Diagnostic marker for hepatic cancer development in chronic hepatic disease

Abstract

The method of the present invention for evaluating a subject's risk of hepatic cancer development includes: (1) a step for measuring the GDF15 level of the subject; and (2) a step for relating the GDF15 level to the risk of hepatic cancer development. When the GDF15 level measured in step (1) is higher than a preset cutoff value, then it is indicated that the subject has a high risk of hepatic caner development. When the GDF15 level is lower than the cutoff value, then it is indicated that the subject has a low risk of hepatic cancer development. The present invention further provides a kit for measuring the GDF15 level of a subject to be used in the method of the present invention, and a diagnostic reagent containing anti-GDF15 specific antibody for evaluating a risk of hepatic cancer development in a subject.

Description

Diagnostic markers for hepatocarcinogenesis in chronic liver disease

[0001] The present invention relates to a method for assessing the risk of developing a disease using a diagnostic marker for the disease, and more particularly to a method for assessing the risk of developing hepatocarcinogenesis in chronic liver disease.

[0002] Various chronic liver diseases, such as chronic hepatitis B and C, and non-alcoholic steatohepatitis, can cause liver cancer (Non-Patent Documents 1 and 2). Hepatocellular damage is observed in various chronic liver diseases, and oxidative stress is involved in liver carcinogenesis due to persistent hepatocellular damage (Non-Patent Documents 3 and 4). Chronic liver diseases are classified into viral chronic liver diseases and non-viral chronic liver diseases. Viral chronic liver diseases are mainly classified into hepatitis C caused by hepatitis C virus and hepatitis B caused by hepatitis B virus. Non-viral chronic liver diseases are classified into fatty liver disease and autoimmune liver disease.

[0003] Hepatitis C: The emergence of direct-acting antivirals (DAAs), a novel treatment for hepatitis C, has led to the elimination of hepatitis C virus (HCV) in many cases. Even after hepatitis C virus elimination, or more precisely, even after achieving sustained virological response (SVR), there is still a risk of hepatocarcinogenesis (HCC) (Non-Patent Document 5). Therefore, many hepatitis C cases undergo periodic imaging and blood tests after completing treatment. However, with the recent spread of DAAs treatment, the number of cases achieving SVR has increased, and uniform periodic testing for all SVR-achieving cases presents a medical economic challenge. Therefore, it is necessary to stratify HCC risk and prioritize testing for populations at high risk. Tumor markers used in blood tests include AFP (alpha-fetoprotein), the FIB-4 index (also written as Fib-4, FIB4, or Fib4; Non-Patent Documents 6-8), which is calculated from platelets and age, and other factors, and are used to predict cancer development. However, their diagnostic ability is insufficient, and more precise stratification techniques are needed.

[0004] As of 2019, there were an estimated 296 million HBs antigen-positive hepatitis B patients worldwide, with approximately 820,000 dying annually from cirrhosis or liver cancer caused by the hepatitis B virus (HBV). While the risk of liver cancer can be reduced by lowering serum HBV DNA levels with nucleoside analogues (NUCs), cases of liver cancer have been observed even in patients with low HBV DNA levels (Non-Patent Document 9). The incidence of liver cancer after HBs antigen disappearance is 0.0368 / year. On the other hand, the incidence of liver cancer in patients with persistently positive HBs antigen levels is 0.1957 / year, a significantly higher incidence (Non-Patent Document 10). Factors that increase the risk of liver cancer include age (over 40 years old), sex (male), high viral load, alcohol consumption, family history of liver cancer, co-infection with HCV, HDV, and / or HIV, progression of liver fibrosis, decreased platelet count reflecting progression of liver fibrosis, genotype C, and core promoter mutations (Non-Patent Document 11). However, even when these factors are taken into consideration, it is difficult to predict the onset of liver cancer under NUC administration, and new liver carcinogenesis markers are clinically important.

[0005] Fatty Liver Disease Most non-viral liver diseases are fatty liver disease (so-called fatty liver). Fatty liver disease refers to a disease in which fat accumulates in 5% or more of hepatocytes. Fatty liver disease is further classified into non-alcoholic fatty liver disease (NAFLD) and secondary fatty liver. Here, non-alcoholic refers to an alcohol intake of less than 30 g / day for men and less than 20 g / day for women, calculated as ethanol. NAFLD is further classified into non-alcoholic fatty liver (NAFL), which is not accompanied by histological hepatocellular damage, and non-alcoholic steatohepatitis (NASH), which is accompanied by histological hepatocellular damage and inflammation. Secondary fatty liver is classified into alcohol-related, drug-related (amiodarone, methotrexate, tamoxifen, steroids, valproic acid, antiretroviral drugs, etc.), disease-related (hepatitis C (genotype 3), Wilson disease, lipodystrophy, starvation, parenteral nutrition, Reye syndrome, acute fatty liver of pregnancy, HELLP syndrome), congenital metabolic disorders (abetalipoproteinemia, hemochromatosis, alpha-1 antitrypsin deficiency, lecithin cholesterol acyltransferase deficiency, lysosomal lipase deficiency, etc.), and others (post-pancreaticoduodenectomy).

[0006] It is estimated that, assuming the population of Japan is 127 million, approximately 30% of that population, or 30-40 million people, suffer from NAFLD, and approximately 10% of those suffer from NASH. This number of patients is greater than that of hepatitis C (2 million), hepatitis B (1 million), and alcoholic hepatitis (2 million), and is roughly equivalent to or exceeds the total number of viral hepatitis patients.

[0007] While some NAFL cases progress slowly to fibrosis, some NASH cases progress to cirrhosis and hepatocarcinogenesis. However, it is known that there is a reciprocal transition between the two (Non-Patent Documents 12 and 13). A comparative study of prognosis has shown that the progression of fibrosis is more important for prognosis than whether the condition is NAFL or NASH (Non-Patent Document 14). Another comparative study of prognosis has shown that cases with advanced liver fibrosis have a poor prognosis (Non-Patent Document 15). A study of the causes of primary liver cancer in Japan from 1995 to 2015 revealed an increase in hepatocarcinogenesis due to non-viral liver disease (Non-Patent Document 16).

[0008] The usefulness of the FIB-4 index as a predictive marker for the development of liver cancer from non-viral liver diseases has also been reported (Gastroenterology. 2018 Dec;155(6):1828-1837.e2), but its diagnostic ability is insufficient, and further stratification techniques are needed.

[0009] Oxidative stress induces C>A / G>T gene mutations (Non-Patent Document 17), and this mutation pattern is observed more frequently in liver cancer than in other cancers (Non-Patent Document 18), suggesting the involvement of oxidative stress in liver carcinogenesis. Growth differentiation factor 15 (GDF15), a member of the TGF-β superfamily, is upregulated in response to oxidative and mitochondrial stress (Non-Patent Documents 19-22). GDF15 levels are elevated in cases of advanced liver fibrosis, and high GDF15 levels have been reported as a poor prognostic factor in liver cancer cases (Non-Patent Document 23). However, it is not known whether GDF15 is associated with chronic liver diseases, including liver cancer development after SVR in hepatitis C, liver cancer development under NUC administration in hepatitis B, and liver cancer development from NASH.

[0010] Non-patented publication 1: Sagnelli E, et al. Infection. 2020 Feb;48(1):7-17. Non-patented publication 2: Zhang CH, et al. Liver Int. 2022 doi: 10.1111 / liv.15251. Non-patented publication 3: Hikita H, et al. J Hepatol. 2012 Jul;57(1):92-100. Non-patented publication 4: Hikita H, et al. Cancer Prev Res (Phila). 2015 Aug;8(8):693-701. Non-patented publication 5: Janjua NZ et al., J Hepatol 2017;66:504-513. Non-patented publication 6: Watanabe T et al., J Med Virol 2020;92:3507-3515. Non-patented publication 7: Kanwal F, Singal AG, Gastroenterology 2019;157:54-64. Non-patented publication 8: Nagata H et al., J Hepatol 2017;67:933-939. Non-patented publication 9: Liaw YF, et al. N Engl J Med. 2004;351:1521-1531. Non-patented publication 10: Simonetti J, et al. Hepatology. 2010;51:1531-1537. Non-patented publication 11: Yim HJ, Lok AS. HEPATOLOGY 2006;43:S173-S181. Non-patented publication 12: Tokushige, K, et al. Hepatology Research. 2021;51:1013-1025. Non-patented reference 13: Yoshiji, H. et al. J Gastroenterol, 2021;56:593-619. Non-patented reference 14: Angulo P, et al. Gastroenterology. 2015;149:389-97. Non-patented reference 15: Hagstrom H, et al. J Hepatol. 2017;67:1265-1273. Non-patented reference 16: Tateishi R, et al. J Gastroenterol. 2019;54:367-376. Non-patented reference 17: van Loon B, et al.DNA Repair (Amst) 2010;9:604-16 Non-patent document 18: Guichard C, et al. Nat Genet 2012;44:694-8. Non-patent document 19: Han ES, et al. Physiol Genomics. 2008;34:112-26. Non-patent document 20: Tsai VWW et al., Cell Metab 2018;28:353-368. Non-patent document 21: Kim J et al., Nat Metab 2021;3:410-427. Non-patent document 22: Kang SG et al., iScience 2021;24:102181 Non-patent document 23: Myojin Y et al., Gastroenterology 2021;160:1741-1754.

[0011] An object of the present invention is to provide a novel biomarker for predicting the development of liver cancer from chronic liver disease.A further object of the present invention is to provide a novel technology for risk assessment and / or stratification of liver carcinogenesis from chronic liver disease, including, but not limited to, liver carcinogenesis after SVR in hepatitis C, liver carcinogenesis under NUC administration in hepatitis B, and liver carcinogenesis from NAFLD, comprising the novel biomarker.

[0012] In order to achieve the above-mentioned object, the inventors have conducted further research and discovered that carcinogenesis from chronic liver disease can be predicted based on the level of GDF15 protein in serum or the level of GDF15 transcripts in liver tissue, thereby completing the present invention.

[0013] The present invention provides a method for assessing a subject's risk of developing liver cancer, comprising the steps of: (1) measuring the GDF15 level in the subject; and (2) correlating the GDF15 level with the risk of developing liver cancer.

[0014] In the method of the present invention, the subject may be at least one type of subject selected from the group consisting of subjects who have achieved sustained negativity (SVR) for hepatitis C virus (HCV), subjects receiving NUC for hepatitis B, and subjects who have developed NAFLD.

[0015] In the method of the present invention for assessing a subject's risk of developing liver cancer, if the subject's GDF15 level is equal to or higher than a predetermined cutoff value, it is an indicator that the subject has a high risk of developing liver cancer, and if it is below the cutoff value, it is an indicator that the subject has a low risk of developing liver cancer.

[0016] In the method of the present invention for assessing a subject's risk of developing liver cancer, the GDF15 level measured in step (1) can be the level of GDF15 protein in serum or plasma and / or the level of GDF15 transcript in liver tissue or circulating blood.

[0017] In the method of the present invention for assessing the risk of developing liver cancer in a subject, the cutoff value may be set based on statistical analysis or ROC analysis of the GDF15 level.

[0018] In the method of the present invention for assessing the risk of developing liver cancer in a subject, the level of GDF15 measured in step (1) may be the level of GDF15 protein in serum.

[0019] In the method of the present invention for assessing a subject's risk of developing liver cancer, the cutoff value may be the median GDF15 protein level of the subject for each individual chronic liver disease, or may be determined from the ROC curve for each individual chronic liver disease.

[0020] In the method for assessing a subject's risk of developing liver cancer of the present invention, the cutoff value for the serum GDF15 protein level can be about 1400 pg / mL for hepatitis C patients who have achieved SVR, about 845 pg / mL for hepatitis B patients receiving NUC, and about 2000 pg / mL for NAFLD patients.

[0021] In the method of the present invention for assessing the risk of developing liver cancer in a subject, the level of GDF15 protein in the serum can be determined by ELISA.

[0022] In the method of the present invention, in the step of assessing the risk of developing liver cancer, the cutoff values ​​of AFP and FIB-4 index can be further combined for assessment.

[0023] In the method of the present invention, the cutoff values ​​of the AFP and FIB-4 index may be about 5 ng / mL and about 3.25, respectively, for subjects who have achieved SVR for hepatitis C and subjects receiving NUC for hepatitis B. For subjects who have developed NAFLD, the cutoff values ​​may be about 5 ng / mL and about 2.67, respectively.

[0024] The present invention provides a kit for measuring GDF15 levels in a subject for use in the method of the present invention, which comprises an anti-GDF15 antibody and / or a primer pair or probe for specifically detecting a GDF15 transcript.

[0025] The present invention provides a diagnostic agent for assessing a subject's risk of developing liver cancer by the method of the present invention. The diagnostic agent of the present invention comprises an anti-GDF15 antibody and / or a primer pair and a probe for specifically detecting a GDF15 transcript.

[0026] The present invention provides a method for using GDF15 as a biomarker for assessing a subject's risk of developing liver cancer, comprising the steps of: (1) measuring the subject's GDF15 level; and (2) correlating the GDF15 level with the risk of developing liver cancer.

[0027] In the use of GDF15 of the present invention as a biomarker for assessing a subject's risk of developing liver cancer, the subject may be at least one type of subject selected from the group consisting of subjects who have achieved sustained negativity (SVR) for hepatitis C virus (HCV), subjects receiving NUC for hepatitis B, and subjects who have developed NAFLD.

[0028] The present invention provides a method for screening for liver cancer in stratified subjects, comprising testing subjects assessed as having a high risk of liver cancer by the method of the present invention for assessing the risk of liver cancer in a subject at a higher frequency than subjects assessed as having a low risk of liver cancer to determine whether or not they have liver cancer. The presence or absence of liver cancer may be determined based on findings from ultrasound, contrast-enhanced CT images, MRI, and / or liver biopsy tissue. For subjects assessed as having a high risk of liver cancer, the presence or absence of liver cancer may be determined based on findings from ultrasound, contrast-enhanced CT images, and MRI. For subjects assessed as having a low risk of liver cancer, the test for determining whether or not they have liver cancer may not be performed, or only blood tumor marker tests, including those used in the method of the present invention for assessing the risk of liver cancer in a subject, may be performed at regular intervals. Here, examples of blood tumor markers include, in addition to GDF15, AFP (alpha-fetoprotein), AFP-L3 (LCA (lentil lectin) strong binding fraction), and PIVKA-II (protein induced by vitamin K absence or antagonist II), but are not limited to these.

[0029] The present invention provides a method for preventing liver cancer, which comprises administering a prophylactic agent for liver cancer to a subject who has been evaluated as having a high risk of developing liver cancer by the liver cancer screening method of the present invention.The present invention also provides a pharmaceutical agent for preventing the development of liver cancer in a subject who has been evaluated as having a high risk of developing liver cancer by the liver cancer screening method of the present invention.

[0030] In the method for screening for liver cancer for hepatitis C of the present invention, the subject is required to maintain a blood pressure of 25 kg / m ２ The number of people eligible for the program can be limited to those with a BMI below 100.

[0031] According to the present invention, it is possible to evaluate the risk of developing liver cancer with high accuracy based on the GDF15 level of a subject, thereby enabling the implementation of a liver cancer screening method for stratified subjects, which involves examining the presence or absence of liver cancer by tests with different frequencies and / or contents depending on the risk of developing liver cancer.

[0032] Table showing case details of the derivation and validation cohorts of hepatitis C patients who achieved SVR. Pyramid graphs showing the distribution of GDF15 levels in stored serum collected at each time point for the derivation cohort of hepatitis C patients who achieved SVR. The vertical axis represents serum GDF15 levels (pg / mL). The left graph shows the distribution of GDF15 levels in serum collected before treatment (Pretreatment), the middle graph shows the distribution of GDF15 levels in serum collected at the end of treatment (End of Treatment), and the right graph shows the distribution of GDF15 levels in serum collected 24 weeks after SVR was achieved (Post 24 weeks). An asterisk (*) indicates a significant difference of p<0.0001 by Turkey Kramer test between the three graphs in Figure 2-1. This graph shows the correlation between GDF15 mRNA levels in preserved liver tissue before DAA treatment and GDF15 levels in preserved serum of hepatitis C patients who achieved SVR. The vertical axis represents GDF15 levels in serum (pg / mL), and the horizontal axis represents relative GDF15 mRNA levels (arbitrary units, AU). This table shows case details for the high GDF15 group and the low GDF15 group of hepatitis C patients who achieved SVR. This pyramid graph shows the distribution of serum GDF15 levels by fibrosis score group in hepatitis C patients who achieved SVR. The vertical axis represents serum GDF15 levels (pg / mL). From the left, the distribution of serum GDF15 levels for each fibrosis score group (F0-F4) is shown. An asterisk (*) indicates a significant difference of p<0.0001 between the graphs in Figure 4-1 based on a test for linear trend after one-way ANOVA. Graph showing the correlation between serum GDF15 levels and FIB-4 index values ​​in hepatitis C patients who achieved SVR. The vertical axis represents serum GDF15 levels (pg / mL), and the horizontal axis represents FIB-4 index values. Graph showing the correlation between serum GDF15 levels and various parameters in hepatitis C patients who achieved SVR.The horizontal axis represents serum GDF15 levels (pg / mL), and the vertical axis represents age (A), hemoglobin (B), platelet count (C), AST (D), ALT (E), γGTP (F), eGFR (G), albumin (H), prothrombin time (I), AFP (J), and ALBI score (K). A table showing details of derivation cohort cases of hepatitis C patients who achieved SVR. A graph showing the change in liver cancer incidence over time in the derivation cohort of hepatitis C patients who achieved SVR. The vertical axis represents the cumulative liver cancer incidence rate, and the horizontal axis represents the observation period (months). A pyramid graph showing the distribution of serum GDF15 levels according to the presence or absence of liver cancer onset at three time points: before treatment (Pretreatment), at the end of treatment (End of Treatment), and 24 weeks after SVR was achieved (Post 24 weeks) in hepatitis C patients who achieved SVR. The vertical axis shows serum GDF15 levels (pg / mL). From the left, the distribution of serum GDF15 levels for groups with and without liver cancer at three time points: pre-treatment (Pretreatment), end-of-treatment (End-of-Treatment), and 24 weeks after SVR was achieved (Post-24 weeks) is shown. An asterisk (*) indicates a significant difference of p<0.005 by Turkey Kramer test between the graphs for the presence and absence of liver cancer at the time of Figure 6-2. Graph showing the changes in serum GDF15 levels one year before and at the time of liver cancer onset for each case of liver cancer in hepatitis C patients who achieved SVR. The vertical axis shows serum GDF15 levels (pg / mL), and the horizontal axis shows the observation period, including one year before liver cancer onset (-1 year) and the time of liver cancer onset (End of Observation). Graph showing cumulative liver cancer incidence in the high GDF15 group and low GDF15 group of hepatitis C patients who achieved SVR. The vertical axis represents cumulative liver cancer incidence, and the horizontal axis represents observation period (months). The arrows labeled "GDF15 HIGH" and "GDF15 LOW" indicate graphs showing changes in cumulative liver cancer incidence over time in the high GDF15 group and low GDF15 group, respectively. Table showing hazard ratios after liver cancer onset in a derived cohort of hepatitis C patients who achieved SVR. Graphs showing ROC curves for predicting liver cancer onset using GDF15 (A), AFP (B), and FIB-4 index (C) in hepatitis C patients who achieved SVR.(D) shows the area under the curve (AUC) of the ROC curve. (E) shows the cutoff value, sensitivity, and specificity calculated from the ROC curve of liver cancer onset prediction for GDF15, AFP, and FIB-4 index. A is a graph showing the change over time in the cumulative liver cancer incidence rate in the high AFP group and the low AFP group in hepatitis C patients who achieved SVR. B is a graph showing the change over time in the cumulative liver cancer incidence rate in the high FIB-4 index group and the low FIB-4 index group in hepatitis C patients who achieved SVR. In both A and B, the vertical axis represents the cumulative liver cancer incidence rate, and the horizontal axis represents the observation period (months). The arrows labeled "AFP HIGH," "AFP LOW," "FIB4-index HIGH," and "FIB4-index LOW" indicate graphs showing the time-dependent changes in cumulative liver cancer incidence in the high AFP group, low AFP group, high FIB-4 index group, and low FIB-4 index group, respectively. The graphs show the time-dependent changes in cumulative liver cancer incidence in the high-risk group (3 points), medium-risk group (1-2 points), and low-risk group (0 points) in hepatitis C patients who achieved SVR, stratified by a scoring system in which high values ​​of GDF15, AFP, and FIB-4 index were assigned a score of 1. The vertical axis represents the cumulative liver cancer incidence, and the horizontal axis represents the observation period (months). The arrows labeled "High," "Middle," and "Low" indicate graphs showing the time-dependent changes in cumulative liver cancer incidence in the high-risk group, medium-risk group, and low-risk group, respectively. Table showing details of cases in the validation cohort of hepatitis C patients who achieved SVR. Graph showing the change over time in cumulative liver cancer incidence in the validation cohort of hepatitis C patients who achieved SVR. The vertical axis represents the cumulative liver cancer incidence, and the horizontal axis represents the observation period (months). A is a graph showing the change over time in cumulative liver cancer incidence in the high GDF15 group and the low GDF15 group in hepatitis C patients who achieved SVR. B is a graph showing the change over time in cumulative liver cancer incidence in the high AFP group and the low AFP group in hepatitis C patients who achieved SVR. C is a graph showing the change over time in cumulative liver cancer incidence in the high FIB-4 index group and the low FIB-4 index group in hepatitis C patients who achieved SVR. In all of A, B, and C, the vertical axis represents the cumulative liver cancer incidence, and the horizontal axis represents the observation period (months).The arrows labeled "GDF15 HIGH," "GDF15 LOW," "AFP HIGH," "AFP LOW," "FIB4-index HIGH," and "FIB4-index LOW" indicate graphs showing the time-dependent changes in cumulative liver cancer incidence in the high GDF15 group, low GDF15 group, high AFP group, low AFP group, high FIB-4 index group, and low FIB-4 index group, respectively. The graphs show the time-dependent changes in cumulative liver cancer incidence in the high-risk group (3 points), medium-risk group (1-2 points), and low-risk group (0 points) in hepatitis C patients who achieved SVR according to the scoring system of the present invention. The vertical axis represents the cumulative liver cancer incidence, and the horizontal axis represents the observation period (months). The arrows labeled "High," "Middle," and "Low" indicate graphs showing the time-dependent changes in cumulative liver cancer incidence in the high-risk group, medium-risk group, and low-risk group, respectively. Graph showing the time course of cumulative liver cancer incidence rates in the high-risk group (2 points), medium-risk group (1 point), and low-risk group (0 point) according to the scoring system of the present invention for a derived cohort of hepatitis C patients who achieved SVR. The vertical axis represents the cumulative liver cancer incidence rate, and the horizontal axis represents the observation period (weeks). The "High", "Middle", and "Low" arrows indicate the time course of cumulative liver cancer incidence rates in the high-risk group, medium-risk group, and low-risk group, respectively. Graph showing the time course of cumulative liver cancer incidence rates in the high- and low-GDF15 level groups in groups with low values ​​of both the known markers AFP and FIB-4 index for a derived cohort of hepatitis C patients who achieved SVR. The vertical axis represents the cumulative liver cancer incidence rate, and the horizontal axis represents the observation period (weeks). The "High" arrow represents a graph showing the change over time in the cumulative incidence of liver cancer in a group with low AFP and FIB-4 index but high GDF15 levels. The "Low" arrow represents a graph showing the change over time in the cumulative incidence of liver cancer in a group with low AFP and FIB-4 index and low GDF15 levels. This graph shows the change over time in the cumulative incidence of liver cancer in a group with high and low GDF15 levels in a derived cohort of hepatitis C patients who achieved SVR, in a group with high values ​​of the known markers AFP and FIB-4 index. The vertical axis represents the cumulative incidence of liver cancer, and the horizontal axis represents the observation period (weeks).The "High" arrow represents a graph showing the change in cumulative liver cancer incidence over time in a group with high AFP and FIB-4 index values ​​and high GDF15 levels. The "Low" arrow represents a graph showing the change in cumulative liver cancer incidence over time in a group with high AFP and FIB-4 index values ​​but low GDF15 levels. Scatter plot of preserved serum from hepatitis B cases receiving NUC, selected according to criteria. Black circles represent non-carcinogenic cases, and white circles represent carcinogenic cases. The vertical axis represents the GDF15 concentration (ng / mL) in the preserved serum. A table showing patient background data for the entire cohort of hepatitis B cases receiving NUC. A graph showing the change in liver cancer incidence over time in the entire cohort of hepatitis B cases receiving NUC. The vertical axis represents the carcinogenicity rate, and the horizontal axis represents the observation period (days). This table shows the patient background of the entire cohort of hepatitis B cases receiving NUC, grouped by the median serum concentration of GDF15 (0.833 ng / mL). This table shows the patient background of the entire cohort of hepatitis B cases receiving NUC, grouped by the presence or absence of liver cancer. This graph shows the results of a receiver operating characteristic (ROC) curve analysis of the presence or absence of carcinogenesis over time for GDF15, Fib4, AFP, and Plt 5 years after each stored serum point in the entire cohort of hepatitis B cases receiving NUC. The vertical axis of each graph represents sensitivity or true positive rate, and the horizontal axis represents false positive rate (1-specificity), and AUC represents the area under the ROC curve of each graph. This graph shows the results of a receiver operating characteristic (ROC) curve analysis of the presence or absence of carcinogenesis over time for GDF15, Fib4, AFP, and Plt over 10 years from the respective stored serum points in the entire cohort of hepatitis B cases receiving NUC. The vertical axis of each graph represents sensitivity or true positivity, and the horizontal axis represents false positivity (1-specificity), with AUC representing the area under the ROC curve for each graph. This graph shows the change in the incidence of liver cancer over time in the entire cohort of hepatitis B cases receiving NUC, using the cutoff value (0.845 ng / mL) calculated from the ROC curve. This table shows the results of univariate / multivariate analysis of factors contributing to carcinogenesis using the Cox proportional hazards model.Graph showing the change in liver cancer incidence over time, with cases grouped by score, with one point assigned to cases with cutoff values ​​of 5 ng / mL and 0.845 ng / mL or more for two markers, AFP and GDF15. Scatter plot of GDF15 serum concentrations in patients with NAFL or NASH. Black circles represent cases without liver cancer, and white circles represent cases with liver cancer. Five of the six cases of liver cancer were primary hepatocellular carcinoma (HCC), but the one case indicated by the arrow was cholangiocarcinoma (CCC). Table showing the patient background of patients with NAFL or NASH. Table showing the patient background of patients grouped into NAFL and NASH. Graph showing the change in liver cancer incidence over time in the entire cohort of patients with NAFL or NASH. Scatter plot of GDF15 serum concentrations for cases grouped according to liver fibrosis Brunt Stage 0 to 4. Scatter plot examining the correlation between GDF15 serum concentrations and Fib-4 index in patients with NAFL or NASH. Table showing hazard ratios for various attributes, blood markers, Fib-4 index, etc., of patients with NAFL or NASH. Graph showing the results of receiver operating characteristic (ROC) curve analysis of GDF15 and Fib-4 index over time to determine the presence or absence of carcinogenesis 5 years after each stored serum point. Graph showing the change in liver cancer incidence over time using a cutoff value of 2.00 ng / mL. Graph showing the change in liver cancer incidence over time using a cutoff value of 1.35 ng / mL. Patient background of 183 cases at Ogaki Municipal Hospital. Patient background of 170 cases in the cohort of Example 3, in which the observation period was extended. Scatter plot of preserved serum from 183 patients with NAFL or NASH at Ogaki Municipal Hospital. Black circles represent non-cancer cases, and gray circles represent cancer cases. The vertical axis represents the concentration of GDF15 (ng / mL) in the preserved serum. All nine cases of liver cancer that developed were primary hepatocellular carcinoma (HCC). Scatter plot of preserved serum from 170 cases at Ogaki Municipal Hospital, in which the observation period was extended. Black circles represent non-cancer cases, and gray circles represent cancer cases. The vertical axis represents the concentration of GDF15 (ng / mL) in the preserved serum. Seven of the eight cases of liver cancer that developed were primary hepatocellular carcinoma (HCC), but the one case indicated by the arrow was cholangiocellular carcinoma (CCC). Liver cancer incidence rate among 183 cases at Ogaki Municipal Hospital. Liver cancer incidence in the cohort of Example 3 with an extended observation period.Graph showing the results of a time-course ROC (Receiver Operating Characteristic) curve analysis performed on a total of 353 cases from the Ogaki Municipal Hospital cohort and the cohort of Example 3 with an extended observation period to determine whether or not carcinogenesis occurred 5 years from the stored serum points. Graph showing the results of a time-course ROC (Receiver Operating Characteristic) curve analysis performed on a total of 353 cases from the Ogaki Municipal Hospital cohort and the cohort of Example 3 with an extended observation period to determine whether or not carcinogenesis occurred 7 years from the stored serum points. Graph showing the change in liver cancer incidence rate over time for a total of 353 cases from the Ogaki Municipal Hospital cohort and the cohort of Example 3 with an extended observation period, with a cutoff value of 2.00 ng / mL. Graph showing the change in liver cancer incidence rate over time for a total of 353 cases from the Ogaki Municipal Hospital cohort and the cohort of Example 3 with an extended observation period, with a cutoff value of 1.74 ng / mL.

[0033] Definition: In the present invention, GDF15 is a cytokine belonging to the transforming growth factor (TGF) beta superfamily, a protein consisting of 308 amino acids in total. It is highly expressed in the placenta but weakly expressed in normal tissues other than the placenta. It is rapidly upregulated during inflammation. The amino acid sequence of human GDF15 protein and the nucleotide sequence of GDF15 mRNA have been published as NCBI Reference Sequences: NP_004855.2 and NM_004864.4, respectively, and can be isolated by methods known per se.

[0034] In the present invention, the GDF15 level refers to the level of GDF15 protein and / or GDF15 transcript. The GDF15 protein and GDF15 transcript levels refer to the content of GDF15 protein and GDF15 mRNA, respectively, in a certain amount of sample. In the present invention, the biological species of the GDF15 protein and GDF15 transcript are the same as the biological species of the subject. The levels of GDF15 protein and GDF15 transcript can be expressed in a manner known to those skilled in the art according to the measurement method described below, for example, they may be expressed as the concentration of GDF15 protein and GDF15 transcript, or may be expressed as a relative value based on the measured value of a standard sample.

[0035] Immunological techniques based on antibodies specific to GDF15 protein are used to measure GDF15 protein levels. Methods that can be used to measure GDF15 protein levels include antibody arrays, flow cytometry analysis, radioimmunoassays (RIAs), ELISA (Engvall E, Methods in Enzymol. 1980;70:419-439), Western blotting, immunohistochemical staining, enzyme immunoassays (EIA), fluorescent immunoassays (FIAs), immunochromatography, immunoturbidimetry, and immunonephelometry. However, ELISA is preferred from the standpoints of sensitivity and ease of implementation. Details of ELISA are provided in the Examples.

[0036] Methods that can be used to measure GDF15 transcript levels include Northern blotting, RNase protection assay, reverse transcription polymerase chain reaction (RT-PCR) (Weis JH et al., Trends in Genetics 1992;8:263-264), and quantitative real-time RT-PCR (Held CA et al., Genome Research 1996;6:986-994). However, quantitative real-time RT-PCR is preferred from the viewpoints of sensitivity and ease of implementation. Details of quantitative real-time RT-PCR will be described in the Examples.

[0037] The subject in the present invention may be any mammal, but is preferably a mammal with chronic liver disease. Examples of mammals include rodents such as mice, rats, hamsters, and guinea pigs, laboratory animals such as rabbits, pets such as dogs and cats, livestock such as cows, pigs, goats, horses, and sheep, primates such as monkeys, orangutans, and chimpanzees, and humans, with humans being particularly preferred. Here, chronic liver disease includes, but is not limited to, viral hepatitis and fatty liver disease. Viral hepatitis includes, but is not limited to, hepatitis C and hepatitis B. Fatty liver disease includes, but is not limited to, non-alcoholic fatty liver disease (NAFLD) and secondary fatty liver. NAFLD includes, but is not limited to, non-alcoholic fatty liver (NAFL) and non-alcoholic steatohepatitis (NASH). The subject in the present invention includes patients with chronic liver disease who need to be evaluated for the risk of developing liver cancer. Subjects of the present invention include, but are not limited to, subjects after SVR for hepatitis C, subjects receiving NUC for hepatitis B, and subjects suffering from NASH.

[0038] In the present invention, chronic liver disease and tests for determining whether or not chronic liver disease has led to the development of liver cancer are described in detail in, for example, Non-Patent Documents 3 and 4.

[0039] In the present invention, sustained HCV negativity (SVR), treatment to achieve SVR, and tests to check for the onset of liver cancer after achieving SVR are based on the Hepatitis C Treatment Guidelines (edited by the Hepatitis Clinical Practice Guidelines Committee of the Japan Society of Hepatology, 8th edition, published in July 2020 (https: / / www.jsh.or.jp / lib / files / medical / guidelines / jsh_guidlines / C_v8_20201005.pdf) and the English version (Hepatology Research 2020; 50: 791-816.) and the Liver Cancer Treatment Guidelines (edited by the Japan Society of Hepatology, 2017 edition, published in October 2017 (https: / / www.jsh.or.jp / medical / guidelines / jsh_guidlines / medical / examination_jp_2017.html), English version (https: / / www.jsh.or.jp / English / examination_en / guidelines_hepatocellular_carcinoma_2017.html)), Ghany MG, and Morgan TR. (Hepatitis C Guidance 2019 Update: American Association for the Study of Liver Diseases-Infectious Diseases Society of America Recommendations for Testing, Managing, and Treating Hepatitis C Virus Infection. Hepatology 2020;71:686-721.), Clinical Practice Guidelines Panel (EASL recommendations on treatment of hepatitis C: Final update of the series. J Hepatol 2020;73:1170-1218.) according to the definitions of authoritative hepatitis and / or liver cancer experts.

[0040] In the present invention, the treatment of hepatitis B by administering NUC follows the definitions of authoritative hepatitis experts, including but not limited to the Hepatitis B Treatment Guidelines (Version 3.4) compiled by the Japan Society of Hepatology, May 2021 (https: / / www.jsh.or.jp / lib / files / medical / guidelines / jsh_guidlines / B_v3.4.pdf) and its English version (Hepatology Research, 2020; 50: 892-923.).

[0041] In the present invention, fatty liver disease, non-alcoholic fatty liver disease (NAFLD), non-alcoholic fatty liver (NAFL), and non-alcoholic steatohepatitis (NASH) are defined in accordance with authoritative hepatitis expert definitions, including, but not limited to, the NAFLD / NASH Clinical Practice Guidelines 2020 (edited by the Japanese Society of Gastroenterology and the Japan Society of Hepatology, revised 2nd edition, published November 2020, Nanzando (https: / / www.jsge.or.jp / guideline / guideline / pdf / nafldnash2020.pdf)) and its English version (Tokushige, K. et al. HepatologyResearch.2021;51:1013-1025. and Tokushige, K. et al. Journal of Gastroenterology 2021;56: 951-963.).

[0042] In the present invention, tests to determine whether or not liver cancer has developed from chronic viral hepatitis, fatty liver disease, or other liver diseases are in accordance with definitions by authoritative liver cancer experts, including but not limited to the Liver Cancer Treatment Guidelines (edited by the Japan Society of Hepatology, 2017 edition, supra).

[0043] In the present invention, assessing the risk of developing liver cancer means performing evaluation and testing according to certain criteria. Specifically, it means assessing whether or not a subject with chronic liver disease but not developing liver cancer, including but not limited to subjects who have not developed liver cancer after achieving SVR for HCV, subjects who have not developed liver cancer while undergoing treatment with NUC for HBV, and subjects who have not developed liver cancer but have NAFL, NASH, or other fatty liver diseases, is at risk of developing liver cancer in the future, and includes assessing whether or not the subject is at risk of developing liver cancer, and assessing the level of the risk. In particular, one purpose of assessing the risk of developing liver cancer in the present invention is to stratify subjects according to the level of liver cancer risk, and to allocate liver cancer screening resources to subjects with a high risk of developing liver cancer rather than to subjects with a low risk of developing liver cancer.

[0044] In the present invention, the risk of developing liver cancer is associated with the GDF15 level of a subject. Specifically, whether the GDF15 level of a subject is higher or lower than a predetermined cutoff value is an indicator of whether the subject has a high or low risk of developing liver cancer.

[0045] The cutoff value in the present invention is predetermined by preparing a database tracking individual GDF15 levels and the presence or absence of liver cancer in an evaluation population of chronic liver disease patients, and then conducting statistical analysis or ROC analysis of the GDF15 level data for liver cancer patients and non-liver cancer patients. When the cutoff value is determined by statistical analysis, for example, the median, arithmetic mean, or other average value of the GDF15 level data for the evaluation population can be used. When the cutoff value is determined by ROC analysis, for example, the cutoff value based on ROC analysis can be the GDF15 level at the point on the ROC curve where the distance between the point on the vertical axis (sensitivity or true positive) of the ROC curve graph that is 1.0 and the point on the horizontal axis (1 - specificity) that is 0.0 is the shortest, or the cutoff value can be derived from the Youden index of the ROC curve (Cancer 1950;3:32-35.). Once established, the database of the evaluation population of chronic liver disease patients may be used without any changes to set a cutoff value in the method of assessing the risk of liver cancer in subjects of the present invention. Alternatively, new chronic liver disease patients, including the subjects of the present invention, may be incorporated into the evaluation population, and the database of the evaluation population of chronic liver disease patients may be updated as appropriate and used to set a cutoff value in the method of assessing the risk of liver cancer in subjects of the present invention. As described in the examples of the present specification, the cutoff value for predicting liver cancer development determined by the ROC curve in the observational study of the present inventors was found to be within a numerical range of 90% or more and 110% of the median GDF15 level in the serum of the subjects in the observational study of the present inventors. Therefore, the median may be used as the cutoff value.

[0046] In the method of the present invention for assessing a subject's risk of developing liver cancer, the cutoff value for the serum GDF15 protein level can be about 1400 pg / mL for hepatitis C patients who have achieved SVR. The cutoff value for GDF15 can be about 845 pg / mL for hepatitis B patients receiving NUC. The cutoff value for GDF15 can be about 2000 pg / mL for NAFLD patients. In the method of the present invention, the step of assessing the risk of developing liver cancer can further combine cutoff values ​​for AFP and FIB-4 index. In the method of the present invention, the cutoff values ​​for AFP and FIB-4 index can be about 5 ng / mL and about 3.25, respectively, for hepatitis C patients who have achieved SVR.

[0047] Hereinafter, the methods of the present invention for assessing a subject's risk of developing liver cancer will be explained separately as a method based on the subject's GDF15 protein level and a method based on the subject's GDF15 transcript level.

[0048] 1. Method for assessing the risk of developing liver cancer in a subject suffering from chronic liver disease based on the subject's GDF15 protein level One embodiment of the present invention provides a method for assessing the risk of developing liver cancer in a subject suffering from chronic liver disease, comprising: (1) measuring the GDF15 protein level in the subject; and (2) correlating the GDF15 protein level with the risk of developing liver cancer.

[0049] The GDF15 protein level of a subject is measured using a serum or plasma sample from the subject. The subject's plasma and serum can be prepared by a method known per se from a peripheral blood sample collected from the subject. These may be appropriately diluted with a known buffer or the like depending on the method for measuring the GDF15 protein level. For example, they may be diluted 75 to 100 times or more using a dilution buffer or the like included in a GDF15 protein human enzyme-linked immunosorbent assay (ELISA) kit (#DGD150, R&D systems, Minneapolis, MN). If it takes a long time from blood collection to measurement, plasma and / or serum that has been frozen and stored can be used for measurement.

[0050] The level of GDF15 protein in step (1) can be measured by an immunological technique using an antibody that specifically recognizes GDF15 protein (i.e., a GDF15-specific antibody). Examples of immunological techniques include antibody array, flow cytometry analysis, radioimmunoassay (RIA), ELISA (Methods in Enzymol. 70: 419-439 (1980)), Western blotting, immunohistostaining, enzyme immunoassay (EIA), fluorescence immunoassay (FIA), immunochromatography, immunoturbidimetry, and immunonephelometry. ELISA is preferred from the viewpoints of sensitivity and ease of implementation.

[0051] "Specific recognition" of antigen X by an antibody means that in an antigen-antibody reaction, the affinity of the antibody for antigen X is stronger than the affinity for antigens other than antigen X. In this specification, an antibody that specifically recognizes antigen X may be abbreviated as "anti-X antibody" or "X-specific antibody."

[0052] The GDF15-specific antibody may be either a polyclonal antibody or a monoclonal antibody, or a binding fragment thereof.

[0053] The antibody may be directly or indirectly labeled with a labeling substance. Examples of the labeling substance include fluorescent substances (e.g., FITC, rhodamine), radioactive substances (e.g., ３２ P. ３５ S.１４ C. ３ H), enzymes (e.g., alkaline phosphatase, peroxidase), colored particles (e.g., metal colloid particles, colored latex), biotin, etc.

[0054] The antibody can be used in a soluble state without any other binding, or it can be bound to a solid phase. Examples of "solid phases" include plates (e.g., microwell plates), tubes, beads (e.g., plastic beads, magnetic beads), chromatography supports (e.g., water-absorbent substrates such as nitrocellulose membranes, Sepharose), membranes (e.g., nitrocellulose membranes, PVDF membranes), gels (e.g., polyacrylamide gels), and metal membranes (e.g., gold membranes). Among these, plates, beads, chromatography supports, and membranes are preferably used, with plates being most preferred due to their ease of handling. Examples of the bond include covalent bonds, ionic bonds, and physical adsorption, and are not particularly limited. However, covalent bonds and / or physical adsorption are preferred because they provide sufficient binding strength. The bond to the solid phase may be direct or indirect, using a known substance. In addition, in order to suppress nonspecific adsorption and nonspecific reactions, it is common practice to bring a phosphate buffer solution of bovine serum albumin (BSA) or bovine milk protein into contact with the solid phase, thereby blocking the portions of the solid phase surface that are not coated with the antibody with the BSA or bovine milk protein.

[0055] The contact of a GDF15-specific antibody with plasma or serum from a subject is not particularly limited in its mode, order, or specific method, as long as it allows the antibody to interact with GDF15 in the plasma or serum. Contact can be achieved, for example, by adding plasma or serum to a plate on which the antibody is immobilized. Alternatively, for example, proteins in the plasma or serum can be separated by SDS-PAGE or other means, transferred to a membrane, and immobilized thereon, followed by contact with the antibody.

[0056] The time for maintaining such contact is not particularly limited as long as it is long enough for the antibody and GDF15 contained in the plasma or serum derived from the subject to bind to form a complex, but is typically several seconds to several tens of hours. The temperature conditions for contact are typically 4°C to 50°C, preferably 4°C to 37°C, and most preferably room temperature of about 15°C to 30°C. Furthermore, the pH conditions for the reaction are preferably 5.0 to 9.0, and particularly preferably a neutral range of 6.0 to 8.0.

[0057] In measuring the level of GDF15 protein, the absolute value of the GDF15 protein concentration can be easily measured, for example, in pg / mL units, by using commercially available GDF15 protein and quantifying the ELISA results of its dilution series using fluorescence, color development, or other reactions.

[0058] Next, in step (2), the GDF15 protein level in the plasma and / or serum of the subject measured in step (1) is correlated with the risk of developing liver cancer. Correlating the GDF15 protein level in the plasma and / or serum with the risk of developing liver cancer means determining whether the subject's data suggest (or indicate) the risk of developing liver cancer.

[0059] The correlation between the subject's data and the risk of developing liver cancer is usually carried out by comparing the subject's data with the data of patients with chronic liver disease other than the subject.As shown in the examples of the present invention, it has been revealed that the group whose GDF15 protein level is higher than the predetermined cutoff value (high GDF15 group) has a higher risk of developing liver cancer than the group whose GDF15 protein level is lower than the predetermined cutoff value (low GDF15 group).This makes it possible to evaluate the risk of developing liver cancer in the subject.

[0060] The method for assessing the risk of developing liver cancer of the present invention may include a step of determining the risk of developing liver cancer based on a cutoff value of GDF15 protein level, wherein if the GDF15 protein level of the subject is equal to or higher than the cutoff value, the risk of developing liver cancer of the subject can be determined to be high, and if the GDF15 protein level of the subject is lower than the cutoff value, the risk of developing liver cancer of the subject can be determined to be low. Furthermore, the cutoff value can be the median value of the GDF15 protein level of the subject group.

[0061] In this specification, the conjunction "about" used to modify a numerical value means a numerical range of 90% or more and 110% or less of the numerical value. For example, "1400 pg / mL" refers to a numerical range of 1260 pg / mL or more and 1540 pg / mL or less.

[0062] In the method of the present invention, the cutoff value determined as the median serum GDF15 protein level in hepatitis C patients who achieved SVR may be about 1400 pg / mL. In the Examples of the present specification, the cutoff value in hepatitis C patients who achieved SVR, determined as the serum GDF15 concentration at which the Youden index is maximized by the ROC curve, is 1448 pg / mL, which is within a numerical range of 90% or more and 110% or less of the cutoff value of 1400 pg / mL determined as the median.

[0063] In the method of the present invention, in the step of assessing the risk of developing liver cancer, the cutoff values ​​of AFP and / or FIB-4 index can be further combined for assessment. For hepatitis C patients who have achieved SVR, assessment can be made using AFP and the FIB-4 index in addition to GDF15. For hepatitis B patients undergoing treatment with NUC, assessment can be made using AFP in addition to GDF15. This is because methods for stratifying the risk of developing liver cancer in subjects who have achieved SVR after DAA treatment using AFP and the FIB-4 index as markers have been known for some time. In conventional techniques known to those skilled in the art, the cutoff values ​​for the AFP and the FIB-4 index have been set at 5 ng / mL and 3.25, respectively.

[0064] 2. Method for assessing the risk of liver cancer in a subject suffering from chronic liver disease based on the GDF15 transcript level in the subject One embodiment of the present invention provides a method for assessing the risk of liver cancer in a subject suffering from chronic liver disease, comprising: (1) measuring the GDF15 transcript level in the subject; and (2) correlating the GDF15 transcript level with the risk of liver cancer.

[0065] The GDF15 transcript level of a subject is measured using a biological tissue, serum, or plasma sample from the subject. The biological tissue can be obtained by collecting a portion of the subject's biological tissue using methods including, but not limited to, percutaneous biopsy (needle biopsy), endoscopic biopsy, and surgical biopsy. In the method of the present invention for assessing a subject's risk of developing liver cancer, the biological tissue is preferably liver tissue. Circulating GDF15 transcripts or portions thereof contained in serum or plasma samples can also be measured. To measure the GDF15 transcript level of a subject, RNA can be isolated from the biological sample using conventional methods. General methods for extracting RNA are well known in the art and are disclosed in molecular biology experimental protocol books such as "Molecular Cloning: A Laboratory Manual" by Sambrook, J. and Russell, D.W. (3rd ed., Cold Spring Harbor Laboratory Press, 2001). Specifically, RNA isolation can be performed using a commercially available purification kit such as an RNeasy column (Qiagen, Hulsterweg, Germany) according to the manufacturer's instructions.

[0066] To measure the GDF15 transcript level from isolated RNA, for example, reverse transcription polymerase chain reaction (RT-PCR), quantitative real-time RT-qPCR, etc. can be used. Complementary DNA is prepared from the RNA by reverse transcription using GDF15-specific primers. Using the complementary DNA as a template, a reaction solution containing a GDF15-specific PCR primer pair and a fluorescently labeled probe is amplified in a real-time PCR system, and the fluorescence can be quantified. Specifically, analysis can be performed by quantitative real-time reverse transcription polymerase chain reaction using Thunderbird qPCR Master Mix (Toyobo, Osaka, Japan) and TaqMan probes (human GDF15, Hs00171132_m1, human beta actin, Hs 9999902_m3, Applied Biosystems, Waltham, MA). In measuring GDF15 transcript levels, the absolute concentration of GDF15 transcript in RNA can be quantified by measuring the concentration of pre-synthesized GDF15 mRNA or a purified RNA portion thereof, amplifying the resulting dilution series using a real-time PCR device, and quantifying the fluorescence. Alternatively, the relative value of the measured GDF15 transcript in RNA derived from a subject sample to be measured relative to the measured GDF15 transcript in a control RNA of the same concentration can be quantified. The relative value of the measured GDF15 transcript can be expressed in arbitrary units (AU).

[0067] The primer pair and probe for specifically detecting the GDF15 transcript used in step (1) can be synthesized based on the nucleotide sequence of human GDF15 mRNA published as NCBI Reference Sequence: NM_004864.4. The base lengths of the primers and probes are not particularly limited. The primers may contain either a partial nucleotide sequence of the nucleotide sequence of human GDF15 mRNA or a partial nucleotide sequence complementary to the nucleotide sequence of human GDF15 mRNA in the forward primer and the other in the reverse primer. The base lengths of the primers may be 10 to 50 nucleotides, preferably 15 to 30 nucleotides. The probe contains a partial nucleotide sequence complementary to the nucleotide sequence of human GDF15 mRNA, and the base length of the probe may range from 10 nucleotides to the entire length of the nucleotide sequence complementary to the nucleotide sequence of human GDF15 mRNA, preferably 20 to 150 nucleotides.

[0068] The primer pair and probe used in step (1) for specifically detecting the GDF15 transcript may be natural nucleic acids such as RNA and DNA, or may be a combination of natural nucleic acids with chemically modified nucleic acids or pseudo-nucleic acids, if necessary. Examples of chemically modified nucleic acids and pseudo-nucleic acids include PNA (Peptide Nucleic Acid), LNA (Locked Nucleic Acid; registered trademark), methylphosphonate DNA, phosphorothioate DNA, and 2'-O-methyl RNA. Furthermore, the primers and probes may contain fluorescent substances and / or quenchers, or radioisotopes (e.g., 32 P, 33 P, 35Labeling or modification may be performed using a labeling substance such as Fluorescent Protein I (FITC), Texas Amino Acids (TFA), or a modifying substance such as biotin, (streptavidin), or magnetic beads. The labeling substance is not limited, and commercially available substances can be used. For example, fluorescent substances such as FITC, Texas Amino Acids (TFA), Cy3, Cy5, Cy7, Cyanine 3, Cyanine 5, Cyanine 7, FAM, HEX, VIC, fluorescamine and its derivatives, and rhodamine and its derivatives can be used. Quencher substances such as AMRA, DABCYL, BHQ-1, BHQ-2, or BHQ-3 can be used. The labeling position of the labeling substance in the primer and probe can be determined appropriately depending on the properties of the modifying substance and the intended use. Generally, modification is performed at the 5' or 3' end. Furthermore, a single primer and probe molecule may be labeled with more than one type of labeling substance. The design of the nucleotide sequences of primers and probes and the selection of labeling substances are well known and are disclosed in molecular biology experimental protocol books such as Molecular Cloning: A Laboratory Manual by Sambrook, J and Russell, DW (3rd ed., Cold Spring Harbor Laboratory Press, 2001).

[0069] Next, in step (2), the GDF15 transcript level of the subject measured in step (1) is correlated with the risk of developing liver cancer. Correlating the GDF15 transcript level with the risk of developing liver cancer means determining whether the subject's data suggest (or indicate) the risk of developing liver cancer.

[0070] The correlation between the subject's data and the risk of developing liver cancer is usually carried out by comparing the subject's data with the data of patients other than the subject who suffer from chronic liver disease.As shown in the examples of the present invention, it has been revealed that the group whose GDF15 transcript level is higher than the predetermined cutoff value (high GDF15 group) has a higher risk of developing liver cancer than the group whose GDF15 transcript level is lower than the predetermined cutoff value (low GDF15 group).This makes it possible to evaluate the risk of developing liver cancer in the subject.

[0071] The method for assessing the risk of developing liver cancer of the present invention may include a step of determining the risk of developing liver cancer based on a cutoff value of GDF15 transcript level, wherein if the GDF15 transcript level of the subject is equal to or higher than the cutoff value, the risk of developing liver cancer of the subject can be determined to be high, and if the GDF15 transcript level of the subject is lower than the cutoff value, the risk of developing liver cancer of the subject can be determined to be low. Furthermore, the cutoff value can be the median value of the GDF15 transcript level of the subject group.

[0072] The method for assessing the risk of developing liver cancer of the present invention may include a step of determining the risk of developing liver cancer based on a cutoff value of GDF15 transcript level, wherein if the GDF15 transcript level of the subject is equal to or higher than the cutoff value, the risk of developing liver cancer of the subject can be determined to be high, and if the GDF15 transcript level of the subject is lower than the cutoff value, the risk of developing liver cancer of the subject can be determined to be low. Furthermore, the cutoff value can be the median value of the GDF15 transcript level of the subject group.

[0073] In the method of the present invention, the step of assessing the risk of developing liver cancer can further combine cutoff values ​​of AFP and FIB-4 index for assessment. This is because methods for stratifying the risk of developing liver cancer in subjects who have achieved SVR after DAA treatment using AFP and FIB-4 index as markers have been known. In conventional techniques known to those skilled in the art, the cutoff values ​​for the AFP and FIB-4 index have been set at 5 ng / mL and 3.25, respectively.

[0074] 3. Kit for Measuring GDF15 Levels in a Subject, Used in the Method of the Present Invention The present invention provides a kit for measuring GDF15 levels in a subject, used in the method of the present invention. The kit of the present invention comprises an anti-GDF15-specific antibody and / or a primer pair and probe for specifically detecting GDF15 transcripts. The anti-GDF15-specific antibody contained in the kit of the present invention is as described in section "1. Method for assessing the risk of liver cancer in a subject suffering from chronic liver disease based on the GDF15 protein level in the subject" herein. The primer pair and probe for specifically detecting GDF15 transcripts contained in the kit of the present invention are as described in section "2. Method for assessing the risk of liver cancer in a subject suffering from chronic liver disease based on the GDF15 transcript level in the subject" herein.

[0075] 4. Diagnostic Agent for Assessing the Liver Cancer Risk of a Subject with Chronic Liver Disease Using the Method of the Present Invention The present invention provides a diagnostic agent for assessing the liver cancer risk of a subject with chronic liver disease using the method of the present invention. The diagnostic agent of the present invention comprises an anti-GDF15 antibody and / or a primer pair and / or probe for specifically detecting GDF15 transcripts. The anti-GDF15-specific antibody contained in the diagnostic agent of the present invention is as described in section "1. Method for Assessing the Liver Cancer Risk of a Subject with Chronic Liver Disease Based on the GDF15 Protein Level of the Subject" herein. The primer pair and / or probe for specifically detecting GDF15 transcripts contained in the diagnostic agent of the present invention is as described in section "2. Method for Assessing the Liver Cancer Risk of a Subject with Chronic Liver Disease Based on the GDF15 Transcript Level of the Subject" herein.

[0076] 5. Use of GDF15 as a biomarker for assessing a subject's risk of developing liver cancer The present invention provides the use of GDF15 as a biomarker for assessing a subject's risk of developing liver cancer. The steps involved in the use of GDF15 of the present invention are as described in the sections "1. Method for assessing a subject's risk of developing liver cancer based on the subject's GDF15 protein level" and "2. Method for assessing a subject's risk of developing liver cancer based on the subject's GDF15 transcript level" of this specification.

[0077] All documents mentioned herein are incorporated by reference in their entirety.

[0078] The following examples of the present invention are for illustrative purposes only and do not limit the technical scope of the present invention. The technical scope of the present invention is limited only by the claims. The present invention may be modified, for example, by adding, deleting, or substituting components of the present invention, provided that the modifications do not depart from the spirit of the present invention.

[0079] Example 1: Assessment of Liver Cancer Risk in Subjects Achieving Sustained Hepatitis C Virus (HCV) Negative Response (SVR) A. Materials and Methods (1) Study Population Patients with hepatitis C were enrolled at baseline from 26 medical institutions participating in the Osaka Liver Forum and received interferon-free DAA treatment in accordance with the guidelines of the Japan Society of Hepatology (Hepatol Res 2020;50:791-816). Patients with hepatitis B virus or human immunodeficiency virus co-infection, decompensated cirrhosis, or other liver diseases (e.g., autoimmune hepatitis or primary biliary cholangitis), post-liver transplant patients, or patients under the age of 20 were excluded from enrollment. A total of 2,840 hepatitis C patients had been enrolled and completed DAA treatment by December 2017. Of the 2,840 patients, those who did not achieve SVR, those for whom baseline serum samples were unavailable, and those with a history of liver cancer treatment were excluded, leaving 1,609 patients from 20 institutions for this study. Of these, 823 patients underwent liver biopsy before DAA treatment. Histological analysis was performed using the Metavir score.

[0080] (2) Clinical Research Review All patients participating in this study provided written informed consent. The design of this study complies with the Declaration of Helsinki. The patient information and sample collection protocols of this study were approved by the Osaka University Hospital Clinical Research Ethics Review Committee and the ethics committees of each institution (IRB 14148, 14419, 15080, 15325, 16314, 16494, 12449), and the analysis protocol was approved by the Osaka University Hospital Institutional Review Board (IRB No. 17032).

[0081] (3) Antiviral Treatment and SVR DAA treatment was performed using the following protocols: asunaprevir and daclatasvir for 24 weeks, sofosbuvir and ledipasvir for 12 weeks, ombitasvir and paritaprevir, combined with ritonavir for 12 weeks, sofosbuvir and ribavirin for 12 weeks, and elbasvir and grazoprevir for 12 weeks. In the present invention, SVR refers to undetectable HCV RNA levels 24 weeks after the end of treatment. All patients were treated in accordance with the Japan Society of Hepatology guidelines for the treatment of chronic HCV infection.

[0082] (4) Follow-up and Liver Cancer Surveillance. Before the start of DAA treatment, all patients underwent ultrasound, CT, and / or MRI to exclude cases of liver cancer. Patients undergoing DAA treatment underwent blood tests every two weeks, including hematology, biochemistry, and virology. Post-treatment patients underwent liver cancer surveillance using ultrasound and / or CT / MRI every six months. Diagnosis was made using typical contrast-enhanced CT images and / or MRI, in accordance with recommendations from the European Association for the Study of the Liver - European Organisation for Research and Treatment of Cancer (EASL-EORTC) and the American Association for the Study of Liver Diseases (AASLD) (J Hepatol 2012;56:908-943 and Hepatology 2018;67:358-380). If imaging was insufficient to diagnose liver cancer, targeted biopsy of the tumor was performed and a histological diagnosis was made. The start date of follow-up was the end date of DAA treatment. The endpoint was the date of liver cancer onset or the date of the last follow-up liver cancer surveillance imaging examination. The endpoint for overall survival was the date of death from any cause or the date of the last follow-up.

[0083] (5) Serological Testing. Serum samples from enrolled patients were stored in a −80°C freezer at Osaka University at the time points specified in the prospective study protocol. Serum GDF15 concentrations were measured using a human enzyme-linked immunosorbent assay (ELISA) kit (#DGD150, R&D Systems, Minneapolis, MN) according to the manufacturer's protocol. Absorbance was measured using a Varioscan LUX (Thermo Scientific, Waltham, MA).

[0084] (6) mRNA Expression Analysis. Liver tissue RNA was extracted using an RNeasy column (Qiagen, Hulsterweg, Germany) and reverse transcribed to complementary DNA. Messenger RNA expression was analyzed by quantitative real-time reverse transcription polymerase chain reaction using Thunderbird qPCR Master Mix (Toyobo, Osaka, Japan) and TaqMan probes (human GDF15, Hs00171132_m1; human beta actin, Hs9999902_m3; Applied Biosystems, Waltham, MA). Target gene expression was normalized to beta-actin.

[0085] (7) Statistical Analysis Statistical analysis for comparison of parametric and nonparametric values ​​was performed using the Student t-test and Mann-Whitney U test, respectively. One-way ANOVA followed by the Turkey-Kramer post-hoc test or the Kruskal-Wallis test was performed for parametric and nonparametric multiple comparisons, respectively. For liver cancer incidence analysis, the end date of DAA was used as the index. Cases were followed up until liver cancer onset, death, or the last day of liver cancer surveillance before December 31, 2020, whichever came first, and Kaplan-Meier curves were plotted. The log-rank test was used to compare liver cancer incidence between the two groups. Logistic regression analysis was used to analyze liver cancer prediction, and the Cox proportional hazards model was used to compare liver cancer risk. Prism version 8.4.2 for Windows (Graph Pad PRISM RRID; SCR_014242) was used for the analysis.

[0086] B. Results (1) Serum GDF15 levels were higher in patients who developed liver cancer after DAA treatment than in patients who did not develop liver cancer. Patients with no history of liver cancer treatment were divided into a derivation cohort in which serum samples were stored at both the end of treatment and 24 weeks after the end of treatment, and a validation cohort in which serum samples were not stored at either time point. Details of the cases in the derivation and validation cohorts are shown in Figure 1.

[0087] The derivation cohort contained serum samples stored at three time points: before treatment (Pre or Pre-Treatment), at the end of treatment (EOT), and 24 weeks after SVR was achieved (p24w or Post 24 weeks). Analysis of GDF15 levels in the stored serum at these three time points revealed that serum GDF15 levels after DAA treatment were lower than those before treatment (Figure 2-1). Furthermore, in the 55 cases in which liver tissue was cryopreserved before DAA treatment, a weak correlation was observed between serum GDF15 levels and GDF15 expression in the liver (Figure 2-2).

[0088] Patients were divided into two groups based on a median pretreatment serum GDF15 level of 1400 pg / mL. As shown in Figure 3, the high-GDF15 and low-GDF15 groups were older and had lower platelet counts, higher AST, ALT, GGT, triglycerides, fasting blood glucose, HbA1c, AFP, FIB-4 index, and ALBI scores, and lower HCV RNA levels, hemoglobin, eGFR, and albumin than the low-GDF15 group. Serum GDF15 levels correlated with fibrosis scores (Figure 4-1). Serum GDF15 levels correlated with FIB-4 index values ​​(Figure 4-2). As shown in Figure 4-3, serum GDF15 levels were also correlated with older age (Figure 4-3A), high AST (Figure 4-3D), low eGFR (Figure 4-3G), low albumin (Figure 4-3H), and high ALBI score (Figure 4-3K).

[0089] In the derivation cohort, 49 cases developed liver cancer after DAA treatment within the observation period of the present invention. Details are shown in Figure 5. The incidence of liver cancer in the derivation cohort was 1.59% at 1 year, 2.85% at 2 years, and 5.02% at 3 years (Figure 6-1). Serum GDF15 levels were higher in patients who developed liver cancer after treatment than in patients who did not develop liver cancer after treatment at all three time points: pre-treatment, end-of-treatment, and 24 weeks after achieving SVR (Figure 6-2). GDF15 levels at the time of liver cancer development did not change significantly compared with GDF15 levels one year before liver cancer development (Figure 6-3). This suggests that serum GDF15 levels, unlike tumor markers such as AFP and PIVKA2, may reflect the severity of liver disease at that time point in each case.

[0090] The 1-, 2-, and 3-year cumulative incidence rates of liver cancer were 2.80%, 4.44%, and 8.31%, respectively, in the high GDF15 group, and 0.47%, 1.12%, and 1.93%, respectively, in the low GDF15 group. The 1-, 2-, and 3-year cumulative incidence rates of liver cancer were significantly lower in the low GDF15 group than in the high GDF15 group (Figure 6-4).

[0091] (2) Serum GDF15 levels may be a novel biomarker for predicting the development of liver cancer after DAA treatment. To examine pretreatment serum GDF15 levels as a biomarker for predicting liver cancer development, pretreatment variables associated with liver cancer development were analyzed using the Cox Hazards model. Patients with an FIB-4 index greater than 3.25 (>3.25) were considered to have advanced liver fibrosis based on previous findings (Gastroenterology 2017;153:996-1005.e1001 and Hepatology 2007;46:32-36). Other variables were assigned to two groups based on their median or previous findings (Clin Gastroenterol Hepatol 2014;12:1186-1195). Univariate analysis revealed that older age, low platelet count, high AST, high ALT, low albumin, low prothrombin activity, high AFP, high serum GDF15, high FIB-4 index, and high ALBI score were associated with an increased risk of developing liver cancer (Figure 7).

[0092] The following parameters were selected for the multivariate Cox regression model: gender, AFP as a tumor marker, GDF15 level, age, FIB-4 index as the degree of fibrosis calculated from AST, ALT and platelets, and ALBI score calculated from albumin and bilirubin. Among these parameters, AFP (J Med Virol 2020;92:3507-3515. and J Hepatol 2017;67:933-939.), FIB-4 index (Gastroenterology 2017;153:996-1005.e1001., J Med Virol 2020;92:3507-3515. and J Hepatol 2017;68:25-32.), and ALBI score (Dig Liver Dis 2019;51:681-688.) are known to be risk factors for liver cancer after HCV clearance. High GDF15 levels (HR 2.52, 95% CI 1.17-6.09), high AFP levels (HR 2.26, 95% CI 1.16-4.69), and high FIB-4 index levels (HR 2.40, 95% CI 1.18-5.24) were independently associated with increased risk of liver cancer (Figure 7). No significant differences were observed in the predictive power of AFP, FIB-4 index, and GDF15 levels (Figure 8). The cumulative liver cancer incidence rate over 3 years was 8-9% in the high AFP or high FIB-4 index group and 2-3% in the low AFP or low FIB-4 index group (Figure 9A and B).

[0093] The cutoff values ​​for GDF15, AFP, and FIB-4 index for predicting liver cancer were determined using receiver operating characteristic curves (ROC) with the Youden index (Cancer 1950;3:32-35). The cutoff values ​​for serum GDF15, AFP, and FIB-4 index were 1448 pg / mL, 6.020 ng / mL, and 3.025, respectively (Fig. 8E). These cutoff values ​​were similar to the median serum GDF15 levels before treatment, 1400 pg / mL, 5 ng / mL, and 3.25 (Fig. 7).

[0094] To stratify liver cancer risk, a total score was calculated for each patient, with high values ​​of GDF15, AFP, and FIB-4 index each assigned a score of 1. A score of 0 was assigned to the low-risk group, a score of 1 or 2 to the intermediate-risk group, and a score of 3 to the high-risk group. The 1-, 2-, and 3-year cumulative liver cancer risks were 0%, 0.40%, and 0.40% for the low-risk group, 1.18%, 1.89%, and 4.44% for the intermediate-risk group, and 4.95%, 8.26%, and 13.2% for the high-risk group, respectively (Figure 10).

[0095] Of the 248 low-risk cases, one developed liver cancer. The BMI of this case was 31.4 kg / m ２ In Japan, the BMI is 30 kg / m ２ In the present invention, the population with a BMI of 30 kg / m ２ The proportion of cases exceeding 30 kg / m was 2.5% in both the derivation and validation cohorts. ２ A future challenge will be to determine whether GDF15 is as effective in predicting the risk of developing liver cancer in cases exceeding this level as it is in cases with a normal BMI.

[0096] (3) A scoring system using GDF15 levels, AFP, and FIB-4 index can stratify liver cancer risk in the validation cohort. In a derivation cohort study, liver cancer risk was stratified using a scoring system using GDF15 levels, AFP, and FIB-4 index. This scoring system was validated using a validation cohort of 751 cases for which only pre-DAA treatment serum samples were available. In the validation cohort, 39 cases developed liver cancer after DAA treatment during the observation period (Figure 11). The liver cancer incidence rate in the validation cohort was 1.95% at 1 year, 4.54% at 2 years, and 5.82% at 3 years (Figure 12). No significant difference in cumulative liver cancer incidence was observed between the derivation cohort and the validation cohort (p=0.57). The low GDF15 group, low AFP group, and low FIB-4 index group had significantly lower cumulative liver cancer incidence rates than the high GDF15 group, high AFP group, and high FIB-4 index group, respectively (FIGS. 13A-C).

[0097] This scoring system clearly stratified the risk of developing liver cancer by narrowing down the high-risk and low-risk groups (Figure 14). The 1-, 2-, and 3-year cumulative liver cancer risks were 6.12%, 13.43%, and 14.2%, respectively, in the high-risk group (score 3, N = 183), and 1.0%, 2.91%, and 5.52%, respectively, in the intermediate-risk group (score 1-2, N = 322). Importantly, no liver cancer developed in the low-risk group (score 0, N = 236) (Figure 14).

[0098] To clarify the significance of post-SVR hepatocarcinogenesis risk stratification using a scoring system that combines the novel biomarker GDF15 with known markers AFP and FIB-4 index, the results of post-SVR hepatocarcinogenesis risk stratification for the derivation cohort of the present invention were compared below with those using only known markers AFP and FIB-4 index.

[0099] For the derivation cohort, high values ​​of the known markers AFP and FIB-4 index were assigned a score of 1, and a total score for each case was calculated. A score of 0 was assigned to the low-risk group, a score of 1 to the medium-risk group, and a score of 2 to the high-risk group. As shown in Figure 15-1, the degree of stratification of the Kaplan-Meier curves by risk group was rough, and even the low-risk group had a liver cancer incidence rate comparable to that of the medium-risk group.

[0100] When the low-risk group in a scoring system using only the known markers AFP and FIB-4 index was stratified into a group with low AFP and FIB-4 indexes but high GDF15 levels, and a group with low AFP and FIB-4 indexes and low GDF15 levels, as shown in Figure 15-2, the incidence of liver cancer in the group with low AFP and FIB-4 indexes and low GDF15 levels was 0. Therefore, the results of the present invention for the first time demonstrated that by stratifying liver cancer risk after SVR using a scoring system that uses the novel biomarker GDF15 in combination with the known markers AFP and FIB-4 index, there were no cases of liver cancer in the group with low GDF15, low AFP, and low FIB-4 index, demonstrating that the risk of liver cancer after DAA treatment is extremely low.

[0101] In contrast, in a group with low AFP and FIB-4 index but high GDF15 levels, the incidence of liver cancer was approximately 7% per year, confirming that stratification by low AFP and low FIB-4 index alone does not sufficiently reduce the risk of liver cancer after DAA treatment.

[0102] When the high-risk group based on a scoring system using only the known markers AFP and FIB-4 index was stratified into a group with high GDF15 levels and a group with low GDF15 levels, as shown in Figure 15-3, the incidence of liver cancer after DAA treatment significantly differed between a group with high AFP and FIB-4 index but low GDF15 levels and a group with high AFP and FIB-4 index and high GDF15 levels. This indicates that stratification by GDF15 contributes more to predicting the incidence of liver cancer after DAA treatment than stratification by the known markers AFP and FIB-4 index. The above analysis demonstrated the usefulness of a scoring system that includes the novel biomarker GDF15.

[0103] It should be noted that the present example was a retrospective study using stored serum, and bias related to serum availability cannot be ruled out. To address this concern, a comparison of the derivation cohort and validation cohort demonstrated no significant difference, at least in the incidence of liver cancer. Furthermore, Myojin, Y. et al. (Aliment Pharmacol Ther. 2022;55:422-433) analyzed the derivation cohort (964 cases) and validation cohort (642 cases) in which cases from the derivation cohort and validation cohort of this example were randomly assigned 3:2. The cutoff values ​​determined by the ROC curves for serum GDF15, AFP, and FIB-4 index to maximize the Youden index were 1350 pg / mL, 5 ng / mL, and 3.25, respectively. These values ​​were similar to the median serum GDF15 levels before treatment in this example.

[0104] Example 2: Assessment of the risk of developing liver cancer in subjects undergoing NUC treatment for hepatitis B virus (HBV) who have not developed liver cancer A. Materials and Methods (1) Study Population The study subjects in this example were nucleic acid analog (NUC)-administered patients with stored serum available for long-term follow-up. The selection criteria for stored serum were that the patient had a history of NUC administration for 8 months or more at the stored serum point (the oldest serum point if multiple serum points exist), and that the serum HBV DNA level at the stored serum point was less than 3.0 log IU / ml. However, patients with a history of liver cancer at the time of the serum point and patients with liver diseases other than hepatitis B were excluded.

[0105] (2) Clinical Research Review The design of this invention complies with the Declaration of Helsinki. The patient information and sample collection and analysis protocols of this invention have been approved by the Osaka University Hospital Clinical Research Ethics Review Committee (IRB 17032), and permission has been obtained from each facility, including Osaka University Hospital.

[0106] (3) Antiviral treatment The protocol for nucleoside analogue (NUC) treatment was based on the Hepatitis B Treatment Guidelines (1st edition) compiled by the Japan Society of Hepatology (April 2013 - May 2021) (3.4th edition) (https: / / www.jsh.or.jp / lib / files / medical / guidelines / jsh_guidlines / B_v3.4.pdf) and the English version (Hepatology Research, 2020; 50: 892-923.). Specifically, after starting NUC treatment, oral administration of NUC available at that time was continued.

[0107] (4) Follow-up and liver cancer surveillance Follow-up and liver cancer surveillance for patients undergoing treatment with NUC for HBV were conducted in accordance with the follow-up and liver cancer surveillance for patients undergoing antiviral treatment for HCV.

[0108] (5) Serological Tests and Statistical Analysis Serological tests and statistical analyses for HBV patients undergoing treatment with NUC were performed in the same manner as for HCV patients undergoing antiviral treatment.

[0109] B. Results (1) For HBV, serum GDF15 levels were higher in patients who developed liver cancer during treatment with NUC than in patients who did not develop liver cancer. Scatter plots of serum GDF15 concentrations in patients treated with NUC, patient background data, and graphs showing changes in liver cancer incidence over time for the entire cohort are shown in Figures 16, 17, and 18, respectively. As shown in Figures 16 and 17, the median and 25%-75% interval of GDF serum concentrations for the entire cohort were 0.833 ng / mL and 0.555-1.206 ng / mL, respectively.

[0110] Figure 19 is a table in which patient backgrounds are grouped by the median GDF15 serum concentration (0.833 ng / mL). Figure 20 is a table in which patient backgrounds are grouped by the presence or absence of liver cancer onset. Figures 21 and 22 are graphs showing the results of receiver operating characteristic (ROC) curve analysis of GDF15, Fib4, AFP, and Plt for the presence or absence of carcinogenesis 5 and 10 years after the stored serum point, respectively. The vertical axis of each graph represents sensitivity or true positive rate, the horizontal axis represents false positive rate (1-specificity), and AUC represents the area under the ROC curve for each graph. Figure 23 is a graph showing the change in liver cancer incidence over time using the cutoff value (0.845 ng / mL) determined from the ROC curve as the maximum value of the Youden index. Figure 24 shows the results of univariate / multivariate analysis of factors contributing to carcinogenesis using the Cox proportional hazards model. Figure 25 shows a graph showing the time course of liver cancer incidence, with cases grouped by score, plotted, and scored as 1 for cases with cutoff values ​​of 5 ng / mL and 0.845 ng / mL or higher for the two markers AFP and GDF15, which showed favorable results in multivariate analysis. Therefore, stratification using a scoring system using the novel biomarker GDF15 alone or in combination with the known marker AFP was shown to be useful for predicting liver cancer development in HBV patients undergoing NUC treatment.

[0111] These results demonstrate that the serum concentration of GDF15 is a useful marker for predicting liver cancer development, even in subjects undergoing treatment with NUC for hepatitis B virus (HBV) who have not developed liver cancer.

[0112] Example 3 Assessment of the risk of developing liver cancer in subjects with NAFL or NASH but without liver cancer A. Materials and methods (1) Case population to be investigated The subjects to be investigated in this example were patients diagnosed with NAFLD by liver biopsy between 2012 and 2020, and for whom stored serum samples from the liver biopsy were available for follow-up. However, patients with a history of liver cancer at the serum point and patients with other liver diseases besides NAFLD were excluded.

[0113] (2) Clinical Research Review All patients participating in this study provided written informed consent. The design of this study complies with the Declaration of Helsinki. The patient information and sample collection and analysis protocols of this study were approved by the Osaka University Hospital Clinical Research Ethics Review Committee (IRB 17032, 19551), and permission to conduct the study has been obtained at each facility, including Osaka University Hospital.

[0114] (3) Treatment of NAFL or NASH Non-alcoholic fatty liver (NAFL) and non-alcoholic steatohepatitis (NASH) were treated in accordance with the guidelines available at the time, such as the NAFLD / NASH Clinical Practice Guidelines 2020 (edited by the Japanese Society of Gastroenterology and the Japan Society of Hepatology, revised 2nd edition, published in November 2020, Nanzando (https: / / www.jsge.or.jp / guideline / guideline / pdf / nafldnash2020.pdf)) or its English version (Tokushige, K. et al. HepatologyResearch.2021;51:1013-1025. and Tokushige, K. et al. Journal of Gastroenterology 2021;56: 951-963.).

[0115] (4) Follow-up and liver cancer surveillance Follow-up and liver cancer surveillance for patients with NAFL or NASH were conducted in accordance with the follow-up and liver cancer surveillance for patients undergoing antiviral treatment for HCV.

[0116] (5) Serological Tests and Statistical Analysis Serological tests and statistical analyses of patients with NAFL or NASH were performed in the same manner as those of patients receiving antiviral treatment for HCV.

[0117] B. Results Figure 26 is a scatter plot of GDF15 serum concentrations in patients with NAFL or NASH. In the scatter plot in Figure 26, black circles represent cases without liver cancer, and white circles represent cases with liver cancer. Five of the six cases of liver cancer were primary hepatocellular carcinoma (HCC), while the one case indicated by the arrow was cholangiocarcinoma (CCC). Figure 27 is a table showing the patient background of patients with NAFL or NASH. Figure 28 is a table showing the patient background of patients grouped into NAFL and NASH. As shown in Figure 28, the median and 25%-75% interval of GDF serum concentrations for the entire cohort of patients with NAFL were 1.07 ng / mL and 0.69-1.49 ng / mL, respectively, and the median and 25%-75% interval of GDF serum concentrations for the entire cohort of patients with NASH were 1.41 ng / mL and 0.98-1.94 ng / mL, respectively. Figure 29 is a graph showing the change in the incidence of liver cancer over time for the entire cohort of patients with AFL or NASH.

[0118] Figure 30 is a scatter plot of GDF15 serum concentrations for cases grouped according to Brunt Stage 0 to 4. The tendency for GDF15 serum concentrations to increase in each group as the Brunt Stage progressed from 0 to 4 was verified using the Jonckheere-Terpstra trend test, with a significance level of P = 0.003. Figure 30 therefore demonstrates a significant correlation between GDF15 serum concentrations and fibrosis.

[0119] 31 is a scatter plot examining the correlation between serum GDF15 concentrations and Fib-4 index in patients with NAFL or NASH. Significant difference P is less than 0.0001, coefficient of determination R ２ was 0.245, indicating a significant correlation.

[0120] Figure 32 is a table showing hazard ratios for various attributes, blood markers, Fib-4 index, etc. of patients suffering from NAFL or NASH. As is clear from Figure 32, the P value for GDF15 is less than 0.0001, which is significantly lower than any of the other attributes, blood markers, Fib-4 index, etc.

[0121] Figure 33 shows the results of a time-course ROC (Receiver Operating Characteristic) curve analysis of the presence or absence of carcinogenesis 5 years after the stored serum point for GDF15 and Fib-4 index. The vertical axis of each graph represents sensitivity or true positive, and the horizontal axis represents false positive (1 - specificity), and AUC represents the area under the ROC curve for each graph. In a cohort of NAFL or NASH patients, 5 of 76 cases were followed up for 5 years and developed liver cancer. Comparison of AUCs showed that GDF15 had a higher ability to predict liver cancer development within 5 years than the Fib-4 index.

[0122] Graphs 34-1 and 34-2 show the time course of liver cancer incidence rates with cutoff values ​​of 2.00 ng / mL and 1.35 ng / mL, respectively. 2.00 ng / mL is the cutoff value determined from the ROC curve of carcinogenesis within 5 years of a cohort of NAFL or NASH patients, where the Youden index is at its maximum. 1.35 ng / mL is the cutoff value determined from the ROC curve of the derivation cohort of hepatitis C patients who achieved SVR in Example 1, where the Youden index is at its maximum. It has been shown that stratification with a cutoff value of 2.00 ng / mL significantly contributes to the prediction of liver cancer incidence rates in NAFL and NASH.

[0123] The above results demonstrate that the serum concentration of GDF15 is a useful marker for predicting hepatocarcinogenesis also in subjects suffering from NAFLD, including NAFL and NASH.

[0124] Example 4: Study with additional cohort from Ogaki Municipal Hospital Further study was conducted with 183 additional cases from Ogaki Municipal Hospital.

[0125] A. Materials and Methods (1) Case Population to be Surveyed The survey subjects in this example were patients diagnosed with NAFLD by liver biopsy between 2005 and 2020, and had serum samples stored at the time of liver biopsy available for follow-up. However, patients with a history of liver cancer at the serum point in time and patients with other liver diseases besides NAFLD were excluded.

[0126] (2) Clinical Research Review All patients participating in this study provided written informed consent. The design of this study complies with the Declaration of Helsinki. The patient information and sample collection and analysis protocols of this study were approved by the Osaka University Hospital Clinical Research Ethics Review Committee (IRB 17032), and permission to conduct the study has been obtained from Osaka University Hospital and Ogaki Municipal Hospital.

[0127] (3) Treatment of NAFLD Treatment of NAFLD was carried out in the same manner as in Example 3.

[0128] (4) Follow-up and liver cancer surveillance Follow-up and liver cancer surveillance for patients with NAFLD were conducted in accordance with the follow-up and liver cancer surveillance for patients undergoing antiviral treatment for HCV.

[0129] (5) Serological Tests and Statistical Analysis Serological tests and statistical analyses of patients with NAFLD were performed in the same manner as those of patients receiving antiviral treatment for HCV.

[0130] B. Results Figures 35-1 and 35-2 show the patient background of the cohorts at Ogaki City Hospital and Example 3 with an extended observation period, respectively.

[0131] Figure 36-1 is a scatter plot of GDF15 serum concentrations in patients with NAFLD at Ogaki Municipal Hospital. In the scatter plot in Figure 36-1, black circles represent cases without liver cancer, and gray circles represent cases with liver cancer. All nine cases of liver cancer were primary hepatocellular carcinoma (HCC). Figure 36-2 is a scatter plot of GDF15 serum concentrations in patients with NAFLD in the cohort of Example 3, in which the observation period was extended. In the scatter plot in Figure 36-2, black circles represent cases without liver cancer, and gray circles represent cases with liver cancer. Seven of the eight cases of liver cancer were primary hepatocellular carcinoma (HCC), while the one case indicated by the arrow was cholangiocellular carcinoma (CCC).

[0132] Figures 37-1 and 37-2 show the changes over time in the incidence of liver cancer in the Ogaki Municipal Hospital cohort and the cohort of Example 3 in which the observation period was extended, respectively.

[0133] Figures 38-1, 38-2, and 38-3 are graphs showing the results of a receiver operating characteristic (ROC) curve analysis of the presence or absence of carcinogenesis over time, 5 and 7 years after the stored serum points, for the Ogaki City Hospital cohort and the cohort from Example 3 with an extended observation period. The vertical axis of each graph represents sensitivity or true positivity, the horizontal axis represents false positivity (1 - specificity), and AUC represents the area under the ROC curve for each graph.

[0134] Figures 39-1 and 39-2 show graphs showing the time course of liver cancer incidence rates for the Ogaki Municipal Hospital cohort and the cohort of Example 3 with an extended observation period, with cutoff values ​​of 2.00 ng / mL and 1.74 ng / mL for GDF15, respectively. 2.00 ng / mL is the cutoff value determined as the maximum value of the Youden index from the ROC curve of a cohort of NAFLD patients over a 5-year observation period. 1.74 ng / mL is the cutoff value determined as the maximum value of the Youden index from the ROC curve of a cohort of NAFLD patients over a 7-year observation period. Stratification using cutoff values ​​of 2.00 ng / mL and 1.74 ng / mL has been shown to significantly contribute to the prediction of liver cancer incidence rates in NAFLD patients.

[0135] From the above results, it was proven that the serum concentration of GDF15 is a useful marker for predicting liver carcinogenesis, even when the Ogaki Municipal Hospital cohort and the cohort of Example 3, in which the observation period was extended, were examined together.

[0136] This application is based on patent application No. 2021-121873 filed in Japan on July 26, 2021, and patent application No. 2022-084132 filed in Japan on May 23, 2022, the entire contents of which are incorporated herein by reference.

[0137] The present invention makes it possible to evaluate the risk of liver cancer in subjects with higher accuracy based on GDF15 levels. Furthermore, subjects can be stratified according to the assessed level of liver cancer risk, so that subjects with a high risk of liver cancer undergo liver cancer screening tests more frequently than subjects with a low risk of liver cancer. Therefore, it is possible to reduce both the burden of testing on individual subjects and the medical economic loss to society as a whole.

Claims

1. (1) A step of measuring the GDF15 level of a subject; (2) A method for evaluating the risk of liver cancer development in a subject, comprising a step of associating the GDF15 level with the risk of liver cancer development.

2. The method according to claim 1, wherein the subject is at least one type of subject selected from the group consisting of a subject who has achieved sustained virological response (SVR) to hepatitis C virus (HCV), a subject who is receiving nucleos(t)ide analogue (NUC) therapy for hepatitis B, and a subject who has developed non-alcoholic fatty liver disease (NAFLD).

3. The method according to claim 1 or 2, wherein when the GDF15 level of the subject is equal to or higher than a preset cut-off value, it serves as an indicator that the subject has a high risk of liver cancer development, and when it is lower than the cut-off value, it serves as an indicator that the subject has a low risk of liver cancer development.

4. The method according to claim 1 or 2, wherein the GDF15 level is the level of GDF15 protein in serum or plasma and / or the level of GDF15 transcript in liver tissue or circulating blood.

5. The method according to claim 3, wherein the cut-off value is set based on statistical analysis or ROC analysis of the GDF15 level.

6. The method according to claim 5, wherein the cut-off value is the median value of the GDF15 level of the subject or the numerical value of the GDF15 level at which the Youden index of the ROC curve of the subject reaches the maximum value.

7. The method according to claim 6, wherein the subject is a subject who has achieved SVR for HCV, and the cut-off value of the GDF15 level is about 1400 pg / mL of the serum concentration of GDF15 protein; or the subject is a subject who is receiving NUC therapy for hepatitis B, and the cut-off value of the GDF15 level is about 845 pg / mL of the serum concentration of GDF15 protein; or the subject is a subject who has developed NAFLD, and the cut-off value of the GDF15 level is about 2000 pg / mL of the serum concentration of GDF15 protein.

8. The method according to claim 1 or 2, wherein the level of GDF15 protein in the serum is determined by the ELISA method.

9. The method according to claim 7, wherein the subject is a subject who has achieved SVR for HCV, the cut-off value of the GDF15 level in the serum is about 1400 pg / mL, and in the step of determining the risk of liver cancer development, the cut-off values of AFP and the FIB-4 index are further combined for determination.

10. ​ The method according to claim 9, wherein the cut-off values of the AFP and Fib-4 index are 5 ng / mL and 3.25, respectively.

11. A kit for measuring the GDF15 level of a subject for use in the method according to claim 1 or 2, comprising an anti-GDF15 specific antibody and / or a primer pair or a probe for specifically detecting a GDF15 transcript.

12. A diagnostic agent for use in the method according to claim 1 or 2, comprising an anti-GDF15 specific antibody and / or a primer pair or a probe for specifically detecting a GDF15 transcript.

13. (1) Measuring the GDF15 level of a subject; (2) Associating the GDF15 level with the risk of developing liver cancer. Use of GDF15 as a biomarker for evaluating the risk of developing liver cancer in a subject.

14. The use according to claim 13, wherein the subject is at least one subject selected from the group consisting of a subject who has achieved sustained virological response (SVR) to hepatitis C virus (HCV), a subject who is receiving NUC administration in hepatitis B, and a subject who has developed non-alcoholic fatty liver disease (NAFLD).