Diagnostic model of hepatocellular carcinoma microvascular invasion and construction method and application thereof

Abstract

The invention discloses a diagnosis model for hepatocellular carcinoma microvascular invasion and a construction method and application thereof, and belongs to the technical field of medical examination. Metabolite in the blood of a hepatocellular carcinoma patient is taken as a cut-in, preoperative clinical tumor indexes are combined, it is found that PIVKA-II, GGT, L-Serine, Sarcosine and tumor sizes have important significance on formation of hepatocellular carcinoma MVI, a diagnosis model based on multiple omics and multiple dimensions is established based on the diagnosis indexes, the diagnosis model is used for diagnosis of the preoperative MVI of the hepatocellular carcinoma patient, and the diagnosis model is used for diagnosis of the preoperative MVI of the hepatocellular carcinoma patient. The defects of existing diagnosis means in the aspect of preoperative diagnosis are overcome, and a more timely and accurate basis is provided for clinical decision making.

Description

Technical Field

[0001] The invention belongs to the technical field of medical testing, and in particular relates to a diagnostic model for microvascular invasion of hepatocellular carcinoma and a construction method and use thereof. Background Art

[0002] Hepatocellular carcinoma (HCC) is a common type of primary liver cancer and the fourth leading cause of cancer-related death worldwide. Surgical operations, mainly liver resection and transplantation, are still effective treatments. Despite this, the postoperative recurrence rate is still high, with a 5-year recurrence rate of 50-70% after resection and a 5-year recurrence rate of 30-35% after transplantation. Hepatocellular carcinoma is characterized by easy metastasis and high recurrence. During its development, it is easy to invade blood vessels and form microvascular invasion (MVI). Microvascular invasion is an independent risk factor for poor prognosis of HCC. The recurrence rate of HCC patients with MVI is significantly higher than that of HCC patients without MVI, with a 5-year recurrence rate of up to 60-70%. Therefore, clarifying the MVI status before surgery is of great significance for the formulation of treatment plans and the improvement of patient prognosis.

[0003] At present, the "gold standard" for the diagnosis of MVI relies on pathological examination of tumor lesions, but the results of pathological examination are easily affected by the sampling site (such as the temporal and spatial heterogeneity of the tumor), and cannot provide clinical information on the presence of MVI before surgery. Imaging diagnosis of MVI is currently a common method, and the status of MVI is judged by the recognition of characteristic images, but the extraction of characteristic images depends on experienced clinicians, and the result judgment is relatively subjective and time-consuming. Secondly, the detection rate of MVI by imaging methods is not good, such as the detection rate of enhanced computed tomography (CT) is 30-50%, the detection rate of enhanced magnetic resonance imaging (MRI) is 40-60%, and the detection rate of ultrasound contrast is 20-40%. Therefore, new and effective preoperative MVI diagnostic methods are urgently needed to make up for the shortcomings of existing diagnostic methods in preoperative diagnosis and provide more timely and accurate basis for clinical decision-making.

[0004] More and more studies have revealed the role of proteomics and metabolomics in the occurrence and development of HCC. The present invention detects the changes in metabolites in the blood of HCC patients with and without MVI, screens out metabolites related to MVI, and combines preoperative clinical tumor indicators to establish a highly sensitive and specific preoperative MVI diagnostic model. The preoperative MVI diagnostic model provided by the present invention obtains multidimensional data in a non-invasive form, accurately predicts the MVI status of hepatocellular carcinoma patients, and provides a new idea for preoperative diagnosis of MVI. Summary of the invention

[0005] The purpose of the present invention is to overcome the lack of existing means for preoperative MVI diagnosis, as well as the defects of insufficient sensitivity and specificity, and to provide a diagnostic model that can improve the effective detection rate of MVI. Specifically, the present invention uses metabolites in the blood of patients with hepatocellular carcinoma as an entry point, combined with preoperative clinical tumor indicators, and finds that PIVKA-II, GGT, L-serine (L-Serine), sarcosine (Sarcosine) and tumor size are of great significance to the formation of MVI in hepatocellular carcinoma, and based on the above diagnostic indicators, a multi-omics, multi-dimensional diagnostic model is established for the diagnosis of preoperative MVI in patients with hepatocellular carcinoma.

[0006] In the first aspect of the present invention, the present invention provides a use of a diagnostic indicator or a reagent for detecting a diagnostic indicator in the preparation of a product for preoperative diagnosis of microvascular invasion of hepatocellular carcinoma, wherein the diagnostic indicator is PIVKA-II, GGT, L-Serine, Sarcosine and tumor size.

[0007] The products include but are not limited to models, kits, membrane strips, chips or detection platforms.

[0008] If not particularly limited, the reagents for detecting diagnostic indicators described in the present invention are reagents or means known to those skilled in the art for detecting PIVKA-II, GGT, L-Serine, Sarcosine and tumor size.

[0009] In the second aspect of the present invention, the present invention provides a method for obtaining diagnostic indicators of microvascular invasion of hepatocellular carcinoma, the method comprising: grouping hepatocellular carcinoma patients according to the presence or absence of MVI; performing difference analysis and enrichment pathway analysis on the patient's metabolome data to screen for metabolites related to MVI; using univariate and multivariate binary logistic regression analysis to screen for diagnostic indicators of microvascular invasion of hepatocellular carcinoma with a threshold of P<0.05.

[0010] Preferably, the method for obtaining diagnostic indicators of microvascular invasion of hepatocellular carcinoma comprises the following steps:

[0011] (1) HCC patients were grouped according to whether they had MVI or not, and the metabolite composition in blood samples of HCC patients was detected using UHPLC-MS to obtain metabolome data;

[0012] (2) Perform differential analysis and enrichment pathway analysis on the metabolome data to screen out metabolites related to MVI;

[0013] (3) Collect preoperative clinical tumor indicators of HCC patients;

[0014] (4) Univariate analysis was used to screen out indicators associated with MVI in HCC patients, with P < 0.05 as the threshold;

[0015] (5) Multivariate analysis was performed on the candidate indicators obtained in step (4), with P<0.05 as the threshold, to screen out diagnostic indicators, wherein the diagnostic indicators consisted of PIVKA-II, GGT, L-Serine / Sarcosine and tumor size.

[0016] In a specific embodiment of the present invention, the method for obtaining diagnostic indicators of microvascular invasion of hepatocellular carcinoma comprises the following steps:

[0017] (1) HCC patients were grouped according to whether they had MVI or not, and the metabolite composition in blood samples of HCC patients was detected using UHPLC-MS to obtain metabolome data;

[0018] (2) OPLS-DA was used to analyze metabolites and MVI status, and differential metabolites with VIP>1 and P<0.05 were screened. Then, enrichment pathway analysis based on the KEGG database was performed on the differential metabolites to screen metabolites with an impact value >0.2.

[0019] (3) Preoperative clinical indicators of the patients were collected, including alpha-fetoprotein, protein induced by vitamin K absence or antagonist-II (PIVKA-II), albumin, total bilirubin, alanine aminotransferase, aspartate aminotransferase, gamma-glutamyl transferase (GGT), hepatitis B surface antigen, tumor size, tumor number, and Chinese liver cancer staging classification;

[0020] (4) The effects of individual factors obtained in steps (2) and (3) on MVI were analyzed using univariate binary logistic regression to screen indicators with P < 0.05 that were related to MVI;

[0021] (5) Multivariate binary logistic regression was used to analyze the effects of multiple factors on MVI obtained in step (4), and indicators with P < 0.05 were selected as diagnostic indicators.

[0022] In the third aspect of the present invention, the present invention provides a method for constructing a diagnostic model for microvascular invasion of hepatocellular carcinoma, the method comprising using the diagnostic indicators obtained by the screening method described in the second aspect of the present invention as variables, and constructing a diagnostic model based on the linear coefficients obtained by multifactor binary logistic regression analysis.

[0023] The hepatocellular carcinoma microvascular invasion diagnostic model of the present invention is prepared by the following method:

[0024] HCC patients were grouped according to whether they had MVI or not. The metabolome data of patients were subjected to differential analysis and enrichment pathway analysis to screen MVI-related metabolites. Univariate and multivariate binary logistic regression analysis was used to screen the diagnostic indicators of microvascular invasion of hepatocellular carcinoma with a threshold of P < 0.05. The diagnostic indicators were used as variables, and a diagnostic model was constructed based on the linear coefficients obtained by multivariate binary logistic regression analysis.

[0025] Preferably, the hepatocellular carcinoma microvascular invasion diagnostic model is prepared by the following method:

[0026] (1) HCC patients were grouped according to whether they had MVI or not, and the metabolite composition in blood samples of HCC patients was detected using UHPLC-MS to obtain metabolome data;

[0027] (2) Perform differential analysis and enrichment pathway analysis on the metabolome data to screen out metabolites related to MVI;

[0028] (3) Collect preoperative clinical tumor indicators of HCC patients;

[0029] (4) Univariate analysis was used to screen out indicators associated with MVI in HCC patients, with P < 0.05 as the threshold;

[0030] (5) using multivariate analysis to screen the candidate indicators obtained in step (4) with a threshold of P < 0.05 to obtain diagnostic indicators, wherein the diagnostic indicators consist of PIVKA-II, GGT, L-Serine / Sarcosine, and tumor size;

[0031] (6) Using the diagnostic indicators described in step (5) as variables, a diagnostic model is constructed based on the linear coefficients obtained by multivariate binary logistic regression analysis.

[0032] In a specific embodiment of the present invention, the hepatocellular carcinoma microvascular invasion diagnostic model is prepared by the following method:

[0033] (1) HCC patients were grouped according to whether they had MVI or not, and the metabolite composition in blood samples of HCC patients was detected using UHPLC-MS to obtain metabolome data;

[0034] (2) OPLS-DA was used to analyze metabolites and MVI status, and differential metabolites with VIP>1 and P<0.05 were screened. Then, enrichment pathway analysis based on the KEGG database was performed on the differential metabolites to screen metabolites with an impact value >0.2.

[0035] (3) Collect the patients' preoperative clinical indicators, including alpha-fetoprotein, PIVKA-II, albumin, total bilirubin, alanine aminotransferase, aspartate aminotransferase, GGT, hepatitis B surface antigen, tumor size, tumor number, and Chinese liver cancer staging classification;

[0036] (4) The effects of individual factors obtained in steps (2) and (3) on MVI were analyzed using univariate binary logistic regression to screen indicators with P < 0.05 that were related to MVI;

[0037] (5) Multivariate binary logistic regression was used to analyze the effects of multiple factors obtained in step (4) on MVI, and indicators with P < 0.05 were selected as diagnostic indicators;

[0038] (6) Using the diagnostic indicators described in step (5) as variables, a diagnostic model is constructed based on the linear coefficients obtained by multivariate binary logistic regression analysis.

[0039] In a fourth aspect of the present invention, the present invention provides a diagnostic model for microvascular invasion of hepatocellular carcinoma, wherein the diagnostic model is constructed by the method described in the third aspect of the present invention.

[0040] In a specific embodiment of the present invention, the diagnostic model of microvascular invasion of hepatocellular carcinoma is:

[0041]

[0042] The diagnostic performance was evaluated with a cut-off value of 8.344. When the Y value > 8.344, it means that the patient with hepatocellular carcinoma has MVI; when Y < 8.344, it means that the patient with hepatocellular carcinoma has not had MVI.

[0043] The model provided by the present invention can diagnose whether a patient has MVI before surgery, which helps to select an appropriate treatment plan, thereby effectively reducing the probability of postoperative metastasis and recurrence in patients, improving the quality of life of patients, and prolonging their survival time, providing a feasible solution for the clinical application of predicting the prognosis of hepatocellular carcinoma based on microvascular invasion.

[0044] The model provided by the present invention has good diagnostic efficacy for the MVI status of HCC patients before surgery. The AUC area of ​​the model verified by the experiment is 0.896, which is significantly higher than that of a single indicator, such as AUC PIVKA-II =0.787, AUC GGT =0.721, AUC L-Serine / Sarcosine = 0.737 and AUC 肿瘤大小 =0.614. The clinical sensitivity and specificity of the model were 80.0% and 87.2%, respectively, which were significantly higher than the current diagnostic methods.

[0045] In a fifth aspect of the present invention, the present invention provides a use of the model described in the fourth aspect of the present invention in at least one of the following:

[0046] (1) Use in the preparation of products for preoperative MVI diagnosis or auxiliary diagnosis of hepatocellular carcinoma;

[0047] (2) Use in the preparation of products for prognosis assessment or auxiliary assessment of hepatocellular carcinoma.

[0048] The model for MVI diagnosis of hepatocellular carcinoma and the construction method thereof provided by the present invention have the following technical advantages:

[0049] In view of the high metastasis and recurrence of hepatocellular carcinoma, the preoperative prediction of microvascular invasion is of great significance for the selection of clinical treatment plans and the improvement of patient prognosis. The present invention first uses ultra high performance liquid chromatography-mass spectrometer (UHPLC-MS) to detect the composition of metabolites in blood samples of hepatocellular carcinoma patients to obtain metabolome data; then uses orthogonal partial least-squares discriminant analysis (OPLS-DA) to analyze metabolites and MVI status, and uses variable importance in the projection (VIP)>1 and P<0.05 as thresholds to screen out differential metabolites; based on the Kyoto Encyclopedia of Genes and Genomes (Kyoto Encyclopedia of Genes and Genomes), the metabolites were analyzed by orthogonal partial least-squares discriminant analysis (OPLS-DA). The enrichment pathway analysis of the screened metabolites was performed using the KEGG) database, and the impact value > 0.2 was set as the threshold to screen out the metabolites with the most significant impact on MVI; then, the candidate metabolites and clinical tumor characteristics were combined to screen out the influencing factors related to MVI by univariate analysis; the indicators with P < 0.05 in the univariate analysis were included in the multivariate analysis, and the indicators with P < 0.05, namely the combination of PIVKA-II, GGT, L-Serine / Sarcosine and tumor size, were selected as the diagnostic indicators for diagnosing MVI of hepatocellular carcinoma; further, the diagnostic indicator was used as a variable, and a diagnostic model was constructed according to the linear coefficients obtained by multivariate binary logistic regression analysis.

[0050] The diagnostic model provided by the present invention involves multi-omics and multi-dimensional markers. Relevant markers are obtained through blood and images in a non-invasive manner, and incorporated into the model for scoring calculation, thereby achieving safe and non-invasive preoperative MVI diagnosis, which is of great significance for the formulation of treatment plans and the improvement of patient prognosis. BRIEF DESCRIPTION OF THE DRAWINGS

[0051] Figure 1 Differential analysis based on OPLS-DA (A) and enriched pathway analysis based on the KEGG database (B).

[0052] Figure 2 Levels of L-Serine (A) and Sarcosine (B) in HCC patients in the MVI positive and negative groups.

[0053] Figure 3 The AUC area of ​​the diagnostic model (A) and the comparison of the AUC areas of different indicators for diagnosing MVI (B). DETAILED DESCRIPTION

[0054] The technical solutions in the embodiments of the present invention are described clearly and completely below. Obviously, the described embodiments are only some embodiments of the present invention, not all. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0055] This study was approved by the Human Subjects Ethics Committee of Zhongshan Hospital of Fudan University (approval number: B2024-125R). The study protocol followed the guidelines established by the Declaration of Helsinki, and written informed consent was obtained from all subjects.

[0056] 1. Selection of research subjects

[0057] Methods: A retrospective analysis was performed on newly diagnosed HCC patients who underwent liver cancer resection in our hospital from May to October 2023. The inclusion criteria were: (1) diagnosed with hepatocellular carcinoma based on histopathological examination; (2) age >18 years. The exclusion criteria were as follows: (1) any history of malignant tumors; (2) previous treatment history; (3) hypertension, diabetes, or other metabolic diseases. A total of 119 HCC patients were included, and preoperative EDTA-K2 anticoagulated whole blood samples were collected. Whole blood samples were centrifuged at 3000 rpm for 10 min at 4°C to separate plasma and stored at -80°C for later use.

[0058] According to the postoperative histopathological diagnosis of the patients, the samples were divided into a non-MVI group and an MVI group.

[0059] 2. Clinical data collection

[0060] Clinical data included general demographic information, such as gender and age, and preoperative clinical tumor indicators, such as alpha-fetoprotein, PIVKA-II, albumin, total bilirubin, alanine aminotransferase, aspartate aminotransferase, GGT, hepatitis B surface antigen, tumor size, tumor number, and Chinese liver cancer staging scheme grade.

[0061] 3. Detection of metabolites in plasma of HCC patients using UHPLC-MS technology

[0062] 100 μL of plasma was mixed with 400 μL of extract containing deuterated internal standard (methanol:acetonitrile = 1:1). The mixture was vortexed for 30 seconds, sonicated in a 4°C water bath for 10 minutes, then incubated at -40°C for 1 hour to complete protein precipitation, and finally centrifuged at 13000 rpm for 15 minutes at 4°C, and the supernatant was taken for use. Ultra-high performance liquid chromatography (Thermo Fisher) equipped with an ACQUITY UPLC BEH Amide column (WATERS, 2.1 mm × 50 mm, 1.7 μm) and mass spectrometry (Orbitrap Exploris 120, Thermo Fisher) were used for separation. The mobile phase consisted of an aqueous solution of 25 mmol / L ammonium acetate and 25 mmol / L ammonium hydroxide (PH = 9.75) (phase A) and acetonitrile (phase B). The autosampler temperature was 4°C and the injection volume was 2 μL. The mass spectrometer acquired data in data-independent mode. The conditions of the electrospray ion source were set as follows: sheath gas flow rate 50 arb, auxiliary gas flow rate 15 arb, capillary temperature 320°C, full scan mass spectrometry resolution 60000, mass spectrometry / mass spectrometry resolution 15000, collision energy: selective protonated neutral loss energy of 20 / 30 / 40, and spray voltages of 3.8 kV (positive ion mode) and -3.4 kV (negative ion mode).

[0063] 4. Screening of metabolites related to MVI

[0064] OPLS-DA was used to analyze metabolites and MVI status, and differential metabolites were screened according to VIP>1 and P<0.05 ( Figure 1 A). Enrichment pathway analysis of the screened metabolites was performed based on the KEGG database, and the metabolites with the most significant impact on MVI, L-Serine and Sarcosine, were screened out according to the impact value > 0.2 ( Figure 1 B). Plasma L-Serine in HCC patients in the MVI-positive group was higher than that in the MVI-negative group (P=0.019, Figure 2 A), while the plasma Sarcosine level in HCC patients in the MVI-positive group was lower than that in the MVI-negative group (P = 0.005, Figure 2 B).

[0065] 5. Construction of a diagnostic model for microvascular invasion of hepatocellular carcinoma

[0066] Combined with candidate metabolites and preoperative clinical tumor indicators, univariate binary logistic regression analysis was performed based on whether the patient had MVI, and the influencing factors related to MVI were screened out. The indicators with P < 0.05 in the univariate analysis were selected and included in the multivariate analysis. Multivariate binary logistic regression analysis was performed based on whether the sample had MVI, and the indicators with P < 0.05 were selected as diagnostic indicators, including PIVKA-II, GGT, L-Serine / Sarcosine and tumor size (Table 1).

[0067] Table 1 Univariate and multivariate analysis of risk factors associated with MVI

[0068]

[0069] The above diagnostic indicators were used as variables, and a diagnostic model was constructed based on the linear coefficients obtained by multivariate binary logistic regression analysis. The diagnostic model is as follows:

[0070]

[0071] The diagnostic model uses 8.344 as the cut-off value for diagnostic efficacy evaluation. When the Y value>8.344, it indicates that the hepatocellular carcinoma patient has MVI; when Y<8.344, it indicates that the hepatocellular carcinoma patient has not had MVI.

[0072] The above model has good diagnostic efficacy for the MVI status of HCC patients before surgery, with an AUC of 0.896 ( Figure 3 A), significantly higher than a single indicator such as AUC PIVKA-II =0.787, AUC GGT =0.721, AUC L-Serine / Sarcosine = 0.737 and AUC 肿瘤大小 =0.614( Figure 3 B). The clinical sensitivity and specificity were 80.0% and 87.2% respectively, which were significantly higher than the current diagnostic methods.

[0073] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or replace some or all of the technical features therein with equivalents. However, these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. Use of diagnostic indicators or reagents for detecting diagnostic indicators in the preparation of products for preoperative diagnosis of microvascular invasion of hepatocellular carcinoma, wherein the diagnostic indicators are PIVKA-II, GGT, L-Serine, Sarcosine and tumor size.

2. A method for obtaining diagnostic indicators of microvascular invasion of hepatocellular carcinoma, the method comprising: Hepatocellular carcinoma patients were grouped according to whether they had MVI; Differential analysis and enrichment pathway analysis were performed on the patient metabolome data to screen for MVI-related metabolites; Univariate and multivariate binary Logistic regression analysis was used to screen the diagnostic indicators of microvascular invasion of hepatocellular carcinoma with P<0.05 as the threshold.

3. The method according to claim 2, characterized in that The method for obtaining diagnostic indicators of microvascular invasion of hepatocellular carcinoma comprises the following steps: (1) HCC patients were grouped according to whether they had MVI or not, and the metabolite composition in blood samples of HCC patients was detected using UHPLC-MS to obtain metabolome data; (2) Perform differential analysis and enrichment pathway analysis on the metabolome data to screen out metabolites related to MVI; (3) Collect preoperative clinical tumor indicators of HCC patients; (4) Univariate analysis was used to screen out indicators associated with MVI in HCC patients, with P < 0.05 as the threshold; (5) Multivariate analysis was performed on the candidate indicators obtained in step (4), with P<0.05 as the threshold, to screen out diagnostic indicators, wherein the diagnostic indicators consisted of PIVKA-II, GGT, L-Serine / Sarcosine and tumor size.

4. The method according to claim 2, characterized in that: The method for obtaining diagnostic indicators of microvascular invasion of hepatocellular carcinoma comprises the following steps: (1) HCC patients were grouped according to whether they had MVI or not, and the metabolite composition in blood samples of HCC patients was detected using UHPLC-MS to obtain metabolome data; (2) OPLS-DA was used to analyze metabolites and MVI status, and differential metabolites with VIP>1 and P<0.05 were screened. Then, enrichment pathway analysis based on the KEGG database was performed on the differential metabolites to screen metabolites with an impact value >0.

2. (3) Collect the patients' preoperative clinical indicators, including alpha-fetoprotein, PIVKA-II, albumin, total bilirubin, alanine aminotransferase, aspartate aminotransferase, GGT, hepatitis B surface antigen, tumor size, tumor number, and Chinese liver cancer staging classification; (4) The effects of individual factors obtained in steps (2) and (3) on MVI were analyzed using univariate binary logistic regression to screen indicators with P < 0.05 that were related to MVI; (5) Multivariate binary logistic regression was used to analyze the effects of multiple factors on MVI obtained in step (4), and indicators with P < 0.05 were selected as diagnostic indicators.

5. A method for constructing a diagnostic model for microvascular invasion of hepatocellular carcinoma, the method comprising using the diagnostic index obtained by the screening method according to any one of claims 2 to 4 as a variable, and constructing a diagnostic model based on the linear coefficients obtained by multivariate binary logistic regression analysis.

6. The method according to claim 5, characterized in that The hepatocellular carcinoma microvascular invasion diagnostic model is prepared by the following method: (1) HCC patients were grouped according to whether they had MVI or not, and the metabolite composition in blood samples of HCC patients was detected using UHPLC-MS to obtain metabolome data; (2) Perform differential analysis and enrichment pathway analysis on the metabolome data to screen out metabolites related to MVI; (3) Collect preoperative clinical tumor indicators of HCC patients; (4) Univariate analysis was used to screen out indicators associated with MVI in HCC patients, with P < 0.05 as the threshold; (5) using multivariate analysis to screen the candidate indicators obtained in step (4) with a threshold of P < 0.05 to obtain diagnostic indicators, wherein the diagnostic indicators consist of PIVKA-II, GGT, L-Serine / Sarcosine, and tumor size; (6) Using the diagnostic indicators described in step (5) as variables, a diagnostic model is constructed based on the linear coefficients obtained by multivariate binary logistic regression analysis.

7. The method according to claim 6, characterized in that The hepatocellular carcinoma microvascular invasion diagnostic model is prepared by the following method: (1) HCC patients were grouped according to whether they had MVI or not, and the metabolite composition in blood samples of HCC patients was detected using UHPLC-MS to obtain metabolome data; (2) OPLS-DA was used to analyze metabolites and MVI status, and differential metabolites with VIP>1 and P<0.05 were screened. Then, enrichment pathway analysis based on the KEGG database was performed on the differential metabolites to screen metabolites with an impact value >0.

2. (3) Collect the patients' preoperative clinical indicators, including alpha-fetoprotein, PIVKA-II, albumin, total bilirubin, alanine aminotransferase, aspartate aminotransferase, GGT, hepatitis B surface antigen, tumor size, tumor number, and Chinese liver cancer staging classification; (4) The effects of individual factors obtained in steps (2) and (3) on MVI were analyzed using univariate binary logistic regression to screen indicators with P < 0.05 that were related to MVI; (5) Multivariate binary logistic regression was used to analyze the effects of multiple factors obtained in step (4) on MVI, and indicators with P < 0.05 were selected as diagnostic indicators; (6) Using the diagnostic indicators described in step (5) as variables, a diagnostic model is constructed based on the linear coefficients obtained by multivariate binary logistic regression analysis.

8. A diagnostic model for microvascular invasion of hepatocellular carcinoma, characterized in that the diagnostic model is constructed by the method according to any one of claims 5 to 7.

9. The diagnostic model according to claim 8, characterized in that The diagnostic model for microvascular invasion of hepatocellular carcinoma is:

10. Use of the model according to any one of claims 8 or 9 in at least one of the following: (1) Use in the preparation of products for preoperative MVI diagnosis or auxiliary diagnosis of hepatocellular carcinoma; (2) Use in the preparation of products for prognosis assessment or auxiliary assessment of hepatocellular carcinoma.

Images