A liver cancer biomarker based on single-cell sequencing and its application

Abstract

This application discloses a liver cancer biomarker based on single-cell sequencing and its application. The liver cancer biomarker based on single-cell sequencing of this application consists of nine genes or proteins expressed by these nine genes: CLDN6, CNN3, CYP27A1, CYP2C9, FGB, HMGCS2, S100A10, S100A9, and SQSTM1. This liver cancer biomarker based on single-cell sequencing can be used to predict the overall survival, cancer status, and prognostic outcome of liver cancer patients. Detecting the liver cancer biomarker of this application through single-cell sequencing can detect tissue heterogeneity and identify gene expression characteristics within different cells.

Description

Technical Field

[0001] This application relates to the field of liver cancer detection technology, specifically to a liver cancer biomarker based on single-cell sequencing and its application. Background Technology

[0002] According to statistics from the Global Cancer Network in 2020, there were 905,677 new cases of liver cancer and 830,180 deaths from liver cancer worldwide in 2020. Liver cancer typically occurs in patients with chronic liver diseases such as hepatitis B and C. Liver cancer metastasis, which refers to the spread and invasion of liver cancer from its primary site to other parts of the body, is a leading cause of death among liver cancer patients. Despite advances in treatment and diagnosis, liver cancer still has a high mortality rate due to its high recurrence rate and the tendency for metastasis to occur after surgical resection. Studies show that liver cancer is one of the malignant tumors with extremely high incidence and mortality rates worldwide.

[0003] Existing liver cancer detection methods and prediction models are all based on results obtained from conventional RNA sequencing, which cannot detect tissue heterogeneity or identify gene expression characteristics within different cells. Currently, there are no liver cancer detection or prediction models based on single-cell sequencing.

[0004] Therefore, how to achieve single-cell sequencing detection or prediction of liver cancer is of great significance for detecting tissue heterogeneity, identifying gene expression characteristics in different cells, and conducting in-depth research and elucidation of the occurrence and development of liver cancer. Summary of the Invention

[0005] The purpose of this application is to provide a novel liver cancer biomarker based on single-cell sequencing and its application.

[0006] To achieve the above objectives, this application adopts the following technical solution:

[0007] The first aspect of this application discloses a liver cancer biomarker based on single-cell sequencing, which consists of nine genes or proteins expressed by these nine genes: CLDN6, CNN3, CYP27A1, CYP2C9, FGB, HMGCS2, S100A10, S100A9, and SQSTM1.

[0008] It should be noted that this application identified and classified the genomic characteristics and marker genes of primary and metastatic hepatocellular carcinoma using scRNA-seq mapping; and obtained transcriptomic and clinical data of liver cancer patients from The Cancer Genome Atlas (TCGA) and the Cancer Genome Consortium (ICGC) databases; identified metastasis-related genes using bioinformatics technology; and further, obtained nine key angiogenesis-related genes as liver cancer biomarkers based on single-cell sequencing through training and validation, namely CLDN6, CNN3, CYP27A1, CYP2C9, FGB, HMGCS2, S100A10, S100A9, and SQSTM1. By detecting the expression levels of these nine genes in the liver cancer biomarkers of this application, the overall survival of liver cancer patients can be effectively predicted, which has important reference value for liver cancer prognostic risk assessment and liver cancer detection. Furthermore, the nine angiogenesis-related genes in this application are identified from transcriptome data and clinical data to identify metastasis-related genes, enabling single-cell sequencing detection and prediction of liver cancer, detecting tissue heterogeneity, and identifying gene expression characteristics within different cells.

[0009] It should also be noted that the key to the liver cancer biomarkers in this application lies in achieving liver cancer detection, prediction of liver cancer patient survival, or assessment of liver cancer prognosis risk by detecting the expression levels of nine genes. It is understood that the expression levels of these nine genes can also be analyzed and determined by detecting the proteins they express. Therefore, the proteins expressed by these nine genes can also serve as liver cancer biomarkers in this application.

[0010] The second aspect of this application discloses the use of the liver cancer biomarkers of this application in the preparation of reagent kits or devices for liver cancer detection, prediction of survival of liver cancer patients, or assessment of prognostic risks of liver cancer.

[0011] It is understood that by detecting the liver cancer biomarkers of this application, it is possible to determine liver cancer, the survival status of liver cancer patients, or the prognostic risk of liver cancer; therefore, the liver cancer biomarkers of this application can be used to prepare reagent kits or devices for liver cancer detection, prediction of liver cancer patient survival, or assessment of liver cancer prognostic risk. Specifically, for example, corresponding detection reagents or dedicated devices can be prepared based on the liver cancer biomarkers of this application.

[0012] A third aspect of this application discloses a kit for liver cancer detection, prediction of liver cancer patient survival, or liver cancer risk assessment, the kit comprising reagents for detecting liver cancer biomarkers of this application.

[0013] In one implementation of this application, the reagents in the kit are primers and / or probes for detecting nine genes: CLDN6, CNN3, CYP27A1, CYP2C9, FGB, HMGCS2, S100A10, S100A9, and SQSTM1.

[0014] It is understood that, given that this application discloses that the nine genes can serve as biomarkers for liver cancer, those skilled in the art can design corresponding primers and / or probes for these nine genes based on existing technologies to detect the expression levels of these nine genes.

[0015] Preferably, the kit also includes enzymes and / or reaction solutions for primers and / or probes to detect genes.

[0016] It is understandable that if the kit uses primers and / or probes to detect nine genes, the kit may further include the corresponding enzymes and / or reaction solutions for ease of use.

[0017] In one implementation of this application, the reagents in the kit are nucleic acids and / or polypeptides for detecting proteins expressed by nine genes: CLDN6, CNN3, CYP27A1, CYP2C9, FGB, HMGCS2, S100A10, S100A9, and SQSTM1.

[0018] Preferably, the kit also includes reaction solutions for detecting proteins in nucleic acids and / or peptides.

[0019] It should be noted that when the liver cancer markers of this application are the proteins expressed by the nine genes of this application, the corresponding nucleic acids and / or peptides can be used to detect these proteins; therefore, the reagents in the kit of this application can also be nucleic acids and / or peptides for detecting the proteins expressed by the nine genes of this application.

[0020] The fourth aspect of this application discloses the application of a genomic combination as a biomarker for liver cancer, the genomic combination consisting of nine genes: CLDN6, CNN3, CYP27A1, CYP2C9, FGB, HMGCS2, S100A10, S100A9 and SQSTM1.

[0021] In one implementation of this application, the application includes liver cancer detection, prediction of liver cancer patient survival, or assessment of liver cancer prognosis risk based on the expression levels of the nine genes in this application.

[0022] In one implementation of this application, the prognostic risk scoring formula for assessing the prognostic risk of liver cancer based on the expression levels of nine genes is as follows:

[0023] Prognostic risk score = (0.1250 × CLDN6 gene expression level) + (0.0075 × CNN3 gene expression level) + (-0.0014 × CYP27A1 gene expression level) + (-0.0016 × CYP2C9 gene expression level) + (-0.0001 × FGB gene expression level) + (-0.0005 × HMGCS2 gene expression level) + (0.0016 × S100A10 gene expression level) + (0.0008 × S100A9 gene expression level) + (0.0024 × SQSTM1 gene expression level).

[0024] Preferably, a prognostic risk score greater than or equal to the median indicates high risk, while a score lower than the median indicates low risk. The median is the median value derived from a model built based on the expression profiles of nine genes and patient survival time information of several known liver cancer patients. For example, in one implementation of this application, the specific median value is 0.888356928.

[0025] It should be noted that the risk score in this application is actually a predicted probability. High risk and low risk are only a preliminary classification to facilitate clinical application and differentiation.

[0026] The fifth aspect of this application discloses the application of a protein combination as a biomarker for liver cancer, the protein combination consisting of proteins expressed by nine genes: CLDN6, CNN3, CYP27A1, CYP2C9, FGB, HMGCS2, S100A10, S100A9 and SQSTM1.

[0027] Due to the adoption of the above technical solutions, the beneficial effects of this application are as follows:

[0028] This application presents liver cancer biomarkers based on single-cell sequencing, which can be used to predict the overall survival, cancer status, and prognostic outcome of liver cancer patients. By detecting the liver cancer biomarkers of this application through single-cell sequencing, it is possible to detect heterogeneity within tissues and identify gene expression characteristics within different cells. Attached Figure Description

[0029] Figure 1 This is the ROC curve of the liver cancer risk model in this application for predicting the survival time of patients in the TCGA-LIHC dataset;

[0030] Figure 2 This is the ROC curve of the liver cancer risk model in this application for predicting patient survival time in the ICGC-LIRI-JP dataset;

[0031] Figure 3The Kaplan-Meier survival curves for low-risk and high-risk hepatocellular carcinoma patients in the TCGA-LIHC dataset in this application embodiment are shown.

[0032] Figure 4 The Kaplan-Meier survival curves for low-risk and high-risk hepatocellular carcinoma patients in the ICGC-LIRI-JP dataset in this application embodiment are shown.

[0033] Figure 5 The results of the correlation analysis between the prognostic risk model score and the survival rate of hepatocellular carcinoma patients in the embodiments of this application are shown.

[0034] Figure 6 This is a graph showing the results of univariate Cox regression analysis of independent prognostic factors in the embodiments of this application;

[0035] Figure 7 This is a graph showing the results of independent prognostic factor analysis in a multivariate Cox regression analysis in the embodiments of this application;

[0036] Figure 8 The nomogram risk score is used as the prognostic prediction model in the embodiments of this application.

[0037] Figure 9 This is the time-dependent ROC curve of the prognostic prediction model nomograph in the embodiments of this application. Detailed Implementation

[0038] Existing liver cancer biomarkers and their predictive models are all based on results obtained from conventional RNA sequencing. This application, however, is based on results obtained from single-cell sequencing, solving the problem that conventional RNA sequencing cannot detect tissue heterogeneity and cannot identify gene expression characteristics within different cells. There are currently no reports on predictive models for liver cancer.

[0039] ScRNA-seq is a next-generation sequencing (NGS) technology that enables high-resolution characterization of single cells. ScRNA-seq has been applied to various cancer studies, including melanoma and clear cell renal cell carcinoma. In this study, scRNA-seq mapping was used to identify and classify genomic features and marker genes of primary and metastatic hepatocellular carcinoma. Furthermore, transcriptomic and clinical data from hepatocellular carcinoma patients were obtained from The Cancer Genome Atlas (TCGA) and the Cancer Genome Consortium (ICGC) databases. Further, metastasis-related genes were identified using bioinformatics techniques. Then, nine key angiogenesis-related genes were trained and validated as predictive models. In one implementation of this application, the TCGA and ICGC cohorts can be further combined to predict overall survival in hepatocellular carcinoma patients.

[0040] The present invention will be further described in detail below with reference to specific embodiments and accompanying drawings. The following embodiments are only for further illustration and should not be construed as limiting the present application. Unless otherwise specified, the instruments and materials used in the following embodiments are all commonly used laboratory equipment.

[0041] Example

[0042] I. Materials and Methods

[0043] 1. Acquisition of angiogenesis-related genes

[0044] A search for angiogenesis-related genes was conducted in the GeneCards database (https: / / www.genecards.org / ) using the term "angangiogenesis". A total of 4923 angiogenesis-related genes were obtained in this study, which will be used for further research and analysis.

[0045] 2. Single-cell sequencing data processing

[0046] In this example, the original single-cell transcriptome dataset of 43,228 cell samples from 10 hepatocellular carcinoma patients in the GSE149614 dataset was obtained from the Gene Expression Omnibus (GEO) database (www.ncbi.nlm.nih.gov / geo / ).

[0047] For detailed information, please refer to the following website:

[0048] https: / / www.ncbi.nlm.nih.gov / geo / query / acc.cgi?acc=GSE149614

[0049] These cells comprise cells from three relevant sites: the primary tumor (T), the venous thrombus (PVTT, P), and the metastatic lymph nodes (L). In this case, these cells were divided into two groups:

[0050] Group 1: The non-metastatic group includes the primary tumor (T);

[0051] Group 2: The metastatic group includes venous tumor thrombus (PVTT, P) and metastatic lymph nodes (L).

[0052] This example primarily utilizes the Seurat R package for single-cell sequencing data analysis, employing the t-SNE algorithm for dimensionality reduction analysis. Specific parameters are: nFeature>600, gene<4500, and mitochondrial genes<20%. Differentially expressed genes between the metastatic and non-metastatic groups are selected using the absolute value of the logarithm of gene expression ratios |logFC|=0.5 and adjPval=0.01 as cutoff values, serving as a candidate gene set for subsequent research.

[0053] 3. Identification of hepatocellular carcinoma metastasis-related marker genes in hepatocellular carcinoma patients.

[0054] Candidate gene sets were identified using scRNA-seq analysis, followed by further analysis of characteristic genes associated with patient survival within the candidate gene set. In this case, 377 hepatocellular carcinoma samples from TCGA-LIHC (portal.gdc.cancer.gov / ) were downloaded, and patient transcriptomes were extracted. From the candidate gene set obtained through single-cell sequencing, LASSO regression and multivariate Cox regression models were performed using the glmnet R package to identify marker genes associated with hepatocellular carcinoma prognosis. Furthermore, survival analysis was performed using the survival and survivvalroc R packages. Based on the above calculation results, a model was established to score patients.

[0055] The risk score is calculated using the formula: risk score = ∑(Coefgenes * Expression of each gene), where Coefgenes represents the regression coefficients calculated by the multivariate Cox regression model, and Expression of each gene represents the expression level of each gene.

[0056] The established model was then validated using expression profiles and patient survival information from 260 hepatocellular carcinoma samples from ICGC-LIRI-JP (https: / / dcc.icgc.org / projects / LIRI-JP) and 225 hepatocellular carcinoma samples from GSE14520 (https: / / www.ncbi.nlm.nih.gov / geo / query / acc.cgi?acc=GSE14520). Hepatocellular carcinoma patients in the dataset were divided into high-risk and low-risk groups using the median risk score as the cutoff value. Kaplan-Meier (KM) survival curves were plotted using the survival toolkit to evaluate the overall survival of hepatocellular carcinoma patients in the high- and low-risk groups. A p-value < 0.05 was considered statistically significant.

[0057] 4. Assess the predictive ability of hepatocellular carcinoma risk models.

[0058] The survival ROC toolkit in R software was used to plot time-dependent receiver operating characteristic (ROC) curves, and the area under the ROC curve (AUC) was used to evaluate the predictive ability of the Cox risk assessment model for 1-year, 3-year, and 5-year overall survival in hepatocellular carcinoma (HCC) patients. Univariate and multivariate Cox regression analyses were performed to assess the role of the model's risk score in predicting overall survival, cancer status, and prognostic outcomes in HCC patients. A p-value < 0.05 was considered statistically significant.

[0059] 5. Independent prognostic analysis:

[0060] Univariate and multivariate Cox regression analyses were performed on the prognostic prediction model using the `survival` and `forestplot` packages in R4.0.3 software, and forest plots were generated to evaluate the relationship between various clinical variables (including age, gender, histologic grade, pathologic stage, T stage, and cancer status) and the previously established prognostic risk score with patient prognosis, thereby determining whether the model can serve as an independent prognostic indicator. Univariate and multivariate Cox regression analyses were used to assess the role of the model's risk score in overall survival, prognostic outcome prediction, and early clinical screening for OC patients. A p-value < 0.05 was considered statistically significant.

[0061] II. Results and Analysis

[0062] 1. Screening for genes related to hepatocellular carcinoma metastasis at the single-cell level

[0063] To screen for genes related to hepatocellular carcinoma metastasis at the single-cell level, this example obtains a single-cell dataset (GSE14520) of hepatocellular carcinoma samples from the GEO database.

[0064] https: / / www.ncbi.nlm.nih.gov / geo / query / acc.cgi?acc=GSE14520

[0065] The non-metastatic hepatocellular carcinoma group consisted of the primary tumor (T), containing 34,414 cells; the metastatic group consisted of the portal vein tumor thrombus (P), containing 5,971 cells; and the metastatic lymph nodes (L), containing 3,843 cells. The total number of cell samples from both groups was 44,228.

[0066] Dimensionality reduction analysis of single-cell sequencing data was performed using the t-SNE algorithm, revealing significant differences in cell distribution between the non-metastatic group (T) and the metastatic group (P+L). This study analyzed differentially expressed genes between the non-metastatic group (T) and the metastatic group (P+L), using |logFoldChange|>0.5 and adjPval<0.01 as parameters. A total of 319 differentially expressed genes (DEGs) were identified, which are metastasis-related genes in hepatocellular carcinoma.

[0067] 2. Screening for angiogenesis-related genes associated with hepatocellular carcinoma metastasis.

[0068] The 319 differentially expressed genes (DEGs) obtained from the single-cell analysis were intersected with the angiogenesis-related genes obtained from the GeneCards database to obtain 179 angiogenesis-related genes (ARGs) associated with hepatocellular carcinoma metastasis, as shown in Table 1.

[0069] Table 1. 179 angiogenesis-related genes associated with HCC metastasis

[0070]

[0071]

[0072]

[0073]

[0074] 3. Establishment of a prognostic risk model for hepatocellular carcinoma patients

[0075] Based on the results of previous single-cell sequencing data analysis, univariate Cox regression analysis was performed on 179 angiogenesis-related genes (ARGs) screened for hepatocellular carcinoma metastasis. The results showed that 67 ARGs were significantly associated with the overall survival of hepatocellular carcinoma patients in TCGA-LIHC (P<0.05), as shown in Table 2.

[0076] Table 2. 67 genes significantly associated with overall survival of HCC patients in TCGA-LIHC.

[0077]

[0078]

[0079]

[0080] Furthermore, the LASSO algorithm was used to analyze the 67 genes that were significantly associated with the overall survival of hepatocellular carcinoma patients in TCGA-LIHC. Kaplan-Meier survival curves showed that nine genes, CLDN6, CNN3, CYP27A1, CYP2C9, FGB, HMGCS2, S100A10, S100A9, and SQSTM1, were significantly associated with the overall survival of hepatocellular carcinoma patients (P<0.05).

[0081] The risk score was further calculated using the formula risk score = ∑(Coefgenes*Expressionof each gene), where Coefgenes represents the regression coefficients calculated by the multivariate Cox regression model. The calculation results are shown in Table 3.

[0082] Table 3. Results of multivariate Cox regression analysis of nine genes.

[0083] CLDN6 0.125047191 1.133201929 1.049278144 1.22383814 0.001446307 CNN3 0.007456419 1.007484288 1.002063668 1.012934229 0.006750177 CYP27A1 -0.001426825 0.998574193 0.996844365 1.000307022 0.106756203 CYP2C9 -0.001567062 0.998434165 0.996456834 1.00041542 0.121302334 FGB -8.84E-05 0.999911559 0.999810389 1.000012738 0.086670813 HMGCS2 -0.000546439 0.99945371 0.998871043 1.000036717 0.066274698 S100A10 0.001589603 1.001590867 0.999902615 1.00328197 0.064773839 S100A9 0.000786269 1.000786578 1.000280342 1.001293071 0.002320837 SQSTM1 0.002412746 1.002415659 1.001179106 1.00365374 0.000127557

[0084] Therefore, the prognostic risk score for each hepatocellular carcinoma patient is calculated as follows: (0.1250 × CLDN6 gene expression level) + (0.0075 × CNN3 gene expression level) + (-0.0014 × CYP27A1 gene expression level) + (-0.0016 × CYP2C9 gene expression level) + (-0.0001 × FGB gene expression level) + (-0.0005 × HMGCS2 gene expression level) + (0.0016 × S100A10 gene expression level) + (0.0008 × S100A9 gene expression level) + (0.0024 × SQSTM1 gene expression level).

[0085] In this case, the median prognostic risk score was 0.888356928. Those with scores above the median were considered high-risk, and those below the median were considered low-risk.

[0086] In this example, prognostic risk scoring was performed on 341 samples, and the results are shown in Table 4.

[0087] Table 4 Prognostic Risk Score for Patients with Hepatocellular Carcinoma

[0088]

[0089]

[0090]

[0091]

[0092] 4. Plotting the survival curve (Kaplan-Meier) and ROC curve

[0093] Based on a median risk score of 0.888356928, hepatocellular carcinoma patients with scores above the median in Table 4 were divided into a high-risk group (n=170) and a low-risk group (n=171) with scores below the median. Survival curves were plotted using R language to analyze the overall survival, risk score distribution, and patient survival status. Further ROC analysis revealed that the risk score model had a considerably high predictive ability for the 1-, 3-, and 5-year overall survival of hepatocellular carcinoma patients in the TCGA-LIHC and ICGC-LIRI-JP cohorts. Figure 1 and Figure 2 As shown. Figure 1 ROC curves for predicting patient survival time in the TCGA-LIHC dataset using a liver cancer risk model; Figure 2 The ROC curves for predicting patient survival time in the ICGC-LIRI-JP dataset using a liver cancer risk model are shown. Figure 1 and Figure 2 The results showed that the predictive model established using the nine genes in this case can be used to predict the prognosis of patients with hepatocellular carcinoma, and has important clinical value.

[0094] Furthermore, in the TCGA-LIHC cohort, patients in the low-risk group had significantly better prognoses than those in the high-risk group (P<0.05), such as... Figure 3 As shown; while in ICGC-LIRI-JP, as Figure 4 As shown, the same result was obtained in the dataset. Figure 3 The results showed that the Kaplan-Meier survival curves indicated that the overall survival of patients with hepatocellular carcinoma in the low-risk group was significantly higher than that of patients with hepatocellular carcinoma in the high-risk group in the TCGA-LIHC dataset (P<0.05). Figure 4 The results showed that the overall survival of patients with hepatocellular carcinoma in the low-risk group in the ICGC-LIRI-JP dataset was significantly higher than that of patients with hepatocellular carcinoma in the high-risk group (P<0.05).

[0095] Furthermore, a study was conducted on the correlation between prognostic risk model scores and survival time in the TCGA-LIHC cohort of hepatocellular carcinoma patients. The results are as follows: Figure 5 As shown. Figure 5 The results showed that, through correlation analysis of survival time of hepatocellular carcinoma patients in high- and low-risk model scoring groups, it could be seen that as the risk score increased, the survival time of hepatocellular carcinoma patients gradually decreased.

[0096] 5. Evaluation of the prognostic prediction model nomogram

[0097] Clinical sample data downloaded from the TCGA database were processed, and samples with incomplete information were removed, resulting in a total of 341 hepatocellular carcinoma patients with complete clinical data. The results of univariate Cox regression analysis are as follows: Figure 6 As shown, the results of the multivariate Cox regression analysis are as follows: Figure 7 As shown. Figure 6 and Figure 7 The results showed that tumor status (cancer_status) and prognostic risk score were independent prognostic factors. Therefore, in this case, tumor status (cancer_status) and prognostic risk score were used to construct a nomogram, a prognostic prediction model for hepatocellular carcinoma patients.

[0098] The nomogram risk score, a prognostic prediction model, was calculated for 341 patients with hepatocellular carcinoma using the risk scoring formula. The results are as follows: Figure 8 As shown in Table 5, the median risk score of the prognostic prediction model nomogram is 6.194666116.

[0099] Table 5. Nomogram Risk Scores for Prognostic Prediction Models

[0100]

[0101]

[0102]

[0103]

[0104]

[0105]

[0106]

[0107] In Table 5, survival status is represented by 0 (dead) and 1 (alive); tumor status is represented by 0 (no tumor) and 1 (tumor); risk score refers to the score calculated according to the risk score formula of nine genes, and nomogram risk score refers to the risk score of the prognostic prediction model nomogram.

[0108] Figure 8 The results in Table 5 show that patients were divided into high-risk and low-risk groups based on the median risk score. The high-risk group consisted of 171 patients, and the low-risk group consisted of 170 patients.

[0109] The time-dependent ROC curve analysis results are as follows: Figure 9 As shown, Figure 9 The results showed that the model had a good effect in predicting the prognostic survival of patients with hepatocellular carcinoma, with AUCs of 0.745, 0.747 and 0.784 at 1, 3 and 5 years, respectively.

[0110] The above study shows that nine genes in this case—CLDN6, CNN3, CYP27A1, CYP2C9, FGB, HMGCS2, S100A10, S100A9, and SQSTM1—are significantly associated with the overall survival of patients with hepatocellular carcinoma (P<0.05). Furthermore, the established prognostic risk score (risk score = (0.1250 × CLDN6 gene expression level) + (0.0075 × CNN3 gene expression level) + (-0.0014 × CYP27A1 gene expression level) + (-0.0016 × CYP2C9 gene expression level) + (-0.0001 × FGB gene expression level) + (-0.0005 × HMGCS2 gene expression level) + (0.0016 × S100A10 gene expression level) + (0.0008 × S100A9 gene expression level) + (0.0024 × SQSTM1 gene expression level)) can predict the prognosis of hepatocellular carcinoma patients. This prognostic risk score plays an important role and has reference value in predicting the overall survival, cancer status, and prognostic outcome of hepatocellular carcinoma patients.

[0111] The above examples illustrate the present invention only to aid in understanding it and are not intended to limit the scope of the invention. Those skilled in the art can make various simple deductions, modifications, or substitutions based on the principles of this invention.

Claims

1. A liver cancer biomarker based on single-cell sequencing for predicting the survival of liver cancer patients, characterized in that: The liver cancer markers consist of nine genes or proteins expressed by these nine genes: CLDN6, CNN3, CYP27A1, CYP2C9, FGB, HMGCS2, S100A10, S100A9, and SQSTM1.

2. The use of the liver cancer biomarker according to claim 1 in the preparation of a kit or device for predicting the survival of liver cancer patients.

3. A kit for predicting the survival of patients with liver cancer, characterized in that: Includes reagents for detecting the liver cancer markers of claim 1; The reagents are primers and / or probes for detecting nine genes: CLDN6, CNN3, CYP27A1, CYP2C9, FGB, HMGCS2, S100A10, S100A9, and SQSTM1.

4. The reagent kit according to claim 3, characterized in that: It also includes enzymes and / or reaction solutions for primers and / or probes to detect genes.

5. A kit for predicting the survival of patients with liver cancer, characterized in that: Includes reagents for detecting the liver cancer markers of claim 1; The reagent is a nucleic acid and / or polypeptide for detecting proteins expressed by nine genes: CLDN6, CNN3, CYP27A1, CYP2C9, FGB, HMGCS2, S100A10, S100A9, and SQSTM1.

Images