System for predicting prognosis of patient with primary liver cancer

Abstract

The invention discloses a system for predicting the prognosis of a patient with primary liver cancer and a method for predicting the prognosis of the patient with primary liver cancer based on the system. According to the application, the prognosis risk score of the patient is calculated based on the expression levels of four CD8 + T cell related genes KCTD17, TNFRSF4, SLC16A3 and C5ORF30, and then the prognosis risk (high / low) of the patient is judged. On the basis, a visual Nomogram model for predicting the prognosis (survival probability of 1 year, 2 years, 3 years, 4 years and / or 5 years after definite diagnosis) of the primary liver cancer patient is constructed in combination with the age and Stage stage of the liver cancer patient. Through verification of an external verification set, the system for predicting the prognosis of the patient with primary liver cancer has high accuracy in predicting the survival probability of the patient with primary liver cancer after definite diagnosis. According to the prediction system and method disclosed by the invention, the disease trend can be judged, and useful decision information is provided for a clinical treatment scheme.

Description

Technical Field

[0001] The present invention belongs to the technical field of medical care informatics, and specifically relates to a system for predicting the prognosis of patients with primary liver cancer based on the expression level of CD8+T cell-related genes and a method for predicting the prognosis of patients with primary liver cancer based on the system. Background Art

[0003] In recent years, more and more studies have revealed that the immune microenvironment plays an important role in the occurrence, development and treatment response of tumors; in particular, CD8+T cells play a core role in tumor immune surveillance. They can identify and eliminate tumor cells, so the infiltration level of CD8+T cells and the expression level of related genes are often regarded as important indicators of tumor immune response.

[0004] Studies have shown that high-density CD8+T cell infiltration is usually associated with a better prognosis of tumors. By detecting the expression of CD8+T cell-related genes, the immune status of the tumor microenvironment can be more accurately evaluated, and then the prognosis of the tumor can be predicted. The activity of CD8+T cells is regulated by a variety of genes. For example, the expression of genes such as IFNG, GZMA, PRF1, and PDCD1 directly affects the effector function of CD8+T cells. IFNG can enhance tumor immune response, GZMA and PRF1 are involved in the killing of tumor cells, and the expression of PDCD1 is associated with immunosuppression. Therefore, the expression levels of these key genes can fully reflect the intensity and effectiveness of tumor immune response.

[0005] However, in medical practice, it is found that despite the presence of CD8+T cells, tumors are still able to escape immune surveillance. This immune escape is usually associated with the activation of immunosuppressive factors and immune checkpoints, resulting in the inhibition of CD8+T cell function. In recent years, immune checkpoint inhibitors (such as PD-1 / PD-L1 inhibitors) have become an important means of cancer immunotherapy. Therefore, screening out key genes related to the inhibition of CD8+T cell function will help identify high-risk patients and provide valuable reference for clinical prognosis evaluation and immunotherapy decision-making.

[0006] The Chinese invention patent application with publication number CN117809843A (publication date April 2, 2024) discloses a system for predicting the prognosis of liver cancer based on CD8+T cell-related genes. The CD8+T cell-related genes included in the risk assessment include: CCDC88C, CD7, FYN, GATA3, IL18RAP, ITGB7, MCOLN2 and TAF4B; the FPKM value of each gene is assigned different weights to calculate the risk score of liver cancer patients, and the prognosis of the patient is judged by comparing it with the reference value.

[0007] The occurrence and development of liver cancer is an extremely complex process that has not yet been fully understood. Therefore, it is still clinically meaningful to develop a more accurate and simpler system for predicting the prognosis of patients with primary liver cancer. Summary of the invention

[0008] In view of the problems existing in the prior art, the present invention provides a system for predicting the prognosis of patients with primary liver cancer based on the expression level of CD8+T cell-related genes and a method for predicting the prognosis of patients with primary liver cancer based on the system.

[0009] Therefore, the present invention adopts the following technical scheme.

[0010] A system for predicting the prognosis of patients with primary liver cancer, based on the expression levels of CD8+T cell-related genes in liver cancer tissue that has been separated from the patient, includes the following modules:

[0011] Data collection module: configured to obtain the patient's age, liver cancer stage data, and FPKM values ​​of genes KCTD17, TNFRSF4, SLC16A3, and C5ORF30 in liver cancer tissues;

[0012] The prognostic risk discrimination module is configured to substitute the FPKM values ​​of the genes KCTD17, TNFRSF4, SLC16A3 and C5ORF30 obtained by the data collection module into the risk scoring formula to calculate the prognostic risk score of the patient, and judge the risk level according to the calculated prognostic risk score, wherein a prognostic risk score ≥ 0.05 indicates a high risk, and a prognostic risk score < 0.05 indicates a low risk;

[0013] Prognostic risk score = (KCTD17) × 0.011 + (TNFRSF4) × 0.252 + (SLC16A3) × 0.634 + (C5ORF30) × 0.436,

[0014] Wherein, "(gene name)" represents the FPKM value of the gene obtained by the data collection module;

[0015] The prognosis judgment module is configured to use the established visual Nomogram model for predicting the prognosis of patients with primary liver cancer, assign points to the prognosis risk obtained by the prognosis risk discrimination module and the patient age and liver cancer stage collected by the data collection module, calculate the total score, and obtain the corresponding prognosis judgment result according to the total score; in the visual Nomogram model, the scoring rules for patient age, liver cancer stage and prognosis risk are:

[0016] Age: 10 years old has a score of 0, 18 years old has a score of 7, and starting from 18 years old, the score increases by 0.88 for each additional year of age;

[0017] Stage: Stage I score is 0, Stage II score is 27.96, Stage III score is 70.20, Stage IV score is 100;

[0018] Prognostic risk: low risk score is 0 and high risk score is 34.58.

[0019] Preferably, the prognosis refers to the 1-year survival probability, 2-year survival probability, 3-year survival probability, 4-year survival probability and / or 5-year survival probability of the patient after being diagnosed with primary liver cancer.

[0020] Another object of the present invention is to provide a method for predicting the prognosis of a patient with primary liver cancer; the method is based on the system for predicting the prognosis of a patient with primary liver cancer described in the present invention, and comprises the following steps:

[0021] I. Data Collection Steps

[0022] Obtaining the age of the primary liver cancer patient, liver cancer stage data, and FPKM values ​​of genes KCTD17, TNFRSF4, SLC16A3, and C5ORF30 in liver cancer tissue;

[0023] II. Data Input Steps

[0024] Inputting the data collected in step 1 into the data collection module;

[0025] III. Prognostic risk determination steps

[0026] The prognostic risk discrimination module is used to calculate the prognostic risk score of the patient, and the risk is judged according to the calculated prognostic risk score, wherein a prognostic risk score ≥ 0.05 is high risk, and a prognostic risk score < 0.05 is low risk;

[0027] Wherein, the prognostic risk score is calculated by the following formula:

[0028] Prognostic risk score = (KCTD17) × 0.011 + (TNFRSF4) × 0.252 + (SLC16A3) × 0.634 + (C5ORF30) × 0.436,

[0029] Wherein, "(gene name)" represents the FPKM value of the gene obtained by the data collection module;

[0030] IV. Prognostic steps:

[0031] The prognosis judgment module is used to use the established visual Nomogram model for predicting the prognosis of patients with primary liver cancer, and the patient's age, liver cancer stage and prognostic risk are respectively scored, the total score is calculated, and the prognosis of the patient is judged according to the total score.

[0032] Preferably, the prognosis refers to the 1-year survival probability, 2-year survival probability, 3-year survival probability, 4-year survival probability and / or 5-year survival probability of the patient after being diagnosed with primary liver cancer.

[0033] The FPKM values ​​of the genes KCTD17, TNFRSF4, SLC16A3 and C5ORF30 can be obtained by whole transcriptome sequencing using RNA-Seq or qRT-PCR detection.

[0034] This application is based on the expression levels of 4 CD8+T cell-related genes to calculate the patient's prognostic risk score, and then judge the patient's prognostic risk (high / low). On this basis, combined with the age and Stage staging of liver cancer patients, a visual Nomogram model for predicting the prognosis of patients with primary liver cancer (1 year, 2 years, 3 years, 4 years and / or 5 years after diagnosis) was constructed. The prediction model provided in this application is used to predict the 1-year, 2-year and 3-year survival rate curve areas (AUC values) of liver cancer patients in the external validation set, which are 0.63, 0.69 and 0.73, respectively, which are higher than the prediction model disclosed in CN117809843A. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] The present invention will be further described below in conjunction with the accompanying drawings.

[0036] Figure 1 : Heat map of differentially expressed genes between liver cancer samples and normal samples obtained in Study Example 1. In the figure, "N" represents the normal sample group, and "T" represents the liver cancer sample group.

[0037] Figure 2 : Volcano plot of differentially expressed genes between liver cancer samples and normal samples obtained in Study Example 1.

[0038] Figure 3 : In Study Example 1, the Scale independence graph (left) and mean connectivity graph (right) obtained by WGCNA analysis of the TCGA dataset, where the horizontal axes of the left and right graphs represent the soft threshold (power value).

[0039] Figure 4 : In study example 1, the gene module clustering diagram obtained by WGCNA analysis of the TCGA dataset.

[0040] Figure 5: In Study Example 1, the correlation diagram between the module genes obtained by WGCNA analysis and the immune cell infiltration phenotype.

[0041] Figure 6 : Intersection diagram of differentially expressed genes and WGCNA analysis results in Study Example 1.

[0042] Figure 7 : In study example 1, 14 genes with significant differences in expression of CD8+T cells related to liver cancer were initially screened by Cox univariate analysis.

[0043] Figure 8 : LASSO regression cross-validation plot of the initially screened CD8+T cell differentially expressed genes associated with liver cancer in Study Example 1.

[0044] Fig. 9 : LASSO regression path diagram of the CD8+T cell differentially expressed genes associated with liver cancer that were initially screened in Study 1.

[0045] Fig.10 : Risk score triplet plot of the training set drawn in Study Example 2.

[0046] Fig.11 : Risk score triplet plot of the validation set drawn in Study Example 2.

[0047] Fig.10 and Fig.11 In , each point represents a patient.

[0048] Fig.12 : KM curve of the training set drawn in Study Example 2.

[0049] Fig.13 : The KM curve of the validation set drawn in Study Example 2.

[0050] Fig.14 : ROC curve of the training set drawn in Study Example 2.

[0051] Fig.15 : ROC curve of the validation set drawn in Study Example 2.

[0052] Fig.16 : Heat map of clinical indicator distribution in the entire set of high-risk and low-risk groups drawn in Study Case 2.

[0053] Fig.17 : Distribution diagram of the prognostic risk score in different clinical subgroups drawn in Study 2; in the figure, the vertical axis represents the prognostic risk score, and the horizontal axis represents T stage (A), Stage stage (B), Grade (C), age (D) and gender (E).

[0054] Fig.18 : In study case 3, the results of Cox univariate regression analysis of age, gender, Grade, Stage and prognostic risk score.

[0055] Fig.19 : In study case 3, the results of multivariate regression analysis of age, gender, Grade, Stage and prognostic risk score.

[0056] Fig. 20 : Nomogram visualization constructed in Study Example 3.

[0057] Fig.21 : The ROC curve of the Nomogram model drawn in Study Example 3 predicting the 1-year, 3-year and 5-year survival probability of 369 primary liver cancer patients taken from the TCGA database in Study Example 1.

[0058] Fig. 22 : The calibration curve of the Nomogram model drawn in Study Example 3 predicts the 1-year, 3-year and 5-year survival probability of 369 primary liver cancer patients taken from the TCGA database in Study Example 1.

[0059] Fig.23 : DCA curve diagram drawn in Study Example 3. In the figure:

[0060] “Nomogram” represents the DCA curve of the Nomogram model constructed by the present invention; “Risk Score” represents the DCA curve of the prognostic risk score established by the present invention; “All” represents the DCA curve assuming that all patients are at high risk; “None” is the horizontal reference line.

[0061] Fig.24 : Comparative study example 1, the prognostic risk scoring system of the present invention predicted the ROC curve of the external validation set 1-year, 2-year and 3-year survival probability. DETAILED DESCRIPTION

[0062] The present invention is described below with reference to specific examples. It will be appreciated by those skilled in the art that these examples are only used to illustrate the present invention and are not intended to limit the scope of the present invention in any way.

[0063] The experimental methods in the following examples are all conventional methods unless otherwise specified. The raw materials, reagents, etc. used in the following examples are all commercially available products unless otherwise specified. Research Example 1 Screening of CD8+T cell key genes and construction of a prognostic risk discrimination module 1. Research subjects

[0064] Gene expression information and clinical characteristics of patients with primary liver cancer were obtained through the TCGA database. The clinical characteristics of the 369 patients with primary liver cancer included in this study are shown in Table 1.

[0065] Table 1 Clinical characteristics of patients with primary liver cancer in the TCGA database

[0066]

[0067]

[0068] 2. Preliminary screening of differentially expressed genes in CD8+T cells related to liver cancer

[0069] Differential gene analysis was performed on 369 primary liver cancer samples and 50 normal samples, and 2625 differentially expressed genes were obtained based on |log2FC|>1.5 and P value<0.05. Figure 1 Heatmap and Figure 2 The volcano plot visualizes the differentially expressed genes.

[0070] Then, immune cell infiltration analysis and WGCNA analysis (Weighted Gene Co-Expression Network Analysis) were used to obtain the gene module with the greatest correlation with CD8+ T cells in liver cancer patients in the TCGA database. Figure 3 shown. Figure 3 The network topology analysis results under different power values ​​are shown.

[0071] When the power value is set to 7, the connectivity between genes satisfies the scale-free network distribution (i.e., when the scale-free R2 = 0.9). After similar genes are clustered into modules and the modules with high correlation are merged, a total of 8 gene groups are formed, which are distinguished by different colors, namely black, blue, brown, green, grey, red, turquoise, and yellow, and the number of genes in each module is greater than 60. The results are shown in Figure 4-Figure 5 shown. Figure 4 A gene module clustering plot is shown. Figure 5 The correlation between the module genes and the immune cell infiltration phenotype is shown. Finally, after the intersection of the differential gene results and the WGCNA results, 59 CD8+T cell differentially expressed genes related to liver cancer were obtained. Figure 6 and Table 2.

[0072] Table 2 Differentially expressed genes related to CD8+ T cells in liver cancer tissues

[0073]

[0074]

[0075] 3. Screening of key genes of CB8+T cells related to liver cancer

[0076] 369 liver cancer patients obtained from the TCGA database were divided into training set and validation set in a ratio of 7:3. 59 CD8+T cell differentially expressed genes related to liver cancer were initially screened. Cox univariate analysis of the training set showed that 14 differentially expressed genes related to CD8+T cells were significantly correlated with the prognosis of patients with primary liver cancer (P<0.05). Figure 7 As shown. Further LASSO regression was performed on these 14 genes, and the results are shown in Figure 8 and Fig. 9 . Figure 8 shows a LASSO regression cross validation plot, Fig. 9 The LASSO regression path plot is shown. Figure 8 The parameter λ.min corresponding to the left dashed line in the LASSO regression cross-validation diagram is 4, which is the best fitting λ value, that is, the 4 liver cancer-related CD8+T cell genes are closely related to the overall survival of liver cancer patients. They are KCTD17, TNFRSF4, SLC16A3 and C5ORF30. A formula for calculating the prognostic risk score of liver cancer patients was constructed:

[0077] Prognostic risk score = (KCTD17) × 0.011 + (TNFRSF4) × 0.252 + (SLC16A3) × 0.634 + (C5ORF30) × 0.436,

[0078] Wherein, "(gene name)" represents the expression level (FPKM value) of the gene obtained by the data collection module.

[0079] Study 2: Evaluation of the prognostic risk score constructed in Study 1

[0080] 1. Relationship between prognostic risk score and survival time and survival status

[0081] The prognostic risk scores of each patient in the training set and validation set were calculated, and 0.05 was used as the cut-off value to divide the patients into a high-risk group (prognostic risk score ≥ 0.05) and a low-risk group (prognostic risk score < 0.05). The risk score triplet was plotted for the prognostic risk scores of the training set and validation set, respectively, and the survival time and survival status were plotted as shown in Fig.10 and Fig.11 . Fig.10 The risk score triptych of the training set is shown. Fig.11Risk score triplets for the validation set are shown. Fig.10 and Fig.11 All indicate that the mortality rate increases with the increase of the risk score, and the survival period is shortened accordingly. The specific key genes involved in the prognostic risk score of the present invention are significant risk indicators.

[0082] 2. KM survival analysis

[0083] In order to evaluate the correlation between the prognostic risk score and the patient's survival time, the total survival time of the patients in the high-risk score group and the low-risk score group in the training set and the verification set was statistically analyzed, and the KM curve was drawn. The results are shown in Fig.12 and Fig.13 shown.

[0084] Fig.12 and Fig.13 The results showed that the overall survival rate of patients in the high-risk score group was significantly lower than that in the low-risk score group, whether in the training set or the validation set (P<0.001).

[0085] 3. ROC Curve

[0086] The receiver operating characteristic (ROC) curves of the training set and the validation set were plotted respectively, and the area under the ROC curve (AUC) was used to measure the prediction accuracy of the prognostic risk score. The results are shown in Fig.14 and Fig.15 . Fig.14 It is shown that in the training set, the AUC values ​​of 1-year, 2-year, and 3-year survival are 0.78, 0.77, and 0.80, respectively. Fig.15 It is shown that in the validation set, the AUC values ​​at the corresponding time points are 0.76, 0.75 and 0.74 respectively. The AUC values ​​of 1 year, 2 years and 3 years in the training set and validation set are all greater than 0.5, indicating that the prognostic risk score established by the present invention has a significant correlation with the survival time of patients within 3 years after diagnosis.

[0087] 4. Correlation between prognostic risk score and clinical characteristics

[0088] The training set and the validation set were combined into a whole set, and the patients were divided into a high-risk group and a low-risk group according to the prognostic risk score cut-off value of 0.05. The distribution diagrams of the two groups in terms of clinical indicators (M stage, N stage, age, gender, T stage, Stage stage, Grade grading) were drawn. The results are shown in Fig.16 . Fig.16 The results showed that there were significant differences in the distribution of Grade, Stage and T stages between the high-risk and low-risk groups.

[0089] At the same time, the distribution of prognostic risk scores in different clinical subgroups was compared. Fig.17 AE. Fig.17A shows that, for T stages, the prognostic risk score of T1 is significantly lower than that of other stages. Fig.17 B shows that, for Stage grading, the prognostic risk score of Stage I is significantly lower than that of Stage II. Fig.17 Figure C shows that for Grade stage, the prognostic risk score of G1 stage is significantly lower than that of other stages, and there are also significant differences in the prognostic risk scores between G2, G3 and G4. However, there is no significant difference in the prognostic risk score among HCC patients of different ages and genders ( Fig.17 D and E). In summary, the prognostic risk score established by the present invention can significantly distinguish early and middle-late stage primary liver cancer patients. Study Example 3 Establishment and verification of a visual Nomogram model based on the prognostic risk score of the present invention 1. Screening of indicators included in the Nomogram

[0090] In order to construct a visual nomogram model for predicting the prognosis of patients with primary liver cancer, the prognostic risk score was combined with various clinical indicators based on the above research cases. Cox univariate and multivariate regression analysis was performed, incorporating relevant indicators such as age, gender, grade, stage, and prognostic risk score. The results are shown in Fig.18 and 19 . Fig.18 and Fig.19 All showed that only the prognostic risk score and Stage stage were independent predictors of overall survival in patients with primary liver cancer (P<0.05).

[0091] 2. Construction of Visual Nomogram

[0092] The data of patient age, stage and prognostic risk were imported into the open source data analysis software R version 4.0.4 (http: / / www.rproject.org / ) to obtain the score of each factor and the 1-year, 2-year, 3-year, 4-year and 5-year overall survival probability of primary liver cancer patients corresponding to the total score. Based on this, a visualization of the Nomogram was drawn, including the scores of the three related factors, the total score axis and the 1-year, 2-year, 3-year, 4-year and 5-year overall survival probability axis, see Fig. 20 shown.

[0093] The score for an 18-year-old is 7. Starting from 18, the score increases by 0.88 for each additional year of age.

[0094] Stage: Stage I score is 0, Stage II score is 27.96, Stage III score is 70.20, Stage IV score is 100;

[0095] Prognostic risk: low risk score is 0 and high risk score is 34.58.

[0096] like Fig. 20 As shown in the figure, when using the constructed visual Nomogram model, find the corresponding score of each factor on the scoring axis, then add the scores of all factors, find the corresponding point on the total score axis, draw a straight line perpendicular to the total score axis through the point and intersect with the following 1-year survival probability axis (Probability of 1), 2-year survival probability axis (Probability of 2), 3-year survival probability axis (Probability of 3), 4-year survival probability axis (Probability of 4) and / or 5-year survival probability axis (Probability 5), the value of the intersection is the 1-year survival probability, 2-year survival probability, 3-year survival probability, 4-year survival probability and / or 5-year survival probability.

[0097] 3. Evaluation of the Visual Nomogram Model

[0098] Using the constructed visualization nomogram model, the 1-year, 3-year, and 5-year survival rates of 369 patients with primary liver cancer selected from the TCGA database were predicted in Study 1, and ROC curves and calibration curves were drawn. Fig.21 and Fig. 22 As shown. Among them, Fig.21 The ROC curve shown is used to evaluate the discrimination of the model. Fig. 22 The calibration curve shown was used to assess the accuracy of the model predictions.

[0099] Fig.21 It is shown that the areas under the curve (AUC values) for predicting the 1-year, 3-year and 5-year survival rates of patients with primary liver cancer are 0.76, 0.83 and 0.85, respectively, all greater than 0.5, indicating that the Nomogram model constructed by the present invention can accurately predict the 1-year, 3-year and 5-year survival probabilities of patients.

[0100] The more consistent the survival probability predicted by the model is with the actual observed survival probability, the closer the slope of the calibration curve is to 1. Fig. 22 It is shown that the calibration curve passes through the origin and the slope is close to 1, indicating that the Nomogram model constructed by the present invention can accurately predict the prognosis of patients with primary liver cancer.

[0101] In addition, the DCA (Decision Curve Analysis) of the Nomogram model constructed by the present invention is plotted to evaluate the clinical practicality of the model, and the DCA curve of the prognostic risk score of the present invention is also plotted, see Fig.23 shown. Fig.23This indicates that the Nomogram model constructed by the present invention has greater clinical benefits than simply using the prognostic risk score to predict the survival rate of patients with primary liver cancer.

[0102] Comparative Study Example 1 Comparison between the prognostic risk score constructed by the present invention and the prediction system of CN117809843A

[0103] Microarray data and clinical information of 242 primary hepatocellular carcinoma (HCC) samples were obtained from the ICGC database, and 231 patients with complete clinical information were screened. The clinical characteristics of the patients are shown in Table 3.

[0104] Table 3 Clinical characteristics of patients in the external validation set

[0105]

[0106] The 231 patients were used as an external validation set, and the prognostic risk score of the present invention was used to predict the 1-year, 2-year and 3-year survival rates of the external validation set after diagnosis, and the ROC curve was drawn. Fig.24 .like Fig.24 As shown, the areas under the curve (AUC values) for predicting the 1-year, 2-year and 3-year survival rates of primary liver cancer patients in the external validation set are 0.63, 0.69 and 0.73, respectively, all greater than 0.5, indicating that the Nomogram model constructed by the present invention has a high prediction accuracy.

[0107] The system established in CN117809843A was then used to predict the 1-year, 2-year, and 3-year survival rates of primary liver cancer patients in the external validation set, and the ROC curve was drawn and the area under the curve (AUC) was calculated. The results are shown in Table 4.

[0108] Table 4 Comparison of different prediction models (systems)

[0109] Contained variables 1-year AUC 2-year AUC 3-year AUC CN117809843A 8 genes 0.58 0.62 0.65 Nomogram model of the present invention Genetic risk score + clinical factors 0.63 0.69 0.73

[0110] Table 4 shows that the Nomogram model constructed by the present invention is used to predict the external validation set, and the areas under the ROC curve of the 1-year, 2-year and 3-year survival probabilities are greater than those of the prediction system of CN117809843A, indicating that the Nomogram model of the present invention has a better prediction effect.

[0111] In summary, this application is based on the expression levels of four CD8+T cell-related genes KCTD17, TNFRSF4, SLC16A3 and C5ORF30, and calculates the patient's prognostic risk score to further determine the patient's prognostic risk (high / low). On this basis, combined with the age and Stage staging of liver cancer patients, a visual Nomogram model for predicting the prognosis of patients with primary liver cancer (1 year, 2 years, 3 years, 4 years and / or 5 years of survival probability after diagnosis) was constructed. The prediction system and method disclosed in this application are helpful in judging the trend of the disease and providing useful decision-making information for clinical treatment plans.

Claims

1. A system for predicting the prognosis of patients with primary liver cancer, based on the expression levels of CD8+T cell-related genes in liver cancer tissue that has been separated from the patient, comprising the following modules: Data collection module: configured to obtain the patient's age, liver cancer stage data, and FPKM values ​​of genes KCTD17, TNFRSF4, SLC16A3, and C5ORF30 in liver cancer tissues; The prognostic risk discrimination module is configured to substitute the FPKM values ​​of the genes KCTD17, TNFRSF4, SLC16A3 and C5ORF30 obtained by the data collection module into the risk scoring formula to calculate the prognostic risk score of the patient, and judge the risk level according to the calculated prognostic risk score, wherein a prognostic risk score ≥ 0.05 indicates a high risk, and a prognostic risk score < 0.05 indicates a low risk; Prognostic risk score = (KCTD17) × 0.011 + (TNFRSF4) × 0.252 + (SLC16A3) × 0.634 + (C5ORF30) × 0.436, in, "(Gene name)" represents the FPKM value of the gene obtained by the data collection module; The prognosis judgment module is configured to use the established visual Nomogram model for predicting the prognosis of patients with primary liver cancer, assign points to the prognosis risk obtained by the prognosis risk discrimination module and the patient age and liver cancer stage collected by the data collection module, calculate the total score, and obtain the corresponding prognosis judgment result according to the total score; in the visual Nomogram model, the scoring rules for patient age, liver cancer stage and prognosis risk are: Age: 10 years old has a score of 0, 18 years old has a score of 7, and starting from 18 years old, the score increases by 0.88 for each additional year of age; Stage: Stage I score is 0, Stage II score is 27.96, Stage III score is 70.20, Stage IV score is 100; Prognostic risk: low risk score is 0 and high risk score is 34.

58.

2. The system according to claim 1, characterized in that The prognosis refers to the 1-year survival probability, 2-year survival probability, 3-year survival probability, 4-year survival probability and / or 5-year survival probability of a patient after being diagnosed with primary liver cancer.

3. A method for predicting the prognosis of a patient with primary liver cancer; the method is based on the system for predicting the prognosis of a patient with primary liver cancer according to claim 1, comprising the following steps: I. Data Collection Steps Obtaining the age of the primary liver cancer patient, liver cancer stage data, and FPKM values ​​of genes KCTD17, TNFRSF4, SLC16A3, and C5ORF30 in liver cancer tissue; II. Data Input Steps Inputting the data collected in step 1 into the data collection module; III. Prognostic risk determination steps The prognostic risk discrimination module is used to calculate the prognostic risk score of the patient, and the risk is judged according to the calculated prognostic risk score, wherein a prognostic risk score ≥ 0.05 is high risk, and a prognostic risk score < 0.05 is low risk; Wherein, the prognostic risk score is calculated by the following formula: Prognostic risk score = (KCTD17) × 0.011 + (TNFRSF4) × 0.252 + (SLC16A3) × 0.634 + (C5ORF30) × 0.436, Wherein, "(gene name)" represents the FPKM value of the gene obtained by the data collection module; IV. Prognostic steps: The prognosis judgment module is used to use the established visual Nomogram model for predicting the prognosis of patients with primary liver cancer, and the patient's age, liver cancer stage and prognostic risk are respectively scored, the total score is calculated, and the prognosis of the patient is judged according to the total score.

4. The method according to claim 3, characterized in that The prognosis refers to the 1-year survival probability, 2-year survival probability, 3-year survival probability, 4-year survival probability and / or 5-year survival probability of a patient after being diagnosed with primary liver cancer.

Images