Method for constructing cancer evaluation model, cancer evaluation system, medium and electronic equipment

Abstract

The invention discloses a method for constructing a cancer assessment model, a cancer assessment system, a medium and electronic equipment. According to the evaluation model developed by the invention, the diagnosis efficiency of the hepatitis B virus related liver cancer is remarkably improved, and the evaluation model particularly shows excellent prediction performance in early-stage liver cancer and serum marker negative patients. Compared with a single marker and an existing scoring model, the cancer evaluation model constructed by the method has huge advantages in comprehensive indexes such as AUC, sensitivity and specificity. Although the introduced GP73 is a negative coefficient in the model, the diagnosis performance is effectively improved through the synergistic effect of the GP73 and other markers. The evaluation model constructed by the invention provides a reliable tool for non-invasive prediction and early screening of liver cancer, and has important clinical application value.

Description

Technical Field

[0001] The present invention relates to the field of disease assessment, and in particular to a cancer assessment model and a construction method thereof, as well as a cancer assessment system, a computer-readable storage medium and an electronic device based on the model. Background Art

[0002] Hepatocellular carcinoma (HCC) is one of the most common malignant tumors in the world. Although ultrasound, magnetic resonance imaging (MRI) and liver biopsy have greatly improved the accuracy of HCC diagnosis, these technologies have certain shortcomings, such as high cost, invasiveness, and the inability to detect small tumors <2 cm in time, which greatly limits the application of these technologies. At present, a variety of molecular markers, such as alpha-fetoprotein (AFP), alpha-fetoprotein isomer L3 (AFP-L3), and abnormal prothrombin (DCP), have been widely used in the clinical diagnosis of HCC. However, the performance of single molecular markers such as detection accuracy is still insufficient. Many studies have shown that the combined use of multiple markers can significantly improve the sensitivity and specificity of liver cancer diagnosis. At present, the GALAD (consisting of gender, age, AFP, AFP-L3% and DCP) multi-marker diagnostic model for HCC, which was first developed by German scholars, has shown good diagnostic efficacy in multiple centers, different countries and populations, making the GALAD model highly universal. Based on this, domestic scholars developed the C-GALAD scoring system which is more suitable for Chinese patients, but its specificity and sensitivity in diagnosing liver cancer are limited.

[0003] The information in the background technology is only for illustrating the general background of the present invention and should not be regarded as admitting or suggesting in any form that such information constitutes the prior art known to a person skilled in the art. Summary of the invention

[0004] In order to solve at least some of the technical problems existing in the prior art, the present invention provides a method for constructing a cancer assessment model based on a combination of multiple indicators. The method of the present invention not only analyzes the correlation between each indicator and the disease, but also analyzes the correlation between each indicator, excludes indicators with close correlation, and incorporates new indicators through specific multi-factor analysis, thereby greatly improving the sensitivity and accuracy of the assessment model. Specifically, the present invention includes the following contents.

[0005] A first aspect of the present invention provides a method for constructing a cancer assessment model, comprising the following steps: (1) Group the samples according to the symptoms; (2) Obtain multiple indicators that differ between different groups, analyze the correlation between each indicator and the disease, and the correlation between each indicator, and obtain the initial screening indicators; (3) A cancer assessment model is obtained by performing multivariate analysis on the initial screening indicators based on the following formula: logit(P)=ln[P / (1-P)]=β0+β1x1+β2x2+......+βmxm; Among them, P represents the probability that the disease is A, 1-P represents the probability that the disease is B, β0 represents the intercept term, and β1, β2, and βm each represent the coefficient of the corresponding indicator.

[0006] In certain embodiments, according to the method for constructing a cancer assessment model of the present invention, the diseases include chronic hepatitis B, HBV-related cirrhosis and newly diagnosed liver cancer patients.

[0007] In certain embodiments, according to the method for constructing a cancer assessment model of the present invention, the indicators include basic characteristics of the patient and serological indicators.

[0008] In certain embodiments, according to the method for constructing a cancer assessment model of the present invention, indicators for which there are significant differences among the disease groups and there is no correlation or low correlation between the indicators are selected as primary screening indicators.

[0009] A second aspect of the present invention provides a cancer assessment model, wherein the assessment model is constructed by the method described in the first aspect.

[0010] In certain embodiments, according to the cancer assessment model of the present invention, the assessment model formula is: logit(P)=-10.837+0.102×age+2.754×sex(male=1, female=0)-2.019×log GP73+0.922×log(AFP)+2.468×log(DCP).

[0011] A third aspect of the present invention provides a cancer assessment system, comprising: A data collection unit configured to obtain indicator data of a subject; A data processing unit comprises the cancer assessment model described in the second aspect and is configured to use the cancer assessment model to process the indicator data acquired in the data acquisition unit to obtain a prediction result.

[0012] In certain embodiments, according to the cancer assessment system of the present invention, the indicator data includes the age, gender, and amounts of AFP, DCP, and GP73 of the subject.

[0013] In a fourth aspect of the present invention, a computer-readable storage medium is provided, which stores a computer program, wherein the computer program includes the cancer assessment model described in the second aspect, and when executed by a processor, implements the following method to assess cancer: Obtaining the index data of the subject from the data collection unit; The cancer assessment model is used to process the indicator data acquired in the data acquisition unit to obtain a prediction result.

[0014] According to a fifth aspect of the present invention, there is provided an electronic device, comprising: a processor and a memory, wherein: The memory is used to store executable instructions of the processor; The processor is configured to perform the following method by executing the executable instructions: The indicator data of the subject is obtained from the data collection unit, and the cancer assessment model described in the second aspect is used to process the indicator data obtained from the data collection unit to obtain a prediction result.

[0015] The present invention provides a method for constructing a cancer assessment model and a cancer assessment system, medium and electronic device. The developed assessment model significantly improves the diagnostic efficacy of hepatitis B virus-related liver cancer, especially showing excellent predictive performance in early liver cancer and patients with negative serum markers. Compared with single markers (AFP, DCP, GP73) and existing scoring models (GALAD, C-GALAD), the cancer assessment model constructed by the present invention has huge advantages in comprehensive indicators such as AUC, sensitivity and specificity. Although GP73 is a negative coefficient in the model, its synergistic effect with other markers effectively improves the diagnostic performance. Studies have shown that the model developed by the present invention provides a reliable tool for non-invasive prediction and early screening of liver cancer, and has important clinical application value. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Figure 1 Heat map of the correlation between DCP, AFP, AFP-L3%, and GP73.

[0017] Figure 2 Comparison of the values ​​of DCP, AFP, GP73, G-GADA, C-GALAD, and GALAD score in diagnosing liver cancer (A) and early liver cancer (A).

[0018] Figure 3 Comparison of the diagnostic value of DCP, AFP, GP73, G-GADA, C-GALAD, and GALAD scores in the differential diagnosis of cirrhosis and liver cancer. Among them, (A) liver cancer, (B) early liver cancer.

[0019] Figure 4Comparison of the diagnostic value of DCP, AFP, GP73, G-GADA, C-GALAD, and GALAD scores in the differential diagnosis of chronic hepatitis B and liver cancer. Among them, (A) liver cancer, (B) early liver cancer.

[0020] Figure 5 Performance of DCP, AFP, GP73, G-GADA, C-GALAD, and GALAD score in the diagnosis of liver cancer in patients with negative AFP and DCP. DETAILED DESCRIPTION

[0021] Various exemplary embodiments of the present invention will now be described in detail. This detailed description should not be considered as limiting the present invention, but should be understood as a more detailed description of certain aspects, features, and embodiments of the present invention.

[0022] It should be understood that the terms described in the present invention are only for describing special embodiments and are not intended to limit the present invention. In addition, for the numerical range in the present invention, it should be understood that the upper and lower limits of the scope and each intermediate value therebetween are specifically disclosed. Each smaller range between the intermediate value in any stated value or stated range and any other stated value or intermediate value in the described range is also included in the present invention. The upper and lower limits of these smaller ranges can be independently included or excluded in the scope.

[0023] Unless otherwise indicated, all technical and scientific terms used herein have the same meanings as commonly understood by those skilled in the art to which the invention belongs. Although the present invention describes only preferred methods and materials, any methods and materials similar or equivalent to those described herein may also be used in the implementation or testing of the present invention. All documents mentioned in this specification are incorporated by reference to disclose and describe the methods and / or materials related to the documents. In the event of a conflict with any incorporated document, the content of this specification shall prevail.

[0024] Methods for constructing cancer assessment models One aspect of the present invention provides a method for constructing a cancer assessment model, comprising the following steps: (1) Group the samples according to the symptoms; (2) Obtain multiple indicators that differ between different groups, analyze the correlation between each indicator and the disease, and the correlation between each indicator, and obtain the initial screening indicators; (3) A cancer assessment model is obtained by performing a multivariate analysis of the initial screening indicators based on the following formula: logit(P)=ln[P / (1-P)]=β0+β1x1+β2x2+......+βmxm.

[0025] In step (1) of the present invention, the samples are grouped according to patients with chronic hepatitis B, HBV-related liver cirrhosis and newly diagnosed liver cancer, and patients who have chronic hepatitis B, HBV-related liver cirrhosis and newly diagnosed liver cancer are accurately diagnosed using standards known in the art, and liver cancer patients can be further divided into early and late stages. The sample type is not particularly limited and can be a fluid sample (blood sample) or a tissue sample from a patient.

[0026] In step (2) of the present invention, multiple indicators that differ between different groups are obtained, including but not limited to basic characteristics of patients (such as indicators such as the patient's gender, age, etc.) and serological indicators (such as serological indicators such as DCP, AFP, GP73, AFP-L3%). Then, the correlation between each indicator and the disease and the correlation between each indicator are analyzed to obtain the initial screening indicator. In a specific embodiment, the correlation between each indicator and the disease refers to the significant correlation between gender, age, DCP, AFP, GP73, AFP-L3% and the disease. After in-depth research on the correlation between the indicators, it was found that AFP-L3% was correlated with AFP, DCP, GP73, and gender. At this time, this indicator was discarded. At the same time, GP73 had no significant correlation with DCP and showed a weak correlation with AFP and AFP-L3%, and this indicator was used as the initial screening indicator, which significantly improved the problem of low diagnostic value (lower AUC value) caused by using GP73 as an indicator or reference factor. In a preferred embodiment, the initial screening indicators include gender, age, DCP, AFP, and GP73.

[0027] In step (3) of the present invention, gender, age, DCP, AFP, and GP73 are used as variables based on logit(P)=ln[P / (1-P)]=β0+β1x1+β2x2+......+βmxm, wherein P represents the probability that the disease is A, 1-P represents the probability that the disease is B, β0 represents the intercept term, and β1, β2, and βm each represent the coefficient of the corresponding indicator.

[0028] In the present invention, the formula of the cancer assessment model finally obtained is: logit(P)=-10.837+0.102×age+2.754×sex (male=1, female=0)-2.019×log GP73+0.922×log(AFP)+2.468×log(DCP).

[0029] Cancer assessment model and system including the same One aspect of the present invention provides a cancer assessment model constructed by the above method, wherein the formula is: logit(P)=-10.837+0.102×age+2.754×sex (male=1, female=0)-2.019×log GP73+0.922×log(AFP)+2.468×log(DCP).

[0030] In another aspect of the present invention, a cancer assessment system (also referred to as a "cancer assessment device") is provided, comprising: A data collection unit configured to obtain indicator data of a subject; A data processing unit includes the above-mentioned cancer assessment model and is configured to use the cancer assessment model to process the indicator data obtained in the data acquisition unit to obtain a prediction result.

[0031] In the present invention, the data acquisition unit can be any suitable device or instrument capable of collecting the age, gender and amount of AFP, DCP and GP73 of the subject, such as an electronic information storage unit containing basic patient information, medical records, follow-up data, etc., or a kit or instrument capable of measuring or quantifying AFP, DCP and GP73. An exemplary data acquisition unit includes a multi-mode input port. The multi-mode input port can receive information data including the age, gender and amount of AFP, DCP and GP73 of the subject from different sources, including data from online and offline sources. Another exemplary data acquisition unit includes a port connected to a serological indicator measuring instrument (e.g., a chemiluminescent immunoassay analyzer).

[0032] In the present invention, the data processing unit may further include an evaluation unit. The form of the evaluation unit is not limited, as long as it includes the evaluation model of the present invention. Exemplarily, the evaluation unit is a processor. Preferably, the evaluation unit is communicatively connected with the data acquisition unit, so that the data in the data acquisition unit can be retrieved and input into the evaluation model for calculation.

[0033] The present invention further provides a computer storage medium (readable storage medium) or cloud for use by an instruction execution system, device or equipment (such as a computer-based system, a system including a processor, or other system that can fetch instructions from an instruction execution system, device or equipment and execute instructions), or in combination with these instruction execution systems, devices or equipment. For the purposes of the present invention, "computer-readable medium" can be any device that can contain, store, communicate, propagate or transmit a program for use by an instruction execution system, device or equipment, or in combination with these instruction execution systems, devices or equipment. More specific examples of computer-readable media include the following: an electrical connection with one or more wirings (electronic devices), a portable computer disk box (magnetic device), a random access memory (RAM), a read-only memory (ROM), an erasable and programmable read-only memory (EPROM or flash memory), a fiber optic device, and a portable compact disk read-only memory (CDROM). In addition, the computer-readable medium may even be paper or other suitable medium on which the program is printed, since the program may be obtained electronically, for example, by optically scanning the paper or other medium and then editing, interpreting or processing in other suitable ways if necessary, and then stored in a computer memory.

[0034] It should be understood that the various parts of the present invention can be implemented by hardware, software, firmware or a combination thereof. In the above-mentioned embodiments, a plurality of steps or methods can be implemented by software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented by hardware, as in another embodiment, it can be implemented by any one of the following technologies known in the art or their combination: a discrete logic circuit having a logic gate circuit for implementing a logic function for a data signal, a dedicated integrated circuit having a suitable combination of logic gate circuits, a programmable gate array (PGA), a field programmable gate array (FPGA), etc.

[0035] A person of ordinary skill in the art will appreciate that all or part of the steps in the method for implementing the above-mentioned implementation scheme may be completed by instructing related hardware through a program, and the program may be stored in a computer-readable storage medium, which, when executed, includes one of the steps of the method embodiment or a combination thereof.

[0036] In addition, each functional unit in each embodiment of the present invention can be integrated into a processing module, or each unit can exist physically separately, or two or more units can be integrated into one module. The above-mentioned integrated module can be implemented in the form of hardware or in the form of a software functional module. If the integrated module is implemented in the form of a software functional module and sold or used as an independent product, it can also be stored in a computer-readable storage medium. The above-mentioned storage medium can be a read-only memory, a disk or an optical disk, etc.

[0037] Example 1. Materials and Methods 1. Research subjects A total of 332 non-liver cancer cases and 70 patients with untreated liver cancer who were admitted to the Mengchao Hepatobiliary Hospital of Fujian Medical University from June 2015 to June 2020 were selected. The 332 non-liver cancer patients included 209 patients with chronic hepatitis B and 123 patients with HBV-related cirrhosis. The 70 liver cancer patients all met the criteria of the "Guidelines for the Diagnosis and Treatment of Primary Liver Cancer" issued by the Chinese Society of Clinical Oncology and were confirmed by histopathology or imaging. Early HCC was defined as a single tumor with a diameter of no more than 5 cm or a total diameter of multiple tumors of no more than 3 cm. Advanced HCC was defined as conditions that exceeded the Milan criteria, such as a single tumor larger than 5 cm or a total diameter of multiple tumors of more than 3 cm. The 70 liver cancer patients included 34 early liver cancer cases and 36 mid-to-late liver cancer cases, 7 cases with distant metastasis and 63 cases without distant metastasis, 23 cases with microvascular invasion and 47 cases without microvascular invasion.

[0038] 2. Inclusion and Exclusion Criteria 2.1 Inclusion criteria: (1) The subjects must meet the following diagnostic criteria: The diagnosis of primary liver cancer is based on the "Guidelines for the Diagnosis and Treatment of Primary Liver Cancer (2011, 2017 Editions)" and is confirmed by liver tissue biopsy or cytology. The diagnosis of cirrhosis and chronic hepatitis is based on the "Guidelines for the Prevention and Treatment of Chronic Hepatitis B (2015, 2019 Editions)". According to the Milan criteria, liver cancer can be further divided into early and late stages. In addition, the following conditions must be met: a. Aged 18 years and above, regardless of gender; b. HBsAg and / or HBV DNA positive for more than 6 months; (2) Patients with liver cancer must be diagnosed for the first time; (3) The research subjects must not have severe liver or kidney dysfunction; (4) Patients must voluntarily participate in the study and be able to fully understand and sign the informed consent form.

[0039] 2.2 Exclusion criteria: (1) Diagnosed with other viral infections (such as HIV) before enrollment; (2) a history of major organ transplantation, warfarin therapy, or surgery, ablation, radiotherapy, or chemotherapy for liver cancer, liver metastasis, or intrahepatic bile duct cancer; (3) have a history of pancreatitis, or currently have clinical or laboratory evidence supporting the diagnosis of pancreatitis, or show symptoms and laboratory abnormalities related to pancreatitis; (4) no frozen serum samples were available, or the amount of serum was insufficient to complete the biomarker assay; (5) unable to understand or refused to sign the informed consent form, or unable to comply with the requirements of the research protocol; (6) Other circumstances that the researchers consider unsuitable for participation in the study. The research team strictly implemented the above diagnostic, inclusion, and exclusion criteria during the study. This study has been approved by the Ethics Committee of Mengchao Hepatobiliary Hospital of Fujian Medical University (approval number: 2017-014-01). All patients were informed of the study and voluntarily signed the research consent form.

[0040] 3. Methods 3.1 Specimen collection and testing In this example, peripheral blood samples were collected from patients who met the study criteria. The collected blood samples were centrifuged at 1000 r / min for 10 minutes to separate the serum. The serum was immediately frozen at -80°C after packaging until testing. Serum AFP, DCP and GP73 levels were measured by chemiluminescent immunoassay (MAGLUMI ® AFP (CLIA), MAGLUMI ® DCP (CLIA) and MAGLUMI ® GP73 (CLIA)) was measured, and the detection process was strictly carried out in accordance with the specific storage conditions and usage guidelines in the kit instructions. The critical values ​​are age>50, DCP≤40 mAU / mL, and AFP<20 ng / mL. Since there is no unified standard for the reference range of GP73 in clinical practice, this example is mainly used to compare and analyze the relationship between its level and liver cancer diagnosis.

[0041] 3.2 Observation indicators GALAD model value calculation: logit(P)=-10.08+0.09×[age]+1.67×[sex (male=1, female=0)]+2.34×log10[AFP]+0.04×AFP-L3(%)+1.33×log10 [DCP]; C-GALAD model value calculation: logit(P)=-11.501+0.733×[sex (male=1, female=0)]+0.099×[age]+0.073×[AFP-L3(%)]+0.840×log[AFP(μg / L)]+2.346×log[DCP(mAU / mL)].

[0042] 4. Statistical Analysis According to the distribution characteristics of each variable, appropriate descriptive statistical methods were selected, and the normality of the data was evaluated by the Kolmogorov-Smirnov (KS) test. For normally distributed quantitative data, mean and standard deviation (X ± SD) were used for description; for non-normally distributed variables, median and quartile [M (P25, P75)] were used. Categorical variables were expressed as percentages, and intergroup differences were compared by the χ2 test. Statistical analysis and charting were performed using SPSS 26.0 software, GraphPad Prism 9.0, and R language 3.4.4. This embodiment uses MedCalc software to draw the receiver operating characteristic curve (ROC), calculate the area under the ROC curve, determine the optimal cutoff value by the Youden index, and calculate sensitivity and specificity. Correlation analysis was performed using the Spearson correlation method. All statistical tests were bilateral tests. P < 0.05 was considered to be statistically significant.

[0043] 2. Results 1. Comparison of baseline characteristics of subjects In this example, there were significant differences in age, gender and biomarkers (DCP, AFP, GP73, AFP-L3%) between different groups (CHB, LC, HCC) (P<0.05). The HCC group was mainly male, with an average age of 54 years old. The serum DCP, AFP and AFP-L3% of the patients were significantly higher than those of the CHB and LC groups, and the GP73 level was also significantly different between the CHB group and the LC and HCC groups (P<0.05), but there was no statistical difference between the LC and HCC groups. These results provide potential distinguishing variables for the diagnostic model (Table 1).

[0044] Table 1 Analysis of baseline characteristics of the study population * indicates a difference compared with the CHB group, and # indicates a difference compared with the LC group.

[0045] 2. Correlation analysis between serological indicators and clinical characteristics of liver cancer According to the cutoff values ​​of serological indicators, liver cancer patients were divided into groups: based on the DCP cutoff value of 40 mAU / mL, they were divided into DCP low expression group (n=14) and high expression group (n=56); based on the AFP cutoff value of 20 ng / mL, they were divided into AFP low expression group (n=30) and high expression group (n=40).

[0046] Further analysis showed that serum DCP level in patients with liver cancer was significantly correlated with tumor size and Milan stage (P<0.05); serum AFP level was significantly correlated with tumor size, microvascular invasion and Milan stage (P<0.05); serum AFP-L3% level was significantly correlated with patient age, gender, tumor size, distant metastasis, microvascular invasion and Milan stage (P<0.05); serum GP73 level was significantly correlated with tumor size, tumor number, microvascular invasion and Milan stage (P<0.05). The specific results are shown in Table 2.

[0047] Table 2 Correlation between serum DCP, AFP, AFP-L3% and GP73 levels and clinical and tumor characteristics in patients with liver cancer 3. Correlation analysis between clinical and serological indicators of patients Correlation analysis was performed on the patient's age and serum DCP, AFP, AFP-L3% and GP73 levels. The results showed that AFP was weakly correlated with the patient's gender, DCP, and GP73 (P<0.05), and strongly correlated with AFP-L3% (P<0.05); there was no significant correlation between DCP and GP73, but it was weakly correlated with the patient's gender and AFP-L3% (P<0.05); there was also a weak correlation between AFP-L3% and GP73 and the patient's gender (P<0.05). In summary, AFP-L3% is correlated with the other three serological indicators and the patient's gender ( Figure 1 ).

[0048] 4. Establishment of G-GADA model The patient's age, gender, DCP, AFP, AFP-L3% and GP73 were used as independent variables, and whether or not the patient had liver cancer (non-liver cancer = 0, liver cancer = 1) was used as the dependent variable, and included in the following formula: logit(P) = ln[P / (1-P)] = β0+β1x1+β2x2+β3x3+β4x4+β5x5. Among them, P represents the probability of liver cancer, 1-P represents the probability of non-liver cancer, β0 represents the intercept term, and β1, β2, β3, β4, and β5 each represent the coefficient of the corresponding independent variable. After step-by-step screening, the variables finally included in the model included age, sex, DCP, AFP, and GP73, and the following equation was established: logit(P)=−10.837+0.102×age+2.754×sex(male=1, female=0)-2.019×log GP73+0.922×log(AFP)+2.468×log(DCP) (Table 3). The significant difference between this model and the GALAD model is that GP73 replaces AFP-L3%. As an easily detectable biomarker, GP73 provides greater convenience and potential advantages for the extensive clinical application of the model. To reflect this improvement while retaining its relevance to the GALAD model, this model is named the “G-GADA model”, which is an optimized version of the GALAD model that introduces GP73, an innovative marker. This naming method not only highlights the innovation in variable selection, but also emphasizes the advantages of the model in clinical application.

[0049] Table 3 Multivariate analysis of patient age, gender, DCP, AFP, AFP-L3%, and GP73 5. Comparison of the diagnostic value of the G-GADA model with other indicators 5.1 Efficacy of DCP, AFP, GP73, G-GADA, C-GALAD and GALAD score in diagnosing liver cancer and early liver cancer ROC curve was used to evaluate the diagnostic value of DCP, AFP, GP73, G-GADA, C-GALAD and GALAD score for liver cancer. The results showed that the AUCs of single serum markers DCP, AFP and GP73 were 0.888, 0.708 and 0.532, respectively. In comparison, the AUC of the G-GADA model reached 0.941, which was significantly better than the single index and C-GALAD and GALAD (AUC: 0.915, 0.890) scoring models (Z G-GADA - C-GALAD=2.558, Z G-GADA - GALAD=3.593; p =0.0105, p=0.0003). In the diagnosis of early liver cancer, the G-GADA model also showed excellent diagnostic ability. The AUC of the G-GADA model reached 0.908, which was significantly better than the single index and C-GALAD and GALAD (AUC: 0.848, 0.814) scoring models (Z G-GADA - C-GALAD = 4.580, Z G-GADA - GALAD = 4.388; p< 0.0001, p= 0.0001). In general, whether in the diagnosis of liver cancer or early liver cancer, the G-GADA model performed best in comprehensive performance indicators such as AUC, sensitivity and specificity, showing higher diagnostic efficiency and its clinical application potential in liver cancer screening, as shown in Table 4. Figure 2 .

[0050] Table 4 Comparison of the value of DCP, AFP, GP73, G-GADA, C-GALAD, and GALAD scores in diagnosing liver cancer and early liver cancer 5.2 Efficacy of DCP, AFP, GP73, G-GADA, C-GALAD and GALAD score in differential diagnosis of liver cancer and cirrhosis Through further subgroup analysis, the performance of the G-GADA model and other indicators in the differential diagnosis of cirrhosis and liver cancer and early liver cancer was evaluated. In the diagnosis of cirrhosis and liver cancer, the AUC values ​​of single serum markers DCP, AFP and GP73 were 0.891, 0.725 and 0.520, respectively. The AUC value of the G-GADA model was 0.916, the sensitivity was 88.57%, and the Youden index reached 0.7475, which was significantly better than other single markers and C-GALAD (AUC: 0.885) and GALAD (AUC: 0.866) scoring models (Z G-GADA - C-GALAD = 2.113, Z G-GADA - GALAD = 2.633; p< 0.0346, p= 0.0085). In the differentiation of cirrhosis and early liver cancer, the AUC values ​​of single serum markers AFP, DCP and GP73 were 0.601, 0.815 and 0.654, respectively. The AUC value of the G-GADA model was 0.870, the specificity, positive predictive value and negative predictive value were 86.18%, 62.29% and 94.43%, respectively, and the Youden index was 0.6853, which was significantly better than the single index and C-GALAD (AUC: 0.797) and GALAD (AUC: 0.773) scoring models (Z G-GADA - C-GALAD = 3.576, Z G-GADA - GALAD = 3.300;p< 0.0003, p= 0.0010). Overall, the G-GADA model showed the best comprehensive diagnostic performance. This result further verified that the G-GADA model has a high efficacy in differential diagnosis of liver cancer, as shown in Table 5. Figure 3 .

[0051] Table 5 Comparison of the diagnostic value of DCP, AFP, GP73, G-GADA, C-GALAD and GALAD score in the differential diagnosis of cirrhosis and liver cancer 5.3 Efficacy of DCP, AFP, GP73, G-GADA, C-GALAD and GALAD score in differential diagnosis of liver cancer and chronic hepatitis B In the diagnosis of chronic hepatitis B and liver cancer, the G-GADA model showed excellent diagnostic performance, with an AUC of 0.955, a sensitivity of 88.57% and a specificity of 95.22%, and a Youden index of 0.8379. In contrast, among single serum markers, DCP had an AUC of 0.887 and outstanding specificity (97.61%), but its comprehensive diagnostic ability was still inferior to that of the G-GADA model. The AUCs of the C-GALAD and GALAD models were 0.933 and 0.905, respectively, and their specificity, positive predictive value, and Youden were lower than those of G-GADA, with statistically significant differences (Z G-GADA - C-GALAD=2.324, Z G-GADA - GALAD=3.582; p= 0.0201, p= 0.0003). In the diagnosis of chronic hepatitis B and early liver cancer, the G-GADA model also performed best, with an AUC value of 0.931. In comparison, the AUCs of the C-GALAD and GALAD models were 0.878 and 0.838, respectively, and were lower than those of G-GADA in terms of specificity, positive predictive value, negative predictive value and comprehensive index (Z G-GADA - C-GALAD = 3.947, Z G-GADA -GALAD = 4.266; p= 0.0001, p< 0.0001). Overall, the G-GADA model performed best in diagnosing liver cancer in the chronic hepatitis B population, demonstrating its important clinical value in early screening and accurate diagnosis of liver cancer. See Table 6, Figure 4 .

[0052] Table 6 Comparison of the diagnostic value of DCP, AFP, GP73, G-GADA, C-GALAD, and GALAD scores in differential diagnosis of chronic hepatitis B and liver cancer 5.4 Performance of DCP, AFP, GP73, G-GADA, C-GALAD and GALAD score in the diagnosis of liver cancer in patients with negative AFP and DCP In the diagnosis of liver cancer in patients with negative AFP (<20 ng / mL), although there was no statistical difference in AUC between the G-GADA model and the C-GALAD and GALAD models (G-GADA: 0.900, C-GALAD: 0.879, GALAD: 0.870) (p=0.0568, p=0.0789), G-GADA was superior to other models in terms of specificity (93.01%), positive predictive value (55.94%), negative predictive value (97.67%), and Youden index (0.7301). In contrast, DCP performed best among the single serum markers, with an AUC of 0.849 and a high specificity (86.76%), but its overall performance was significantly lower than that of G-GADA. In patients with negative DCP (≤40 mAU / mL), the G-GADA model showed a good balance between sensitivity (78.57%) and specificity (66.89%). Compared with the higher specificity of C-GALAD (87.71%) and the higher specificity of GALAD (92.86%), the G-GADA model has more advantages in comprehensive diagnostic performance, effectively making up for the imbalance between sensitivity and specificity between C-GALAD and GALAD. Overall, in patients with negative AFP or DCP, the G-GADA model showed stronger comprehensive diagnostic capabilities in comparison with the C-GALAD and GALAD models. This further highlights the application value of G-GADA in clinical practice, especially for patients with negative serum markers, see Table 7, Figure 5 .

[0053] Table 7 Comparison of the diagnostic value of DCP, AFP, GP73, G-GADA, C-GALAD, and GALAD scores in the differential diagnosis of chronic hepatitis B and liver cancer Discussion At present, the diagnosis of hepatocellular carcinoma (HCC) mainly relies on imaging examinations such as ultrasound, supplemented by serum tumor markers (AFP, AFP-L3 and DCP). However, imaging is not sensitive enough for early HCC and is affected by subjective factors of the operator. Computed tomography (CT) or magnetic resonance imaging (MRI) examinations have radiation risks, high costs, and false positive signals due to non-specificity. Many scholars have proposed that a combined diagnostic mathematical model of multiple biochemical indicators should be established to improve the diagnostic accuracy of HCC. Best et al. constructed the GALAD scoring model based on AFP, AFP-L3, DCP levels, age and gender, and proved that its diagnostic efficacy is higher than that of a single indicator. Feng Wenxing et al. showed that the AUC of the GALAD model for diagnosing HCC is better than that of ultrasound examination. Liu et al. developed the C-GALAD model based on Chinese patient data, and its diagnostic performance is better than that of traditional models. The GALAD model includes three biomarkers: AFP, AFP-L3, and DCP, and two demographic characteristics: age and gender. AFP is generally upregulated in liver cell regeneration, primary liver cancer cells and embryonic carcinoma tissues, and can be used as a biomarker for diagnosing HCC. Some scholars believe that its diagnostic performance is good, but there are still 30% AFP-negative patients diagnosed with liver cancer. AFP-L3 is a subtype of AFP, which is secreted only by HCC cells and is the main form of AFP in the serum of HCC patients. Its level is related to tumor differentiation and can effectively reflect the growth rate and tumor volume of HCC tumors. The present invention also confirms that it is related to tumor size, whether there is distal metastasis, microvascular infiltration and tumor stage. Under normal circumstances, DCP is usually not detected in the serum of healthy individuals because it is mainly produced by malignant tumor liver cells. Therefore, DCP is considered to be a highly specific HCC biomarker and is widely recommended for liver cancer screening, especially in patients with AFP negative. It shows potential advantages. However, the clinical application of DCP in HCC is still controversial. In recent years, GP73 has been widely used in the detection of HCC as a new biomarker. In normal liver, GP73 is mainly expressed in bile duct epithelial cells, and less in hepatocytes. Studies have shown that the AUC value of GP73 in chronic hepatitis B, chronic hepatitis C, non-alcoholic fatty liver disease, autoimmune hepatitis and the resulting cirrhosis is as high as 0.89. Therefore, serum GP73 has a good diagnostic value for hepatitis and cirrhosis caused by different etiologies. Studies have also shown that the sensitivity of GP73 in diagnosing hepatocellular carcinoma is 74.6%, the specificity is 97.4%, and the positive predictive value is 82.7%. Its serum expression is higher than that of the control group and adjacent tissues. Therefore, GP73 is not only a good serum marker for liver fibrosis and cirrhosis, but is also widely used to distinguish healthy people, benign liver lesions and liver cancer.

[0054] The present invention has been found through research that the expression level of GP73 in patients with chronic hepatitis B, cirrhosis and liver cancer shows a nonlinear trend, among which the GP73 of LC patients is not significantly different from that of HCC patients, while the lowest in patients with chronic hepatitis B. This trend also reflects the pathophysiological changes in the course of disease progression. Cirrhosis is an important risk factor for HCC, and the occurrence of liver cancer is a cumulative process with increasing age, usually going through the stages from chronic hepatitis B to cirrhosis and then to liver cancer. In the present invention, although significant age differences were observed between the CHB group and the HCC group, the age differences between the LC group and the HCC group were not obvious. This result may suggest that GP73 reflects more the degree of liver fibrosis or chronic inflammation rather than a direct marker in the occurrence of liver cancer. However, in the multivariate analysis model (C-GADA), the coefficient of GP73 is negative (OR<1), suggesting that its increased level may be negatively correlated with the probability of liver cancer diagnosis. This phenomenon may be due to the interaction effect between multiple variables, especially the high expression characteristics of GP73 in patients with cirrhosis, which may weaken its independent diagnostic value in distinguishing liver cancer from non-liver cancer patients. This result emphasizes the importance of considering the interaction between variables in the multivariate model, and suggests that the clinical application of GP73 needs to be comprehensively evaluated in combination with the patient's specific pathological stage and other biomarkers, so as to more accurately reflect the diagnostic value of liver cancer. In addition, the partial functional overlap between GP73 and markers such as AFP in liver cancer may reduce its independent discriminatory power. The results of the present invention show that the combination of AFP, DCP and GP73, three HCC-specific biomarkers and two risk factors as independent risk factors for liver cancer diagnosis, produces a synergistic effect, and further shows that although GP73 performs generally as a single variable, it has a synergistic effect in the joint model and optimizes the overall diagnostic performance.

[0055] The present invention comprehensively evaluates the application value of the G-GADA model in the diagnosis of liver cancer and early liver cancer through ROC curve analysis. The results show that the G-GADA model has significant advantages in diagnostic performance. Specifically, in the diagnosis of liver cancer, the AUC of G-GADA reaches 0.941, which is significantly higher than the single serum markers DCP, AFP and GP73, as well as the scoring models GALAD and C-GALAD (all P < 0.05). This shows that G-GADA not only improves the accuracy of diagnosis by combining multiple indicators, but also has higher clinical value in practical applications. It is worth noting that in the diagnosis of early liver cancer, the AUC of the G-GADA model reaches 0.908, and the specificity is as high as 91.87%, which is superior to other diagnostic models in sensitivity and specificity. This result is particularly critical because the timely diagnosis of early liver cancer is crucial to improving patient prognosis, and the performance of single serum markers such as DCP, AFP and GP73 in the diagnosis of early liver cancer is relatively limited. The G-GADA model effectively compensates for the shortcomings of a single marker through multi-parameter integration, further improving the accuracy of early liver cancer screening. In addition, the stability and advantages of G-GADA were further verified in subgroup analysis. In the subgroups of chronic hepatitis B and liver cancer, and cirrhosis and liver cancer, the AUC of G-GADA was significantly better than that of single markers and GALAD and C-GALAD scoring models; in addition, the G-GADA model also showed good performance in the diagnosis of AFP and DCP-negative patients. In patients with AFP levels below 20 ng / mL, the AUC of G-GADA reached 0.900 and the specificity reached 93.01%. This result shows that even in the case of low AFP levels, G-GADA can still provide high diagnostic accuracy (Youden index 0.7301), suggesting the application potential of this model in AFP-negative patients. At the same time, in patients with DCP≤40, the AUC of G-GADA was also higher than other models, further verifying the applicability of this model in patients with low DCP levels. Overall, the G-GADA model showed stable and superior diagnostic performance in subgroups with different disease states and marker levels, further demonstrating the clinical practicality and wide applicability of the model.

[0056] The research results of the present invention show that the G-GADA model effectively achieves the accurate distinction between liver cancer and non-liver cancer patients by integrating multiple markers such as AFP, DCP and GP73, and optimizes the diagnostic contribution of GP73 through staged analysis, providing more accurate and reliable support for early screening of liver cancer.

[0057] IV. Conclusion The G-GADA model developed by the present invention significantly improves the diagnostic efficacy of HBV-related liver cancer by introducing the new serum marker GP73, especially showing excellent predictive performance in early liver cancer and serum marker-negative patients. Compared with single markers (AFP, DCP, GP73) and existing scoring models (GALAD, C-GALAD), G-GADA has certain advantages in comprehensive indicators such as AUC, sensitivity and specificity. Although GP73 is a negative coefficient in the model, it works synergistically with other markers to effectively improve diagnostic performance. The G-GADA model of the present invention provides a reliable tool for non-invasive prediction and early screening of liver cancer, and has important clinical application value.

[0058] Although the present invention has been described with reference to exemplary embodiments, it should be understood that the present invention is not limited to the disclosed exemplary embodiments. Various adjustments or changes may be made to the exemplary embodiments of the present specification without departing from the scope or spirit of the present invention. The scope of the claims should be based on the broadest interpretation to cover all modifications and equivalent structures and functions.

Claims

1. A method for constructing a cancer assessment model, characterized in that: The following steps are involved: (1) Group the samples according to the symptoms; (2) Obtain multiple indicators that differ between different groups, analyze the correlation between each indicator and the disease, and the correlation between each indicator, and obtain the initial screening indicators; (3) A cancer assessment model is obtained by performing multivariate analysis on the initial screening indicators based on the following formula: logit(P)=ln[P / (1-P)]=β0+β1x1+β2x2+......+βmxm; Among them, P represents the probability that the disease is A, 1-P represents the probability that the disease is B, β0 represents the intercept term, and β1, β2, and βm each represent the coefficient of the corresponding indicator.

2. The method for constructing a cancer assessment model according to claim 1, characterized in that: The conditions include chronic hepatitis B, HBV-related cirrhosis, and patients with newly diagnosed liver cancer.

3. The method for constructing a cancer assessment model according to claim 1, characterized in that: The indicators include basic characteristics of patients and serological indicators.

4. The method for constructing a cancer assessment model according to claim 1, characterized in that: Indicators with significant differences among the symptom groups and no correlation or low correlation among the indicators were selected as initial screening indicators.

5. A cancer assessment model, characterized in that: It is constructed by the method described in any one of claims 1 to 4.

6. The cancer assessment model according to claim 5, characterized in that The formula is: logit(P)=-10.837+0.102×age+2.754×sex(male=1, female=0)-2.019×log GP73+0.922×log(AFP)+2.468×log(DCP).

7. A cancer assessment system, characterized in that: include: A data collection unit configured to obtain indicator data of a subject; A data processing unit, comprising the cancer assessment model according to claim 5, and configured to use the cancer assessment model to process the indicator data obtained in the data acquisition unit to obtain a prediction result.

8. The cancer assessment system according to claim 7, characterized in that: The index data include the age, gender and the amounts of AFP, DCP and GP73 of the subject.

9. A computer-readable storage medium, characterized in that: The computer program stores a computer program, which includes the cancer assessment model according to claim 5, and when executed by a processor, implements the following method to assess cancer: Obtaining the index data of the subject from the data collection unit; The cancer assessment model is used to process the indicator data acquired in the data acquisition unit to obtain a prediction result.

10. An electronic device, characterized in that: include: A processor and a memory, wherein: The memory is used to store executable instructions of the processor; The processor is configured to perform the following method by executing the executable instructions: The indicator data of the subject is obtained from the data collection unit, and the cancer assessment model according to claim 5 is used to process the indicator data obtained from the data collection unit to obtain a prediction result.

Images