Modeling method of hepatocellular carcinoma risk assessment model and hepatocellular carcinoma risk assessment system

Abstract

The invention belongs to the field of medical treatment, and discloses a modeling method of a hepatocellular carcinoma risk assessment model, and biomarkers comprise an aMAP liver cancer risk score composed of age, gender, total bilirubin, albumin and platelets, abnormal prothrombin and alpha fetoprotein. And performing logistic regression multi-factor analysis through statistical software, and constructing a diagnosis mode for diagnosing the early hepatocellular carcinoma. According to the constructed diagnosis mode, the highest diagnosis value can be obtained when early HBV related HCC is diagnosed, the area under a diagnosis curve is close to 90%, and the probability that HCC is found in the early stage is improved to a large extent. According to the invention, the combination of the tumor marker and the risk score can effectively improve the disease diagnosis and timely early warning efficiency, and the successful development of the diagnosis model provides a method and strategy reference for the construction of other disease diagnosis or early warning models.

Description

Technical Field

[0001] The present invention relates to the medical field, and in particular to a modeling method of a hepatocellular carcinoma risk assessment model and a hepatocellular carcinoma risk assessment system. Background Art

[0002] Primary liver cancer is currently the fourth most common malignant tumor and the second leading cause of death from cancer in my country, seriously endangering the lives and health of the Chinese people. Among them, hepatocellular carcinoma (HCC) accounts for about 75% to 85%. Due to its insidious onset and unclear early symptoms, the early diagnosis rate of HCC in my country is extremely low. As many as 70% to 80% of patients are already in the middle and late stages at the time of diagnosis, and have lost the opportunity for radical treatment such as liver resection and liver transplantation, and the prognosis is extremely poor. Therefore, early diagnosis and early warning of HCC are of great significance to improving the prognosis and survival of patients.

[0003] Imaging detection technology is a widely used clinical detection technology for diagnosing early HCC, such as ultrasound, CT and MRI. However, its accuracy is often limited by the technical ability and experience of the operator. Serum markers are also widely used in the early diagnosis of HCC. Alpha-fetoprotein (AFP) is currently recognized and the most commonly used HCC serum marker in clinical practice. However, the diagnostic efficacy of AFP is still not ideal, and the overall sensitivity for HCC at all stages is less than 70%. In recent years, vitamin K deficiency or antagonist-II-induced protein (PIVKA-II, also known as abnormal prothrombin, DCP), a new HCC serum marker, has received increasing clinical attention and has been widely used in many countries and regions such as China, South Korea, and Japan. It was first approved for auxiliary diagnosis of HCC in China.

[0004] The diagnostic model that integrates multiple serological tumor markers and clinical characteristics has better HCC diagnostic efficacy than a single biomarker. Summary of the invention

[0005] The present invention aims to study a biomarker combination with better sensitivity and specificity for diagnosing / early warning HCC and a new diagnostic model.

[0006] The above purpose can be achieved by implementing the following technical solutions:

[0007] A method for modeling a hepatocellular carcinoma risk assessment model comprises the following steps:

[0008] Step 1: randomly grouping the patient sample data to obtain a training set and a validation set; the patient sample data includes liver cancer patient data and non-liver cancer patient data; the patient sample data includes: patient serum AFP content, patient serum DCP content and patient aMAP score;

[0009] Step 2: Using the training set and the validation set, a model is established based on the patient's serum alpha-fetoprotein level, the patient's serum abnormal prothrombin level, and the patient's aMAP score to obtain a hepatocellular carcinoma risk assessment model.

[0010] Optionally, the non-liver cancer patient is a patient with chronic liver disease and / or a patient with liver cirrhosis.

[0011] Optionally, the model is a logistic model.

[0012] Optionally, the aMAP score is calculated using an aMAP scoring model based on the test subject's gender, age, albumin index, total bilirubin index, and platelet index.

[0013] A hepatocellular carcinoma risk assessment system, comprising:

[0014] A data acquisition module is used to obtain test information of the test object;

[0015] an aMAP score calculation module, configured to calculate an aMAP score using an aMAP score model according to the test information;

[0016] An evaluation value calculation module, used to calculate the evaluation value using an evaluation value model according to the serum AFP content, the patient's serum DCP content and the aMAP score in the test information; the evaluation value model is established by the above-mentioned modeling method;

[0017] A result output module is used to compare the evaluation value with an evaluation threshold value, and output a hepatocellular carcinoma risk evaluation result according to the comparison result.

[0018] Optionally, the test information of the test subject includes the test subject's gender, age, ALB, TBIL, PLT, serum AFP content, and patient serum DCP content.

[0019] Optionally, the formula of the aMAP scoring model is as follows:

[0020] aMAP=({0.06×age+0.89×sex(male:1,female:0)+0.48×[(log 10 TBIL×0.66)+(ALB×-0.085)]-0.01×PLT)}+7.4) / 14.77×100;

[0021] Among them, ALB is the albumin index, TBIL is the total bilirubin index, and PLT is the platelet index.

[0022] An electronic device comprises a memory, a processor and a computer program stored in the memory and executable on the processor, wherein the processor implements the steps of the modeling method of the above-mentioned hepatocellular carcinoma risk assessment model when executing the program.

[0023] A computer-readable storage medium stores a computer program, which, when executed by a processor, implements the steps of the modeling method of the above-mentioned hepatocellular carcinoma risk assessment model.

[0024] The present invention has the following beneficial effects:

[0025] The diagnostic model constructed by the present invention can obtain the highest diagnostic value when diagnosing early HBV-related HCC, and the area under the diagnostic curve is close to 90%, which greatly increases the probability of HCC being discovered in the early stage. The present invention shows that the combination of tumor markers and risk scores can effectively improve the effectiveness of disease diagnosis and timely early warning. The successful development of this type of diagnostic model provides a reference for methods and strategies for the construction of other disease diagnosis or early warning models.

[0026] The detection indexes required by the model established by the present invention are derived from serum, which is convenient for sampling, easy for detection, and cheap compared with liquid biopsy technology. Compared with existing HCC diagnostic scores (such as GAAD), it has higher sensitivity and specificity.

[0027] Since the model established by the present invention incorporates the platelet count index that can reflect the coagulation function, and the albumin and total bilirubin indexes that reflect the liver function, the score has excellent HCC short-term warning performance and helps the early screening and diagnosis of HCC. BRIEF DESCRIPTION OF THE DRAWINGS

[0028] In order to more clearly illustrate the specific implementation methods of the present invention or the technical solutions in the prior art, the drawings required for use in the specific implementation methods or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are some implementation methods of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0029] Figure 1 The figure shows the comparison of the effects of the model of the present invention, aMAP-3 and GAAD when SEN is fixed at 80%.

[0030] Figure 2 The figure shows the comparison of the effects of the model of the present invention, aMAP-3 and GAAD when the SPE is fixed at 90%. DETAILED DESCRIPTION

[0031] Now, various exemplary embodiments of the present invention are described in detail, and this detailed description should not be considered as a limitation of the present invention, but should be understood as a more detailed description of certain aspects, characteristics and embodiments of the present invention. It should be understood that the terms described in the present invention are only for describing specific embodiments and are not used to limit the present invention.

[0032] In addition, for the numerical range in the present invention, it is understood that each intermediate value between the upper and lower limits of the range is also specifically disclosed. The intermediate value in any stated value or stated range, and each smaller range between any other stated value or intermediate value in the 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.

[0033] Unless otherwise specified, all technical and scientific terms used herein have the same meaning as commonly understood by those skilled in the art to which the invention pertains. Although only preferred methods and materials are described herein, any methods and materials similar or equivalent to those described herein may also be used in the practice or testing of the present invention.

[0034] The words “include,” “including,” “have,” “contain,” etc. used in this document are open-ended terms, meaning including but not limited to.

[0035] The present invention provides a modeling method for a hepatocellular carcinoma risk assessment model, comprising the following steps:

[0036] Step 1: randomly grouping the patient sample data to obtain a training set and a validation set; the patient sample data includes liver cancer patient data and non-liver cancer patient data; the patient sample data includes: patient serum AFP content, patient serum DCP content and patient aMAP score;

[0037] Step 2: Using the training set and the validation set, a model is established based on the patient's serum AFP level, the patient's serum DCP level, and the patient's aMAP score, thereby obtaining a hepatocellular carcinoma risk assessment model.

[0038] The patient sample data is obtained in the following manner:

[0039] Access to clinical datasets;

[0040] In the present embodiment, the follow-up data of 2889 patients with chronic liver disease who received long-term antiviral treatment at the time of diagnosis of HCC or the last follow-up were retrospectively collected, including gender, age, ALB, TBIL, PLT, etc.

[0041] A total of 2,022 patients were randomly divided into the training set at a ratio of 7 to 3 for the construction of the HCC diagnosis model. Among them, there were 175 HCC patients and 1,847 non-HCC patients. The validation set included 64 HCC patients and 803 non-HCC patients.

[0042] At the same time, we collected the patients' clinical data 6 months and 12 months before diagnosis / last follow-up to evaluate the short-term early warning effectiveness of the constructed model for liver cancer.

[0043] HCC was diagnosed if any of the following conditions were met: 1) confirmed by liver biopsy or surgical resection with histopathological confirmation; 2) detection of lesions meeting HCC criteria by two or more imaging techniques (transabdominal ultrasound imaging, abdominal three-phase computed tomography scan, liver magnetic resonance imaging study, or liver angiography); 3) detection of lesions meeting HCC criteria by one imaging technique with AFP greater than 400 μg / L.

[0044] Clinical features

[0045] The test results of HCC patients and non-HCC patients (divided into chronic liver disease and cirrhosis) at the time of diagnosis / last follow-up are expressed as median (interquartile range). Data analysis showed that compared with non-HCC patients, HCC patients had higher age, male proportion, liver fibrosis scan, TBIL, AFP, DCP, alanine aminotransferase, aspartate aminotransferase and aMAP scores. Among HCC patients, the proportion of early liver cancer (BCLC stage 0 or A) was 65.7% (97 / 178). The specific clinical characteristics of the 2889 study population are shown in Table 1.

[0046] Table 1 Clinical characteristics of subjects at diagnosis / last follow-up

[0047]

[0048]

[0049] Calculation of aMAP score;

[0050] The aMAP score was calculated as follows: aMAP = ({0.06 × age + 0.89 × sex (male: 1, female: 0) + 0.48 × [(log 10 TBIL×0.66)+(ALB×-0.085)]-0.01×PLT)}+7.4) / 14.77×100.

[0051] The method for establishing the model in step 2) is:

[0052] Establishment of logistic regression model: Logistic regression belongs to probabilistic nonlinear regression. In addition to being used for influencing factor analysis, prediction and discrimination are also important applications of logistic regression model. Univariate logistic regression analysis was used to screen variable features, and aMAP score, DCP binary characteristics and AFP were obtained as risk factors for liver cancer, and they were statistically significant. Furthermore, the multivariate logistic regression method was used to construct the model. The dependent variable of this binary logistic regression model is a binary variable, and the values ​​are coded as 0 and 1. As shown below, 1 represents a positive result, which means the patient has liver cancer; 0 represents a negative result, which means the patient does not have liver cancer.

[0053]

[0054] There are three known factors that affect the value of Y, namely, aMAP liver cancer risk score (denoted as X 1 ), DCP binary classification (denoted as X 2 ) and AFP (denoted as X 3 ). At this time, the probability of Y = 1 is recorded as π, and the probability of Y = 0 is 1-π. β 0 is a constant term, β 1 , β 2 , β 3 are the regression coefficients corresponding to the above three influencing factors. E, as a mathematical constant, is the base of the natural logarithm function, which is equal to 2.718. The relationship between π and the linear combination of the three influencing factors is as follows:

[0055]

[0056] Since the probability π ranges from [0,1], the influencing factor X 1 , X 2 and X 3 The range of the linear combination of is (-∞,+∞). Therefore, based on the regression relationship between the probability π and the linear combination of the above three influencing factors, π is further logit transformed and represented by y.

[0057]

[0058] y=logit(π)

[0059] Finally, β 0 is a constant -8.150, β 1 , β 2 , β 3 They are 0.069, 2.976 and 2.376 respectively. The model established by the above method is:

[0060] y=0.069*aMAP score+2.976*DCP dichotomy (≥28.4ng / mL: 1; <28.4ng / mL: 0)+2.376*log 10 (AFP) -8.150. When the value of y is ≥ -2.60, the probability π is greater than one-half, and the subject is determined to be diagnosed with liver cancer.

[0061] Evaluation of model liver cancer diagnosis / short-term warning performance;

[0062] The embodiment of the present invention evaluates the performance of the aMAP-3 score through indicators such as area under the curve (AUC), sensitivity (SEN) and specificity (SPE). Combined with Table 2, it can be seen that compared with a single tumor marker (such as AFP or DCP) and the existing GAAD liver cancer diagnostic score (composed of age, gender, DCP and AFP), the aMAP-3 score has the best liver cancer diagnostic efficacy, with an AUC range of 0.890-0.922. In the validation set, the diagnostic performance of the aMAP-3 score remained stable, with an AUC range of 0.830-0.900. More importantly, the aMAP-3 score has a more outstanding short-term warning performance for liver cancer 6 months and 12 months before diagnosis / last follow-up, and always shows an AUC higher than the GAAD liver cancer diagnostic score.

[0063] Table 2. AUC of each tumor marker and its different combinations at diagnosis / last follow-up, first 6 months, and first 12 months

[0064]

[0065]

[0066] *P<0.05 compared with GAAD score

[0067] In addition, if Figure 1 , 2 As shown in the results, when the fixed SEN was 80%, the aMAP-3 score had higher SPE, positive predictive value, negative predictive value and accuracy at all three time points compared with the GAAD score. Consistently, when the fixed SPE was 90%, the aMAP-3 score still showed superior HCC diagnosis / short-term warning efficacy.

[0068] In summary, the present invention shows that the combination of tumor markers and risk scores can effectively improve the effectiveness of disease diagnosis and timely early warning. The successful development of this type of diagnostic model provides a reference for methods and strategies for the construction of other disease diagnosis or early warning models.

[0069] Based on the above model, a hepatocellular carcinoma risk assessment system is provided, including:

[0070] A data acquisition module is used to obtain test information of the test object;

[0071] an aMAP score calculation module, configured to calculate an aMAP score using an aMAP score model according to the test information;

[0072] An evaluation value calculation module, used to calculate the evaluation value using an evaluation value model according to the serum AFP content, the patient's serum DCP content and the aMAP score in the test information; the evaluation value model is established using the modeling method described in any one of claims 1 to 4;

[0073] A result output module is used to compare the evaluation value with an evaluation threshold value, and output a hepatocellular carcinoma risk evaluation result according to the comparison result.

[0074] The test information of the test subject includes the test subject's gender, age, ALB, TBIL, PLT, serum AFP content, and patient serum DCP content.

[0075] The formula of the aMAP scoring model is as follows:

[0076] aMAP=({0.06×age+0.89×sex(male:1,female:0)+0.48×[(log10TBIL×0.66)+(ALB×-0.085)]-0.01×PLT)}+7.4) / 14.77×100

[0077] The above system can be run on a computer device, which includes: one or more processors, memory, and interfaces for connecting various components, including high-speed interfaces and low-speed interfaces. The various components are connected to each other using different buses for communication, and can be installed on a common motherboard or installed in other ways as needed. The processor can process instructions executed in the computer device, including instructions stored in or on the memory to display graphical information of the GUI on an external input / output device (such as a display device coupled to the interface).

[0078] In some optional embodiments, if desired, multiple processors and / or multiple buses can be used together with multiple memories and multiple memories. Similarly, multiple computer devices can be connected, and each device provides part of the necessary operations (for example, as a server array, a group of blade servers, or a multi-processor system).

[0079] The processor may be a central processing unit, a network processor or a combination thereof. The processor may further include a hardware chip. The hardware chip may be a dedicated integrated circuit, a programmable logic device or a combination thereof. The programmable logic device may be a complex programmable logic device, a field programmable gate array, a general purpose array logic or any combination thereof.

[0080] The memory stores instructions executable by at least one processor, so that the at least one processor executes the method for preparing a coated conductor of a substrate shown in the above embodiment.

[0081] The memory may include a program storage area and a data storage area, wherein the program storage area may store an operating system, an application required for at least one function; the data storage area may store data created according to the use of the computer device, etc. In addition, the memory may include a high-speed random access memory, and may also include a non-transient memory, such as at least one disk storage device, a flash memory device, or other non-transient solid-state storage device. In some optional embodiments, the memory may include a memory remotely arranged relative to the processor, and these remote memories may be connected to the computer device via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0082] The memory may include volatile memory, such as random access memory; the memory may also include non-volatile memory, such as flash memory, hard disk or solid state drive; the memory may also include a combination of the above types of memory.

[0083] The computer device also includes an input device and an output device. The processor, memory, input device and output device can be connected via a bus or other methods.

[0084] The computer device also includes a communication interface, which is used for the computer device to communicate with other devices or a communication network.

[0085] An embodiment of the present invention also provides a computer-readable storage medium, and the above-mentioned method according to the embodiment of the present invention can be implemented in hardware, firmware, or can be implemented as a computer code that can be recorded in a storage medium, or can be implemented by downloading through a network and originally stored in a remote storage medium or a non-temporary machine-readable storage medium and will be stored in a local storage medium, so that the method described herein can be stored in such software processing on a storage medium using a general-purpose computer, a dedicated processor, or programmable or dedicated hardware.

[0086] The storage medium may be a magnetic disk, an optical disk, a read-only storage memory, a random access memory, a flash memory, a hard disk or a solid state drive, etc.; further, the storage medium may also include a combination of the above-mentioned types of memories. It is understood that a computer, a processor, a microprocessor controller or programmable hardware includes a storage component that can store or receive software or computer code. When the software or computer code is accessed and executed by the computer, processor or hardware, the method shown in the above embodiment is implemented.

[0087] Embodiments of the present application may also provide a computer program product, including computer program instructions, which, when executed by a processor, cause the processor to perform the steps in the above method. The computer program product may be written in any combination of one or more programming languages ​​to perform program codes for performing the operations of the disclosed embodiments, wherein the programming languages ​​include object-oriented programming languages, such as Java, C++, etc., and conventional procedural programming languages, such as "C" language 10 or similar programming languages. The program code may be executed entirely on a user computing device, partially on a user computing device, as an independent software package, partially on a user computing device, partially on a remote computing device, or entirely on a remote computing device or server.

[0088] The above embodiments are only used to illustrate the technical solutions of the embodiments of the present invention, rather than to limit them. Although the embodiments of the present invention are 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 of the technical features therein by equivalents. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A modeling method for a hepatocellular carcinoma risk assessment model, characterized in that: The steps include: Step 1: randomly grouping the patient sample data to obtain a training set and a validation set; the patient sample data includes liver cancer patient data and non-liver cancer patient data; the patient sample data includes: patient serum alpha-fetoprotein content, patient serum abnormal prothrombin content and patient aMAP score; Step 2: Using the training set and the validation set, a model is established based on the patient's serum alpha-fetoprotein level, the patient's serum abnormal prothrombin level, and the patient's aMAP score to obtain a hepatocellular carcinoma risk assessment model.

2. The method for modeling a hepatocellular carcinoma risk assessment model according to claim 1, characterized in that: The non-liver cancer patients are patients with chronic liver disease and / or patients with liver cirrhosis.

3. The method for modeling a hepatocellular carcinoma risk assessment model according to claim 1, characterized in that: The model is a logistic model.

4. The method for modeling a hepatocellular carcinoma risk assessment model according to claim 1, characterized in that: The aMAP score is calculated using the aMAP scoring model based on the test subject's gender, age, albumin index, total bilirubin index, and platelet index.

5. A hepatocellular carcinoma risk assessment system, characterized in that: include: A data acquisition module is used to obtain test information of the test object; an aMAP score calculation module, configured to calculate an aMAP score using an aMAP score model according to the test information; An evaluation value calculation module, used to calculate the evaluation value using an evaluation value model according to the serum alpha-fetoprotein content, serum abnormal prothrombin content and aMAP score in the test information; the evaluation value model is established using the modeling method described in any one of claims 1 to 4; A result output module is used to compare the evaluation value with an evaluation threshold value, and output a hepatocellular carcinoma risk evaluation result according to the comparison result.

6. The hepatocellular carcinoma risk assessment system according to claim 5, characterized in that: The test information of the test subject includes the test subject's gender, age, albumin index, total bilirubin index, platelet index, serum alpha-fetoprotein content, and serum abnormal prothrombin content.

7. The hepatocellular carcinoma risk assessment system according to claim 5, characterized in that: The formula of the aMAP scoring model is as follows: aMAP=({0.06×age+0.89×sex(male:1,female:0)+0.48×[(log 10 TBIL×0.66)+(ALB×-0.085)]-0.01×PLT)}+7.4) / 14.77×100; Among them, ALB is the albumin index, TBIL is the total bilirubin index, and PLT is the platelet index.

8. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein: When the processor executes the program, the steps of the method for modeling a hepatocellular carcinoma risk assessment model as described in any one of claims 1 to 6 are implemented.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method for modeling a hepatocellular carcinoma risk assessment model as claimed in any one of claims 1 to 6 are implemented.

Images