Application of device for predicting recurrence risk of hepatocellular carcinoma patient after radical resection in prediction of recurrence risk of hepatocellular carcinoma

Abstract

The invention discloses application of a device for predicting the recurrence risk of a hepatocellular carcinoma patient after radical resection in predicting the recurrence risk of hepatocellular carcinoma. By measuring the content of specific proteins BCAT1, FKBP10 and SOAT1 in hepatocellular carcinoma tissues, the three proteins can be used for evaluating the recurrence risk of hepatocellular carcinoma. Experiments prove that the combined analysis of the three proteins can accurately predict the recurrence risk of the hepatocellular carcinoma patients in the early stage (within 2 years) and the late stage (within 5 years), and shows high sensitivity and specificity. In addition, the method can effectively predict the actual relapse risk of the hepatocellular carcinoma patients which are judged to be low in relapse risk through conventional clinical indexes. The invention provides a novel and more accurate means for risk assessment and patient management of hepatocellular carcinoma, and has a wide application prospect.

Description

Technical Field

[0001] The present invention relates to the application of a device for predicting the recurrence risk after radical resection of hepatocellular carcinoma patients in the field of bioinformatics in predicting the recurrence risk of hepatocellular carcinoma. Background Art

[0002] Hepatocellular carcinoma (HCC) is the sixth most common malignant tumor globally and the fourth leading cause of cancer-related deaths. Due to its insidious onset, high malignancy, and low cure rate, liver cancer is one of the malignant tumors that seriously threaten human health and quality of life. Surgery is the preferred and most effective method for treating liver cancer. Unfortunately, about 70% of patients experience tumor recurrence within 5 years after surgery, and 70% of the recurrence events are early recurrences within 2 years after surgery. Understanding and predicting tumor recurrence is very important for improving the surgical prognosis. Currently, clinically, traditional methods are used to evaluate the recurrence risk of patients based on clinicopathological factors (such as tumor size, number, microvascular cancer thrombus, degree of differentiation, and alpha-fetoprotein level, etc.) to provide a basis for treatment decisions. However, the heterogeneity of tumor biology greatly affects the accuracy of these systems. For example, some patients with single tumor nodules less than 5 cm in diameter and without microvascular invasion are considered to have a low recurrence risk, but still some patients experience recurrence or even death after surgery. Therefore, the scientific problem of diagnosing liver cancer recurrence, how to effectively predict tumor recurrence after resection, and how to prevent tumor recurrence after liver cancer surgery are of great clinical significance for improving the efficacy of liver cancer.

[0003] Finding sensitive recurrence prediction markers based on the molecular characteristics of hepatocellular carcinoma and exploring and establishing a detection method with simple operation, wide application, low cost, high qualitative sensitivity, and accurate localization has become an urgent need to improve clinical efficacy. Summary of the Invention

[0004] The technical problem to be solved by the present invention is how to predict the recurrence risk after radical resection of hepatocellular carcinoma patients.

[0005] To solve the above technical problem, the present invention first provides a device for predicting the recurrence risk after radical resection of hepatocellular carcinoma patients, including substances for detecting the protein contents of BCAT1, FKBP10, and SOAT1.

[0006] In the above device, the substances for detecting the protein contents of BCAT1, FKBP10, and SOAT1 may be composed of a substance for detecting the protein content of BCAT1, a substance for detecting the protein content of FKBP10, and a substance for detecting the protein content of SOAT1.

[0007] Specifically, the substances for detecting the BCAT1 protein content may include reagents and / or instruments for detecting the BCAT1 protein content by immunohistochemical staining; the substances for detecting the FKBP10 protein content include reagents and / or instruments for detecting the FKBP10 protein content by immunohistochemical staining; the substances for detecting the SOAT1 protein content include reagents and / or instruments for detecting the SOAT1 protein content by immunohistochemical staining.

[0008] The substances for detecting the BCAT1 content may include (or may be) BCAT1 antibodies; the substances for detecting the FKBP10 content may include (or may be) FKBP10 antibodies; the substances for detecting the SOAT1 content may include (or may be) SOAT1 antibodies.

[0009] The above device may further include a protein content data processing module, which is used to convert the BCAT1, FKBP10, and SOAT1 protein contents in the isolated hepatocellular carcinoma tissue from the patient with hepatocellular carcinoma to be predicted into the predicted scores of the patient with hepatocellular carcinoma to be predicted, and predict the recurrence risk after radical resection of the hepatocellular carcinoma patient according to the predicted scores of the patient with hepatocellular carcinoma to be predicted.

[0010] In an embodiment of the present invention, the predicted score of the patient with hepatocellular carcinoma to be predicted is the immunohistochemical scores of BCAT1, FKBP10, and SOAT1. The patient with hepatocellular carcinoma to be predicted who satisfies at least one of "BCAT1 immunohistochemical score ≥ 8, FKBP10 immunohistochemical score ≥ 10, SOAT1 immunohistochemical score ≥ 6" has a very high early (2 years after radical resection) and / or late (5 years after radical resection) recurrence risk; the patient with hepatocellular carcinoma to be predicted who satisfies "BCAT1 immunohistochemical score < 8, FKBP10 immunohistochemical score < 10, and SOAT1 immunohistochemical score < 6" has a low early (2 years after radical resection) or late (5 years after radical resection) recurrence risk.

[0011] Among them, the immunohistochemical score is the product of the percentage of positive cells and the staining intensity of the immunohistochemical staining of the hepatocellular carcinoma tissue.

[0012] In the above device, the patient with hepatocellular carcinoma may be a patient diagnosed with hepatocellular carcinoma. Specifically, the patient with hepatocellular carcinoma may also be a patient with a low recurrence risk of clinical hepatocellular carcinoma, that is, a patient with a single tumor hepatocellular carcinoma with a single tumor less than or equal to 5 cm and no microvascular invasion.

[0013] In the above device, the recurrence risk after radical resection may be the recurrence risk within 2 years or 5 years after radical resection.

[0014] The application of the device for predicting the recurrence risk after radical resection of hepatocellular carcinoma patients in the preparation of products for predicting the recurrence risk after radical resection of hepatocellular carcinoma patients also falls within the protection scope of the present invention.

[0015] In the above application, the hepatocellular carcinoma patients can be patients diagnosed with hepatocellular carcinoma. Specifically, the hepatocellular carcinoma patients can also be patients with low recurrence risk of clinical hepatocellular carcinoma, that is, patients with single tumor hepatocellular carcinoma with a single tumor less than or equal to 5 cm and no microvascular invasion.

[0016] In the above application, the recurrence risk after radical resection can be the recurrence risk within 2 years or 5 years after radical resection.

[0017] In the present invention, the contents of BCAT1, FKBP10 and SOAT1 can be the contents of BCAT1, FKBP10 and SOAT1 in hepatocellular carcinoma tissues.

[0018] In the present invention, the radical resection refers to the complete resection of the hepatocellular carcinoma tissue observed by the naked eye during the operation, with negative resection margins under the microscope of the tissue section after the operation, and no residual lesions detected by imaging examination within two months after the operation.

[0019] BCAT1: NP_005495.2, update date: May 13, 2023; BCAT1 gene: NG_008170.2, update date: October 31, 2022. FKBP10: NP_068758, update date: July 18, 2023; FKBP10 gene: NG_015860.1, update date: March 25, 2023. SOAT1: NP_003092.4, update date: December 25, 2022; SOAT1 gene: NG_030638.1, update date: February 18, 2021.

[0020] Experimental results show that BCAT1, FKBP10 and SOAT1 can jointly predict the recurrence risk of hepatocellular carcinoma patients. The AUC for jointly predicting early (2-year) recurrence is 0.634, the sensitivity is 0.466, and the specificity is 0.802. The AUC for jointly predicting late (5-year) recurrence is 0.623, the sensitivity is 0.348, and the specificity is 0.897. Furthermore, it can also predict the recurrence risk of hepatocellular carcinoma patients with low recurrence risk. The AUC for jointly predicting early (2-year) recurrence is 0.677, the sensitivity is 0.474, and the specificity is 0.880. The AUC for jointly predicting late (5-year) recurrence is 0.642, the sensitivity is 0.387, and the specificity is 0.897. The present invention has good application prospects.

[0021] The present invention will be further described in detail below in conjunction with specific embodiments. The provided embodiments are only for clarifying the present invention and not for limiting the scope of the present invention. The following embodiments can be used as a guide for those of ordinary skill in the art to make further improvements and do not constitute any limitation to the present invention in any way. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] Figure 1 are representative pictures of the high and low expressions of three proteins, BCAT1, FKBP10, and SOAT1, in immunohistochemical detection of liver cancer tissues.

[0023] Figure 2 is the recurrence Kaplan-Meier curve of BCAT1, FKBP10, and SOAT1 three proteins drawn according to the optimal cut-off value in the immunohistochemical training set.

[0024] Figure 3 Analysis of the recurrence situation predicted by the three-protein classifier combination in the training set. (A) The recurrence Kaplan-Meier curve of 157 patients in the training set grouped according to the three-protein classifier combination. The meaning of high expression is that the immunohistochemical scores of the three proteins satisfy at least one of "BCAT1≥8, FKBP10≥10, SOAT1≥6", and the meaning of low expression is that each protein in the three-protein classifier is low expressed, that is, "BCAT1<8 and FKBP10<10 and SOAT1<6"; (B) The ROC curves of the training set patients predicting early (2-year) recurrence according to the three proteins alone and in combination; (C) The ROC curves of the training set patients predicting late (5-year) recurrence according to the three proteins alone and in combination.

[0025] Figure 4 is the analysis of the recurrence situation predicted by the three-protein classifier combination in the validation set. (A) The recurrence Kaplan-Meier curve of 139 patients in the validation set grouped according to the three-protein classifier combination; The meaning of high expression is that the immunohistochemical scores of the three proteins satisfy at least one of "BCAT1≥8, FKBP10≥10, SOAT1≥6", and the meaning of low expression is that each protein in the three-protein classifier is low expressed, that is, "BCAT1<8 and FKBP10<10 and SOAT1<6"; (B) The ROC curves of the validation set patients predicting early (2-year) recurrence according to the three proteins alone and in combination; (C) The ROC curves of the validation set patients predicting late (5-year) recurrence according to the three proteins alone and in combination.

[0026] Figure 5Analysis of the predicted recurrence of the triple-protein classifier combination in patients with low clinical recurrence risk in the training set. (A) Recurrence Kaplan-Meier curves of 157 training set patients stratified by recurrence risk according to clinical characteristics; the stratification is defined as follows: patients with a single tumor less than or equal to 5 cm and no microvascular invasion are considered in the clinically low recurrence risk (lowrisk) group, patients with a single tumor greater than 5 cm and no microvascular invasion are considered in the clinically intermediate recurrence risk (Intermediaterisk) group, and patients with a single tumor with microvascular invasion or multiple tumors are considered in the clinically high recurrence risk (Highrisk) group; (B) Kaplan-Meier curves of recurrence predicted by the triple-protein classifier combination for 70 patients in the clinically low recurrence risk group, where high expression means that the immunohistochemical scores of the three proteins satisfy at least one of "BCAT1≥8, FKBP10≥10, SOAT1≥6", and low expression means that each protein in the triple-protein classifier is low expressed, that is, "BCAT1<8 and FKBP10<10 and SOAT1<6"; (C) ROC curves for predicting early (2-year) recurrence by the three proteins alone and in combination for 70 patients with clinically low recurrence risk; (D) ROC curves for predicting late (5-year) recurrence by the three proteins alone and in combination for 70 patients with clinically low recurrence risk.

[0027] Figure 6 Analysis of the predicted recurrence of the triple-protein classifier combination in patients with low clinical recurrence risk in the validation set. (A) Recurrence Kaplan-Meier curves of 139 validation set patients stratified by recurrence risk according to clinical characteristics, the stratification is defined as follows: patients with a single tumor less than or equal to 5 cm and no microvascular invasion are considered in the clinically low recurrence risk (lowrisk) group, patients with a single tumor greater than 5 cm and no microvascular invasion are considered in the clinically intermediate recurrence risk (Intermediaterisk) group, and patients with a single tumor with microvascular invasion or multiple tumors are considered in the clinically high recurrence risk (Highrisk) group; (B) Kaplan-Meier curves of recurrence predicted by the triple-protein classifier combination for 69 patients with clinically low recurrence risk, where high expression means that the immunohistochemical scores of the three proteins satisfy at least one of "BCAT1≥8, FKBP10≥10, SOAT1≥6", and low expression means that each protein in the triple-protein classifier is low expressed, that is, "BCAT1<8 and FKBP10<10 and SOAT1<6"; (C) ROC curves for predicting early (2-year) recurrence by the three proteins alone and in combination for 69 patients with clinically low recurrence risk; (D) ROC curves for predicting late (5-year) recurrence by the three proteins alone and in combination for 69 patients with clinically low recurrence risk. Detailed implementation

[0028] In the experimental methods of the following examples, unless otherwise specified, they are all conventional methods, and are carried out according to the techniques or conditions described in the literature in this field or according to the product instructions. The materials, reagents, instruments, etc. used in the following examples, unless otherwise specified, can be obtained from commercial channels.

[0029] Example 1. BCAT1, FKBP10, and SOAT1 can jointly predict the recurrence risk of patients with hepatocellular carcinoma after radical resection

[0030] I. Tissue sample information

[0031] The hepatocellular carcinoma tissues of HCC patients were used for immunohistochemistry (IHC) experiments, and the hepatocellular carcinoma tissue microarray used was a product of Shanghai CoreBio System Co., Ltd. The tissue microarray cohort was derived from 296 patients pathologically diagnosed with HCC who underwent radical resection and did not receive neoadjuvant therapy between February 2006 and December 2011. The median follow-up time was 4.8 years. The clinicopathological and prognostic data of the patients and the corresponding tissue microarrays were provided by Shanghai CoreBio System Co., Ltd. Among these patients, 255 (86%) were male, 246 (83%) were HBsAg serum positive, and 237 (80%) had solitary tumors. A total of 296 patients were divided into a training set and a validation set according to different chip batches. Two of the four tissue arrays (chip numbers: HLivH180Su08 and HLivH180Su15) were divided into the training set (n = 157), and the other two (chip numbers: T16-855TMAB and TFHCC-02) were used as the validation set (n = 139). The detailed clinical characteristics are shown in Table 1.

[0032] Table 1. Statistical table of clinical characteristics of immunohistochemistry training set, validation set, and all populations

[0033] Clinical characteristics Training set (n = 157) Validation set (n = 139) Total (n = 296) Gender - no. (%) Female 22(14) 19(14) 41(14) Male 135(86) 120(86) 255(86) Age, years - no. (%) ≤ 60 124 (79) 98 (71) 222 (75) > 60 33 (21) 41 (29) 74 (25) HBsAg - no. (%) Negative 29(18) 21(15) 50(17) Positive 128(82) 118(85) 246(83) AFP, ng / mL - no. (%) ≤ 400 98 (62) 91 (65) 189 (64) > 400 59 (38) 48 (35) 107 (36) Liver cirrhosis - no. (%) None 17(11) 9(10) 26(11) Yes 140(89) 78(90) 218(89) NA 52 52 Number of tumors - no. (%) 1 144(92) 93(87) 237(90) ≥ 2 13 (8) 14 (13) 27 (10) NA 0 32 32 Tumor size, cm - no. (%) ≤ 5 98 (62) 89 (64) 187 (63) > 5 59 (38) 50 (36) 109 (37) Microvascular cancer thrombus - no. (%) None 56(73) 94(74) 150(74) Yes 21(27) 33(26) 54(26) NA 80 12 92 TNM stage - no. (%) I 100(64) 99(71) 199(67) II - IV 57(36) 40(29) 97(33) Edmondson - Steiner grade - no. (%) I - II 84(54) 97(71) 181(62) III - IV 73(46) 40(29) 113(38) NA 0 2 2

[0034] In Table 1, NA represents Not Available, no data.

[0035] II. Immunohistochemistry (IHC) experiment

[0036] The immunohistochemical antibodies and kits used are as follows:

[0037] The BCAT1 antibody is a product of Thermo Fisher, and the catalog number is MA5-25892;

[0038] The FKBP10 antibody is a product of Protein Tech, and the catalog number is 12172-1-AP;

[0039] The SOAT1 antibody is a product of Thermo Fisher, with the catalog number PA5-61455;

[0040] The mouse two-step kit is a product of Beijing Zhongshan Golden Bridge Biotechnology Co., Ltd., with the catalog number PV-6002;

[0041] The rabbit two-step kit is a product of Beijing Zhongshan Golden Bridge Biotechnology Co., Ltd., with the catalog number PV-6001.

[0042] The steps of immunohistochemical staining are as follows:

[0043] 1) Baking the slides: Put the tissue microarray into the oven, adjust the temperature to 63 degrees, and bake the wax for 1 hour.

[0044] 2) Deparaffinization: After the slides are baked, take them out of the oven for deparaffinization. Immerse the paraffin sections in the following program: Xylene (1) for 10 minutes → Xylene (2) for 10 minutes.

[0045] 3) Hydration: 100% ethanol (1) for 5 minutes; 100% ethanol (2) for 5 minutes; 95% ethanol for 5 minutes; 90% ethanol for 5 minutes; 85% ethanol for 5 minutes; 80% ethanol for 5 minutes; 70% ethanol for 5 minutes.

[0046] 4) Take out the slides, wash them with tap water for 2 minutes, and soak them in distilled water for 2 minutes.

[0047] 5) Eliminate endogenous peroxidase: Immerse the paraffin sections in 3% hydrogen peroxide and react in the dark for 10 minutes.

[0048] 6) Wash with PBS three times, 5 minutes each time.

[0049] 7) Antigen retrieval: The immunohistochemical experiments for the three antibodies are repaired as follows: BCAT1 is repaired with citrate buffer (pH 6.0) and microwave heating; FKBP10 is repaired with citrate buffer (pH 6.0) and heating with an antigen retrieval instrument; SOAT1 is repaired with TE (Tris-EDTA) buffer (pH 9.0) and microwave heating. Each numbered tissue microarray has 3 replicates, and 1 is used for the repair of 1 antigen.

[0050] 8) Wash with PBS three times, 5 minutes each time.

[0051] 9) BSA blocking: Drop 5% PBST containing BSA on the sections, place them in a wet box and incubate at 37 degrees for 20 minutes.

[0052] 10) Shake off excess serum from the slides and directly add primary antibodies (BCAT, 1:2000 dilution; FKBP10, 1:1500 dilution; SOAT1, 1:500 dilution). Simultaneously, perform a blank control experiment, replacing the primary antibody with PBS and incubate at 37°C for 1 hour. Then, transfer to a refrigerator at 4°C overnight.

[0053] 11) Remove the wet chamber from the refrigerator and rewarm at room temperature for 20-30 minutes. Wash three times with PBS for 3 minutes each. Add the ready-to-use working solution of the mouse secondary antibody (in the mouse two-step kit) to the slides and incubate at 37°C for 30 minutes.

[0054] 12) Wash again with PBS three times, 5 minutes each time.

[0055] 13) DAB color development: Remove the DAB kit from the refrigerator and dilute it to 1x DAB color development solution according to the instructions. Add the diluted DAB solution to the slide and observe the color intensity under a microscope. Stop the reaction when a positive result is satisfactory and the background is appropriate. After the display is complete, place the slide in PBS buffer to stop the reaction.

[0056] 14) Hematoxylin counterstaining and mounting: Add SIGMA Hematoxylin to the slide for 1 minute, then rinse thoroughly with water. Immerse in 0.25% hydrochloric acid alcohol for at least 2 seconds, and rinse with tap water for at least 10 minutes until the slide turns blue.

[0057] 15) Gradient ethanol dehydration:

[0058] 80% ethanol aqueous solution → 90% ethanol aqueous solution → 100% ethanol (1) → 100% ethanol (2), 2 minutes each time.

[0059] 16) Xylene transparency: Xylene (1) → Xylene (2), 5 minutes each time.

[0060] 17) Seal the slide with neutral resin.

[0061] 3. Immunohistochemistry Scoring

[0062] After the immunohistochemistry experiment, the tissue chips were scored independently by two pathologists and one researcher.

[0063] Immunohistochemical scoring criteria: percentage of positive cells (0 points for no positive cells, 1 point for 1% - 24% positive cells, 2 points for 25% - 49%, 3 points for 50% - 74%, 4 points for 75% - 100%); staining intensity (0 points for no staining, 1 point for weak staining, 2 points for moderate staining, 3 points for strong staining). The product of the percentage of positive cells and the staining intensity score is the immunohistochemical score, with the lowest score being 0 and the highest being 12. Among them, the classification method of staining intensity is as follows: all stained tissues are divided into four categories according to staining intensity. Tissues with no cell staining are non-stained, those with the strongest staining are strongly stained, weak staining refers to the category with staining but the lowest staining intensity, and moderate staining refers to the category with staining intensity between weak and strong staining. At least 5 non-repeated fields are observed for each tissue.

[0064] Representative pictures of high and low expression of three proteins, BCAT1, FKBP10, and SOAT1, in immunohistochemical detection of liver cancer tissues are shown as Figure 1 follows.

[0065] IV. Selection of cut-off values for individual proteins and construction of a three-protein classifier

[0066] For each of the proteins BCAT1, FKBP10, and SOAT1, the immunohistochemical score cut-off value with the smallest P value in the training set for the Log-rank test is regarded as its optimal cut-off value, which are 8 points, 6 points, and 10 points respectively. The scoring criteria for high expression of each protein are shown in Table 2. The immunohistochemical scoring criteria for defining a patient as having a high recurrence risk by the three-protein classifier are "BCAT1 ≥ 8 or FKBP10 ≥ 10 or SOAT1 ≥ 6". The recurrence Kaplan-Meier curves of the three proteins plotted according to the optimal cut-off values in the immunohistochemical training set are shown as Figure 2 follows. Then it is tested in the validation set, and the results show that the three proteins can well evaluate the recurrence risk of HCC in both the training set and the validation set.

[0067] Table 2. Scoring criteria for high expression of BCAT1, FKBP10, and SOAT1

[0068]

[0069] Univariate and multivariate Cox regression analyses were performed on the entire population (training set 157 + validation set 139 = 296) respectively (Tables 3 - 4), and the results show that the combination of the three-protein classifier can be used as an independent negative prognostic factor for 2-year recurrence and 5-year recurrence after radical resection of HCC patients.

[0070] Among them, the Kaplan-Meier curve of 157 patients in the training set predicting recurrence according to the combination of the three-protein classifier is shown as Figure 3shown in A; The ROC curves for predicting early (2-year) recurrence by the three proteins alone and in combination in the training set patients are as Figure 3 shown in B. The AUC for combined prediction is 0.598, the sensitivity is 0.353, and the specificity is 0.843. The AUC for BCAT1 alone is 0.557, the sensitivity is 0.147, and the specificity is 0.966. The AUC for FKBP10 alone is 0.545, the sensitivity is 0.191, and the specificity is 0.899. The AUC for SOAT1 alone is 0.527, the sensitivity is 0.132, and the specificity is 0.921. The ROC curves for predicting late (5-year) recurrence by the three proteins alone and in combination in the training set patients are as Figure 3 shown in C. The AUC for combined prediction is 0.623, the sensitivity is 0.348, and the specificity is 0.897. The AUC for BCAT1 alone is 0.560, the sensitivity is 0.135, and the specificity is 0.985. The AUC for FKBP10 alone is 0.559, the sensitivity is 0.191, and the specificity is 0.926. The AUC for SOAT1 alone is 0.551, the sensitivity is 0.146, and the specificity is 0.956.

[0071] The Kaplan-Meier curves for predicting recurrence by the three-protein classifier combination in 139 patients in the validation set are as Figure 4 shown in A; The ROC curves for predicting early (2-year) recurrence by the three proteins alone and in combination in the validation set patients are as Figure 4 shown in B. The AUC for combined prediction is 0.634, the sensitivity is 0.466, and the specificity is 0.802. The AUC for BCAT1 alone is 0.585, the sensitivity is 0.293, and the specificity is 0.877. The AUC for FKBP10 alone is 0.512, the sensitivity is 0.086, and the specificity is 0.938. The AUC for SOAT1 alone is 0.603, the sensitivity is 0.293, and the specificity is 0.914. The ROC curves for predicting late (5-year) recurrence by the three proteins alone and in combination in the validation set patients are as Figure 4 shown in C. The AUC for combined prediction is 0.563, the sensitivity is 0.343, and the specificity is 0.784. The AUC for BCAT1 alone is 0.522, the sensitivity is 0.206, and the specificity is 0.838. The AUC for FKBP10 alone is 0.512, the sensitivity is 0.078, and the specificity is 0.946. The AUC for SOAT1 alone is 0.581, the sensitivity is 0.216, and the specificity is 0.946.

[0072] According to the clinical traditional features such as tumor size, number, and microvascular invasion (MVI), HCC patients were stratified into high, medium, and low recurrence risk groups. The stratification definition is as follows: Patients with a single tumor less than or equal to 5 cm and without microvascular invasion are regarded as clinical low recurrence risk (lowrisk) patients; patients with a single tumor larger than 5 cm and without microvascular invasion are regarded as clinical intermediate recurrence risk (Intermediaterisk) patients; patients with a single tumor with microvascular invasion or multiple tumors are regarded as clinical high recurrence risk (Highrisk) patients. In the training set, there were 70 patients with low recurrence risk, 30 patients with intermediate recurrence risk, and 57 patients with high recurrence risk. The 2-year recurrence rates of the three groups of patients were 31.4%, 46.7%, and 56.1% respectively, and the 5-year recurrence rates were 45.0%, 56.9%, and 72.0% respectively ( Figure 5 as shown in Figure A), and the recurrence rate of the low recurrence risk group (31.4% at 2 years, 45.0% at 5 years) was much lower than that of the medium and high recurrence risk groups. However, even among these patients with the clinical characteristics of a single tumor less than or equal to 5 cm and without microvascular invasion, who were already considered to have a low recurrence risk, patients with poor prognosis could still be further screened out based on the high or low scores of the three-protein classifier. The 2-year recurrence rate of the high-expression group of the three-protein classifier (n = 16) reached 56.3%, and the 5-year recurrence rate reached 75%. The recurrence probability of this part of the patients was comparable to that of the clinical high recurrence risk group of patients ( Figure 5 as shown in Figure B). In the validation set, there were 69 patients with low recurrence risk, 25 patients with intermediate recurrence risk, and 45 patients with high recurrence risk. The 2-year recurrence rates of the three groups of patients were 27.5%, 36.0%, and 66.7% respectively, and the 5-year recurrence rates were 63.8%, 76.0%, and 86.7% respectively ( Figure 6 as shown in Figure A). Similarly, among the 69 clinically low recurrence risk patients, they were further divided into high and low expression groups according to the three-protein classifier. The 2-year recurrence rate of the 15 patients in the high-expression group reached 60.0%, and the 5-year recurrence rate reached 93.3%. The recurrence probability of this part of the patients was also comparable to that of the clinical high recurrence risk group of patients ( Figure 6 as shown in Figure B).

[0073] Further verification showed that the three-protein classifier combination could still be a significant independent negative prognostic factor in the clinical low recurrence risk group. Univariate and multivariate Cox regression analyses were performed on all low recurrence risk populations (training set 70 + validation set 69 = 139) respectively (Table 5-6). The results showed that the three-protein classifier combination could be an independent negative prognostic factor for 2-year recurrence and 5-year recurrence.

[0074] Among them, the Kaplan-Meier curves of predicting recurrence by the three proteins alone and in combination for the 70 clinically low recurrence risk patients in the training set are as Figure 5as shown in B; the ROC curves for predicting early (2-year) recurrence by the three proteins alone and in combination are as Figure 5 shown in C. The AUC for combined prediction is 0.632, the sensitivity is 0.409, and the specificity is 0.854. The AUC for BCAT1 alone is 0.603, the sensitivity is 0.227, and the specificity is 0.979. The AUC for FKBP10 alone is 0.572, the sensitivity is 0.227, and the specificity is 0.917. The AUC for SOAT1 alone is 0.549, the sensitivity is 0.182, and the specificity is 0.917. The ROC curves for predicting late (5-year) recurrence by the three proteins alone and in combination for 70 clinically low recurrence risk patients in the training set are as Figure 5 shown in D. The AUC for combined prediction is 0.642, the sensitivity is 0.387, and the specificity is 0.897. The AUC for BCAT1 alone is 568, the sensitivity is 0.161, and the specificity is 0.974. The AUC for FKBP10 alone is 0.587, the sensitivity is 0.226, and the specificity is 0.949. The AUC for SOAT1 alone is 0.571, the sensitivity is 0.194, and the specificity is 0.959.

[0075] The Kaplan-Meier curves for predicting recurrence by the three proteins alone and in combination for 69 clinically low recurrence risk patients in the validation set are as Figure 6 shown in B; the ROC curves for predicting early (2-year) recurrence by the three proteins alone and in combination are as Figure 6 shown in C. The AUC for combined prediction is 0.677, the sensitivity is 0.474, and the specificity is 0.880. The AUC for BCAT1 alone is 0.628, the sensitivity is 0.316, and the specificity is 0.940. The AUC for FKBP10 alone is 0.526, the sensitivity is 0.053, and the specificity is 1.000. The AUC for SOAT1 alone is 0.592, the sensitivity is 0.263, and the specificity is 0.920. The ROC curves for predicting late (5-year) recurrence by the three proteins alone and in combination for 69 clinically low recurrence updated patients in the training set are as Figure 6 shown in D. The AUC for combined prediction is 0.639, the sensitivity is 0.318, and the specificity is 0.960. The AUC for BCAT1 alone is 0.571, the sensitivity is 0.182, and the specificity is 0.960. The AUC for FKBP10 alone is 0.511, the sensitivity is 0.023, and the specificity is 1.000. The AUC for SOAT1 alone is 0.602, the sensitivity is 0.205, and the specificity is 1.000.

[0076] The method for predicting 2-year recurrence using three proteins is as follows: Immunohistochemical detection of three proteins is performed on tumor FFPE samples of HCC patients (especially those with a low clinical recurrence risk, single tumor less than or equal to 5 cm in size and without microvascular invasion). Scoring is carried out according to the aforementioned scoring rules. If a patient meets the high-expression criteria of the three-protein classifier: conforming to at least one of "BCAT1≥8, FKBP10≥10, SOAT1≥6", it indicates that the patient has a very high early (2-year) and late (5-year) recurrence risk, and it is recommended to perform active postoperative adjuvant treatment; if "BCAT1<8 and FKBP10<10 and SOAT1<6", it indicates that the patient has a low early (2-year) and late (5-year) recurrence risk.

[0077] Table 3. Univariate Cox regression analysis of the entire population based on the combination of three-protein classifier and clinicopathological features

[0078]

[0079]

[0080] Among them, high indicates high expression, meaning that the immunohistochemical scores of the three proteins meet at least one of "BCAT1≥8, FKBP10≥10, SOAT1≥6"; low indicates low expression, meaning that each protein in the three-protein classifier is lowly expressed, that is, "BCAT1<8 and FKBP10<10 and SOAT1<6".

[0081] Table 4. Multivariate Cox regression analysis of the entire population based on the combination of three-protein classifier and clinicopathological features

[0082]

[0083]

[0084] In Table 4, NA represents (Not Available), indicating not included in the multivariate analysis; INF is the abbreviation of "Infinite", indicating infinity. High indicates high expression, meaning that the immunohistochemical scores of the three proteins meet at least one of "BCAT1≥8, FKBP10≥10, SOAT1≥6"; low indicates low expression, meaning that each protein in the three-protein classifier is lowly expressed, that is, "BCAT1<8 and FKBP10<10 and SOAT1<6".

[0085] Table 5. Univariate Cox regression analysis of the low recurrence risk population based on the combination of three-protein classifier and clinicopathological features

[0086]

[0087] Among them, "high" indicates high expression, meaning that the immunohistochemical scores of the three proteins satisfy at least one of "BCAT1≥8, FKBP10≥10, SOAT1≥6"; "low" indicates low expression, meaning that each protein in the three-protein classifier is lowly expressed, that is, "BCAT1<8 and FKBP10<10 and SOAT1<6".

[0088] Table 6. Multivariate Cox regression analysis of the low recurrence risk population based on the combination of 3-protein classifier and clinicopathological features

[0089]

[0090]

[0091] In Table 6, NA (Not Available) indicates not included in the multivariate analysis; "high" indicates high expression, meaning that the immunohistochemical scores of the three proteins satisfy at least one of "BCAT1≥8, FKBP10≥10, SOAT1≥6"; "low" indicates low expression, meaning that each protein in the three-protein classifier is lowly expressed, that is, "BCAT1<8 and FKBP10<10 and SOAT1<6".

[0092] The above has described the present invention in detail. For those skilled in the art, without departing from the purpose and scope of the present invention and without unnecessary experiments, the present invention can be implemented within a relatively wide range under equivalent parameters, concentrations and conditions. Although specific embodiments of the present invention are given, it should be understood that the present invention can be further improved. In short, according to the principle of the present invention, this application intends to cover any changes, uses or improvements to the present invention, including changes made using conventional techniques known in the art that are outside the scope disclosed in this application.

Claims

1. A device for predicting the recurrence risk of hepatocellular carcinoma patients after radical resection, comprising substances for detecting the protein contents of BCAT1, FKBP10 and SOAT1.

2. The device according to claim 1, characterized in that: The substances for detecting the protein contents of BCAT1, FKBP10 and SOAT1 are composed of a substance for detecting the protein content of BCAT1, a substance for detecting the protein content of FKBP10 and a substance for detecting the protein content of SOAT1.

3. The device according to claim 2, wherein: The substance for detecting the protein content of BCAT1 includes reagents and / or instruments for detecting the protein content of BCAT1 by immunohistochemical staining; the substance for detecting the protein content of FKBP10 includes reagents and / or instruments for detecting the protein content of FKBP10 by immunohistochemical staining; the substance for detecting the protein content of SOAT1 includes reagents and / or instruments for detecting the protein content of SOAT1 by immunohistochemical staining.

4. The application according to claim 2 or 3, characterized in that: The substance for detecting the content of BCAT1 includes BCAT1 antibody; the substance for detecting the content of FKBP10 includes FKBP10 antibody; the substance for detecting the content of SOAT1 includes SOAT1 antibody.

5. The device according to any one of claims 1-4, characterized in that: The device further includes a protein content data processing module, which is used to convert the protein contents of BCAT1, FKBP10 and SOAT1 in the isolated hepatocellular carcinoma tissue from the hepatocellular carcinoma patient to be predicted into a prediction score of the hepatocellular carcinoma patient to be predicted, and predict the recurrence risk of the hepatocellular carcinoma patient after radical resection according to the prediction score of the hepatocellular carcinoma patient to be predicted.

6. The device according to any one of claims 1-5, characterized in that: The hepatocellular carcinoma patient is a single-tumor hepatocellular carcinoma patient with a single tumor less than or equal to 5 cm and no microvascular invasion.

7. The device according to any one of claims 1-6, characterized in that: The recurrence risk after radical resection is the recurrence risk within 2 years or 5 years after radical resection.

8. Use of the device according to any one of claims 1-5 in the preparation of a product for predicting the recurrence risk of hepatocellular carcinoma patients after radical resection.

9. The application according to claim 8, characterized in that: The hepatocellular carcinoma patient is a single-tumor hepatocellular carcinoma patient with a single tumor less than or equal to 5 cm and no microvascular invasion.

10. The application according to claim 8 or 9, characterized in that: The recurrence risk after radical resection is the recurrence risk within 2 years or 5 years after radical resection.

Images