阅读说明：本技术 一种指导选择适用于抗pd-1/pd-l1免疫治疗患者的膀胱癌免疫分类系统 (Guidance selection of bladder cancer immune classification system suitable for anti-PD-1/PD-L1 immunotherapy patients ) 是由 孟佳林 张蒙 陆晓凡 周宇杰 卞子辰 金晨 葛秦涛 莫凡 丁杨 张力 郝宗耀 于 2021-01-22 设计创作，主要内容包括：本发明提供一种指导选择适用于抗PD-1/PD-L1免疫治疗患者的膀胱癌免疫分类系统,涉及生物医学技术领域。所述分类系统的确立主要是采用无监督非负矩阵分解(NMF)和最近模板预测(NTP)算法,将膀胱癌患者分为免疫活化亚组、免疫耗竭亚组和非免疫亚组。本发明克服了现有技术的不足,膀胱癌患者进行分类,确定一种新的免疫分子分类器,为膀胱癌患者的免疫治疗策略的临床选择提供方向。(The invention provides a bladder cancer immune classification system for guiding selection of anti-PD-1/PD-L1 immunotherapy patients, and relates to the technical field of biomedicine. The classification system was established primarily by using unsupervised non-Negative Matrix Factorization (NMF) and the Nearest Template Prediction (NTP) algorithm to classify bladder cancer patients into immune-activated, immune-depleted and non-immune subgroups. The invention overcomes the defects of the prior art, classifies the bladder cancer patients, determines a new immune molecular classifier, and provides a direction for clinical selection of immunotherapy strategies of the bladder cancer patients.)

1. A bladder cancer immune classification system for guiding selection of patients suitable for immunotherapy with anti-PD-1/PD-L1, wherein the bladder cancer immune classification system comprises an immune-enhanced group and a non-immune group, wherein the immune-enhanced group is further divided into an immune-activated group and an immune-depleted group, and wherein establishment of the bladder cancer classification system comprises the steps of:

(1) selecting a batch of bladder cancer patients, and determining gene expression profiles, clinical pathological information and overall survival data of the bladder cancer patients;

(2) performing microdissection on the mRNA expression profile of the patient with bladder cancer by adopting an unsupervised nonnegative matrix factorization algorithm, calculating an immune enrichment score through single-sample gene set enrichment analysis, and selecting a module with the strongest enrichment as an immune module;

(3) extracting 150 representative genes with the largest weight in the immune module, and redefining the bladder patients into an immune enhancement group and a non-immune group through consensus clustering;

(5) further defining an immune enhancement group and a non-immune group by using a multi-dimensional random forest for accurate classification;

(6) the immune enhanced group identification was differentiated into immune activated and immune depleted subpopulations by matrix activation signatures of recent template prediction methods.

2. The system for directing the selection of an immune classification of bladder cancer suitable for use in immunotherapy of patients with anti-PD-1/PD-L1 according to claim 1, wherein: the sample gene selected in the step (3) participates in signal paths of T cell activation, antigen processing and presentation and B cell activation, and is related to the activation of a Th1/Th2 cell differentiation path, a T cell receptor signal path, a B cell receptor signal path and a PD-L1 expression/PD-1 checkpoint related path.

3. The system for directing the selection of an immune classification of bladder cancer suitable for use in immunotherapy of patients with anti-PD-1/PD-L1 according to claim 1, wherein: the immune-enhanced patients were highly enriched in T cells, B cells, interferon and CYT signals; also, the immunodepleted subgroup had increased signaling of TITR, Wnt/transforming growth factor-beta, transforming growth factor-beta 1 activation, and C-ECM.

4. Use of the bladder cancer immune classification system of claim 1 to guide the selection of an anti-PD-1/PD-L1 immunotherapeutic patient for selecting an immunotherapeutic regimen for a bladder cancer patient.

5. The use of the guidance of claim 4 for selecting a bladder cancer immune classification system suitable for use in an anti-PD-1/PD-L1 immunotherapy patient, wherein: the immune activated subgroup of patients is suitable for a single anti-PD-1 immunotherapy; an appropriate treatment modality for the immunodepleted subgroup of patients is immune checkpoint blockade plus a transforming growth factor-beta inhibitor or an EP300 inhibitor.

Technical Field

The invention relates to the technical field of biomedicine, in particular to a bladder cancer immune classification system for guiding and selecting patients suitable for anti-PD-1/PD-L1 immunotherapy.

Background

Bladder cancer is a heavy health burden worldwide, particularly in europe and north america. There are about 55 million cases per year, with nearly 20 million people dying from bladder cancer. The incidence of bladder cancer varies widely around the world, with southern europe having the highest incidence and central africa having the lowest incidence. Bladder cancer is the 6 th most common type of cancer in the united states, with about 81000 new cases reported in 2020 and 18000 new deaths. Most patients are diagnosed with early-stage bladder cancer, also known as non-muscle-invasive bladder cancer (NMIBC), and can be treated by transurethral resection (TUR) or bacillus calmette-guerin (BCG) intravesical perfusion therapy or other chemotherapeutic methods, but recurrence is extremely common in NMIBC, with approximately 70% of patients relapsing within 10 years, and one third of patients progressing to muscle-invasive bladder cancer (MIBC). The standard treatment for MIBC is radical cystectomy with or without neoadjuvant or chemotherapy. However, even with treatment, nearly 50% of MIBC patients experience metastasis, relapse, and death within 3 years.

Some studies have investigated diagnostic, prognostic or therapeutic targets for malignancies based on Tumor Microenvironment (TME), as well as diagnostic, prognostic or therapeutic targets for bladder cancer. Bcg was the earliest approved immunotherapy for the treatment of bladder cancer, and it stimulates the immune response, inducing pro-inflammatory cytokines and intercellular cytotoxic effects. At present, bcg remains the standard treatment for NMIBC, reflecting that bladder cancer patients can benefit from immunotherapy. Blockade of the immunodetection site has also been applied to the treatment of bladder cancer. The FDA has approved two PD-1 inhibitors (pembrolizumab and nivolumab) and three PD-L1 inhibitors (atezolizumab, duvalumab and avelumab) for the treatment of bladder cancer. In the IMvigor210 clinical study, altuzumab was used to block PD-L1: the Objective Remission Rate (ORR) of imvisor 210 cohort 1 was only 23%, and the ORR of cohorts 244, 45 was only 15%. The ORR values for Nivolumab and duvalumab were similar, from 17% to 24.4%. Therefore, the immunophenotype of the bladder cancer is fully understood, and the screening of patients suitable for the anti-PD-1/PD-L1 treatment has important guiding significance for the clinical treatment of the bladder cancer.

In tumor tissue, normal cells, blood vessels, and cytokines that surround and support the viability of tumor cells constitute the Tumor Microenvironment (TME). Tumor and TME interact, and tumor cells can alter TME, which can promote tumor growth and spread. At the molecular level, bladder cancer has a variety of heterogeneity, including gene mutations, gene copy number changes, neoantigens, and infiltration of immune cells. Several groups have established a molecular classification of bladder cancer. Mo and the like apply a tumor signal of 18 genes in MIBC patients, can reflect the differentiation of urothelium and predict clinical results; the basal and differentiated groups were the two groups with the highest and lowest risk scores, respectively. Damraver et al developed BASE47, a transcriptional classifier using 47 genes for classifying MIBC tumors as luminal or basal-like subtypes. Robertson et al performed consensus hierarchical clustering of Bayesian NMF on 408 MIBC tumors from TCGA and found five expression subtypes, including three luminal subtypes (luminal-papillary, luminal-infiltrating and luminal), basal/squamous, and neuronal. However, most molecular classifiers focus only on clinical outcome and not on tumor immune microenvironment. Therefore, our goal is to fully understand the immune response of bladder cancer patients with different internal molecular characteristics and to help find the most suitable patient for precise immunotherapy.

Disclosure of Invention

Aiming at the defects of the prior art, the invention provides a bladder cancer immune classification system for guiding and selecting a patient suitable for anti-PD-1/PD-L1 immunotherapy, determines a new immune molecular classifier by classifying the bladder cancer patient, and provides a direction for clinical selection of an immune treatment strategy of the bladder cancer patient.

In order to achieve the above purpose, the technical scheme of the invention is realized by the following technical scheme:

a bladder cancer classification system based on tumor immune microenvironment, comprising an immune-enhanced group and a non-immune group, wherein the immune-enhanced group is further divided into an immune-activated group and an immune-depleted group, the establishment of the bladder cancer classification system comprising the steps of:

(1) selecting a batch of bladder cancer patients, and determining gene expression profiles, clinical pathological information and overall survival data of the bladder cancer patients;

(2) performing microdissection on the mRNA expression profile of the patient with bladder cancer by adopting an unsupervised nonnegative matrix factorization algorithm, calculating an immune enrichment score through single-sample gene set enrichment analysis, and selecting a module with the strongest enrichment as an immune module;

(3) extracting 150 representative genes with the largest weight in the immune module, and redefining the bladder patients into an immune enhancement group and a non-immune group through consensus clustering;

(5) further defining an immune enhancement group and a non-immune group by using a multi-dimensional random forest for accurate classification;

(6) the immune enhanced group identification was differentiated into immune activated and immune depleted subpopulations by matrix activation signatures of recent template prediction methods.

Preferably, the sample gene selected in step (3) is involved in signal pathways of T cell activation, antigen processing and presentation, and B cell activation, and is associated with activation of Th1/Th2 cell differentiation pathway, T cell receptor signal pathway, B cell receptor signal pathway, and PD-L1 expression/PD-1 checkpoint-related pathway.

Preferably, the immune-enhanced patients are highly enriched in T cells, B cells, interferon and CYT signals; also, the immunodepleted subgroup had increased signaling of TITR, Wnt/transforming growth factor-beta, transforming growth factor-beta 1 activation, and C-ECM.

Use of the bladder cancer immune classification system for selecting an immunotherapy regimen for a bladder cancer patient.

Preferably, the immune activated subgroup of patients is suitable for a single anti-PD-1 immunotherapy; an appropriate treatment modality for the immunodepleted subgroup of patients is immune checkpoint blockade plus a transforming growth factor-beta inhibitor or an EP300 inhibitor.

The invention provides a bladder cancer immune classification system for guiding selection of patients suitable for anti-PD-1/PD-L1 immunotherapy, which has the advantages compared with the prior art that:

classification was performed using unsupervised non-Negative Matrix Factorization (NMF) and recent template prediction (NTP) algorithms, immune and non-immune groups were identified from a training cohort of cancer genomic map-bladder urothelial carcinoma (TCGA-BLCA), 150 genes with the greatest differences in expression between the two groups were extracted for recurrent classification in 20 validation cohorts, and matrix activation markers were assessed using the NTP algorithm for immune-activated and immune-depleted subsets. The patients in the immune group showed a high enrichment of immune cells, while the immune depleted subgroup showed characteristics of activated transforming growth factor-beta 1 and C-ECM. Patients of the immune activation subgroup showed lower genetic alterations and better overall survival. The anti-PD-1/PD-L1 immunotherapy is more beneficial to immune activation subgroup patients, and the immune checkpoint blockade therapy plus the transforming growth factor-beta inhibitor or the EP300 inhibitor can achieve better curative effect on immune exhaustion subgroup patients, so that a new immune molecular classifier is determined, and the direction is provided for clinical selection of immunotherapy of bladder cancer patients.

Description of the drawings:

FIG. 1: a flow chart for carrying out the invention;

FIG. 2: immune class relationship graph recognition for the non-Negative Matrix Factorization (NMF) algorithm of the present invention: (A) the method comprises the following steps: 9 modules generated by NMF algorithm, wherein the immune modules gather patients with higher immune concentration fraction; (B) the method comprises the following steps: the heatmap shows the expression of the first 150 sample genes in the immune-enriched and non-immune-enriched clusters, divided by consensus clustering; (C) the method comprises the following steps: a multidimensional scaling (MDS) random forest further modifies the clustering into immune and non-immune categories; (D) distribution of patients in different NMF modules, immune module weights, sample gene clustering.

FIG. 3: heterogeneity maps for the diverse immune characteristics and genetic phenotypes of the non-immune, immune-activated and immune-depleted subgroups of the invention: (A) the method comprises the following steps: three immunophenotypes, and characterized by cytotoxicity scores, tumor infiltrating Tregs, myeloid lineage suppressor cells, tertiary lymphoid structures, tumor-associated extracellular matrix; (B) differences in abundance of tumor infiltrating lymphocytes; (C) differences in PD-L1mRNA expression levels; (D) differences in chromosomal arm level and focal level gene copy number changes (including amplifications and deletions); (E) differences in tumor mutational burden; (F) differences in neoantigens; (G) specific mutant genes in the immune activation subgroup; (H) a specific mutant gene in the immunodepleted subgroup, wherein WT is wild-type; IM-Act is an immune activation subgroup; IM-exh is an immunodepleted subgroup;

figure 4 is a representation of the immunophenotype of the present invention reproduced in the IMLIGN210 cohort, reflecting the different responses to immunotherapy against PD-L1: (A) these three immunophenotypes were reproduced in the IMvigor210 cohort; (B) the distribution of the best overall response confirmed against PD-L1 treatment in these three immunophenotypes; (C) the distribution of complete responses and progressive disease among these three immunophenotypes;

figure 5 is a graph of the response of the present invention to predict immune checkpoint blockade therapy, revealing different overall survival outcomes of three immunophenotypes: (A) predicted outcome of response to anti-PD-L1 treatment; (B) predicted outcome of response to anti-CTAL-4 and anti-PD-1 treatments; (C) the overall survival results for the three immunophenotypes differed.

Wherein: NMF is non-negative matrix decomposition; TCGABLCA is a tumor genomic map-bladder cancer; TIL is tumor infiltrating lymphocytes; CNA is copy number change; TMB is tumor mutation burden; WT is wild type; IM-Act is an immune activation subgroup; IM-exh is an immunodepleted subgroup; non-IM is non-immune; IM-exh is an immunodepleted subgroup.

Detailed Description

In order to make the objects, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention are clearly and completely described below in conjunction with the embodiments of the present invention, and it is obvious that the described embodiments are some embodiments of the present invention, but not all embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.

Example 1:

determination and application of the bladder cancer classification system:

1. preparing materials: 4028 patients with bladder cancer were enrolled, including their gene expression profiles, clinical pathology information, and overall survival data (as shown in FIG. 1); for the TCGA-BLCA cohort, we obtained a tertiary gene expression profile from TCGAdata Portal for 408 patients, retaining at least 50% of the genes expressed by the sample for subsequent analysis; for further external verification queues, gene expression profiles and clinical information of GSE32894, GSE83586, GSE87304, GSE128702, GSE13507, GSE129871, GSE120736, GSE39016, GSE128701, GSE124035, GSE86411, GSE48276, GSE31684, GSE134292, GSE93527 and GSE69795 are selected; TAB-4321 and E-MTAB-1803 cohorts, downloading gene expression profiles and paired clinical profiles from Arrayexpress.

2. The classification system determines:

(1) performing microdissection on mRNA expression profiles of 408 bladder cancer patients in the TCGA-BLCA cohort by adopting an unsupervised non-Negative Matrix Factorization (NMF) algorithm, and analyzing the mRNA expression profiles of 408 bladder cancer patients in the TCGA-BLCA cohort to serve as a training cohort;

(2) obtaining an immune module: presetting corresponding module numbers as 5 to 10, when the total number of modules is 9, enriching patients with higher Immune Enrichment Scores (IES) by the first module, defining as an immune module (figure 2A), and defining 150 genes with the largest weight in the immune module as sample genes reflecting the characteristics of the immune module;

according to the ontology analysis of biological processes, these genes are involved in the signaling pathways of T cell activation, antigen processing and presentation, B cell activation and are associated with the activation of the Th1/Th2 cell differentiation pathway, T cell receptor signaling pathway, B cell receptor signaling pathway, and PD-L1 expression/PD-1 checkpoint-related pathway (all P < 0.05);

(3) a total of 408 bladder patients were redefined as immune-enhanced and non-immune groups by consensus clustering based on 150 sample genes (fig. 2B); and in fig. 2D, the distribution of 408 bladder cancer patients in NMF module, immune module weights, sample gene clustering, final immune categories and immune enrichment scores are shown;

(4) collecting several immune-related characteristic gene lists, helping to identify immune or non-immune groups, and determining an enrichment score of each immune-related characteristic gene list of each patient by ssGSEA;

it was observed that immune cells were increased in the immune group compared to those in the non-immune group, including T cells (as reflected by 13T cell signatures, T cells, CD8+ T cells, and NK cells)), B cells (B cell clusters and b.p.meta markers), macrophages, tertiary lymphoid tissue structures (TLS), cytolytic activity score (CyT), and Interferon (IFN) markers (mean P <0.05, fig. 3A);

(5) GSEA analysis of the activated KEGG signaling pathway revealed that immune cell pathways (including T cell, B cell, natural killer cell and leukocyte associated pathways), immune response pathways (including chemokine signaling pathways, antigen processing presentation pathways, cell adhesion molecule cAMs and complement coagulation cascade pathways) and pro-inflammatory pathways (including Fc-Epsilon-RI, NOD-like receptors and Fc γ R mediated phagocytic pathways) were all activated in the immune group from the results of fig. 2, 3A, the immune enhanced and non-immune groups in the TCGA-BLCA cohort were microdissected and activated immune-related markers and signaling pathways were observed in the immune enhanced group.

3. And (3) clinical verification:

(1) based on the clinical data of TCGA-BLCA, GSE32894 and E-MTAB-1803 queues, the overall survival of the patients belonging to the immune activation subgroup is the best, and the overall survival time of the patients belonging to the immune exhaustion subgroup is the shortest;

(2) immune depletion focuses primarily on depletion of T cells, reflected in changes in inflammation and tissue microenvironment, lymphocytes, and inhibitory signals from cytokines, which may escape immune recognition by blocking immune checkpoints, which is associated with poor overall survival of patients;

(3) predicting potential response to immunotherapy in bladder cancer patients by comparing their mRNA expression profiles to melanoma samples treated with anti-CTLA-4 or anti-PD-1 checkpoints, patients of the immune activating subgroup may benefit from anti-PD-1 therapy, but not patients of the non-immune activating subgroup (including the immune depleting subgroup and the non-immune subgroup);

(4) to further understand the molecular diversity between these three immunophenotypes, we compared CNA, TMB, and gene mutations; patients in the immunopotentiating group showed lower CAN burden, and lower chromosomal arm and focal levels of gene deletion; the deletion copy number of PD-1 and PD-L1 is repeated to confirm the correlation, and CTLA-4 is positively correlated with the infiltration level of immune cells;

(5) in tumors that are highly chromosomally unstable (also called high CNA), antigen presentation via the MHC-class I pathway is inhibited, and high CNA plays a key role in immune evasion; furthermore, patients receiving immune checkpoint blockade therapy would have a long-lasting clinical benefit and gain better survival if they exhibited a lower CAN burden;

(5) extraction of specific mutant gene findings for each subpopulation: the proportion of mutant TP53, TTN, PIC3CA and RB1 in the immune enhanced group was higher than that in the non-immune group;

(6) the median of the cell densities of CD3+ and CD8+ cells of colorectal cancer patients carrying mutant PIK3CA was higher, and the clinical benefit rate of immunotherapy was also higher (50% versus 8.6%); furthermore, the rate of ERBB2 mutations is high in the immunoactive subgroup, ERBB2 amplification or overexpression is a biomarker for breast cancer anti-ERBB 2 targeted therapy, the activated ERBB2 oncogene regulates the recruitment and activation of tumor-infiltrating immune cells and the activity of trastuzumab by inducing CCL2 and PD-1 ligand; the V659E mutation in the ERBB2 gene was associated with altered sensitivity to afatinib and lapatinib treatment in vitro, and the EP300 gene mutation rate was highest in the immunodepleted subgroup;

(7) the importance of CBP/EP300 for regulatory T cells (tregs) is due to the loss of EP300 or CBP in mouse regulatory T cells (tregs) which results in an impaired Treg suppression function; intratumoral Tregs suppress the response of effector T cells to tumor antigens, creating an immunosuppressive microenvironment, which is associated with an adverse prognosis of the tumor.

In summary, a new classifier was defined and validated in 4028 patients with bladder cancer, and they were classified into an immune-activated subgroup, an immune-depleted subgroup, and a non-immune group. The immune activation sub-group of patients could benefit more from a single anti-PD-1 immunotherapy and for the immune depleted sub-group of patients the immune checkpoint blockade therapy plus either the transforming growth factor-beta inhibitor or the EP300 inhibitor showed better efficacy.

Example 2:

the tumor immune microenvironment immunodepletion phenotype is characterized by stromal cell activation:

fibroblasts, Mesenchymal Stromal Cells (MSCs) and extracellular matrix (ECM) are key components of the tumor stroma, supporting and linking tumor cells, particularly in the late stages of the tumor, where genetic and epigenetic changes in tumor cells are driven by the activated stromal components; the MSCs are used as the internal regulatory factor of the tumor, can secrete inhibitory soluble factors and change cell surface markers, thereby inhibiting immune microenvironment. The regulation of PD-L1 by MSCs, which suppress immune processes by reducing the expression of pro-inflammatory factors, including interferon-gamma, tumor necrosis factor-alpha and interleukin-1 beta, or by promoting the expression of type 2 factors, interleukin-10 and interleukin-13, affects the suppression of T cell proliferation and the induction of T regulatory cells (Tregs).

(1) The method comprises the following steps of adopting a predefined matrix activation mark to further divide an immune group into immune activation and immune depletion phenotypes to reflect an immune response state;

(2) in the TCGA-BLCA cohort, 11.0% (45/408) of bladder cancer patients were defined by an activated immunophenotype and an inactivated stromal phenotype, belonging to the immune-activated subgroup, while the other 27.0% (110/408) of patients were belonging to the immune-depleted subgroup, with an activated stromal phenotype (fig. 3A);

extracellular matrix cytokines (C-ECM) secreted by fibroblasts can recruit immunosuppressive cells, transforming growth factor- β is a recognized immunosuppressant in the immune microenvironment, Treg cells and MDSC cells in TME can reflect an immune depletion state;

the results show that the expression profile of TITR, Wnt/TGF-beta, transforming growth factor-beta 1 activation and C-ECM was all higher in the immunodepleted subgroup than in the immunoactive subgroup (all P <0.05, FIG. 3A). TIM-3 and LAG3 were associated with an immunodepleted state, with elevated TIM-3(P ═ 0.008) and LAG3(P ═ 0.218) observed in the immunodepleted subgroup; according to the results of fig. 3A, the immune classes were divided into two subgroups of immune activation and immune depletion, and the immune depletion markers TIM-3 and LAG3 demonstrated an increase in the interstitial enrichment score, TITR, MDSC and Wnt/TGF-expression profile of the immune depleted subgroup.

Example 3:

heterogeneity of genetic phenotype between immune and non-immune groups:

to confirm the infiltration of immune cells in both the immune and non-immune categories, the abundance of Tumor Infiltrating Lymphocytes (TILs) was compared in 408 patients with bladder cancer, pre-estimated by a whole-slice image of the HE staining of TCGA samples.

The results show that the immunized group had TILs more abundant than the non-immunized group (P <0.001, fig. 3B), consistent with the definition of the immunized and non-immunized groups; in addition, the expression level of PD-L1 was observed to be higher in the immune enhanced group than in the non-immune group (P <0.001, fig. 3C).

Due to gene Copy Number Alterations (CNA), Tumor Mutational Burden (TMB) and crosstalk between tumor antigens and tumor immune activation, gene deletion in chromosomal arms and foci is increased averagely (P) in non-immune patients_Armdel<0.001，P_focal-del0.007), but no CNA amplification was evident (P)_arm-Amp＝0.733，P_focal-Amp0.065) (fig. 3D), reflecting that immune infiltration is positively correlated with gene CNA deletion.

The association between immune infiltration and deletion of genes CNA, PD-1, PD-L1 and CTLA-4 at overall depth and arm level, as well as the reduction of immune cell infiltration, in particular CD4+ T cells, neutrophils and dendritic cells, were repeatedly confirmed using the online tool TIMER.

The immunized group had higher TMB than the non-immunized group (P0.001, fig. 3E) with no difference in tumor antigen levels (P0.109, fig. 3F); further comparing specific gene mutations in the immune subsets, ERBB2(P ═ 0.035), KMT2A (P ═ 0.013), PKHD1(P ═ 0.007), and MDN1(P ═ 0.015) mutations were similar to the immune activation subset (fig. 3G), while EP300(P ═ 0.020), HMCN1(P ═ 0.014), AKAP9(P ═ 0.003), and MACF1(P ═ 0.016) mutations were more abundant in the immune depletion subset (fig. 3H).

Taken together, the immune classes are associated with lower copy number deletions, higher TIL abundance, higher TMB and higher PD-L1, but not with tumor antigens, and the specific mutant genes in the immune phenotype are diverse.

Example 4:

therapeutic response in patients of the immune activation subgroup:

1. to assess the response of three newly defined immunophenotypes to immunotherapy in patients with bladder cancer:

(1) gene expression profiles and clinical results were collected for 348 patients from a large phase II trial IMlive 210 cohort, and the clinical therapeutic effect of replacing lizumab in blocking PD-L1 was studied.

(2) Endpoints were observed as responses to anti-PD-L1 treatment, including Complete Response (CR), Partial Response (PR), stationary phase (SD), progressive Phase (PD). Objective Response Rate (ORR) includes patients with CR and PR, and Disease Control Rate (DCR) includes patients with CR, PR and SD.

(3) Patients were first divided into immune-enhanced subgroups (237/348, 65.2%) and non-immune groups (121/348, 34.8%) based on 150 DEGs generated by the TCGA-BLCA training cohort.

(4) Of the immunopotentiating subfamily, 85 (24.4%) negative in the prediction of mesenchymal cell activation were immune activation subgroups, and the remaining 142 (142/348) were immune depletion subgroups (fig. 4A).

Wherein, the immune characteristics of the immune group patients are obviously rich, and comprise a predefined immune enrichment score, an immune cell subset score, an immune signal score and a plurality of immune cell signatures (the mean P is less than 0.05). In the isolation of the immune-activated and immune-depleted subpopulations, MDSC, TITR, Treg cells, Wnt/TGF β, transforming growth factor β -1 and C-ECM features showed a higher enrichment score in the immune-depleted subpopulation, consistent with features in the TCGA-BLCA training cohort.

More than half of the patients in the immune activation subgroup were observed to benefit from anti-PD-L1 treatment with an objective effective rate of 24.66% and a disease control rate of 52.06%; however, most patients in the immunodepleted subgroup developed PD outcomes following anti-PD-L1 treatment (P ═ 0.028, fig. 4B). More importantly, if only the treatment results for CR and PD were noted, it was found that patients in the immune activated subgroup showed the highest efficacy (20.5% vs 15.1% vs 4.8%, fig. 4C).

Taken together with the results of fig. 4, newly defined patients with triple immunophenotype bladder cancer can differentiate the appropriate patient population to receive anti-PD-L1 treatment in the immunoactive subgroup.

2. To verify the immunotherapeutic effect of the immune-activated subpopulation:

SubMap analysis was used to compare the gene expression profiles of responders in the IMLIGN210 cohort, MD Anderson melanoma cohort, and the immune activation subsets TCGA-BLCA, GSE32894, E-MTAB-1803 cohort.

It was found that patients in the immunocompetent subgroup of the three cohorts had a more similar gene profile in the IMLITY 210 cohort than in patients who were not treated with anti-PD-L1, but that in the non-immunocompetent subgroup, the gene profiles were different (all with Bonferroni correction P <0.05) (FIG. 5A).

In addition, consistent results compared to the MDAnderson melanoma cohort were observed, with patients of the immune activation subgroup benefiting more from anti-PD-1 treatment (both Bonferroni corrected P <0.05) (fig. 5B).

Furthermore, it was found that patients in the immune activation subgroup showed a better prognosis, whereas patients in the immune depletion subgroup showed a poorer prognosis (fig. 5C).

In conjunction with the results of fig. 5, patients of the immune activation subgroup could benefit more from anti-PD-1 or anti-PD-L1 treatment, with the longest average overall survival.

It is noted that, herein, relational terms such as first and second, and the like may be used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions. Also, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising an … …" does not exclude the presence of other identical elements in a process, method, article, or apparatus that comprises the element.

The above examples are only intended to illustrate the technical solution of the present invention, but not to limit it; although the present invention has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical solutions described in the foregoing embodiments may still be modified, or some technical features may be equivalently replaced; and such modifications or substitutions do not depart from the spirit and scope of the corresponding technical solutions of the embodiments of the present invention.