阅读说明：本技术 一种用于肝细胞癌预后评估的生物标志物组合及其筛选方法和应用 (Biomarker combination for hepatocellular carcinoma prognosis evaluation and screening method and application thereof ) 是由 刘利平 张强弩 魏腾 严巧婷 余洁玲 鲍世韵 于 2021-08-09 设计创作，主要内容包括：本发明提供了一种用于肝细胞癌预后评估的生物标志物组合及其筛选方法和应用,所述生物标志物组合包括24个溶质转运蛋白超家族成员。本发明通过多个不同中心的大尺度队列整合了溶质转运蛋白超家族成员的mRNA在HCC中的表达数据,获取了一个具有一致性和稳健性的肝细胞癌相关的溶质转运蛋白超家族成员的mRNA表达谱。通过机器学习方法,基于所述生物标志物组合的mRNA表达水平构建了有效的患者预后评估模型,且该评估模型能有效评估HCC患者的预后,具有一定的特异性和敏感性。(The invention provides a biomarker combination for hepatocellular carcinoma prognosis evaluation, and a screening method and application thereof, wherein the biomarker combination comprises 24 solute transporter superfamily members. The invention integrates the expression data of the mRNA of the solute transport protein superfamily member in HCC through a plurality of large-scale queues at different centers, and obtains the mRNA expression profile of the hepatocellular carcinoma-related solute transport protein superfamily member with consistency and robustness. An effective patient prognosis evaluation model is constructed based on the mRNA expression level of the biomarker combination through a machine learning method, and the evaluation model can effectively evaluate the prognosis of HCC patients and has certain specificity and sensitivity.)

1. A biomarker combination for the prognostic assessment of hepatocellular carcinoma, comprising the solute transporter superfamily members SLC10a1, SLC16a2, SLC17a1, SLC17a2, SLC17A3, SLC19A3, SLC1a1, SLC1A3, SLC1a4, SLC22a1, SLC22a7, SLC25a15, SLC25a44, SLC27a2, SLC27A5, SLC29a1, SLC2a2, SLC38a4, SLC38A6, SLC39a1, SLC41A3, SLC46A3, SLC4a2 and SLC4a 4.

2. A method of screening for a biomarker combination according to claim 1, comprising:

(1) collecting mRNA microchip expression data and RNA sequencing data of hepatocellular carcinoma;

(2) integrating the mRNA microchip expression data by using an RRA algorithm, and listing genes with fold change of 1.5 or < -1.5 and P of 0.05 in the obtained result into an HCC RRA list, wherein the genes in the HCC RRA list are hepatocellular carcinoma related genes;

(3) screening members of the solute transporter superfamily from the HCC RRA list to obtain the biomarker panel of claim 1;

(4) and (3) verifying by using the RNA sequencing data to confirm that the expression patterns of the biomarker combinations in the mRNA microchip expression data and the RNA sequencing data are consistent.

3. Use of the biomarker combination according to claim 1 in the preparation of a hepatocellular carcinoma prognosis evaluation model or a hepatocellular carcinoma prognosis evaluation kit.

4. A model for prognosis evaluation of hepatocellular carcinoma, wherein the mRNA level of each gene in the biomarker set of claim 1 is used as an input variable, and the SLC score is used as an output variable, and the prognosis of hepatocellular carcinoma is evaluated according to the SLC score.

5. The hepatocellular carcinoma prognosis evaluation model according to claim 4, wherein said hepatocellular carcinoma prognosis evaluation model is expressed by the following formula:

SLC fraction ═ Σ (Gene expression level × Integrated HR × Gene _ Weight);

wherein the Gene expression level is the mRNA level, the Integrated HR is the integration risk coefficient, and Gene _ Weight is the important coefficient.

6. The hepatocellular carcinoma prognostic evaluation model according to claim 5, wherein the correspondence between each gene and the significant coefficient in the biomarker combination is as follows:

the important coefficient of SLC39a1 is 8.642577153, the important coefficient of SLC25a15 is 8.175604617, the important coefficient of SLC38a 8.175604617 is 8.175604617, the important coefficient of SLC27a 8.175604617 is 8.175604617, the important coefficient of SLC10a 8.175604617 is 8.175604617, the important coefficient of SLC22a 8.175604617 is 8.175604617, the important coefficient of SLC1a 8.175604617 is 8.175604617, the important coefficient of SLC16a 8.175604617 is 8.175604617, the important coefficient of SLC27a 8.175604617 is 8.175604617, the important coefficient of SLC22a 8.175604617 is 8.175604617, the important coefficient of SLC2a 8.175604617 is 8.175604617, the important coefficient of SLC46a 8.175604617 is 8.175604617, the important coefficient of SLC25a 8.175604617 is 8.175604617, the important coefficient of SLC19a 8.175604617, the important coefficient of SLC38a 8.175604617 is 8.175604617, the important coefficient of SLC41a 8.175604617 is 8.175604617, the important coefficient of SLC4a 8.175604617 is 8.175604617, the important coefficient of SLC29a 8.175604617 is 8.175604617, the important coefficient of SLC 72 is 8.175604617, the important coefficient of SLC 8.175604617 is 8.175604617, the important coefficient of SLC17a 8.175604617 is 8.175604617, the important coefficient of SLC 72 is 8.175604617, the important coefficient of SLC17a 8.175604617, the important coefficient of SLC 72 is 8.175604617, the SLC17a 8.175604617, the important coefficient is 8.175604617, the important coefficient of SLC17a 8.175604617, the important coefficient of SLC 72 is 8.175604617, the important coefficient of the SLC 72 is 8.175604617, the SLC 72, the important coefficient of the SLC17a 8.175604617 is 8.175604617.

7. The prognostic assessment model for hepatocellular carcinoma according to claim 5, wherein SLC39A1, SLC38A6, SLC1A4, SLC41A3 and SLC4A2 in said biomarker combination are risk factors;

SLC25A15, SLC27A2, SLC10A1, SLC22A7, SLC16A2, SLC27A5, SLC22A1, SLC2A2, SLC46A3, SLC25A44, SLC19A3, SLC38A4, SLC4A4, SLC29A1, SLC17A1, SLC1A3, SLC17A2, SLC1A1 and SLC17A3 in the biomarker combination are protective factors.

8. The model of claim 7, wherein the biomarker profile has an integrated risk factor of 1 for risk factors and an integrated risk factor of-1 for protective factors.

9. A kit for prognosis evaluation of hepatocellular carcinoma, comprising a reagent for detecting the mRNA level or protein level of each gene in the biomarker set according to claim 1.

10. Use of the biomarker combination according to claim 1, the hepatocellular carcinoma prognosis evaluation model according to any one of claims 4 to 8, or the hepatocellular carcinoma prognosis evaluation kit according to claim 9 for hepatocellular carcinoma prognosis evaluation or development of drugs for hepatocellular carcinoma treatment.

Technical Field

The invention belongs to the field of medical diagnosis and prognosis evaluation, particularly relates to prognosis evaluation of hepatocellular carcinoma, and particularly relates to a biomarker combination for prognosis evaluation of hepatocellular carcinoma, and a screening method and application thereof.

Background

Liver cancer is the sixth most serious cancer in the world and the fourth most lethal cause of cancer in the world. 85% -95% of primary liver cancers are Hepatocellular carcinoma (HCC), and 80% of HCC patients are diagnosed at a middle and advanced stage due to the hidden onset and incomplete early diagnosis measures, so that the operation chance is lost. The mortality rate of patients with HCC in the middle and late stages is up to 80%, the median survival rate is less than 1 year, and the 5-year survival rate is less than 20%. The progress of surgical techniques, chemoradiotherapy techniques, targeted therapeutic drugs and immunotherapy techniques has been made in recent years, and the progress brings new hopes to patients with intermediate and advanced HCC, but the curative effect of the intermediate and advanced HCC is still unsatisfactory.

Prognostic evaluation is a key step in the treatment of HCC patients. Several staging systems have been proposed in the medical community, including the Barcelona Clinical Liver Cancer (BCLC) system; a TNM staging system; chinese university prognosis index; japanese integrated staging, etc. These staging systems have limitations in clinical use. In order to more accurately predict the prognosis (survival) of a liver cancer patient, in addition to considering the liver function, the tumor stage and the physical condition of the patient, the molecular biological characteristics of the patient must be considered at the same time, and a novel prognosis evaluation system based on the molecular biological characteristics will contribute to the individualized treatment and precise medical treatment of the HCC patient.

The solute carrier (SLC) superfamily encodes solute transporters, the second large membrane protein family in the human genome, a class of membrane proteins that constitute cell membrane, organelle membrane transporters. Members of the SLC family are responsible for the transmembrane transport of various substrates, including inorganic ions, amino acids, fatty acids, neurotransmitters and carbohydrates. The physiological activities involved in this family remain: metabolic transformation, energy homeostasis, tissue development, oxidative stress, host defense, and neuromodulation, among others.

It is now clear that members of the SLC family can participate in the development of tumors by controlling the transport of metabolites, etc. Several studies have demonstrated that there is differential expression of some SLC family members in cancerous and paracancerous tissues, and that this differential expression correlates with patient prognosis and clinical characteristics. Most previous studies have focused on the function and mechanism of individual RBPs in HCC cells at the in vitro level. There is currently a lack of research on the systematic and clinical use of SLC family members in HCC. Some studies, while involving the relationship of SLC family members to clinical prognosis, are based on single or small data sets, and thus some studies are inconsistent and contradictory.

Therefore, the method for developing the prognosis evaluation based on the data of the large-scale queue and the multi-center queue and combined with SLC family members has important application value in the HCC research field.

Disclosure of Invention

Aiming at the defects in the prior art, the invention aims to provide a biomarker combination for hepatocellular carcinoma prognosis evaluation, and a screening method and application thereof. The biomarker combination comprises 24 members of a solute transporter superfamily related to hepatocellular carcinoma, and a prognosis evaluation model constructed by utilizing the biomarkers has better sensitivity and specificity when evaluating the prognosis condition of the hepatocellular carcinoma.

In order to achieve the purpose, the invention adopts the following technical scheme:

in a first aspect, the present invention provides a biomarker panel for the prognostic assessment of hepatocellular carcinoma, comprising 24 members of the solute transporter superfamily, each being:

SLC10A1, SLC16A2, SLC17A1, SLC17A2, SLC17A3, SLC19A3, SLC1A1, SLC1A3, SLC1A4, SLC22A1, SLC22A7, SLC25A15, SLC25A44, SLC27A2, SLC27A5, SLC29A1, SLC2A2, SLC38A4, SLC38A6, SLC39A1, SLC41A3, SLC46A3, SLC4A2 and SLC4A 4.

According to the invention, expression data of mRNA of SLC family members in HCC is integrated through mRNA microchip expression data in the existing database, 24 SLC genes related to hepatocellular carcinoma are screened out, a consistent and robust mRNA expression profile of the SLC family members related to HCC is obtained, and verification is carried out by combining RNA sequencing data in the Cancer Genome database (the Cancer Genome Atlas, TCGA) and the International Cancer Genome alliance (International Cancer Genome Consortium).

The mRNA expression of the 24 SLC family members has consistent expression patterns in 9 HCC queues, so that the 24 SLC family members are identified as HCC-related SLC family members, can be used for prognosis evaluation of hepatocellular carcinoma, and provide a new mode for development and research of a hepatocellular carcinoma prognosis evaluation method.

By machine learning techniques, we constructed an effective patient prognosis evaluation system based on the mRNA expression levels of the above identified HCC-associated SLC family members: the SLC scoring system (SLC score system) has the clinical application value verified in different data sets, and is considered to be capable of effectively evaluating the prognosis of HCC patients and has certain specificity and sensitivity.

In a second aspect, the present invention provides a method of screening a biomarker combination as defined in the first aspect, the method comprising:

(1) collecting mRNA microchip expression data and RNA sequencing data of hepatocellular carcinoma;

(2) integrating the mRNA microchip expression data by using an RRA algorithm, and listing genes with fold change of 1.5 or < -1.5 and P of 0.05 in the obtained result into an HCC RRA list, wherein the genes in the HCC RRA list are hepatocellular carcinoma related genes;

(3) screening out solute transporter superfamily members from the HCC RRA list to obtain the biomarker combination of the first aspect;

(4) and (3) verifying by using the RNA sequencing data to confirm that the expression patterns of the biomarker combinations in the mRNA microchip expression data and the RNA sequencing data are consistent.

In a third aspect, the present invention provides an application of the biomarker combination according to the first aspect in the preparation of a hepatocellular carcinoma prognosis evaluation model or a hepatocellular carcinoma prognosis evaluation kit.

In a fourth aspect, the present invention provides a hepatocellular carcinoma prognosis evaluation model, in which the mRNA levels of the respective genes in the biomarker combinations of the first aspect are used as input variables, the SLC score is used as an output variable, and the prognosis of hepatocellular carcinoma is evaluated according to the SLC score.

In the present invention, we have constructed an effective patient prognosis evaluation model, or SLC score system, based on the above identified mRNA expression levels of the SLC family members associated with HCC, by machine learning techniques.

The clinical application value of the scoring system is verified in different data sets, so that the SLC scoring system provided by the invention can effectively evaluate the prognosis of HCC patients, and has certain specificity and sensitivity.

As a preferred embodiment of the present invention, the hepatocellular carcinoma prognosis evaluation model is represented by the following formula:

SLC fraction ═ Σ (Gene expression level × Integrated HR × Gene _ Weight);

wherein the Gene expression level is the mRNA level, the Integrated HR is the integration risk coefficient, and Gene _ Weight is the important coefficient.

As a preferred technical scheme of the invention, the corresponding relation between each gene and the important coefficient in the biomarker combination is as follows:

the important coefficient of SLC39a1 is 8.642577153, the important coefficient of SLC25a15 is 8.175604617, the important coefficient of SLC38a 8.175604617 is 8.175604617, the important coefficient of SLC27a 8.175604617 is 8.175604617, the important coefficient of SLC10a 8.175604617 is 8.175604617, the important coefficient of SLC22a 8.175604617 is 8.175604617, the important coefficient of SLC1a 8.175604617 is 8.175604617, the important coefficient of SLC16a 8.175604617 is 8.175604617, the important coefficient of SLC27a 8.175604617 is 8.175604617, the important coefficient of SLC22a 8.175604617 is 8.175604617, the important coefficient of SLC2a 8.175604617 is 8.175604617, the important coefficient of SLC46a 8.175604617 is 8.175604617, the important coefficient of SLC25a 8.175604617 is 8.175604617, the important coefficient of SLC19a 8.175604617, the important coefficient of SLC38a 8.175604617 is 8.175604617, the important coefficient of SLC41a 8.175604617 is 8.175604617, the important coefficient of SLC4a 8.175604617 is 8.175604617, the important coefficient of SLC29a 8.175604617 is 8.175604617, the important coefficient of SLC 72 is 8.175604617, the important coefficient of SLC 8.175604617 is 8.175604617, the important coefficient of SLC17a 8.175604617 is 8.175604617, the important coefficient of SLC 72 is 8.175604617, the important coefficient of SLC17a 8.175604617, the important coefficient of SLC 72 is 8.175604617, the SLC17a 8.175604617, the important coefficient is 8.175604617, the important coefficient of SLC17a 8.175604617, the important coefficient of SLC 72 is 8.175604617, the important coefficient of the SLC 72 is 8.175604617, the SLC 72, the important coefficient of the SLC17a 8.175604617 is 8.175604617.

As a preferred technical solution of the present invention, the combination of biomarkers has SLC39A1, SLC38A6, SLC1A4, SLC41A3 and SLC4A2 as risk factors.

SLC25A15, SLC27A2, SLC10A1, SLC22A7, SLC16A2, SLC27A5, SLC22A1, SLC2A2, SLC46A3, SLC25A44, SLC19A3, SLC38A4, SLC4A4, SLC29A1, SLC17A1, SLC1A3, SLC17A2, SLC1A1 and SLC17A3 in the biomarker combination are protective factors.

Preferably, the risk factor for integration of the combination of biomarkers is 1 and the risk factor for integration of the protective factor is-1.

In a fifth aspect, the present invention provides a kit for prognosis evaluation of hepatocellular carcinoma, wherein the kit comprises a reagent for detecting the mRNA level or protein level of each gene in the biomarker combination according to the first aspect.

In a sixth aspect, the present invention provides a biomarker combination according to the first aspect, a hepatocellular carcinoma prognosis evaluation model according to the third aspect, or an application of the hepatocellular carcinoma prognosis evaluation kit according to the fourth aspect in hepatocellular carcinoma prognosis evaluation or development of a drug for hepatocellular carcinoma treatment.

Compared with the prior art, the invention has the beneficial effects that:

(1) the invention provides a biomarker combination for hepatocellular carcinoma prognosis evaluation, which comprises the following components in sequence according to importance ranking: SLC39a1, SLC25a15, SLC38A6, SLC27a2, SLC10a1, SLC22a7, SLC1a4, SLC16a2, SLC27A5, SLC22a1, SLC2a2, SLC46A3, SLC25a44, SLC19A3, SLC38a4, SLC41A3, SLC4a2, SLC4a4, SLC29a1, SLC17a1, SLC1A3, SLC17a2, SLC1a1 and SLC17 A3; wherein SLC39A1, SLC38A6, SLC1A4, SLC41A3 and SLC4A2 are risk factors, and the rest are protective factors; the expression levels of the biomarker combination in hepatocellular carcinoma and normal tissues are obviously different, so that the biomarker combination can be used for prognosis evaluation of hepatocellular carcinoma;

(2) the invention also provides a hepatocellular carcinoma prognosis evaluation model, in the prognosis evaluation model, the mRNA level of each gene in the biomarker combination is used as an input variable, the SLC score is used as an output variable, and the prognosis of hepatocellular carcinoma is evaluated according to the SLC score, wherein the higher the SLC score is, the worse the prognosis condition of a patient is; therefore, the hepatocellular carcinoma prognosis evaluation model can help doctors to predict the prognosis condition of patients in advance, and has the effect of assisting clinical treatment.

Drawings

FIG. 1 is the mRNA expression of 24 HCC-associated SLC members screened and identified in example 1 in a 9 HCC microchip cohort.

FIG. 2A is a graph showing mRNA expression in TCGA-LIHC sequencing data set of 24 SLC-associated members identified by screening in example 1.

FIG. 2B is a schematic diagram showing the mRNA expression of 24 HCC-associated SLC members screened and identified in example 1 in the ICGC-LIRI-JP sequencing data set.

FIG. 3A is a graph of the difference in mRNA expression in TCGA-LIHC sequencing data set between 24 HCC-associated SLC members in example 2 in liver cancer tissue and peri-cancer tissue.

FIG. 3B is a graph showing the difference between the expression level of mRNA of 24 HCC-associated SLC members in the ICGC-LIRI-JP sequencing data set in the liver cancer tissue and the peri-cancer tissue in example 2.

FIG. 4 is a graph showing the correlation between the mRNA levels of the 24 HCC-associated SLC genes and the overall survival of the patient, as simulated in example 2 using the COX proportional hazards model.

FIG. 5 is a ranking chart of the importance (kini coefficient) of the overall survival of HCC evaluated by 24 HCC-related SLC genes obtained by the random forest algorithm in example 3.

Figure 6A is a graph of the SLC scoring system referred to in this invention after assessing overall survival of HCC patients in the GSE14520 cohort.

Figure 6B is a graph of the SLC scoring system mentioned in the present invention after assessing overall survival of HCC patients in the TCGA-LIHC cohort.

FIG. 6C is a graph of the SLC scoring system proposed in the present invention after assessing overall survival of HCC patients in the ICGC-LIRI-JP cohort.

Figure 7A is a graph of the SLC scoring system of the present invention after assessing disease-free survival of HCC patients in the GSE14520 cohort.

FIG. 7B is a graph of the SLC scoring system of the present invention after assessing disease-free survival of HCC patients in the TCGA-LIHC cohort.

FIG. 8A is a ROC plot obtained after evaluation of HCC patients in the GSE14520 cohort by the SLC scoring system of the present invention.

FIG. 8B is a ROC plot obtained after evaluation of HCC patients in the TCGA-LIHC cohort by the SLC scoring system of the present invention.

Detailed Description

The technical solutions of the present invention are further described in the following embodiments with reference to the drawings, but the following examples are only simple examples of the present invention and do not represent or limit the scope of the present invention, which is defined by the claims.

In the following examples, reagents and consumables used were obtained from conventional reagent manufacturers in the field unless otherwise specified; unless otherwise indicated, all experimental methods and technical means are conventional in the art.

Example 1

This example was used to screen and identify HCC-associated SLC family members.

The method comprises the following specific steps:

(1) microchip (microarray mRNA) Expression data obtained from Gene Expression Omnibus (https:// www.ncbi.nlm.nih.gov/geo /) for HCC cohorts of 9 different clinical centers, including GSE14520, GSE22058, GSE25097, GSE36376, GSE45436, GSE64041, GSE76427, GSE54236 and GSE 63898;

RNA sequencing data, namely TCGA-LIHC and ICGC-LIRI-JP, were obtained from the HCC cohort from the tumor Genome database (TCGA) and the International Cancer Genome Consortium, respectively.

The gene ID of all datasets were converted to the latest HUGO gene signature and mRNA expression data were log2 normalized.

(2) To identify SLC family members with consistent mRNA expression patterns in multiple HCC cohorts, the mRNA microchip expression data cohorts from 9 clinical centers were integrated in this example using the Robust Rank Aggregation (RRA) algorithm, which was performed using R statistics software (version3.6.1) and the robustrank aggregation package;

the genes whose fold change is >1.5 or < -1.5 and P <0.05 in the result are defined as HCC RRA list (RRA list) which contains 1280 genes and which show more significant up-or down-regulation in the 9 HCC cohorts.

There are 24 members of the SLC family in the RRA list, including SLC10a1, SLC16a2, SLC17a1, SLC17a2, SLC17A3, SLC19A3, SLC1a1, SLC1A3, SLC1a4, SLC22a1, SLC22a7, SLC25a15, SLC25a44, SLC27a2, SLC27A5, SLC29a1, SLC2a2, SLC38a4, SLC38A6, SLC39a1, SLC41A3, SLC46A3, SLC4a2 and SLC4a 4.

The mRNA expression of the 24 SLC family members had a consistent expression pattern in the 9 HCC cohorts and were therefore identified as HCC-associated SLC family members.

The difference in mRNA expression between these 24 SLC members in the 9 cohorts, cancer tissue and normal tissue around the cancer is shown in FIG. 1.

(3) The expression conditions of the 24 SLC members in two RNA sequencing data sets of TCGA-LIHC and ICGC-LIRI-JP are also analyzed in the example;

as shown in FIGS. 2A and 2B, the expression pattern of these 24 SLC genes in the two RNA sequencing data sets of TCGA-LIHC and ICGC-LIRI-JP was completely consistent with that in the 9 microchip arrays.

The above experimental results prove that the expression of the 24 HCC-related SLC family members is closely related to hepatocellular carcinoma.

Example 2

In this example, the relationship of 24 SLC genes to TNM staging and overall survival of HCC patients was analyzed.

(1) Relationship between 24 SLC genes and TNM stage of HCC patients

As shown in FIGS. 3A and 3B, in the two RNA sequencing data sets of TCGA-LIHC and ICGC-LIRI-JP, the principal component analysis method suggested that the mRNA expression of 24 SLC genes could effectively distinguish the liver cancer tissue from the peri-normal tissue.

(2) Relationship between 24 SLC genes and overall survival of HCC patients

In this example, the COX proportional risk regression model was used to analyze the relationship between the mRNA levels of 24 SLC genes and the overall survival of HCC patients;

as shown in FIG. 4, some of the 24 SLC genes are risk factors for a poor prognosis of HCC patients, and some are protective factors;

wherein, the risk factors include SLC39A1, SLC38A6, SLC1A4, SLC41A3 and SLC4A2, and the protective factors include SLC25A15, SLC27A2, SLC10A1, SLC22A7, SLC16A2, SLC27A5, SLC22A1, SLC2A2, SLC46A3, SLC25A44, SLC19A3, SLC38A4, SLC4A4, SLC29A1, SLC17A1, SLC1A3, SLC17A2, SLC1A1 and SLC17A 3.

Example 3

In this example, a novel HCC prognosis evaluation model was constructed based on the mRNA levels of the HCC-associated SLC family member genes.

(1) On the basis of the identified 24 HCC-related SLC family members, the importance of the 24 HCC-related SLC family members in evaluating the prognosis of HCC patients is evaluated by using a random forest algorithm;

the random forest model construction is carried out by using R statistical software (version3.6.1) and a randomForest program package, TCGA-LIHC data is used as a training set, and a GSE14520 queue is used as a verification set;

figure 5 shows the importance coefficients (kini coefficients) for the evaluation of HCC patients by 24 SLC family members, as shown in table 1 below:

TABLE 1

(2) A model for prognosis evaluation of HCC based on SLC family member gene is constructed, and the model is a scoring system and is named as SLC scoring system.

The formula for calculation using the formula SLC score (SLC-score) is:

SLC-score＝∑(Gene_score×Gene_Weight)。

wherein Gene _ Weight is an important coefficient of each Gene calculated by a random forest algorithm;

the calculation formula for Gene score for a single Gene is:

Gene_score＝Gene express level×Integrated HR。

wherein, Gene expression level is the mRNA level corresponding to the Gene in the liver cancer tissue of a certain patient;

(3) determination of Integrated HR (Integrated Risk coefficient)

In the embodiment, a COX proportional model is constructed on the basis of 24 SLC genes in three data sets of TCGA-LIHC, GSE14520 and ICGC-LIRI-JP respectively, and 3 models are obtained in total;

the risk factors of each gene in 3 models in each data set were Integrated to obtain the Integrated HR of the gene. The Integrated HR finally takes two values, 1 or-1, indicating that the gene is a risk factor or a protective factor, respectively.

The Integrated HR values for the 24 SLC genes are shown in Table 2 below:

TABLE 2

Thus, this example provides a HCC prognosis evaluation model constructed using mRNA levels of 24 SLC genes.

Example 4

In this example, SLC scoring was performed on patients in three data sets, GSE14520, TCGA-LIHC, and ICGC-LIRI-JP, using the HCC prognosis evaluation model provided in example 3.

The patients in each data set were assigned the four segments Q1 to Q4 with the upper quartile, the middle quartile and the lower quartile as cut-points, with the patients in the Q1 segment having the lowest SLC score and the patients in the Q4 segment having the highest SLC score.

GSE14520 (n-242), whose survival analysis result (number at risk) is shown in table 3 and fig. 6A below;

TABLE 3

TCGA-LIHC (n ═ 364), and the survival analysis results (number at risk) thereof are shown in table 4 below and fig. 6B;

TABLE 4

ICGC-LIRI-JP (n ═ 212), whose survival analysis results (number at risk) are shown in table 5 below and fig. 6C;

TABLE 5

According to Kaplan-Meier survival analysis, the liver cancer patients with larger SLC scores are found to have worse overall survival in three data sets of GSE14520, TCGA-LIHC and ICGC-LIRI-JP.

Meanwhile, GSE14520 (n-242) has disease-free survival results as shown in table 6 below and fig. 7A;

TABLE 6

TCGA-LIHC (n ═ 359), disease-free survival results are shown in table 7 below and fig. 7B;

TABLE 7

Therefore, the disease-free survival rate of the liver cancer patients with larger SLC scores is lower, and the SLC scores can effectively reflect the survival and recurrence conditions of the HCC patients.

In the present invention, the curve of the tested workers using the SLC scoring system for assessing overall survival of patients is shown in fig. 8A and 8B, and the specific values are shown in table 8 below:

TABLE 8

AUC(％)	GSE14520	TCGA-LIHC
			1 year	70.26	67.76
For 3 years	66.84	63.95
			5 years old	65.83	62.08

In conclusion, the scoring system has good specificity and sensitivity in evaluating the prognosis of patients, and has good efficacy (AUC value of area under curve is more than 60%).

The applicant declares that the above description is only a specific embodiment of the present invention, but the scope of the present invention is not limited thereto, and it should be understood by those skilled in the art that any changes or substitutions that can be easily conceived by those skilled in the art within the technical scope of the present invention are within the scope and disclosure of the present invention.