Biomarker combination for diagnosing liver cancer, Logistic model and application thereof

Abstract

The invention provides a biomarker combination for early diagnosis of liver cancer. The biomarker combination is composed of an RNA biomarker and a protein biomarker, the RNA type biomarkers comprise LINC01554, miR-222, SNORD31, circMIRIAF, SNHG8 and RNU5E-1, and the RNA type biomarkers are used for detecting the RNA type biomarkers. The protein type biomarkers comprise SIGLEC5, CXCL10, AFP (Alpha Fetoprotein) and PIVKA-II (Phosphatidylserine Virus Kinase Activator-II) The invention also provides an application of the biomarker combination in preparation of a liver cancer early diagnosis kit, and a Logistic model constructed based on the biomarker combination. The kit and the Logistic model provided by the invention can be suitable for diagnosis of early stage, middle stage and late stage of liver cancer. The Logistic model and the diagnosis standard of the Logistic model have high sensitivity and specificity on the diagnosis of the liver cancer.

Description

Technical Field

[0001] The present invention relates to a series of biomarker combinations composed of RNA and protein for liver cancer diagnosis, a Logistics model constructed based on the biomarker combination, and application thereof in liver cancer diagnosis, and belongs to the field of tumor medicine. Background Art

[0002] Due to the hidden onset of liver cancer and the lack of characteristic clinical manifestations, 60-70% of patients are already in the middle and late stages when diagnosed, and have lost the opportunity for surgery, with a 5-year survival rate of only 12%. Compared with middle and late stage liver cancer, the 5-year survival rate of early stage liver cancer patients exceeds 60%, and the treatment cost is only one-seventh of that of late stage liver cancer. Therefore, "early diagnosis and early treatment" of liver cancer is the key to improving the clinical efficacy of liver cancer and reducing the socioeconomic burden of liver cancer patients.

[0003] Liver associations in various countries and my country's primary liver cancer diagnosis and treatment guidelines point out that high-risk groups should undergo liver cancer screening every 6 months. However, less than 20% of the risk groups undergo regular monitoring. In addition to the patient's personal cognitive factors, the difficulty in popularizing existing detection methods and the low detection rate of liver cancer are the main challenges in liver cancer screening. Liver ultrasound and serum alpha-fetoprotein (AFP) testing are currently the main means of liver cancer screening. Although the overall sensitivity of ultrasound examination for liver cancer diagnosis has reached 84%, its sensitivity for detecting early liver cancer is only 47%. This is largely due to the fact that the effect of ultrasound examination depends on the operator's experience and technology, and the increase in the number of people with fatty liver leads to poor visual field, resulting in a low screening rate for risk groups. Although AFP detection is less restricted by equipment conditions and subjective factors, the function of AFP and its correlation with liver cancer have not yet been fully elucidated, and its sensitivity for diagnosing early liver cancer is also less than 60%. Even if combined with ultrasound for screening, the sensitivity is only 63%, and its role in improving the early diagnosis rate is still limited. Although many researchers have established multi-index scoring models (such as GALAD score) for screening by introducing AFP subtypes and vitamin K deficiency-induced protein PIVKA-II, they can only achieve a diagnostic sensitivity of about 60%. Although CT and MR have high accuracy in diagnosing liver cancer, they are difficult to promote on a large scale in early diagnosis due to radiation factors and equipment conditions.

[0004] Therefore, in recent years, the field of liver cancer screening has focused on liquid biopsy technology around new markers, including DNA methylation markers, free DNA (cfDNA), circulating tumor DNA (ctDNA) and extracellular vesicles. Early screening of liver cancer can be achieved by detecting biological components released by tumor cells in peripheral blood, with a sensitivity of up to 82% in the study cohort. However, existing related studies also face many challenges. Taking ctDNA, which is the most studied, as an example, due to the low content of ctDNA in early cancer, it is difficult to accurately distinguish between true mutations and sequencing errors, and even if there are mutations, more than half of them come from clonal hematopoiesis, increasing the possibility of false positives; and for genes with low mutation frequencies, a small amount of blood samples has a high risk of missed diagnosis. The introduction of second-generation sequencing technology may help improve the accuracy of diagnosis, but it will greatly increase the cost of testing and pose challenges to the application of this screening technology. Therefore, although the FDA has given breakthrough medical device recognition to domestic and foreign companies such as Exact Science, Lemon Health, and PanGenomics, due to the expected limited actual clinical benefits, in the field of liver cancer, the FDA has not yet approved any liquid biopsy technology products for marketing, and related products still need to be verified in large-scale prospective studies.

[0005] RNA testing is another emerging field for early screening of liver cancer, including microRNA (miRNA), mRNA and long non-coding RNA (lncRNA). Academician Fan Jia's team has developed my country's first in vitro diagnostic kit for liver cancer based on 7 miRNAs, with a diagnostic sensitivity of 82%. MiRNA testing has also been included in the national medical service fee items, indicating that the field of RNA testing has broad prospects.

[0006] The above research results are helpful to improve the early diagnosis rate and survival rate of liver cancer patients in my country, but the sensitivity and specificity of liver cancer early diagnosis through a single biomarker are often difficult to balance. Therefore, it is necessary to propose a new liver cancer early diagnosis marker detection matrix and model to better balance the diagnostic sensitivity and specificity and further improve the efficiency of liver cancer early diagnosis. Summary of the invention

[0007] The technical problem to be solved or the primary purpose of the present invention is to use a series of biomarkers to solve the problem of balancing sensitivity and specificity in the early diagnosis of liver cancer.

[0008] Another object of the present invention is to provide a product for early diagnosis of the above-mentioned liver cancer.

[0009] In order to achieve the above object, the present invention adopts the following technical solutions:

[0010] In the first aspect, a biomarker combination for early diagnosis of liver cancer is provided, which is composed of RNA biomarkers and protein biomarkers; the RNA biomarkers include LINC01554, miR-222, SNORD31, circMIRIAF, SNHG8 and RNU5E-1; the protein biomarkers include SIGLEC5, CXCL10, AFP and PIVKA-II.

[0011] The LINC01554 described herein refers to the long non-coding RNA with NCBI sequence number 202299;

[0012] The miR-222 is a microRNA with NCBI sequence number 407007;

[0013] The SNORD31 described herein refers to the small nucleolar RNA with NCBI sequence number 9298;

[0014] The circMIRIAF refers to the circular RNA with NCBI sequence number 5887;

[0015] The SNHG8 refers to the long non-coding RNA with NCBI sequence number 100093630;

[0016] The SIGLEC5 is a protein with NCBI sequence number AAF87846.1;

[0017] The CXCL10 is a protein with NCBI sequence number AAH10954.1;

[0018] The AFP is a protein with NCBI sequence number EAX05681.1;

[0019] The PIVKA-II is a protein with NCBI sequence number NP_001341646.2.

[0020] The present invention applies the biomarker combination to prepare an early diagnosis reagent for liver cancer.

[0021] In the scheme described in the present invention, by using a detection reagent for the biomarker combination to detect the expression level of each biomarker in a sample from a subject, the subject's liver cancer condition can be accurately diagnosed and evaluated at an early stage. Furthermore, the sample from the subject is any one or more of the subject's peripheral venous blood, paracancerous tissue or tumor tissue; preferably peripheral venous blood.

[0022] In a preferred embodiment of the present invention, the early diagnosis reagent for liver cancer is a detection reagent based on RNA capture probe technology, high-throughput sequencing, and enzyme-linked immunosorbent assay technology.

[0023] In the second aspect, the present invention also provides a kit, comprising a reagent for detecting the expression level of a biomarker, and one or more substances selected from the following group: a container, instructions for use, a positive control, a negative control, a buffer, an adjuvant or a solvent; the reagent for detecting the expression level of a biomarker includes a reagent for detecting the expression level of all markers in the biomarker combination described in the first aspect of the present invention. Further, the reagent includes: an RNA probe and a specific antibody; the RNA probe is composed of a specific nucleotide sequence complementary fragment of the six RNA biomarkers in the biomarker combination described in the first aspect of the present invention; the specific antibody is composed of specific antibodies of the four protein biomarkers in the biomarker combination described in the first aspect of the present invention.

[0024] The instruction manual records how to use the kit for testing and how to use the test results to judge the occurrence and progression of liver cancer. The components of the kit can be packaged in the form of an aqueous medium or in a freeze-dried form.

[0025] The third aspect shows that the present invention also provides a Logistic model for early diagnosis of liver cancer based on the combination of biomarkers described in the first aspect of the present invention, comprising:

[0026] 1) obtaining valid data of the individual sample to be diagnosed; the valid data includes the expression level of each biomarker in the biomarker combination described in the first aspect of the present invention in the sample;

[0027] 2) Substitute all expression level data obtained in 1) into the following model calculation formula:

[0028] Risk value=2.479×LINC01554-0.216×miR222+19.078×SNORD31-0.097×circMIRIAF+1.562×SNHG8+0.808×RNU5E1-0.003×SIGLEC5+0.962×CXCL10+0.003×PIVKAII+0.063×AFP;

[0029] Wherein, the LINC01554, miR222, SNORD31, circMIRIAF, SNHG8, RNU5E1, SIGLEC5, CXCL10, PIVKAII and AFP respectively represent their respective expression level values ​​in the samples, and all values ​​have the same unit;

[0030] 3) Based on the calculation results of 2), the liver cancer progression discrimination analysis was performed according to the quantitative evaluation criteria;

[0031] 4) Output early diagnosis results of liver cancer.

[0032] In the Logistic model described in the present invention, the sample described in 1) can be any one or a combination of two or more of the samples of blood, liver, liver cancer and adjacent tissues from the individual to be diagnosed.

[0033] In the Logistic model described in the present invention, the liver cancer progression discriminant analysis described in 3) according to the calculation result of 2) includes: setting 9, preferably 9.58, as the quantitative evaluation standard or threshold of the calculation result to determine whether the patient has liver cancer; that is, when the calculation result of the Risk value described in 2) is greater than 9, preferably greater than 9.58, it indicates that the patient has liver cancer; otherwise, it indicates that the patient does not have liver cancer.

[0034] The present invention is committed to solving the technical challenges in the field of early diagnosis of liver cancer, and provides a biomarker combination consisting of a series of RNA biomarkers and protein biomarkers. By detecting the expression levels of all markers, a detection matrix is ​​formed, and different weight coefficients are assigned to the expression levels of all biomarkers in the matrix, forming a Logistic model for calculating the risk value of liver cancer, and setting an evaluation threshold for the risk value. The present invention applies the biomarker detection matrix, Logistic model and evaluation criteria to the early diagnosis of liver cancer. By detecting the expression levels of all biomarkers in the matrix, and bringing the Logistic model described in the present invention into the calculation of the risk value of liver cancer, and then comparing the risk value with the established threshold, the early diagnosis of liver cancer can be achieved. The biomarker matrix of the present invention provides a powerful tool for the early diagnosis of liver cancer, and can be used for the accurate screening and diagnosis of early liver cancer. The application of this technology will help improve the diagnostic accuracy of liver cancer and provide an important basis for the early treatment of patients.

[0035] In the biomarker matrix described in the present invention, all biomarkers are highly correlated with cancer cell growth and poor prognosis of hepatocellular carcinoma (HCC), but there are differences in their correlations. The expression levels of some biomarkers are positively correlated with liver cancer, while the expression levels of some biomarkers are negatively correlated with liver cancer. Even if they are also positively correlated or negatively correlated, different biomarkers have different degrees of influence on the occurrence and development of liver cancer. Therefore, when only a single or single biomarker is used to carry out early diagnosis of liver cancer in the prior art, there is often a problem of unbalanced sensitivity and specificity of diagnosis, which makes it difficult to diagnose lung cancer as early as possible and accurately. The present invention is different from traditional single marker diagnosis. It establishes a detection matrix of multiple early diagnosis markers based on a series of liver cancer-related RNA and protein markers, which can balance the optimal sensitivity and specificity of tumor detection and construct a more optimized new model for early screening of liver cancer.

[0036] The model building method of the present invention specifically includes:

[0037] Model building steps:

[0038] 1. Clinical liver cancer samples: The information of liver cancer and healthy population cohorts was retrospectively collected from the clinical data database of the Second Affiliated Hospital of Zhejiang University School of Medicine, and their blood, liver, liver cancer and adjacent tissue samples were obtained from the biological sample bank;

[0039] 2. Screen differentially expressed RNA and protein biomarkers (more than 10) through high-throughput transcriptomics and proteomics technologies;

[0040] 3. Combine the patient's clinical information and survival prognosis data to analyze and screen out biomarker profiles with relatively independent value for liver cancer diagnosis and prognostic analysis;

[0041] 4. Use the Logistics regression model to analyze and fit the best diagnostic model.

[0042] Result analysis:

[0043] After fitting different diagnostic models based on different numbers or types of biomarkers, the performance of different models was compared and it was found that the models established solely with biomarkers such as AFP or PIVKA-II that have relatively independent effects on liver cancer were not ideal in terms of sensitivity and specificity, and the two performances were obviously unbalanced; even the models constructed based on multiple biomarkers of the RNA type, or the models constructed based on multiple biomarkers of the protein type, their sensitivity was still not ideal, and they could not achieve a balance between sensitivity and specificity; in contrast, the Logistics model constructed based on a marker combination that included both multiple RNA biomarkers and multiple protein biomarkers showed high performance in both sensitivity and specificity, and the two performances reached an unexpected balance.

[0044] Compared with the prior art, the beneficial effects of the present invention are: based on the mechanism of occurrence and development of liver cancer, a series of RNA and protein biomarkers closely related to liver cancer have been found. By detecting the expression levels of the above markers and using a new diagnostic model for integrated analysis, early diagnosis and accurate evaluation of liver cancer can be achieved. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] Figure 1 This demonstrates the predictive efficacy of a series of iterative liver cancer biomarker diagnostic models of Example 2 of the present invention and Comparative Examples 1 to 4 in early diagnosis of liver cancer in a cohort of liver cancer and healthy subjects. DETAILED DESCRIPTION

[0046] In order to make the purpose and technical solution of the present invention more clear, the present invention is further described in detail below in conjunction with the accompanying drawings. The experimental methods described in the following experiments are all conventional methods unless otherwise specified; if no specific technology or conditions are specified in the experiment, the technology or conditions described in the literature in the field or the product instructions are used; the reagents and materials described can be obtained from commercial channels unless otherwise specified.

[0047] In order to make the purpose and technical solution of the present invention more clear, the present invention is further described in detail below in conjunction with the accompanying drawings. The experimental methods described in the following examples are all conventional methods unless otherwise specified; if no specific techniques or conditions are specified in the examples, they are carried out according to the techniques or conditions described in the literature in the field or according to the product instructions; the reagents and materials, unless otherwise specified, can be obtained from commercial channels.

[0048] The liver cancer early diagnosis product of the present invention is a detection kit and a matching diagnostic model: the kit contains reagents for detecting the expression levels of 6 RNA biomarkers and 4 protein biomarkers, and the diagnostic model includes a liver cancer risk calculation formula and quantitative evaluation criteria for integrated analysis of the content of each marker. Among them, the detection reagent for each RNA biomarker includes a specific RNA probe, and the detection reagent for each protein biomarker includes a specific antibody. RNA probes are complementary fragments of specific nucleotide sequences, and specific antibodies refer to immunoglobulins with high affinity for specific protein biomarkers.

[0049] In addition, the kit also includes a container, instructions for use, positive controls, negative controls, buffers, adjuvants or solvents. The instructions for use describe how to use the kit for testing, how to use the test results for integrated analysis, and how to judge the occurrence and progression of liver cancer. The components of the kit can be packaged in the form of an aqueous medium or in a lyophilized form.

[0050] This kit can be used for accurate screening and diagnosis of early liver cancer.

[0051] The detection methods of the kit include RNA capture probe technology, high-throughput sequencing, and enzyme-linked immunosorbent assay technology, and the analysis methods include Logistics regression model analysis.

[0052] ELISA is an experimental technique for rapidly detecting the content of specific antigens or antibodies in samples. It uses the highly specific antibody-antigen binding in immunology as the reaction basis and has the advantages of being convenient, fast, specific and sensitive, highly stable, not requiring special equipment and reagents, and accurate results. Therefore, it is suitable for rapid screening and accurate diagnosis for the general public.

[0053] High-throughput sequencing technology is a technology that performs parallel sequencing of a large number of nucleic acid molecules in the sample to be tested. It mainly includes steps such as sample preparation, library construction, sequencing by sequencer, and data analysis. First, RNA is extracted from the sample. Secondly, the extracted RNA fragments are connected to sequencing primers to construct a library. Subsequently, with the help of a sequencing instrument, the RNA fragments in the library are subjected to high-throughput sequencing to obtain a large amount of raw sequencing data. Finally, the raw data is subjected to quality control, screening, alignment, assembly, and functional annotation, and finally the expression level results of a series of RNA biomarker molecules are obtained.

[0054] RNA capture probe technology is an experimental technology for detecting the content of specific RNA nucleic acid fragments. It uses a specific nucleotide sequence complementary fragment (probe) to capture the target RNA and couple the probe to the surface of magnetic beads. After the RNA capture probe binds to the specific RNA nucleic acid fragment, the magnetic beads are adsorbed by an external magnetic field to separate the specific RNA nucleic acid fragment, and the obtained high-purity RNA nucleic acid fragment can be used for subsequent high-throughput sequencing detection. The RNA capture probe method can capture high-purity target RNA without DNA enzyme treatment, avoiding the loss of target RNA caused by DNA enzyme treatment, and combined with subsequent high-throughput sequencing detection to ensure the accuracy of the test results. This method is widely used in research fields such as cell biology, biochemistry, and medical genetics.

[0055] The present invention is based on the description of the above embodiments, and specifically lists the following examples and comparative examples.

[0056] Example 1. Early diagnosis kit for liver cancer

[0057] At least the following three groups of reagents are included:

[0058] First reagent module: RNA capture probe reagent

[0059] 6 types of magnetic beads coupled with specific fragments (probes) complementary to the nucleotide sequences of LINC01554, miR-222, SNORD31, circMIRIAF, SNHG8, and RNU5E-1, respectively;

[0060] Also contains: blood sampling tubes, magnetic stand, binding wash solution, eluent and instructions.

[0061] Second reagent module: high-throughput sequencing reagents

[0062] It mainly includes PCR reaction buffer, cluster generation reagent, dNTPs, DNA polymerase, ligase, transposase, reaction termination solution and instruction manual.

[0063] The third reagent module: enzyme-linked immunosorbent assay (ELISA) reagent, containing:

[0064] 4 sets of pre-coated microplates; the wells of the microplates are pre-immobilized with specific antibodies (capture antibodies) for SIGLEC5, CXCL10, AFP and PIVKA-II;

[0065] 4 standards for SIGLEC5, CXCL10, AFP, and PIVKA-II;

[0066] 4 enzyme-labeled secondary antibodies: antibodies homologous to each capture antibody and labeled with horseradish peroxidase;

[0067] Also contains: substrate solution, stop solution, wash buffer, sample diluent, control sample, instructions and sealing film.

[0068] Example 2. Early diagnosis model for liver cancer based on RNA and protein composite biomarkers

[0069] 1) For the clinical population to be diagnosed, collect peripheral venous blood as the sample to be tested.

[0070] 2) Detecting the expression level of biomarkers using the kit described in Example 1

[0071] 2.1) Using the first reagent module (RNA capture probe reagent) and the second reagent module (high-throughput sequencing reagent) to respectively detect the expression levels of LINC01554, miR-222, SNORD31, circMIRIAF, SNHG8 and RNU5E-1 in the sample to be tested; specifically including:

[0072] 1. Sample preparation: Collect no less than 5 mL of peripheral blood into a sampling tube, let stand at room temperature for 15-20 minutes, and centrifuge at 3000 rpm for 10 minutes to obtain serum;

[0073] 2. Add 300 μL of magnetic beads with specific RNA capture probes coupled to the surface into a new test tube, place it on a magnetic rack, remove the supernatant, add 600 μL of 1× binding wash buffer to resuspend, and remove the supernatant on the magnetic rack again;

[0074] 3. Add 600 μL of 1× Binding Wash Buffer to the test tube, and then add 40 μL of the test serum. Mix by gentle inversion, and incubate at room temperature on a vertical rotator for 30 minutes at a slow speed.

[0075] 4. Place the test tube on the magnetic rack, remove the supernatant, add 600 μL 1× binding wash solution (add RNase inhibitor in advance, 1000 U / mL), wash twice, and remove the supernatant;

[0076] 5. Add 200 μL of 1× elution buffer (RNase inhibitor, 1000 U / mL, added in advance) to the test tube, mix by pipetting, let stand at room temperature for 3 minutes, centrifuge at 12000 rpm for 2 minutes, collect the supernatant, and obtain the specific RNA nucleic acid fragment bound to the RNA capture probe;

[0077] 6. Use Nanodrop 2000 spectrophotometer to detect the OD260 and OD280 values ​​of the samples to evaluate the RNA concentration and purity in the samples;

[0078] 7. Perform subsequent high-throughput sequencing on the samples as soon as possible, or store the samples at -80°C.

[0079] 8. Add 200 μL of chloroform to the sample obtained in the previous step, vortex for 30 seconds, and let stand at room temperature for 3 minutes;

[0080] Centrifuge at 14000g at 9.4℃ for 15 minutes. The sample in the tube can be seen to be divided into three layers. Take the top layer of solution where RNA is located and transfer it to a new EP tube.

[0081] 10. Add 500 μL of isopropanol, mix by inversion for 15 seconds, and let stand at room temperature for 10 minutes. Centrifuge at 14000 g for 10 minutes at 4°C, discard the supernatant, add 1 mL of 75% ethanol, resuspend and wash twice;

[0082] 11. After the last centrifugation, discard the supernatant and dry at room temperature until the RNA precipitate becomes transparent;

[0083] 12. Add 30 μL DEPC water to dissolve the RNA precipitate and place the RNA sample on ice for later use;

[0084] 13. Select 200 ng RNA from each sample group, use the TruSeqTM RNA sample preparation kit to prepare the transcriptome library, and enrich and purify the constructed DNA sample by PCR amplification and super agarose gel electrophoresis;

[0085] 14. After quantification by TBS-380 fluorometer, the paired-end RNA-seq sequencing library was sequenced using Illumina NovaSeq 6000 sequencer;

[0086] 15. Use HISAT2 software to align the clean reads with the reference genome in a directed mode, and assemble the data according to the reference method using StringTie software to obtain the expression level results of each RNA sequence.

[0087] 2.2) Using the third reagent module (enzyme-linked immunosorbent assay reagent) to detect the expression levels of SIGLEC5, CXCL10, AFP, and PIVKA-II in the samples respectively; specifically including:

[0088] 1. Add the specific antibody solution (concentration of 0.5-2 μg / mL) diluted with PBS to the experimental well plate, incubate and coat the antibody overnight at 4°C, and wash the well plate with washing buffer (0.05% Tween-20 solution diluted with PBS);

[0089] 2. Add 10 μL of the patient's serum to be tested to the experimental well plate, and add 100 μL of the detection secondary antibody labeled with horseradish peroxidase (HRP), and incubate at 37°C for 60 minutes;

[0090] 3. After incubation, discard the liquid, pat dry with absorbent paper, add 200μL washing solution, let it stand for a while, shake off the washing solution, and pat dry with absorbent paper. Repeat the washing 5 times;

[0091] 4. Add 50 μL of detection substrate and incubate at 37°C in the dark for 15 minutes;

[0092] 5. Determine the OD value of the experimental well plate at a specific wavelength (such as 450nm), and calculate the serum concentration of the corresponding biomarker based on the standard curve formula.

[0093] 3) All RNA and protein biomarker expression level data of each sample obtained in 2) were cleaned to remove missing values, outliers or invalid data, and then subjected to preliminary standardization (converted to uniform units and decimal places) and then entered into the following model calculation formula to calculate the risk value:

[0094] Risk value=2.479×LINC01554-0.216×miR222+19.078×SNORD31-0.097×circMIRIAF+1.562×SNHG8+0.808×RNU5E1-0.003×SIGLEC5+0.962×CXCL10+0.003×PIVKAII+0.063×AFP;

[0095] Define the Logistic regression model, set its loss function to cross entropy loss, and the optimization method to gradient descent;

[0096] Using the standardized RNA and protein biomarker expression levels as input features and liver cancer as the target variable, the model parameters were iteratively optimized to enable the model to accurately predict the probability of liver cancer.

[0097] The parameters of the logistic regression model are tuned by regularization strength, number of iterations, etc. to improve model performance.

[0098] The diagnostic criteria are: when the calculated Risk value is greater than 9.58, it indicates liver cancer.

[0099] The diagnostic sensitivity of the model in this embodiment is 97%, the specificity is 99%, and the area under the ROC curve is 0.992.

[0100] Comparative Example 1. Early diagnosis model of liver cancer based on simple RNA biomarkers

[0101] The method described in reference embodiment 2 is different in that:

[0102] Step 2.2) was not included, that is, only the expression levels of LINC01554, miR-222, SNORD31, circMIRIAF, SNHG8, and RNU5E-1 in the samples were detected;

[0103] The model calculation formula of step 3) is changed to:

[0104] Risk value=4.752×LINC01554-0.004×miR222+7.685×SNORD31-0.082×circMIRIAF+0.694×SNHG8+1.767×RNU5E1;

[0105] The diagnostic criteria are set as follows: when the calculated Risk value is greater than 2.95, it indicates liver cancer;

[0106] Thus, a liver cancer diagnosis model based on pure RNA biomarkers was established;

[0107] The diagnostic sensitivity of the model was 88%, the specificity was 96%, and the area under the ROC curve was 0.927.

[0108] Comparative Example 2. Early diagnosis model for liver cancer based on simple protein biomarkers

[0109] The method described in reference embodiment 2 is different in that:

[0110] Step 2.1) was not included, i.e., only the expression levels of SIGLEC5, CXCL10, AFP, and PIVKA-II in the samples were detected;

[0111] The model calculation formula of step 3) is changed to:

[0112] Risk value=0.086×AFP+0.004×PIVKAII+0.303×CXCL10-0.001×SIGLEC5;

[0113] The diagnostic criteria are: when the calculated Risk value is greater than 3.09, it indicates liver cancer;

[0114] Thus, a liver cancer diagnosis model based on simple protein biomarkers was established;

[0115] The diagnostic sensitivity of the model was 77%, the specificity was 98%, and the area under the ROC curve was 0.903.

[0116] Comparative Example 3. Early diagnosis model of liver cancer using only PIVKAII as a biomarker

[0117] The method described in reference embodiment 2 is different in that:

[0118] Step 2) only detecting the expression level of PIVKA-II in the sample;

[0119] The model calculation formula of step 3) is Risk value = PIVKAII;

[0120] The diagnostic criteria are: when the calculated Risk value is greater than 97.58, it indicates liver cancer;

[0121] Thus, a liver cancer diagnosis model based on a single biomarker, PIVKA-II, was established;

[0122] The diagnostic sensitivity of the model was 66%, the specificity was 84%, and the area under the ROC curve was 0.774.

[0123] Comparative Example 4. Early diagnosis model of liver cancer using only AFP as a biomarker

[0124] The method described in reference embodiment 2 is different in that:

[0125] Step 2) only detecting the expression level of AFP in the sample;

[0126] The model calculation formula of step 3) is Risk value = AFP;

[0127] The diagnostic criteria are: when the calculated Risk value is greater than 11.42, it indicates liver cancer;

[0128] Thus, a liver cancer diagnosis model based on a single biomarker, AFP, was established;

[0129] The diagnostic sensitivity of the model was 59%, the specificity was 94%, and the area under the ROC curve was 0.707.

[0130] Comparison of the diagnostic efficacy of Example 2 of the present invention and Comparative Examples 1-4 shows that Figure 1As shown, the diagnostic model of Example 2 represented by the red line has achieved the best balance in terms of sensitivity and specificity, and the area under the ROC curve is the largest, while the diagnostic models based on multiple biomarkers of a single type represented by the orange and green lines are not sensitive enough, and the diagnostic models based on a single biomarker PIVKA-II and AFP represented by the purple and blue lines have the worst sensitivity.

[0131] The description of the above embodiments is only used to understand the method and core idea of ​​the present invention. It should be pointed out that, for those skilled in the art, several improvements and modifications can be made to the present invention without departing from the principle of the present invention, and these improvements and modifications will also fall within the scope of protection of the claims of the present invention.

Claims

1. A biomarker combination for early diagnosis of liver cancer, characterized in that: It is composed of RNA biomarkers and protein biomarkers; the RNA biomarkers include LINC01554, miR-222, SNORD31, circMIRIAF, SNHG8 and RNU5E-1; the protein biomarkers include SIGLEC5, CXCL10, AFP and PIVKA-II.

2. Use of the biomarker combination according to claim 1 in the preparation of an early diagnosis reagent for liver cancer.

3. The use according to claim 2, characterized in that: The early diagnosis reagent for liver cancer is used to detect the expression level of each substance in the biomarker combination of claim 1 in a sample from a subject.

4. The use according to claim 3, characterized in that: The sample from the subject is any one or a combination of two or more of the subject's peripheral venous blood, paracancerous tissue or tumor tissue; preferably peripheral venous blood.

5. The use according to claim 3, characterized in that: The liver cancer early diagnosis reagent is a detection reagent based on RNA capture probe technology, high-throughput sequencing, and enzyme-linked immunosorbent assay technology.

6. A kit comprising a reagent for detecting the expression level of a biomarker and one or more substances selected from the group consisting of a container, instructions for use, a positive control, a negative control, a buffer, an adjuvant or a solvent; characterized in that: The reagent for detecting the expression level of biomarkers includes: an RNA probe and a specific antibody; the RNA probe is composed of complementary fragments of specific nucleotide sequences of LINC01554, miR-222, SNORD31, circMIRIAF, SNHG8 and RNU5E-1; and the specific antibody is composed of specific antibodies of SIGLEC5, CXCL10, AFP and PIVKA-II.

7. A Logistic model for early diagnosis of liver cancer based on the biomarker combination of claim 1, comprising: 1) obtaining valid data of a sample of an individual to be diagnosed; the valid data comprising the expression level of each biomarker in the biomarker combination of claim 1 in the sample; 2) Substitute all the expression level data obtained in 1) into the following model calculation formula to calculate the risk value: Risk value=2.479×LINC01554-0.216×miR222+19.078×SNORD31- 0.097×circMIRIAF+1.562×SNHG8+0.808×RNU5E1- 0.003×SIGLEC5+0.962×CXCL10+0.003×PIVKAII+0.063×AFP; Wherein, the LINC01554, miR222, SNORD31, circMIRIAF, SNHG8, RNU5E1, SIGLEC5, CXCL10, PIVKAII and AFP respectively represent their respective expression level values ​​in the samples, and all values ​​have the same unit; 3) Based on the calculation results of 2), the liver cancer progression discrimination analysis was performed according to the quantitative evaluation criteria; 4) Output early diagnosis results of liver cancer.

8. The model according to claim 7, characterized in that: 1) The sample is any one or a combination of two or more of the samples of blood, liver, liver cancer and adjacent tissues from the individual to be diagnosed.

9. The model according to claim 7, characterized in that: 3) The liver cancer progression discrimination analysis according to the calculation result of 2) includes: setting 9, preferably 9.58, as the quantitative evaluation standard or threshold of the calculation result to determine whether the patient has liver cancer; that is, when the calculation result of the Risk value of 2) is greater than 9, preferably greater than 9.58, it indicates that the patient has liver cancer; otherwise, it indicates that the patient does not have liver cancer.

Images