-invasively using blood-derived samples such as blood, plasma, serum, or circulating immune cells. In the present invention, by measuring the expression levels of a multi-gene panel in circulating immune cells in the blood, it can be utilized to non-invasively predict the EGFR mutation status and responsiveness to ICI treatment.

[0054] In this specification, "TCR clonality analysis" is an analysis method for selecting a population of T cells that have clonally proliferated in response to the same antigen (clonotype) by analyzing the V(D)J recombination sequence of a T cell receptor (TCR). Since T cells with clonality are highly likely to be tumor-responsive T cells that respond to tumor antigens, selecting tumor-responsive T cells through TCR clonality analysis is useful for more accurately analyzing the population of T cells that participate in the actual anti-tumor immune response.

[0055] In the present specification, the agent for measuring the level of the CXCL13 gene may be a primer or probe that specifically binds to the CXCL13 gene, but is not limited thereto. Analysis methods for this purpose include reverse transcription polymerase chain reaction (RT-PCR), competitive reverse transcription polymerase chain reaction (Competitive RT-PCR), real-time reverse transcription polymerase chain reaction (Real-time RT-PCR), RNase protection assay (RPA), Northern blotting, RNA-sequencing (RNA-seq), nanostring, DNA microarray chip, etc., but are not limited thereto as long as they are methods capable of measuring the expression level of mRNA.

[0056] In the present specification, the agent for measuring the level of the CXCL13 protein may be an antibody or aptamer that specifically binds to the CXCL13 protein, but is not limited thereto. Analytical methods for this purpose include western blotting, ELISA (enzyme-linked immunosorbent assay), radioimmunoassay, radioimmunodiffusion, immunohistochemical staining, immunoprecipitation assay, complete fixation assay, flow cytometry (FACS), protein chip, ligand binding assay, etc., but are not limited thereto as long as they are methods capable of measuring the expression level of the protein.

[0057] In this specification, "diagnostic kit" refers to a diagnostic device capable of determining the status of EGFR mutations by measuring the level of a specific gene or protein in a sample, and is not limited to any form that can confirm the amount of the said gene or protein from a biological sample isolated from a patient. The kit may take the form of a gene amplification kit, a microarray chip, a next-generation sequencing (NGS) panel, a flow cytometry panel, etc., but is not limited thereto.

[0058] Preferred embodiments are presented below to aid in understanding the present invention. However, the following embodiments are provided merely to facilitate a better understanding of the invention, and the scope of the invention is not limited by the following embodiments.

[0059] [Example]

[0060] Experimental method

[0061] As shown in Figure 1, single-cell analysis was performed using lung samples from 31 patients with non-small cell lung cancer (22 patients with EGFR mutations (E19del and L858R) and 9 wild-type patients) and 15 normal controls.

[0062] Single cell multi-omics library creation

[0063] Single-cell samples isolated from non-small cell lung cancer and normal lung samples were centrifuged at 400g for 5 minutes. Subsequently, cell pellets were collected, and cell counts and viability were determined using trypan blue on a Luna-II automated cell counter (Logos Biosystems). After labeling with antibodies (AbSeq Antibody-Oligos; BD Biosciences) for surface proteomic analysis, a total of 40,000 cells were injected into microwell cartridges (BD Rhapsody Express System; BD Biosciences). To analyze the whole transcriptome, surface proteome, T-cell receptor, and B-cell receptor at the single-cell level, the BD Rhapsody WTA Amplification Reagent Kit and the BD Rhapsody TCR / BCR Assays Reagent Kit were prepared according to the manufacturer's instructions. Sequencing was performed on an Illumina NextSeq 2000 using 2 x 100bp paired-end reads with an 8bp single index to produce whole transcriptome and surface proteome data, and using 85 x 215bp paired-end reads to produce T-cell receptor and B-cell receptor data.

[0064] Single-cell multiomics data analysis

[0065] Raw single-cell gene expression results were obtained in the form of a spare triplet table containing cell barcodes, gene identifiers, and Unique Molecular Identifier (UMI) counts. For each sample, the triplet table was reconstructed into a gene x cell UMI count matrix, and missing entries were set to 0. The generated count matrix was analyzed in the R environment. For each sample, a Seurat object was created from the raw UMI counts, and sample-level metadata including sample ID, patient ID, and study group information was included in the object's metadata.

[0066] Quality control indicators were calculated for each cell, including the number of detected genes (nFeature_RNA), total UMI count (nCount_RNA), and the proportion of mitochondrial transcripts (percent.mito). Low-quality cells were removed based on predefined criteria (nFeature_RNA > 500 and percent.mito < 50). The filtered dataset was log-normalized with a scale factor of 10,000 using the LogNormalize method, and 2,000 high-variance genes were selected using a variance-stabilizing transformation (vst).

[0067] To integrate multiple samples and mitigate technical variation, shared integration features were selected, the dataset was scaled, and Principal Component Analysis (PCA) was performed. Integration anchors were identified using Reciprocal PCA (RPCA)-based anchor discovery, thereby generating an integrated assay. The integrated data was scaled by regressively correcting nCount_RNA and percent.mito as technical covariates, and subsequently, PCA and UMAP were performed to reduce dimensionality and visualize the results. Graph-based clustering was conducted using the nearest neighbor graph, and cluster resolution was selected based on cluster stability and biological interpretability.

[0068] Cluster identities were designated based on standard marker gene expression and cluster-specific gene expression. Cluster marker genes were selected to include only positive markers using Seurat's FindAllMarkers function, and the minimum detection rate and log-fold change criteria followed the values ​​defined in the analysis pipeline. Differences in cell composition were evaluated by calculating the number and proportion of cells per cluster for each sample and study group. Differential expression analysis between predefined biological conditions was performed using the FindMarkers function of the Seurat package, and multiplex correction was applied.

[0069] For TCR / BCR repertoire analysis, cell-specific V(D)J information summarized in the VDJ_perCell.csv file generated from the SevenBridges WTA pipeline was integrated into the metadata of the corresponding Seurat object. Only cells with Paired_Chains = TRUE were selected for TCR and BCR repertoire analysis, respectively. Cells possessing the same dominant TCR or BCR chain were defined as a single clonotype, and the number of cells sharing the same dominant chain was calculated. Clonotypes detected in two or more cells were designated as Clonality = TRUE, and these expanded clonotypes were utilized for subsequent analysis.

[0070] In addition to the total transcriptome expression data, antibody-derived tag (ADT) data corresponding to 40 selected surface proteome panels (AbSeq / CITE-seq) were also included. Protein expression counts were included as an additional assay in the Seurat object and co-analyzed with the transcriptome data to be used for cell type classification and subsequent characterization.

[0071] All analyses were performed in R (version 4.3.2). The main packages used included Seurat (v4.4), SeuratObject (v5.0.2), dplyr (v1.1.4), ggplot2 (v3.5.1), clustree (v0.5.1), EnhancedVolcano (v1.20.0), clusterProflier (v4.10.0), and org.Hs.eg.db (v3.18.0).

[0072] Analysis of cellular heterogeneity of CD4 T cells

[0073] The results of the analysis of the cellular heterogeneity of CD4 T cells in the above samples are shown in Figures 2 and 3. As shown in Figures 2 and 3, CD4 T cells can be divided into four subgroups based on their known molecular functions: naive, Treg, effector / memory, and follicular helper T cells. In patients with EGFR mutations, the frequency of follicular helper T cells within the tumor is reduced, and a significant decrease in the cell frequency of follicular helper T cells and CXCL13 expression was observed even when clonal T cells were selectively analyzed through TCR profiling. These results suggest that TLS formation and the immune response against the tumor may be impaired in patients with EGFR mutations.

[0074] CD8 T cell status analysis

[0075] The results of the analysis of the CD8 T cell status of the above samples are shown in Figures 4 and 5. Consistent with the decrease in follicular helper T cells and CXCL13 expression, patients with EGFR mutations showed a significant decrease in dysfunctional CD8 T cells within the tumor, as well as an increase in nive-like cells, when analyzing the tumor response population based on TCR profiling. This indicates a sharp decrease in the number of tumor-responsive T cell populations that can be reactivated by immune checkpoint inhibitors (ICIs) or newly generated tumor-responsive T cells that can become dysfunctional.

[0076] Analysis of the immune microenvironment of EGFR-mutated non-small cell lung cancer using single-cell multi-omics

[0077] (1) Patient group composition and multi-omics experimental design

[0078] To compare how the tumor and immune microenvironment differ according to EGFR mutation types, the patient population was classified into three groups at the start of the analysis: EGFR wild-type (WT), EGFR exon 19 deletion (E19del), and EGFR L858R point mutation (L858R). Tumor and normal lung tissues were secured from each patient to design a sample structure that allowed for the simultaneous analysis of not only inter-patient differences but also tumor changes relative to normal tissues within the same patient. The secured samples were processed while maintaining their tissue state to enable single-cell analysis.

[0079] Subsequently, the experimental design applied a multi-omics pipeline that acquired transcriptomes (mRNA) via Whole Transcriptome Analysis (WTA), obtained barcode sequences corresponding to surface proteins using AbSeq technology with antibody-oligonucleotide conjugates in the same cells, and acquired TCR / BCR sequences (DNA-based V(D)J information) in parallel for clonal tracking of immune cells. The protein panel consisted of approximately 40 markers, and the entire data was designed with a flow that performed preprocessing and alignment / counting along the analysis pipeline to be processed at a large-scale single-cell level.

[0080] Through the above design, a data structure was secured that allows comparison of the EGFR variant group (E19del, L858R) and the WT group within the same experimental frame. In addition, sampling was configured in a way that enables comparison between tumor-normal and between variant types, thereby establishing a foundation for stepwise testing whether changes in cell composition, cell state, and clonal changes differ according to the EGFR variant type.

[0081] (2) UMAP-based major cell types and cell subtype clusters

[0082] The single-cell transcriptome and multi-omics analysis data began with a step of understanding the overall structure by placing all cells at once. To this end, Uniform Manifold Approximation and Projection (UMAP) was performed based on cell-specific gene expression matrices to embed cells with similar transcriptome expression patterns into clusters. Subsequently, major cell types were initially defined through clustering, and cell identity was annotated using a dot plot-type marker expression panel to verify the actual cell type of each cluster.

[0083] In the subsequent steps, since primary major cell types alone are insufficient to adequately explain changes in the tumor microenvironment state, subclustering was performed again within each major cell type. Again, dot plots and heatmaps were used to determine cell subtypes based on which marker combinations each subgroup possessed. The cell types and cell subtypes identified in this experiment are shown in Table 1 below.

[0084] Major cell typesCell subtypesB cellsB_AberrantB_MALTB_PlasmaB_TransB_Folicular_IgGB_Folicular_IgMBreg1Breg2T cellsCD4_TnCD4_TregCD4_TemCD4_TfhCD8_TpdCD8_TcCD8_TnCD8_TdNKEndothelial cellsEC_AerocytesEC_ArterialEC_CapillaryEC_LymphaticEC_VenousEpithelial cellsEpi_ClubEpi_AT1Epi_AT2Epi_CiliatedEpi_GobletEpi_BasalEpi_TumorEpi_Ciliated_likeEpi_Tumor_KRTEpi_Tumor_LMO4Epi_Tumor_S100PEpi_Basal_likeMyeloid cellsMye_MacMye_AlvMMye_cMMye_IntVMMye_cDC2Mye_NeuMye_ncMMye_mDCMye_cDC1Mast_cellStromal cellsFB_PulFB_MyoFB_MatSMCFB_MFAP5PericyteFB_CLDN1Mitotic cellsMitotic_ImmMitotic_Epi

[0085] In the whole-cell analysis, NSCLC tissues were separated into seven cell types, and a total of 53 detailed subtypes were defined through further subclustering. In the dot plot / heatmap, each subtype was distinguished by different marker expression patterns, providing a reference coordinate system that allowed for the quantitative comparison of which subtypes increased or decreased in subsequent steps. Non-small cell lung cancer expression markers were analyzed according to each cell type and cell subtype, and the markers expressed and their expression levels according to cell subtype were plotted (Fig. 6). The cell type-specific markers identified in this experiment were CXCL13; B cell-specific markers MS4A1 and BANK1; T cell-specific markers CXCR4 and IL7R; endothelial cell-specific markers HPGD and AQP1; and epithelial cell-specific markers RAB11FIP1 and SPOCK2; POU2F and CXCL8 are myeloid cell-specific markers; and TNFAIP3 and IGFBP3 are stromal cell-specific markers. Validation of ICI responsiveness prediction using a multi-gene panel through public data analysis.

[0086] Analysis was performed using the NSCLC-related scRNA-seq dataset from the NCBI GEO (Gene Expression Omnibus) public database.

[0087] The major GEO datasets used in the analysis are as follows: (1) GSE207422 - scRNA-seq data by ICI treatment response group in NSCLC patients; (2) GSE150403 - immunoprofiling scRNA-seq data in NSCLC patients; (3) GSE131907 - NSCLC scRNA-seq data related to ICI response.

[0088] The purpose of the analysis consists of two parts. First, the expression levels of a biomarker panel that essentially includes CXCL13 and additionally includes two or more genes selected from a group consisting of MS4A1, BANK1, CXCR4, IL7R, HPGD, AQP1, RAB11FIP1, SPOCK2, POU2F, CXCL8, TNFAIP3, and IGFBP3 were compared between the EGFR mutant group (mut) and the wild-type group (WT). Second, the predictive performance of the panel score regarding ICI responsiveness was quantified as an Area Under the Curve (AUC) value through Receiver Operating Characteristic (ROC) curve analysis.

[0089] The panel score was calculated as the average expression level of CXCL13 and two or more genes selected from the above group (e.g., MS4A1, BANK1, CD79A, etc.).

[0090] The results of the analysis using the GSE207422 dataset are as follows. It was confirmed that the average expression of CXCL13 in the EGFR mutant group was significantly reduced compared to the wild type (EGFR mut: approx. 0.45, WT: approx. 1.20, t-test p-value: 0.002). Furthermore, as a result of the panel score-based ROC curve analysis with MS4A1, BANK1, and CD79A added to CXCL13, the predicted AUC for ICI responsiveness improved to 0.85 compared to the CXCL13 marker alone, demonstrating that diagnostic reliability is significantly improved by combining two or more genes compared to a single marker. In addition, the analysis of the GSE150403 dataset also confirmed low expression of the aforementioned panel scores in the EGFR mut group (p<0.002), confirming that a significant distinction between the ICI responders and non-responders is possible.

[0091] The above verification results support that, compared to the CXCL13 marker alone, a panel combining two or more genes selected from the group consisting of MS4A1, BANK1, CXCR4, IL7R, HPGD, AQP1, RAB11FIP1, SPOCK2, POU2F, CXCL8, TNFAIP3, and IGFBP3 significantly improves the accuracy and reliability of determining EGFR mutation status and predicting ICI responsiveness.

Claims

1. A method for providing information to predict the responsiveness of immune checkpoint inhibitor (ICI) treatment in patients with non-small cell lung cancer, A step of analyzing the proportion of CD8-positive T cells in a subpopulation by performing single-cell multiomics analysis on a patient's lung sample; and The method includes the step of measuring the frequency of dysfunctional CD8 T cells (CD8_Td) and naive-like cells (CD8_Tn) among the CD8-positive T cells. A method for providing information characterized by predicting low therapeutic responsiveness to immune checkpoint inhibitors when the frequency of CD8_Td decreases and the frequency of CD8_Tn increases in patients positive for EGFR mutations.

2. In Claim 1, A method for providing information, characterized in that the subgroup of CD8-positive T cells includes uncontacted T-like cells (CD8_Tn), predysfunctional T cells (CD8_Tpd), dysfunctional T cells (CD8_Td), and cytotoxic T cells (CD8_Tc).

3. In Claim 1, The above single-cell multi-omics analysis is characterized by simultaneously performing whole transcriptome analysis (WTA), surface proteome analysis (AbSeq / CITE-seq), and T-cell receptor / B-cell receptor (TCR / BCR) repertoire analysis at the single-cell level, in a method for providing information.

4. In Claim 3, The above single-cell multi-omics analysis is characterized by capturing a single cell using a microwell cartridge system and preparing a library using the BD Rhapsody WTA Amplification Reagent Kit and the BD Rhapsody TCR / BCR Assays Reagent Kit, a method for providing information.

5. A method for evaluating inhibition of tumor-associated tertiary lymphoid structure (TLS) formation by EGFR mutations in non-small cell lung cancer, A step of measuring follicular helper T cell (CD4_Tfh) frequency and CXCL13 expression in a patient's lung tumor sample through single-cell multiomics analysis; and A method for evaluating inhibition of tumor-associated tertiary lymphoid structure (TLS) formation by EGFR mutations in non-small cell lung cancer, comprising the step of evaluating CD4_Tfh frequency in a selected tumor-responsive T cell population through TCR clonality analysis, and characterized by utilizing the reduction in CD4_Tfh frequency and CXCL13 expression as an indicator of inhibition of TLS formation by EGFR mutations.

6. A method for providing information for predicting immunotherapy responsiveness or EGFR mutation status in patients with non-small cell lung cancer (NSCLC), A step of measuring the expression levels of two or more gene combinations selected from the group consisting of CXCL13 and MS4A1, BANK1, CXCR4, IL7R, HPGD, AQP1, RAB11FIP1, SPOCK2, POU2F, CXCL8, TNFAIP3, and IGFBP3 from a patient's sample; and The method includes the step of calculating a panel score based on the above-mentioned measured expression levels to predict whether EGFR mutations are positive and responsiveness to immune checkpoint inhibitors, and A method for providing information characterized by the panel score being significantly reduced compared to a normal control or EGFR wild type when the EGFR mutation is positive.

7. In Claim 6, A method for providing information characterized in that the above sample is the patient's blood, plasma, serum, or circulating immune cells, and non-invasively determines the EGFR mutation status and responsiveness to immune checkpoint inhibitor treatment, and is performed based on a liquid biopsy. A composition for determining EGFR mutation status in non-small cell lung cancer (NSCLC) or predicting responsiveness to immune checkpoint inhibitor treatment, comprising: 8.CXCL13; and a preparation for measuring mRNA or protein levels of two or more genes selected from the group consisting of MS4A1, BANK1, CXCR4, IL7R, HPGD, AQP1, RAB11FIP1, SPOCK2, POU2F, CXCL8, TNFAIP3, and IGFBP3; A composition for determining the EGFR mutation status of non-small cell lung cancer (NSCLC) or predicting the responsiveness to immune checkpoint inhibitor treatment, characterized in that the above preparation is a primer or probe that specifically binds to the mRNA of the gene, or an antibody or aptamer that specifically binds to the protein encoded by the gene.

9. A method for quantitatively evaluating inhibition of tertiary lymphoid structure (TLS) formation by EGFR mutations using single-cell analysis data of non-small cell lung cancer (NSCLC) patients, A step of calculating panel scores for two or more gene combinations selected from the group consisting of CXCL13; and MS4A1, BANK1, CXCR4, IL7R, HPGD, AQP1, RAB11FIP1, SPOCK2, POU2F, CXCL8, TNFAIP3, and IGFBP3 in single-cell analysis data; A step of comparing the panel scores in an EGFR mutant group (mut) versus a wild-type group (WT); and It includes a step of quantifying ICI responsiveness prediction performance using AUC (Area Under the Curve) values ​​through ROC (Receiver Operating Characteristic) curve analysis, and A method for evaluating inhibition of TLS formation by EGFR mutations, characterized in that a significant decrease in panel scores in the above EGFR mutation group is used as an indicator of TLS formation inhibition.

Images