Application of constructing a liver cancer discrimination tool based on a specific lectin combination and alpha-fetoprotein

Abstract

The present invention belongs to the field of biomedical detection, and relates to the application of a specific lectin combination that binds to IgG and / or IgM in serum and alpha-fetoprotein (AFP) in constructing a test tool for differentiating liver cancer. By using lectin chip technology, the present invention screened out that there are differences in the lectins that bind to IgG (12 kinds) / bind to IgM (11 kinds) between liver cancer patients and healthy people, hepatitis B patients and hepatitis B cirrhosis patients. Based on the specific lectin combination that binds to IgG / IgM in serum and AFP, a test tool for differentiating liver cancer is jointly constructed, and a combined diagnostic model for differentiating liver cancer is established by combining algorithms. The test tool of the present invention can differentiate and diagnose liver cancer, distinguish liver cancer from hepatitis B cirrhosis and chronic hepatitis B diseases. At the same time, for patients with negative AFP, it can be used as a powerful supplement for in vitro diagnosis.

Description

Technical Field

[0001] The present invention belongs to the field of biomedical detection, and relates to the application of constructing a liver cancer discrimination tool based on a specific lectin combination and alpha-fetoprotein, and particularly relates to the application of a test tool for discriminating liver cancer by a specific lectin combination that binds to IgG and / or IgM in serum and alpha-fetoprotein (AFP). Background Art

[0002] Primary liver cancer is one of the common malignant tumors clinically. According to cell typing, it can be divided into hepatocellular carcinoma (HCC), cholangiocarcinoma, and mixed liver cancer. Among them, hepatocellular carcinoma is the main type. Currently, it is considered that the onset of liver cancer is a complex process involving multiple factors and multiple steps, and is affected by both environmental and genetic factors. Among these risk factors, hepatitis B virus (HBV) and hepatitis C virus (HCV) are the main factors for the occurrence of liver cancer.

[0003] Liver cancer is characterized by difficult diagnosis, difficult treatment, easy metastasis and recurrence, and high mortality. Using early diagnosis and screening programs to detect and treat liver cancer cases early can reduce the mortality of liver cancer. So far, alpha-fetoprotein (AFP) is still an important tumor marker for diagnosing primary liver cancer clinically, but due to its poor sensitivity and specificity, its diagnostic value is limited. A randomized controlled study conducted in high-risk populations in China showed that early diagnosis of liver cancer using AFP did not reduce the mortality of liver cancer. Therefore, early diagnosis of liver cancer still requires improved non-invasive biomarkers. Summary of the Invention

[0004] Through a large number of experiments and analyses, the present application detected the binding of IgG and IgM in the sera of liver cancer patients and healthy people to 56 lectins by lectin chip, found differential lectins, established a combined diagnostic model of lectin and AFP detection value using a training group (total 138 cases, including 65 cases of HCC, 20 cases of CHB, 33 cases of HBC, and 20 healthy controls), and then used a validation group (total 69 cases, including 32 cases of HCC, 10 cases of CHB, 17 cases of HBC, and 10 healthy controls) to verify the diagnostic model to determine whether the subject is a liver cancer patient. The combined diagnostic model was evaluated to have good diagnostic ability through ROC curve analysis.

[0005] The object of the present invention is to provide a lectin combination and its combined application with AFP for predicting, diagnosing, and judging the disease progression of hepatocellular carcinoma. Further, mainly by detecting the binding of IgG or IgM in the blood of hepatocellular carcinoma patients to lectins and combining with the AFP level in the patient's blood to predict and monitor hepatocellular carcinoma.

[0006] The object of the present invention is to provide the application of a lectin combination chip for predicting, diagnosing, differentiating, and judging disease progression of hepatocellular carcinoma, Hepatitis B cirrhosis (HBC), and chronic hepatitis B (CHB). Further, it mainly predicts and monitors hepatocellular carcinoma by detecting the binding of IgG or IgM in the blood of hepatocellular carcinoma patients to lectins.

[0007] The technical solution adopted by the present invention is as follows:

[0008] In the first aspect.

[0009] Application of a specific lectin combination that binds to IgG in serum and alpha-fetoprotein (AFP) to construct a test tool for differentiating liver cancer, wherein the specific lectin combination is selected from four lectins: SBA, EEL, MPL, and TL; compared with healthy human samples, the signals of SBA and EEL in liver cancer samples are significantly increased, while the signals of MPL and TL are significantly decreased.

[0010] Application of a specific lectin combination that binds to IgM in serum and alpha-fetoprotein (AFP) to construct a test tool for differentiating liver cancer, wherein the specific lectin combination is selected from two lectins: DSL and LCA; compared with healthy human samples, the signals of DSL and LCA in liver cancer samples are significantly increased.

[0011] Application of a specific lectin combination that binds to IgG in serum to construct a test tool for differentiating liver cancer, wherein the specific lectin combination is selected from eight lectins: PHA-L, SBA, TL, IAA, PHA-E, AIA, VVA, and Black bean crude; compared with Hepatitis B cirrhosis (HBC) samples, the signal of PHA-L in cancer samples is significantly increased; compared with chronic hepatitis B (CHB) samples, the signal of SBA in cancer samples is significantly increased, while the signals of TL, IAA, PHA-E, AIA, VVA, and Black bean crude are significantly decreased.

[0012] Application of a specific lectin combination that binds to IgM in serum to construct a test tool for differentiating liver cancer, wherein the specific lectin combination is selected from nine lectins: SNA-I, MNA-M, PSA, MPL, PHA-L, VVA, CSA, DBA, and SSA; compared with chronic hepatitis B (CHB) samples, the signals of SNA-I, MNA-M, PSA, MPL, PHA-L, VVA, CSA, DBA, and SSA in liver cancer samples are significantly decreased.

[0013] In the second aspect.

[0014] The test tool is a lectin chip, a kit, or a lectin detection intelligent terminal.

[0015] As a further invention, the intelligent terminal for identifying liver cancer based on IgG in serum includes a processor and a program memory, and is characterized in that: the program stored in the program memory, when loaded by the processor, executes the following steps:

[0016] 1) Obtain the lectin test results of the IgG sample in the serum, and the lectin test results reflect the signal expression levels of the binding of lectins SBA, EEL, MPL, and TL to IgG;

[0017] 2) Obtain the AFP test results in the blood;

[0018] 3) Calculate the detection value of the combined diagnosis model of IgG and AFP according to the differential lectins binding to IgG and the AFP test results;

[0019] Differential lectins binding to IgG and AFP combined diagnosis model = -2.277 + 0.510 * AFP + 0.004 * EEL - 0.008 * MPL - 0.015 * TL;

[0020] If the detection value ≥ 2.07, it indicates that the subject to which the serum sample belongs is a liver cancer patient.

[0021] As a further invention, the intelligent terminal for identifying liver cancer based on IgM in serum includes a processor and a program memory, and is characterized in that: the program stored in the program memory, when loaded by the processor, executes the following steps:

[0022] 1) Obtain the lectin test results of the IgM sample in the serum, and the lectin test results reflect the signal expression levels of the binding of lectins DSL and LCA to IgM;

[0023] 2) Obtain the AFP test results in the blood;

[0024] 3) Calculate and construct the detection value of the combined diagnosis model of differential lectins and AFP according to the differential lectins binding to IgM and the AFP test results. Differential lectins binding to IgM and AFP combined diagnosis model = -1.901 + 0.003 * DSL + 0.286 * AFP

[0025] If the detection value ≥ 2.07, it indicates that the subject to which the serum sample belongs is a liver cancer patient.

[0026] In the third aspect.

[0027] A computer-readable storage medium stores a computer program, and the computer program, when loaded by a processor, executes each of the steps listed above.

[0028] In the fourth aspect.

[0029] A human-computer interaction device, including a display screen. When the human-computer interaction device runs, the display screen sequentially displays the following interfaces:

[0030] An input interface for the test results of the binding of serum samples to lectins, where the test results of the binding of serum samples to lectins reflect the binding expression levels of SBA, EEL, MPL, TL corresponding to the lectin with IgG or the binding expression levels of DSL and LCA with IgM;

[0031] An input interface for the AFP detection results in blood;

[0032] An output interface for the discrimination information of the combined diagnosis model; the discrimination information of the combined diagnosis model includes the reference value 2.07 of the combined diagnosis model and the detection value and / or discrimination conclusion; the combined diagnosis model is:

[0033] The combined diagnosis model of differential lectin binding to IgG and AFP = -2.277 + 0.510 * AFP + 0.004 * EEL - 0.008 * MPL - 0.015 * TL or the combined diagnosis model of differential lectin binding to IgM and AFP = -1.901 + 0.003 * DSL + 0.286 * AFP;

[0034] If the detection value ≥ 2.07, it indicates that the subject of the sample is a liver cancer patient; conversely, if the detection value < 2.07, it is a healthy person.

[0035] The fifth aspect.

[0036] A system for differentiating liver cancer based on IgG and / or IgM in serum, including:

[0037] A. A device for obtaining the expression levels of lectins with different bindings to serum IgG and / or IgM, where the lectins correspond to the specific lectin combinations described above;

[0038] B. A device for obtaining the AFP detection value in blood;

[0039] C. An intelligent terminal for differentiating liver cancer based on IgG and / or IgM in serum as described above;

[0040] Among them, the device for obtaining the expression levels of lectins with different bindings to serum IgG and / or IgM includes a lectin chip, an incubation box, and a biochip scanning system, and the specific lectin combinations are set on the lectin chip.

[0041] Beneficial effects.

[0042] When differentiating liver cancer patients from healthy individuals, the diagnostic sensitivity of the combined diagnostic model in the IgG modeling group and the combined diagnostic model in the IgM modeling group is significantly higher than that of AFP alone. The present invention provides a feasible diagnostic mode for diagnosing hepatocellular carcinoma, especially for patients with negative AFP, which can serve as a powerful supplement for in vitro diagnosis.

[0043] Through the significant differences in the signal values detected by the combination of lectins and blood IgG in the application of the present invention, liver cancer can be differentiated and diagnosed, distinguishing liver cancer from hepatitis B cirrhosis and chronic hepatitis B diseases, and providing a feasible diagnostic mode. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] Figure 1 It is a design diagram of the spotting matrix for the lectin chip.

[0045] Figure 2 It is the differential lectins of IgG binding between the HCC group and other groups, and the signal conditions of the differential lectins of IgG binding between hepatocellular carcinoma (HCC) and the sera of patients with hepatitis B cirrhosis (HBC) and chronic hepatitis B (CHB) respectively.

[0046] Figure 3 It is the differential lectins of IgM binding between the HCC group and other groups, and the signal conditions of the differential lectins of IgM binding between hepatocellular carcinoma (HCC) and the sera of patients with hepatitis B cirrhosis (HBC) and chronic hepatitis B (CHB) respectively.

[0047] Figure 4 It is the differential lectins of IgG binding between the HCC group and the healthy group, and the signal conditions of the differential lectins of IgG binding between hepatocellular carcinoma (HCC) and the sera of healthy controls (NC).

[0048] Figure 5 It is the receiver operating characteristic curve (ROC) of the diagnostic model combining IgG-binding lectins and AFP. A: The receiver operating characteristic curve of the diagnostic training group combining IgG-binding lectins and AFP; B: The receiver operating characteristic curve of the diagnostic validation group combining IgG-binding lectins and AFP.

[0049] Figure 6 It is the differential lectins of IgM binding between the HCC group and the healthy group; the signal conditions of the differential lectins of IgM binding between hepatocellular carcinoma (HCC) and the sera of healthy controls (NC).

[0050] Figure 7 It is the receiver operating characteristic curve (ROC) of the diagnostic model combining IgM-binding lectins and AFP. A: The receiver operating characteristic curve of the diagnostic training group combining IgM-binding lectins and AFP; B: The receiver operating characteristic curve of the diagnostic validation group combining IgM-binding lectins and AFP. DETAILED DESCRIPTION OF THE INVENTION

[0051] The preferred embodiments of the present invention are described in detail below. According to conventional conditions, the examples are only for better illustrating the content of the present invention. However, it should not be understood that the content of the present invention is limited to the examples. Any non-essential improvements and adjustments to the implementation scheme based on the above content of the present invention still belong to the protection scope of the present invention.

[0052] The experimental methods described in the following experimental examples are conventional methods unless otherwise specified. The reagents, materials, instruments, etc. used in the following experimental examples can be obtained from commercial channels unless otherwise specified.

[0053] Example 1.

[0054] 1. Research methods

[0055] 1.1. Grouping of training group and validation group.

[0056] A total of 207 sera collected at Shengjing Hospital Affiliated to China Medical University between 2019 and 2020 were collected, including 97 cases of HCC, 50 cases of HBC, 30 cases of CHB, and 30 cases of healthy controls.

[0057] The 207 samples were randomly divided into stratified groups. When formulating the random grouping scheme, the subjects were first grouped by disease; then grouped by gender in each disease; and then grouped by age in men and women, resulting in 16 subgroups. In these 16 subgroups, the subjects were randomly divided into a training group and a validation group at a ratio of 2:1 (random seed number 20201115). The total 207 samples were randomly divided into a training group (a total of 138 cases, including 65 HCC, 20 CHB, 33 HBC, and 20 healthy controls) and a validation group (a total of 69 cases, including 32 HCC, 10 CHB, 17 HBC, and 10 healthy controls), and the gender and age were evenly distributed in the two groups as much as possible.

[0058] 1.2. Experimental steps of lectin chip detection.

[0059] Lectin chip spotting matrix design Figure 1 shown.

[0060] (1) Rewarming: Take the LectinTM lectin chip out of the -80 ℃ refrigerator, place it in a 4 ℃ refrigerator to rewarm for 30 min, and then place it at room temperature to rewarm for 15 min;

[0061] (2) Blocking: After the chip is rewarmed, fix the 14 blocks fence. After fixation, add blocking solution to each block and place it on a side-swinging shaker and block it at room temperature for 3 hours.

[0062] (3)Serum sample incubation: After blocking is completed, pour out the blocking solution, and then quickly add the pre-prepared serum incubation solution. Each chip can incubate 14 serum samples, and the loading volume of each serum sample is 200 µL. Incubate overnight at 4 °C with a side-sway shaker at 20 rpm (the samples are first thawed in a chromatography cabinet at 4 °C and diluted at a ratio of 1:500);

[0063] (4)Washing: After removing the chip fence, place the chip in a chip washing box containing washing solution, and wash it 3 times for 5 minutes each with a horizontal shaker at 80 rpm at room temperature;

[0064] (5)Mouse serum blocking: Transfer the chip to an incubation box containing 3 mL of mouse incubation solution, and incubate it with a side-sway shaker at 40 rpm at room temperature for 1 hour;

[0065] (6)Washing: Take out the chip (note that the upper surface of the chip cannot be touched or scratched), place it in a chip washing box containing washing solution, and wash it 3 times for 10 minutes each with a horizontal shaker at 80 rpm at room temperature. After completion, wash it 2 times with ddH2O for 5 minutes each;

[0066] (7)Secondary antibody incubation: Transfer the chip to an incubation box containing 3 mL of Cy5-Anti-human IgM and Cy3-Anti-human IgG incubation solution, and incubate it with a side-sway shaker at 40 rpm in the dark at room temperature for 1 hour;

[0067] (8)Washing: Take out the chip (note that the upper surface of the chip cannot be touched or scratched), place it in a chip washing box containing washing solution, and wash it 3 times for 10 minutes each with a horizontal shaker at 80 rpm at room temperature. After completion, wash it 2 times with ddH2O for 10 minutes each;

[0068] (9)Drying;

[0069] (10)Scanning: Operate according to the operating specifications and instructions of the scanner; read the signals through a fluorescence scanner. The intensity of the signals is positively correlated with the affinity and quantity of the antibodies. IgG corresponds to the 532 nm green light channel and IgM corresponds to the 635 nm red light channel.

[0070] 1.3 Lectin chip data analysis.

[0071] The original data was read using GenePix Pro v6.0 software. For the extracted data, to avoid interference from abnormal backgrounds, based on the mean and standard deviation of the background values of all samples, a cutoff of mean + 3sd (99% CI) was set, and samples with abnormal background values were excluded. Through calculation, the threshold for the 532 nm channel was 4727, and the threshold for the 635 nm channel was 466. To eliminate the uneven signal caused by inconsistent background values between different points on the same chip, background correction was performed. The implementation method was the difference between the foreground value and the background value of each point, i.e., F - B, and Signal was defined based on this, which was the average of the F - B values of 3 replicated lectins (data with Signal < 1 was corrected to 1). Due to the systematic errors caused by experimental samples and experimental operations, direct data comparison between chips might lead to uncertain results. Therefore, before data comparison, the Signal values of the samples after exclusion were normalized between chips. The implementation method was fixed - point normalization, i.e., normalization of the positive control point Cy3 / Cy5.

[0072] 1.4 Experimental procedures for AFP detection.

[0073] The serum alpha - fetoprotein (AFP) level was detected using the electrochemiluminescence immunoassay double - antibody sandwich method and detected using the Cobas E 602 fully automated immunoanalyzer from Roche, Germany. The specific operation process was carried out strictly in accordance with the instrument and reagent instructions.

[0074] Step I: 10 μl of the specimen, biotinylated anti - AFP monoclonal antibody, and ruthenium (Ru) - labeled anti - AFP monoclonal antibody were mixed to form a sandwich complex.

[0075] Step 2: Streptavidin - coated microparticles were added, and the complex formed above was bound to the microparticles through the reaction between biotin and streptavidin.

[0076] Step 3: The reaction mixture was aspirated into the measurement cell. The microparticles were adsorbed onto the electrode by a magnet, unbound substances were washed away by the washing solution, and chemiluminescence was generated after applying a voltage to the electrode and measured by a photomultiplier tube. The test results were automatically retrieved from the standard curve by the machine. This curve was obtained by the instrument through two - point calibration and was the original standard curve scanned from the reagent barcode into the instrument.

[0077] II. Research results.

[0078] Taking the normalized Signal value as the calculation object, differential lectins that distinguish the disease group from the control group are screened based on statistical methods and analyzed by the Mann-Whitney U rank sum test. When the p-value < 0.05, there is a significant difference between the two; for any lectin, calculate the fold change between the disease group and the control group, that is, the ratio of the disease group to the control group. A fold change ≥ 1.5 is considered a potential difference. Perform univariate and multivariate binary logistic regression analysis, and use the area under the ROC curve to evaluate the diagnostic value of the lectin combined with AFP.

[0079] 2.1. There are significant differential lectins between HCC and other liver diseases.

[0080] The lectin chip results showed a total of significant differences in the binding of various lectins to IgG and / or IgM in serum.

[0081] The difference in the signal of HCC serum IgG binding to lectins compared with other liver diseases.

[0082] Using a lectin chip to detect the binding of IgG in the sera of HCC, HBC, and CHB patient groups to 56 lectins. As Figure 2 shown, compared with the HBC group, the signal of PHA-L binding to IgG in the sera of HCC patients was significantly increased (P = 0.023). Compared with the CHB group, the signal of SBA binding to IgG in the sera of HCC patients was significantly increased (P = 0.0445), and the signals of TL, IAA, PHA-E, AIA, VVA, and Black bean crude binding to IgG were significantly decreased (TL: P = 0.0075; IAA: P = 0.013; PHA-E: P = 0.024; AIA: P = 0.0275; VVA: P = 0.0435; Black bean crude: P = 0.0455).

[0083] The difference in the signal of HCC serum IgM binding to lectins compared with other liver diseases.

[0084] Using a lectin chip to detect the binding of IgM in the sera of HCC, HBC, and CHB patients to 56 lectins. As Figure 3As shown, compared with the CHB group, the binding signals of IgM in the sera of HCC patients to lectins such as SNA-I, MNA-M, PSA, MPL, PHA-L, VVA, CSA, DBA, and SSA were significantly reduced (SNA-I: P = 0.0065; MNA-M: P = 0.019; PSA: P = 0.02; MPL: P = 0.029; PHA-L: P = 0.03; VVA: P = 0.0315; CSA: P = 0.035; DBA: P = 0.036; SSA: P = 0.0405).

[0085] 2.2 Construction of a combined diagnostic model.

[0086] Differences in the binding signals of HCC serum IgG to lectins compared with healthy individuals.

[0087] As Figure 4 shown, the binding of IgG in the sera of HCC patients and healthy individuals to lectins was detected using a lectin microarray. The binding signals of IgG in the sera of HCC patients to SBA and EEL were significantly increased (SBA: P = 0.007; EEL: P = 0.015), and the binding signals to MPL and TL were significantly reduced (MPL: P = 0.0245; TL: P = 0.0345).

[0088] As Figure 5 and Table 1 show, four lectins, SBA, EEL, MPL, and TL, with significant differences in IgG binding in the comparison between the HCC group and the healthy control group were used as independent variables, and the presence or absence of HCC was used as the dependent variable for univariate and multivariate binary Logistic regression analysis. EEL, MPL, and TL were independent influencing factors for the occurrence of liver cancer, P < 0.05. Using the differential lectins that bind to IgG and AFP, a combined diagnostic model for differentiating HCC patients from healthy individuals was established in the training group experimental samples. The combined diagnostic model = -2.277 + 0.510 * AFP + 0.004 * EEL - 0.008 * MPL - 0.015 * TL. The effect of this model in differentiating HCC patients from healthy individuals was significantly better than the single-index detection of AFP (AUC of the combined diagnostic model = 0.96, AUC of AFP = 0.83), and the ROC curve analysis of the above combined diagnostic model was performed in the validation group samples.

[0089] Table 1. Modeling evaluation of the combined diagnosis of HCC using the binding signals of IgG to lectins and AFP.

[0090] 。

[0091] Differences in the binding signals of HCC serum IgM to lectins compared with healthy individuals.

[0092] As Figure 6 shown, the binding of IgM in the sera of HCC patients and healthy individuals to 56 lectins was detected using a lectin microarray. Compared with the healthy control group, the binding signals of IgM in the sera of HCC patients to DSL and LCA were significantly increased (DSL: P = 0.0165; LCA: P = 0.0445).

[0093] As Figure 7 and Table 2 show, DSL and LCA, the two lectins with significant differences in IgM binding in the comparison between the HCC group and the HC group, were used as independent variables, and the presence or absence of HCC was used as the dependent variable for univariate and multivariate binary Logistic regression analysis. DSL was an independent influencing factor for the occurrence of liver cancer, P < 0.05. Using the differential lectin DSL that binds to IgM and AFP, a model for differentiating HCC patients from healthy individuals was established in the training group of experimental samples. The effect of this model in differentiating HCC patients from healthy individuals was significantly better than that of AFP in terms of index detection (AUC of the combined diagnostic model = 0.90, AUC of AFP = 0.83). The combined diagnostic model = -1.901 + 0.003*DSL + 0.286*AFP. And the ROC curve analysis of the above combined diagnostic model was performed on the validation group samples.

[0094] Table 2. Modeling evaluation of the combined diagnosis of HCC by the binding signals of IgM to lectins and AFP.

[0095] .

Claims

1. Application of a specific lectin combination that binds to IgG in serum and alpha-fetoprotein AFP to construct a test tool for differentiating liver cancer, characterized in that: The specific lectin combination is four lectins: SBA, EEL, MPL, and TL; compared with healthy human samples, the signals of SBA and EEL in liver cancer samples are significantly increased, while the signals of MPL and TL are significantly decreased; the test tool is implemented through the diagnostic model -2.277 + 0.510*AFP + 0.004*EEL - 0.008*MPL - 0.015*TL; the liver cancer is hepatocellular carcinoma.

2. Application of a specific lectin combination that binds to IgM in serum and alpha-fetoprotein AFP to construct a test tool for differentiating liver cancer, characterized in that: The specific lectin combination is two lectins: DSL and LCA; compared with healthy human samples, the signals of DSL and LCA in liver cancer samples are significantly increased; the test tool is implemented through the diagnostic model -1.901 + 0.003*DSL + 0.286*AFP; the liver cancer is hepatocellular carcinoma.

3. An intelligent terminal for identifying liver cancer based on IgG in serum, comprising a processor and a program memory, characterized in that: When the program stored in the program memory is loaded by the processor, the following steps are executed: 1) Obtain the lectin test results of IgG samples in serum, and the lectin test results reflect the signal expression levels of the lectins SBA, EEL, MPL, and TL binding to IgG; 2) Obtain the AFP test results in blood; 3) Calculate and construct the detection value of the combined diagnostic model of differential lectins and AFP according to the differential lectins binding to IgG and the AFP test results. The combined diagnostic model of differential lectins binding to IgG and AFP = -2.277 + 0.510*AFP + 0.004*EEL - 0.008*MPL - 0.015*TL. If the detection value ≥ 2.07, it indicates that the subject of the serum sample is a liver cancer patient; The liver cancer is hepatocellular carcinoma.

4. An intelligent terminal for identifying liver cancer based on IgM in serum, comprising a processor and a program memory, characterized in that: When the program stored in the program memory is loaded by the processor, the following steps are executed: 1) Obtain the lectin test results of IgM samples in serum, and the lectin test results reflect the signal expression levels of the lectins DSL and LCA binding to IgM; 2) Obtain the AFP test results in blood; 3) Calculate and construct the detection value of the combined diagnostic model of differential lectins and AFP according to the differential lectins binding to IgM and the AFP test results. The combined diagnostic model of differential lectins binding to IgM and AFP = -1.901 + 0.003*DSL + 0.286*AFP. If the detection value ≥ 2.07, it indicates that the subject of the serum sample is a liver cancer patient; The liver cancer is hepatocellular carcinoma.

5. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is loaded by the processor, it executes each step listed in claim 3 and / or claim 4; the liver cancer is hepatocellular carcinoma.

6. A human-computer interaction device, comprising a display screen, characterized in that: When the human-computer interaction device runs, the display screen sequentially displays the following interfaces: An input interface for the test results of the binding of serum samples to lectins, and the test results of the binding of serum samples to lectins reflect the expression levels of the lectins SBA, EEL, MPL, and TL corresponding to IgG or the expression levels of the lectins DSL and LCA corresponding to IgM; An input interface for the AFP test results in blood; An output interface for the identification information of the combined diagnostic model; the identification information of the combined diagnostic model includes the reference value 2.07 of the combined diagnostic model and the detection value and / or the identification conclusion; the combined diagnostic model is: Differential agglutinin combined with AFP diagnostic model for IgG binding = -2.277 + 0.510 * AFP + 0.004 * EEL - 0.008 * MPL - 0.015 * TL or Differential agglutinin combined with AFP diagnostic model for IgM binding = -1.901 + 0.003 * DSL + 0.286 * AFP; If the detected value ≥ 2.07, it indicates that the subject to which the sample belongs is a liver cancer patient. Conversely, if the detected value < 2.07, it is a healthy person; The liver cancer is hepatocellular carcinoma.

7. A system for differentiating liver cancer based on IgG and / or IgM in serum, characterized in that, Including: A. A device for obtaining the expression level of agglutinin with differential binding to serum IgG and / or IgM, and the agglutinin corresponds to the specific agglutinin combination described in claims 1 and 2; B. A device for obtaining the AFP detection value in blood; C. An intelligent terminal for differentiating liver cancer based on IgG and / or IgM in serum as described in claims 3 and 4; The device for obtaining the expression level of agglutinin with differential binding to serum IgG and / or IgM as described above includes an agglutinin chip, an incubation box, and a biochip scanning system, and the specific agglutinin combination is set on the agglutinin chip; The liver cancer is hepatocellular carcinoma.

Images