WO2016093629A1 - Biomarker for predicting hepatoma-targeted drug response, and use thereof - Google Patents

Abstract

Disclosed in the present application is a marker or a marker combination enabling excellent discrimination of a patient compliant with sorafenib drug from a patient not compliant with the same. The marker according to the present application enables determination of whether a patient is compliant with sorafenib, simply by a noninvasive test using blood, and thus enables customized drug administration based on the determination, thereby reducing drug side effects and being advantageous in the aspect of cost saving.

Description

Biomarker and its use for predicting liver cancer target therapy

It relates to techniques that can predict a patient's response to a targeted therapy for liver cancer.

Liver cancer is one of the cancers with a poor prognosis and ranks fifth among cancers worldwide. In particular, the Asian region, including Korea, which has a high rate of hepatitis virus, has a high incidence and the second highest mortality rate compared to cancer.

However, liver cancer is a disease that appears only when the disease progresses considerably, which often leads to missed the appropriate treatment time, and treatment is very poor in prognosis. In particular, surgical resection is a serious disease that dies within a year. Considering this clinical reality, early diagnosis and prognosis of cancer are the most realistic alternatives to cancer treatment and guidelines for next-generation liver cancer care. .

In particular, liver cancer is known to occur in patients with risk factors such as hepatitis virus, alcohol, and cirrhosis of the liver, and it is clear that the early diagnosis markers will be used for selective surveillance of high risk groups. Indeed, early diagnosis at six-month intervals has been shown to increase survival rates (reduced by 37%) in liver cancer patients.

However, despite efforts for early diagnosis, more than 60% of all liver cancer patients in Korea are diagnosed in the advanced stage. Since these patients are not subject to radical treatment such as surgery or radiofrequency thermal therapy, they are treated based on chemotherapy.

Until now, there has been almost no effective systemic chemotherapy, but since the therapeutic effect of sorafenib, an oral anti-cancer chemotherapy, has recently been demonstrated in the first phase III randomized controlled study of chemotherapy, advanced liver cancer It was recognized as an initial treatment.

However, such sorafenib is limited in that the tumor size decreases only 3.3% and the survival time is about 2-3 months, which is not effective for price. However, the use of sorafenib is expected in advanced liver cancer. In Korea, it is estimated that at least 5% of liver cancer patients are already using sorafenib (more than 1.6 per 100,000 people / year). In particular, according to the recently revised Sorafenib's medical benefit criteria (Stage III and above, Child-Pugh class A, ECOG ECOG: Eastern Cooperative Oncology Group performance status 0-2), As medical aid is used to support the percentage, it is very likely that the barrier to economic entry will be lowered, leading to a sharp increase in the use of sorafenib in the future. The problem of using sorafenib, however, is that it will be a huge burden on the health insurance financing due to the excessively expensive drug price (3 million won per month for non-payment), which requires selective treatment of patients. . In particular, since sorafenib acts on various signal transduction pathways, it is urgent to develop markers that can predict the drug response effect of sorafenib.

Japanese Laid-Open Patent Publication No. 2011-103821 relates to the identification of genes in response to sorafenib, and discloses a method for predicting a reaction by detecting polymorphisms of dihydropyrimidine dehydrogenase, CD247 gene, and the like. .

However, since there is almost no molecular diagnostic market related to the prediction of sorafenib drug response in Korea as well as abroad, it is necessary to develop a biomarker that can effectively predict the effect of sorafenib drug response.

The present application seeks to provide a biomarker capable of predicting a patient's response to sorafenib, a target therapeutic agent for liver cancer.

Consists of C163A, C1QB, CIQC, CATB, CD5L, CH3L1, CO7, FA11, FBLN1, FBLN3, FCG3A, FSTL1, GPX3, IGHG1, IGHG3, IGJ, ISLR, LG3BP, LUM, NRP1, QSOX1, SHBG, SODE and THDE It provides a composition for predicting sorafenib response, comprising a detection reagent of one or more biomarkers selected from the group.

In another embodiment, the present application provides C163A, C1QB, CIQC, CATB, CD5L, CH3L1, CO7, FA11, FBLN1, FBLN3, from a biological sample from a subject in need of administration of sorafenib to provide information necessary for predicting sorafenib drug response. Detecting the concentration of nucleic acid and / or protein of one or more biomarkers selected from the group consisting of FCG3A, FSTL1, GPX3, IGHG1, IGHG3, IGJ, ISLR, LG3BP, LUM, NRP1, QSOX1, SHBG, SODE and THBG ; Comparing the detection result of the concentration of the nucleic acid or protein with the corresponding result of the corresponding marker of the control sample; And comparing with the control sample, when there is a change in nucleic acid or protein concentration of the subject sample, determining the subject as a sorafenib compliant group. Provide a way to judge.

The markers according to the present invention may be used alone or in combination to detect blood samples and are differentially expressed in sorafenib-compliant and non-competent patients, thereby enabling personalized drug prescription of liver cancer patients requiring chemotherapy.

The biomarker according to the present invention can distinguish between patients who are compliant with sorafenib and those who do not with high discrimination, thereby enabling the use of personalized drugs by patients in need of administration of sorafenib. In addition, since the marker according to the present invention acts on various signal transduction pathways, the combination of the multiple markers disclosed herein may contribute to more efficient prognostic prediction, thereby reducing drug side effects and greatly contributing to the reconstruction of health insurance.

Figure 1 schematically shows the MRM analysis process used for biomarker discovery and analysis.

Figure 2 shows the coelution (coelution) of the blood endogenous peptide and the synthetic peptide.

3 shows the Q3 intensity pattern of blood endogenous peptides and synthetic peptides.

4 shows a method for confirming blood endogenous peptide levels.

FIG. 5 shows the results as a dynamic range distribution after measuring blood endogenous peptide levels. FIG.

FIG. 6 shows the concentration in blood for low and high abundance targets.

Figure 7 shows a schematic diagram of the SIS peptide injection concentration determination method.

8 shows a schematic diagram of drug response sample configuration.

9 shows the results of constructing multiple peptide marker panels (accommodation vs. non-compliance before sorafenib treatment).

Figure 10 shows the results of Western blotting analysis for validation of drug response prognostic markers (a) Protein-01. Complement C1q subcomponent subunit C), (b) protein-02. CD5L antigen-like, (c) protein-03. CO7 (Complement component)

Figure 11 shows the results of ELISA analysis for verification of drug response prognostic markers (a) Protein-01. CD5L antigen-like, (b) protein-02. IGHG1 (Ig gamma-1 C region), (C) protein-03. IGHG3 (Ig gamma-3 C region), (d) Protein-04. Igunoglobulin J chain (IGJ), and (e) protein-05. LG3BP (Galectin-3-binding protein).

Figure 12 shows the results of constructing multiple protein marker panels (complied vs. non-conformed groups before sorafenib treatment).

Herein, MRM technology is used to sequentially screen and verify differentially expressed proteins / peptides in normal liver cancer patients and Sorafenib compliance group patients, and show significant differences between groups to determine compliance with Sorafenib. This is based on the discovery of useful proteins / peptides as biomarkers .

In one embodiment, the present disclosure provides a scavenger receptor cysteine-rich type 1 protein M130, C1QB (Complement C1q subcomponent subunit B), CIQC (Complement C1q subcomponent subunit C), CATB (Cathepsin B), CD5L (CD5 antigen-like), CH3L1 (Chitinase-3-like protein 1), CO7 (Complement component C7), FA11 (Coagulation factor XI), FBLN1 (Fibulin-1), FBLN3 (EGF-containing fibulin-like extracellular matrix protein 1), FCG3A (Low affinity immunoglobulin gamma Fc region receptor III-A), FSTL1 (Follistatin-related protein 1), GPX3 (Glutathione peroxidase 3), IGHG1 (Ig gamma-1 chain C region), IGHG3 (Ig gamma-3 chain C region), IGJ ( Immunoglobulin J chain), ISLR (Immunoglobulin superfamily containing leucine-rich repeat protein), LG3BP (Galectin-3-binding protein), LUM (Lumican), NRP1 (Neuropilin-1), QSOX1 (Sulfhydryl oxidase 1), SHBG (Sex hormone -binding globulin), extracellular superoxide dismutase [Cu-Zn] (SODE) and THBG (Thyroxine-binding globulin) Containing a detection reagent of the above biomarkers, seashell penip relates to a reaction composition for the prediction.

Sorafenib {4-[[4-chloro-3- (trifluoromethyl) phenyl] carbamoylamino] phenoxy] -N-methyl-pyridine-2 arboxamide} was developed by Bayer and Onyx Pharmaceutical and sold under the trade name Nexavar. It is a drug based on kinase inhibitors and is used for the treatment of hepatocellular carcinoma and advanced thyroid cancer that is incompatible with radioiodine therapy.

Liver cancer or hepatocellular carcinoma (HCC) as used herein refers to primary malignant tumors that occur in the liver tissue itself in patients with risk factors such as alcohol abuse, viral hepatitis, and metabolic liver disease. According to Line (Korean Society for Liver Cancer, pp17-19, 2014), it can be divided into stage I, II, III, IV A and IV B.

Markers according to the invention can be used to predict whether a liver cancer patient in need of treatment with sorafenib responds to sorafenib.

In one embodiment, the markers according to the present disclosure may be used for the reduction or removal of tumor tissue, removal of tumor cells before or after surgical resection of the tumor, regardless of stage of liver cancer, or even without surgical surgery of the tumor. It can be used to predict whether a patient in need of administration will respond to this drug.

Liver cancer is diagnosed in advanced stages (stage III and above), despite the efforts for early diagnosis, because these patients are not subject to radical treatment such as surgery or radiofrequency ablation. Treatment will be based on chemotherapy. Until recently, almost no systemic chemotherapy has been effective, but it has been used in liver cancer patients since the therapeutic effect of sorafenib, an oral anticancer target drug, has recently been demonstrated in the first phase 3 randomized controlled study of chemotherapy. come.

In this aspect, sorafenib in another embodiment according to the present application is a sorafenib drug in a subject in need of administration that requires chemotherapy that is not subject to curative treatment, such as surgery or radiofrequency thermal therapy, for example, patients with stage III or higher liver cancer. It can be used to predict the response to.

The criteria for evaluating response to sorafenib are based on the best overall survival (BOR) that was seen during the first start of the administration of sorafenib drug and the discontinuation of the drug due to progressive disease (PD) or adverse events. It followed the modified Response Evaluation Criteria in Solid Tumors (mRECIST) guidelines. mRECIST divides the response into four groups with changes in the sum of the tumor diameters as compared to before treatment. Complete response (CR) for complete removal of tumor, partial response (PR) for at least 30% reduction compared to prior treatment, partial response (PR) for 20% to 30% reduction, and at least 20% reduction Increased cases are defined as progressive disease (PD), and herein CR, SD, and PR are defined as drug compliance group and PD as drug non-compliance (or incapacity) group.

As used herein, the term biomarker is a marker that can distinguish between the refractory group and the compliance group for sorafenib before treatment, and includes a protein, a polypeptide derived from the protein, a gene encoding the protein, That fragment.

The biomarker according to the present invention has an increased amount in the blood of the compliant group compared to the samples of the control group such as the non-condensed group, and the protein or nucleic acid sequence is, for example, the UniProt DB (ID) described in Tables 3-1 and 3-2. www.uniprot.org).

In one embodiment according to the present application in particular one or more markers selected from the group consisting of CD5L, IGHG1, IGHG3, IGJ, LG3BP, and QSOX1 having a high AUC value alone are used, wherein the markers are described in the Examples herein. Higher AUC values in both primary and secondary sample groups as shown showed differential expression compared to the control.

Markers according to the invention can be used in one or two or more combinations, for example two, three, four, five combinations. Those skilled in the art will be able to select combinations of markers that meet the desired sensitivity and specificity through methods such as assays using biological samples of subjects, including normal individuals and / or patients, and / or logistic regression assays, such as the methods described herein. Could be.

In one embodiment the combination of markers according to the present application is C163A, C1QB, CIQC, CATB, CH3L1 and any one selected from the group consisting of CD5L, IGHG1, IGHG3, IGJ, LG3BP and QSOX1 having a high AUC value as described above. And combinations of one or more markers selected from the group consisting of, CO7, FA11, FBLN1, FBLN3, FCG3A, FSTL1, GPX3, IGHG1, ISLR, LUM, NRP1, SHBG, SODE and THBG.

In other embodiments, CD5L, IGHG1, IGHG3, IGJ, LG3BP, and QSOX1; FBLN1, LG3BP, CO7, and CD5L; Or in combination with LG3BP, IGHG1, IGHG3, CD5L and IgJ.

The markers according to the invention can be detected at the level of the detection of the presence of nucleic acids, in particular of proteins and / or mRNA, and / or the amount of expression thereof, changes in the amount of expression, difference in the amount of expression through quantitative or qualitative analysis.

Detection herein includes quantitative and / or qualitative analysis, including the detection of presence, absence, and expression level detection. Such methods are well known in the art and those skilled in the art will select appropriate methods for carrying out the present application. Could be.

Detection of such markers according to the present application may be based on functional and / or antigenic characteristics of the marker.

In one embodiment the marker according to the present application can be detected using the detection of the activity or function of the marker, or using a nucleic acid encoding a protein, in particular an agent that specifically interacts at the mRNA level and / or protein level.

In another embodiment the detection of a marker according to the present invention is a mass spectrometry method such as MRM as described herein which detects and quantifies a peptide derived from each marker protein, eg, the corresponding peptide for each marker described in Tables 3-1 and 3-2. It can be carried out through, one or more peptides may be used for one protein. Peptides that can be used for such MRM analysis may not match the antigen recognized by the antibody in the antibody analysis, detection using the peptide and antibody analysis can be used complementary to each other.

In this respect, the detection reagent included in the composition according to the present invention is a reagent capable of detecting the marker according to the present invention through quantitative or qualitative analysis in various ways at the protein or nucleic acid level.

For quantitative and qualitative analysis of the markers according to the present application, various methods for qualitatively or quantitatively detecting known proteins or nucleic acids can be used.

Qualitative or quantitative detection methods at the protein level include, for example, Western blot, ELISA, radioimmunoassay, immunodiffusion, immunoelectrophoresis, tissue immunostaining, immunoprecipitation assay, complement fixation assay, antibody labeled in solution / suspension Methods using binding with, mass spectrometry or protein arrays with antibodies can be used.

Alternatively, as a qualitative or quantitative detection method at the nucleic acid level, a method using a nucleic acid transcription and amplification method, an eTag system, a system based on labeled beads, an array system such as a nucleic acid array, and the like can be used.

Such methods are known and are described, for example, in chip-based capillary electrophoresis: Colyer et al. 1997. J Chromatogr A. 781 (1-2): 271-6; mass spectroscopy: Petricoin et al. 2002. Lancet 359: 572-77; eTag systems: Chan-Hui et al. 2004. Clinical Immunology 111: 162-174; microparticle-enhanced nephelometric immunoassay: Montagne et al. 1992. Eur J Clin Chem Clin Biochem. 30: 217-22, and the like.

In one embodiment according to the present disclosure, the marker may be detected using mass spectrometry, which may be analyzed for example in the manner described in the Examples herein after separating the protein or peptide from the sample. See, eg, Kim, et al. 2010 J Proteome Res. 9: 689-99; Anderson, L et al. 2006. Mol Cell Proteomics 5: 573-88. In one embodiment, multiple reaction monitoring (MRM) techniques using, for example, Triple Quadrupole LC-MS / MS, QTRAP, and the like are used. MRM is a method that can quantitatively and accurately measure multiple substances such as traces of biomarkers in a biological sample. The first mass filter (Q1) is used to detect protons or moions among ion fragments generated from ionization sources. Optionally pass to the crash tube. Proton ions that reach the impingement tube then collide with the internal impingement gas, split and form product or daughter ions to be sent to the second mass filter Q2, where only the characteristic ions are transferred to the detector. In this way, it is a high selectivity and sensitivity analysis method that can detect only the information of the desired component. See, for example, those described in Gillette et al., 2013, Nature Methods 10: 28-34.

In another embodiment, a binding agent or array comprising binding agents that specifically binds to each protein or mRNA from a gene encoding the protein is used.

In another embodiment, a sandwich-type immunoassay such as Enzyme Linked Immuno Sorbent Assay (ELISA) or Radio Immuno Assay (RIA) may be used. This method involves a biological sample on a first antibody bound to a solid substrate such as beads, membranes, slides or microtiterplates made of glass, plastic (eg polystyrene), polysaccharides, nylon or nitrocellulose. After the addition of a label, a label capable of direct or indirect detection may be labeled with a radioactive substance such as ³ H or ¹²⁵ I, a fluorescent substance, a chemiluminescent substance, hapten, biotin, digoxygenin, or the like, or the action of a substrate. Proteins can be detected qualitatively or quantitatively through binding of conjugated antibodies with enzymes such as horseradish peroxidase, alkaline phosphatase, and malate dehydrogenase, which are capable of developing or emitting light.

In other embodiments, immunoelectrophoresis such as Ouchterlony plates, Western blots, Crossed IE, Rocket IE, Fused Rocket IE, Affinity IE, which can simply detect markers through antigen antibody binding, can be used. The immunoassay or method of immunostaining is described in Enzyme Immunoassay, ET Maggio, ed., CRC Press, Boca Raton, Florida, 1980; Gaastra, W., Enzyme-linked immunosorbent assay (ELISA), in Methods in Molecular Biology, Vol. 1, Walker, JM ed., Humana Press, NJ, 1984 and the like. By analyzing the final signal intensity by the above-described immunoassay process, that is, by performing a signal contrast with a normal sample, it is possible to diagnose whether the disease occurs.

Reagents or materials used in such methods are known and include, for example, antibodies, substrates, nucleic acids or peptide aptamers that specifically bind to the marker, or receptors or ligands or cofactors that specifically interact with the marker. And the like can be used. Reagents or materials that specifically interact with or bind to the markers of the present disclosure may be used with chip or nanoparticles.

Markers herein can also be detected quantitatively and / or qualitatively using a variety of methods known at the nucleic acid level, particularly at the mRNA level.

Qualitative or quantitative detection methods at the nucleic acid level include, for example, reverse transcriptase polymerase chain reaction (RT-PCR) / polymerase chain reaction, competitive RT-PCR, for detection of mRNA levels, expression levels or patterns. Real-Time RT-PCR, Nuclease Protection Assays (NPA) For example, RNase, S1 nuclease assays, in situ hybridization, DNA microarrays or chips or Northern blots can be used, such assays are known and also known. It may be carried out using commercially available kits and one skilled in the art will be able to select the appropriate one for the practice herein. For example, Northern blots can be used to determine the size of transcripts present in a cell, have the advantage of using a variety of probes, NPAs are useful for multiple marker analysis, and in situ hybridization can be used for cells of transcripts such as mRNA. Or, it is easy to locate in the tissue, and reverse transcription polymerase chain reaction is useful for detecting a small amount of sample. In addition, an array including a binding agent or a binding agent that specifically binds to a nucleic acid such as mRNA or cRNA from a gene encoding a biomarker protein according to the present application can be used.

Reagents or substances used in the method for detecting the biomarker at the nucleic acid level are known, for example, as a detection reagent in a method for measuring the presence and amount of mRNA by RT-PCR, for example, a polymerase. , Probes and / or primer pairs specific for the mRNA of a marker herein. "Primer" or "probe" refers to a nucleic acid sequence having a free 3 'hydroxyl group capable of complementarily binding to a template and allowing reverse transcriptase or DNA polymerase to initiate replication of the template. do. As used herein, the detection reagent may be labeled with a colorant, luminescent or fluorescent substance as described above for signal detection. In one embodiment, Northern blot or reverse transcription PCR (polymerase chain reaction) is used for mRNA detection. In the latter case, the RNA of a sample is specifically isolated from mRNA, and then cDNA is synthesized therefrom, and then a specific gene or a combination of primers and probes is used to detect a specific gene in the sample. Or it is a method which can determine the expression amount. Such methods are described, for example, in Han, H. et al, 2002. Cancer Res. 62: 2890-6.

The detection reagent included in the composition according to the present application may be labeled directly or indirectly in a sandwich form for detection depending on the specific method used for detection. In the case of the direct labeling method, serum samples used for arrays and the like are labeled with fluorescent labels such as Cy3 and Cy5. In the case of a sandwich, an unlabeled serum sample is first detected by reacting with an array to which a detection reagent is attached, followed by binding to a target protein with a labeled detection antibody. In the case of the sandwich method, the sensitivity and specificity can be increased, and thus the detection can be performed up to pg / mL level. In addition, radioactive materials, coloring materials, magnetic particles and high-density electron particles may be used as the labeling material. Fluorescence luminosity can be used with scanning confocal microscopy, for example Affymetrix, Inc. Or Agilent Technologies, Inc.

The compositions herein may further comprise one or more additional ingredients required for analysis, and may further include, for example, buffers, reagents for sample preparation, blood sampling syringes or negative and / or positive controls.

The composition of the present invention comprising various detection reagents as described above is for ELISA analysis, dip stick rapid kit analysis, MRM analysis kit, microarray, gene amplification, or immunity depending on the assay. It may be provided for analysis and the like, and an appropriate detection reagent may be selected according to the analysis mode.

In one embodiment an ELISA or dipstick rapid kit is used, wherein an antibody that recognizes one or more markers according to the present application is attached to a substrate, such as the surface of a well or glass slide of a multiwell plate or nitrocellulose. Can be. In the case of dipsticks, a technique widely used in the field of point of care test (POCT), in which one or more antibodies recognizing a biomarker according to the present invention is bound to a substrate such as nitrocellulose, which is in contact with a sample such as serum. For example, when one end of a dipstick is immersed in a serum sample, the sample detects the marker in such a manner that the sample moves by the capillary phenomenon and develops color upon binding to the antibody in the substrate.

In another embodiment, an MRM kit is provided based on peptide detection and / or quantification, and the MRM scheme is as described above. The MRM method uses a peptide that selectively recognizes a specific protein, and can more stably detect a marker in a biological sample as compared with a conventional method using an antibody sensitive to the environment such as temperature and humidity. For example, the peptides described in Tables 3-1 and 3-2 herein may be used, and one or two or more peptides may be used in one marker. For example, the peptide corresponding to each protein marker (indicated by a short amino acid) is Scavenger receptor cysteine-rich type 1 protein M130 (C163A)-LVDGVTECSGR; Complement C1q subcomponent subunit B (C1QB) -IAFSATR, LEQGENVFLQATDK; Complement C1q subcomponent subunit C (C1QC) -FQSVFTVTR, TNQVNSGGVLLR, etc.

In other embodiments, it may be provided in the form of an array or chip comprising a microarray. Detection reagents may be attached to the surface of a substrate such as glass or nitrocellulose, and array fabrication techniques are described, for example, in Schena et al., 1996, Proc Natl Acad Sci USA. 93 (20): 10614-9; Schena et al., 1995, Science 270 (5235): 467-70; And U.S. Pat. Nos. 5,599,695, 5,556,752 or 5,631,734. Detection reagents that can be attached to an array include, for example, antibodies, antibody fragments, aptamers, aviders, or peptidomimetics capable of specific binding to a protein.

In another aspect the present invention relates to a kit or system for predicting sorafenib response comprising a reagent for detecting a biomarker. Detection reagents and methods in which such reagents are used are described above. Reagents capable of detecting such markers of the present application may be separately dispensed in a compartment in which the compartment is divided, and in this sense, the present application also relates to an apparatus / apparatus comprising compartmentally containing the marker detection reagent of the present application. The kit may also include additional instructions for use.

In another embodiment, the present disclosure provides C163A, C1QB, CIQC, CATB, CD5L, CH3L1, CO7, FA11, FBLN1, from a biological sample from a subject in need of administration or treatment of sorafenib to provide information necessary for predicting sorafenib response. Concentration or expression of nucleic acids and / or proteins of one or more biomarkers selected from the group consisting of FBLN3, FCG3A, FSTL1, GPX3, IGHG1, IGHG3, IGJ, ISLR, LG3BP, LUM, NRP1, QSOX1, SHBG, SODE and THBG A method for detecting a sorafenib response predictive marker in vitro, comprising detecting an amount.

This method further comprises comparing the results of detecting the concentration or expression of said nucleic acid or protein with a corresponding result of the corresponding biomarker of the control sample; And comparing the control sample with a change in nucleic acid or protein concentration of the subject sample or when there is a change in the presence or absence of the nucleic acid or protein. It includes.

In one embodiment according to the present application, the control sample is a sample from a non-reactive group patient who is previously administered sorafenib but has no therapeutic effect.

In one embodiment according to the present application, when the expression level of the marker according to the present invention is increased compared to the control group (non-conforming group), the subject is determined to be a sorafenib compliant group.

The methods herein include mammals, in particular humans. Human subjects include those deemed necessary, expected or in need of treatment with sorafenib due to liver cancer. Subjects in need of administration or treatment of sorafenib include those who need or are expected to need treatment with sorafenib.

In one embodiment according to the present invention, the method according to the present application is used in patients with liver cancer who are not subject to radical treatment such as surgery or radiofrequency thermal therapy and need to treat chemotherapy based on chemotherapy (advanced stage-stage III or above). However, it is not limited thereto.

The biological sample used in the method according to the present invention may be whole blood, serum or plasma.

As used herein, biological samples refer to substances or mixtures of substances that include one or more components capable of detecting a biomarker and include, but are not limited to, organisms, particularly body fluids, in particular whole blood, plasma, serum or urine.

The biomarker detection method used in the present method, the reagents used in the method, and the data analysis method for the determination may refer to the above and the following. In one embodiment the method according to the invention can be carried out in particular by protein or nucleic acid microarray analysis, nucleic acid amplification, antigen-antibody reactions, or by mass spectrometry including MRM.

Markers, or combinations of markers, used in the methods according to the present disclosure may be referred to above.

When using a combination of markers comprising two or more using the method according to the invention, a dataset can be generated that includes a profile, ie, quantitative information related to the expression of marker proteins in a sample. After obtaining a profile using the marker, the subject's sample is determined to be compliant by comparing the results with the reference or control group. As a control or reference group, reference may be made to the above description.

Known methods can be used to compare marker profiles between control groups and test groups using samples. For example, digital image comparison of expression profiles, comparisons using DB for expression data, or U.S. See patents 6,308,170 and 6,228,575.

Profiles obtained through marker detection according to the present application can be processed using known data analysis methods. Examples include nearest neighbor classifiers, partial-least squares, SVM, AdaBoost, and clustering-based classification methods, for example Ben-Dor et al (2007, J. Comput. Biol. 7: 559-83), Nguyen et al (2002, Bioinformatics 18: 39-50), Wang et al (2003, BMC Bioinformatics 4:60), Liu et al (2001, Genome Inform. Ser. Workshop Genome Inform. 12: 14-23), Yeang et al (2001, Bioinformatics 17 Suppl 1: S316-22) and Xiong (2000, Biotechniques 29 (6): 1264-8, 1270) and the like.

In addition, various statistical processing methods may be used to determine that the result detected through the marker of the present application is significant in determining compliance. As a statistical processing method, a logic regression method is used in one embodiment, and may be referred to Ruczinski, 2003, Journal of Computational and Graphical Statistics 12: 475-512. The method is similar to the CART method in which a classifier is presented as a binary tree, but each node uses a more general Boolean operator associated with the property, compared to the "and" operator generated by CART. Examples of other analysis methods include nearest shrunken centroids (Tibshirani. 2002 PNAS. 99: 6567-72), random forests (Breiman. 2001. Machine Learning 45: 5-32, and MART (Hastie. 2001. The Elements of Statistical Learning, Springer). ).

In one embodiment, statistical processing may determine the level of confidence in the significant difference between the test substance and the control to determine compliance. The raw data used for statistical processing are the values analyzed in duplicate, triple or multiple for each marker.

This statistical analysis method is very useful for making clinically meaningful judgments through statistical processing of biomarkers as well as clinical and genetic data.

Hereinafter, examples are provided to help understand the present invention. However, the following examples are provided only to more easily understand the present invention, and the present invention is not limited to the following examples.

Example

Example 1. Clinical Sample Information Used herein

The experiment was conducted in accordance with the protocol approved by the Medical Research Ethics Review Board of Seoul National University.

To select marker candidates, 60 healthy subjects (male / female = 41/19) and 60 liver cancer patients (male / female = 42 / The initial study was conducted with the sample of 18). Samples are treated based on the best therapeutic response seen during the first start of Sorafenib medication on a BOR (Best Overall Survival) basis and withdrawal of the drug due to progressive disease (PD) or adverse events (S / E). Drug response non-responders were all cases of progression to drug response compliance (Clearance = CR), partial response (PR), and stable (Stable disease = SD). Specified. Through this, the primary sample group was treated with 21 sorafenib drug-response groups (CR-2, PR-3, SD-16) and 44 sorafenib drug-responsive groups (PD-44). The latter pair was used, and the secondary sample group consisted of 20 sorafenib drug-response groups (CR-2, PR-2, SD-16) and 31 sorafenib drug-resistant groups ( Pair samples before / after treatment of PD-44 patients) were used.

Example 2 Data Mining for Target Candidate Selection

The candidate group of the target was selected as follows. Sorafenib drug response protein marker candidates were obtained using the LiverAtlas database, the most comprehensive resource currently associated with liver disease. A total of 50,265 proteins known to be related to liver disease in the LiverAtlas database are selected from among the proteins that can be secreted or secreted into the blood (Uniprot database). A total of 1,683 proteins were selected. Finally, peptide MS / MS using four different peptide MS / MS library sources (NIST Ion-Trap, NIST Q-TOF, ISB human plasma, Home made library) to select only proteins that can be detected by mass spectrometry. As a result, only 960 proteins were selected. Subsequently, detection target group selection was performed as follows. In order to select only targets that can be detected by the actual mass spectrometry equipment, only peptides with proper signal detection through MRM analysis were selected from 60 normal groups and 60 HCC groups. 1316 peptides were selected.

Example 3. Derivation of target marker candidate group through MRM analysis

Example 3-1. Serum Sample Preparation

Serum samples used in the experiment were prepared as follows. Blood was collected using a BD Vacutainer serum separation tube (BD, USA) (Silica clot activator, 10 mL, 16 × 100 mm; Product number = # 367820). Phenylmethyl sulfonyl fluoride (PMSF) was added to the final blood to 1.0 mM. After 10 times of inverting mix, the supernatant was collected by centrifugation (maintaining 4 ° C.) at 3000 rpm for 10 minutes (after color observation, hemolyzed red sample was not used).

Dispense 100 μL into a 1.5 mL Eppendorf tube, place on ice immediately after dispensing, and label the tube (code label corresponding to sample group when labeling (eg NC-01, MC-01, PD-01, etc.) It was then immediately stored at −80 ° C. The procedure was performed on ice and was all finished within 1 hour after blood collection.

Example 3-2. Serum Protein Depletion Treatment

In order to discover the markers present in the blood at low concentrations, a depletion process was first performed to remove 6 proteins (albumin, IgG, IgA, haptoglobin, transferrin, alpha-1-antitrypsin) present in the blood at high-abundant levels. . By removing six large proteins, about 85% of the total protein mass was removed from the blood, and only the proteins corresponding to the remaining 15% protein mass were used for analysis. Depletion uses MARS (Multiple affinity removal system) (Agilent, USA) according to the manufacturer's method to remove the six most abundant proteins from the blood by binding them to the column and eluting small amounts of unbound proteins. Used for analysis.

Example 3-3. Peptides of Serum Proteins

Serum samples obtained after the depletion process were concentrated (w / 3K filter), and protein concentration was quantified by BCA (Bicinchoninic acid) analysis. 100 μg serum samples were taken and then treated with a final concentration of 6M urea / 20mM DTT (Tris pH 8.0), followed by incubation at 37 ° C. for 60 minutes. The final concentration was treated with 50 mM IAA, and then incubated at room temperature for 30 minutes. 100 mM Tris pH 8.0 was treated so that the concentration of urea was 0.6 M or less. Trypsin treatment was performed so that the ratio of trypsin and serum was 1:50, and then incubated at 37 ° C. for 16 hours. The formic acid solution was treated to a final concentration of 5% and then subjected to the following desalting process.

Example 3-4. Desalination of serum proteins

Activation was performed by pouring 1 mL of 60% ACN / 0.1% formic acid three times on an OASIS column (Waters, USA). Equilibration was performed by pouring 1 mL of 0.1% formic acid five times into an OASIS column. Peptide samples were added and washed 5 times with 1 mL of 0.1% formic acid. Peptides were eluted with 1 mL of 40% ACN / 0.1% formic acid and 1 mL of 60% ACN / 0.1% formic acid. It was frozen at −70 ° C. for at least 1 hour and then dried by Speed-vac. The dried peptide samples were dissolved in 50 μl of Sol A buffer (3% ACN / 0.1% formic acid), centrifuged at 15,000 rpm for 60 min, and only 40 μl was transferred to the vials for analysis.

Example 3-5. MRM Analysis

For MRM analysis In the first screening step, in order to select targets that can be reproducibly detected in 537 proteins and 1316 peptides selected as in Example 2 and show differences between the sorafenib conformation and non-compliance groups. Samples from 60 normal patients and 60 liver cancer patients were used. The primary sample was a sample of 21 pairs of patients (pre / post-treatment) and 44 non-treatment groups (pre / post-treatment) who were compliant with Sorafenib drug. As a sample, a 20-pair patient group and a disabled patient group 31-pair sample which comply with the sorafenib drug were confirmed, and 24 protein markers which can distinguish whether the patient adhered or not to sorafenib drug were finally confirmed. The analysis was repeated with MRM three times per set.

MRM analysis was performed using Skyline (http://proteome.gs.washington.edu/software/skyline) for each target protein to select peptide and fragment ions for MRM analysis. Skyline is open source software for MRM method development and analysis (Stergachis AB, et al., 2011, Nat Methods 8: 1041-1043).

In summary, full-length protein sequences were entered into Skyline in FASTA format and then designed as peptides to generate a product ion list, which was monitored by MRM. Peptide filter conditions used for transition selection were as follows: peptide maximum length was 30, minimum length was 6 amino acids and did not include repeated arginine (Arg, R) or lysine (Lys, K). If methionine (Met, M) was also included in the peptide, it was removed due to the possibility of modification. It was also not used when proline came after arginine or lysine, but when histidine (His, H) was included, the charge was changed but used.

Peptides satisfying these conditions were used as the Q1 transition. Quadruple 1 (Q1) served as a filter that can pass only certain Q1 m / z. Precursor ions that passed through the Q1 filter were fragmented by electrical energy in Quadruple 2 (collision cells), which were broken down into product ions. This product ion can pass only certain product ions through Quadruple 3 (Q3), which acts as a filter as in Quadruple 1 (Q1). The ions that passed through Quadruple 3 (Q3) were converted into digital signals at the detector and shown as peak chromatograms. The area of these peaks was analyzed for relative and absolute quantitative analysis. A file containing this information was entered into Analyst (AB SCIEX, USA) for MRM analysis and analyzed using nonscheduled MRM method. The MRM results were generated as wiff and wiff.scan files, which were converted into mzXML format using mzWiff and inputted into Skyline for MRM data processing to obtain peak intensities of MRM transitions. Figure 1 shows the process diagrammatically.

For the purpose of ensuring quantitation, all peptides are known to have a concentration that is normalized to the peak area of the peptide when the MRM is analyzed. This peptide is called an internal standard peptide, which is a stable isotope. It is a peptide having an amino acid containing. In this study, all samples were injected with LNVENPK, a heavy-labeled peptide from beta-galactosidase (lacZ) derived from E. coli, which does not exist in the human proteome. The peak area values of all target peptides from the MRM analysis were normalized to the peak area values of the corresponding internal standard peptides.

Through semi-quantitative MRM analysis as above, it was repeated three times per set to select only proteins / peptides reproducibly detectable by mass spectrometry (MRM-technique), and then at least As a result, only 347 proteins and 754 peptides were selected as reproducible targets.

Then, the p-value and fold-change levels between the groups were checked to select targets with reproducible detection of expression differences between normal and HCC groups. Following a normal distribution (Normality test, Shaipro-Wilk) for a reproducible target (347-protein, 754-peptides) (p-value> 0.05) is an independent T-test parameter. In case of not following the normal distribution (p-value ≤ 0.05), the analysis was performed by the nonparametric Mann-Whitney test. In the case of the independent T-test, the p-value value was determined by dividing it by the equal dispersion (P-value> 0.05) and the non-uniformity (p-value ≤ 0.05) through the Lebenes test. Independent T-test and Mann-Whitney test confirmed that the target protein / peptide showed a significant difference (p-value ≤ 0.05) between the normal group and the HCC group.

In addition, candidate protein marker groups were further selected that showed a pattern of increase or decrease in fold-change levels between the normal and HCC groups of 1.5-fold or more. As a result, 195 proteins and 443 peptides with a p-value of 0.05 or less were identified between the normal group and the HCC group, and 191 proteins and 389 peptides with 1.5 fold-change or more were identified. 492 peptides were selected as target candidates.

To determine whether the selected 227 proteins and 492 peptides were unique peptides on the entire human proteome, unique peptides were identified using NCBI's BlastP search program. Through this, 15 proteins corresponding to 37 non-unique peptides were excluded, and 216 proteins and 460 peptides were finally selected as target candidate groups, which are unique peptides, with significant differences between the final normal group and the HCC group.

Example 4. Confirmation of Blood Endogenous Peptides of Target Candidate Markers

Whether the proteins and peptides selected in Example 3 were present in actual blood was confirmed as follows.

To this end, peptides containing selected peptides in real blood were sampled using a sample of 216 protein-stable-isotope labeled standard (SIS) peptides pooled together with 60 normal groups and 60 HCC groups. Double-checked whether The SIS peptide is a peptide obtained by substituting ¹³ C and ¹⁵ N for ¹² C and ¹⁴ N in amino acids of lysine (Lys, K) or arginine Arg, R) of the peptide C-terminus. It differs from the endogenous peptide in the blood, but because it has the same sequence, the peptide hydrophobicity is the same, so it elutes at the same time (RT) as the peptide in the blood on the chromatogram. (See FIG. 2).

In the MRM analysis, the Q3 intensity pattern of the SIS peptide and the peptide in the blood was analyzed to determine whether the complex blood sample exhibited signal interference by a peptide other than the target peptide. It confirmed (FIG. 3).

To determine whether or not signal interference occurs, the automated detection of inaccurate and imprecise transitions (AuDIT) programs are used to compare the relative generated ion intensities of the peptides in the blood and the SIS peptide (P-value threshold: 0.05). And whether the peak area value of the SIS peptide is constantly detected during repeated measurements (CV threshold: 0.2) was confirmed.

Through this, 123-protein and 231-peptide were finally selected as quantifiable target proteins / peptides without signal interference.

Example 5. Identification of Levels of Blood Endogenous Peptides

As in Example 4, the endogenous level present in the blood was confirmed for the quantifiable blood endogenous target (123-protein protein, 231-peptide). MRM analysis of a sample in which 231-SIS peptide mixture was sequentially diluted to 3-point (20 nM, 200 nM, 2000 nM) in 10 μg of a sample in which 60 normal groups and 60 HCC groups were combined was mixed. The signal for and the complementary synthetic (SIS) peptide signal were identified.

Calculate the relative ratios of the peak area values of peptides in the blood and the peak area values of complementary SIS peptides (3-points) for only the generated ions (Q3) without interference. And, by multiplying the injected SIS-peptide amount, the level of blood in the target peptide was finally confirmed (FIG. 4).

In blood levels of 231 peptides, the lowest concentration found in the serum was ISLR (Immunoglobulin superfamily containing leucine-rich repeat protein) and the concentration was 0.15-fmol / μg. The highest concentration of protein was A2MG (Alpha 2 macroglobulin), and the concentration was found to be 5.57-pmol / μg (Dynamic range: 3.7 × 10 4 orders) (FIG. 5).

As a result of the level measurement in blood, the low-abundance targets determined to be below 20-fmol / μg in blood were 34-protein, 36-peptide, and 20-fmol / μg and 2000-fmol / μg Middle-abundance targets measured between 93-protein, 174-peptide, and high-abundance targets measured above 2000-fmol / μg were 11-protein, 22-peptide. It was confirmed (FIG. 6).

Example 6 Derivation of Liver Cancer Early Diagnosis Marker Candidates through Actual Individual Sample Application

Validated blood endogenous peptides were determined for SIS peptide injection concentrations complementary thereto. For low abundance targets (level in blood <20-fmol / μg), all were injected with 20-fmol SIS-peptides in batch, and middle-abundance targets (20-fmol / μg < For blood level <2000-fmol / μg), the amount of SIS peptide was injected equal to the amount of peptide in the blood, and for high-abundance targets (blood level> 2000-fmol / μg) all 2000-fmol amount of SIS-peptide was injected.

For the ISLR protein with the lowest level in blood, 20-fmol of SIS-peptide was injected up to 13 times higher than the level in the blood, with the highest level of A2MG protein. In the case of injecting 2000-fmol of SIS-peptide, a low amount diluted up to 1/28 times compared to the level in blood was injected (FIG. 7).

When the drug response criterion is treated as a treatment response, the final response is ultimately ultimately due to a small number of CR, PR or SD (if discontinued due to drug side effects) and most are classified as PD. Does not help determine the effectiveness of the drug. Therefore, mRECIST (Modified Response Evaluation Criteria in Solid Tumors) based on the best overall survival (BOR) that occurred during the first chemotherapy and discontinuation of the drug with PD or S / E (Side Effect). The screening was divided into two groups based on the scale.

After drug treatment, the complete disappearance of the tumor (CR) was selected as the response response group, and the other group in which the tumor size was reduced by more than 30% (PR, partial response), with little change Three groups (SD, Stable disease) and more than 20% increased (PD) were selected as non-responders.

A 65-Paired (pre- and post-treatment) liver cancer patient who received sorafenib treatment as the primary sample for the early diagnosis of the final validated target candidate group (123 proteins, 231 peptides) was used (Table 1). Drug response compliance group 21-paired and drug non-response groups 44-Paired samples were used, and the primary sample used 51-Paired samples (Table 2), which was the drug response compliance group 20-paired sample and drug non-response group Markers were derived using 31-Paired samples.

After blinding, the MRM analysis sequence was randomly analyzed so that the experimenter could not identify the patient group, and the analysis was repeated three times per sample. The peak area values of the target peptides thus obtained are normalized to the peak area values of the SIS peptides complementary to each other, and then IBM SPSS statistics (version 21.0) and GraphPad (version 6.00) are obtained. I went through the analysis.

[Table 1] Clinical sample information (primary sample) used for the selection of markers for predicting drug response prognosis

[Table 2] Clinical Sample Information (Secondary Sample) Used to Select Prognostic Markers for Drug Response Prognosis

Example 7 Derivation of a Single Marker Predicting Sorafenib Response

Sorafenib, a liver cancer targeted drug, is useful as an early treatment for advanced liver cancer, but has a limit of less than 5% in tumor size reduction and an increase in survival time of less than 3 months, which is not effective for price. . Predicting the prognosis of sorafenib drug response was analyzed by prioritizing the targets showing differences in expression between the drug-responsive and drug-responsive patients prior to sorafenib treatment. The schematic diagram is shown in FIG.

As a result of the training set, a highly diagnostic (AUC-value> 0.700) target was identified as 32-protein, 44-peptide before the sorafenib treatment. As a result of the secondary analysis (Test set), a target with a high diagnostic difference (AUC-value ≥ 0.700) between the acclimatized patient group and the non-accepted patient group was identified as 46-protein, 7-peptide before treatment with sorafenib.

Common in the results of the training set and the test set, there was a significant difference (P-value ≤ 0.05) between the acclimatized and non-complied groups before the sorafenib treatment and high diagnostic ability (AUC- velue ≥ 0.700) Target was finally identified with 24 proteins and 40 peptides (Tables 3-1 and 3-2).

When comparing the two groups (control vs. case) in Tables 3-1 and 3-2, red shows that the AUC value differs by more than 0.7, increasing the expression level in the case group. In blue, the AUC value is more than 0.7, and the expression level is decreased in the case group. Black indicates that the difference between the two groups is less than or equal to the AUC value of 0.7.

[Table 3-1] Target List of AUC Values Above 0.700 after Analysis of Drug Response Individual Samples

[Table 3-2] Target List of AUC Values Above 0.700 after Analysis of Drug Response Individual Samples

Example 8 Derivation of Multiple Markers for Predicting Sorafenib Response

Multivariate analysis was performed on 24 proteins and 40 peptides with high diagnostic ability in both the acclimatized and non-complied patients before and after sorafenib treatment in the training set and test set results. -variate analysis (MA) was used to build and compare multiple protein marker panels.

As a result of analysis of binding to one panel using logistic regression, one of the statistical methods, five peptide panels (FBLN1 + LG3BP + CO7 + CO7 +) were compared between the drug response compliance group and the non-drug response group before sorafenib treatment. CD5L) was identified as the combination with the highest diagnostic power.

Before the sorafenib treatment, the AUC value of the five peptide panels was 0.980 as compared between the drug response compliance group and the non-compliance group, and 37 of the 41 sorafenib treatment compliance groups were adapted through the five peptide panels. (Accracy 90.2%), 71 out of 75 Sorafenib-treated non-competents (Accuracy 94.7%) can be diagnosed, and the accuracy of diagnosis using the 5-peptide (4-protein) marker panel was 93.1%. (FIG. 9). When the remaining 20-protein (35-peptide) is added to the 4-protein (5-peptide), it was confirmed that it is possible to make a myriad of protein combinations of AUC 0.980 or more.

Example 9 Further Validation with Antibodies to Markers Predicting Liver Cancer Drug Response Prognosis

Markers according to the present application can be detected at the protein level, the detection method at the protein level may include the analysis using the MRM analysis and the antibody used in the present embodiment, only one, that is, the MRM analysis alone provides the desired result according to the present application. You can either get it or use both methods for reconfirmation.

Western Blotting (Protein Level) was performed on 7 proteins with high AUC values among 24-proteins validated by MRM analysis (Peptide Level) for the purpose of predicting hepatic cancer sorafenib prognosis (Table 4).

Since antibodies recognize antigens consisting of some residues in the whole protein, the antibody selection criteria allowed the peptide portion analyzed by MRM to be included (or as close as possible) in the antibody immunogen portion and for plasma / serum. Those with Western blotting results, and the presence of monoclonal antigens were selected first, and the antibodies used were as described in Tables 3-1 and 3-2, and the antibodies were purchased from Santa Cruz Biotechnology, USA.

To perform western blotting, 12 samples were selected per four groups (before and after treatment for drug compliance and non-drug) and randomized to samples used in primary and secondary MRM analysis. Protein Level Validation was conducted to identify markers for predicting liver cancer drug response. SDS-PAGE gel-to-Variation corrections were performed by normalizing to the O.D intensity measured through the integrated samples for all selected groups. As a control, mouse monoclonal antibodies against betaactin and transferrin were used.

The results are described in FIG. These results indicate that the marker according to the present application can be analyzed through various protein analysis methods.

[Table 4] Antibody Information List

ELISA analysis for the purpose of immediately applying / utilizing 7 target proteins with the highest distinction among groups among 24-proteins verified by MRM analysis (Peptide Level) for the purpose of predicting sorafenib prognosis. We added (Protein Level, Native form). For the ELISA analysis, 40 samples were selected per 2-group (prior to drug sorafenib, drug compliance and refractory), and randomized selection of samples used in the first and second MRM analysis was performed. Protein Level Validation (Native Form) was conducted to identify drug response prediction markers.

As a result of ELISA analysis on the seven ELISA targets, five targets (CD5L, IGHG1, IGHG3, IgJ, and LG3BP) that matched the expression pattern of MRM were identified. Among them, three or more of AUC 0.700 [IGHG3 (0.704) were identified. ), CD5L (0.764), and IgJ (0.779)] (FIG. 11).

Example 10 Predicting Drug Response Prognosis ELISA Results, Multiple Marker Verification

Peptides, not proteins, can be measured and used as markers for quantification by MRM.

As a result of ELISA analysis, multi-variate analysis (MA) was performed on five targets (CD5L, IGHG1, IGHG3, IgJ and LG3BP) that matched the expression pattern of MRM, and built a multi-protein marker panel. And compared.

A panel analysis using logistic regression, one of the statistical methods, showed that the AUC value was 0.811 when constructing five protein panels, which was 29 acclimatized among the 40 sorafenib treatment compliance groups. In the group (Accracy 72.5%), 28 out of 40 Sorafenib-treated groups were diagnosed as non-accumulated (Accuracy 70.0%), and the accuracy of diagnosis using the 5-protein marker panel was 71.25% (Fig. 7). 12).

Claims (16)

1. Scavenger receptor cysteine-rich type 1 protein M130, C1QB (Complement C1q subcomponent subunit B), CIQC (Complement C1q subcomponent subunit C), CATB (Cathepsin B), CD5L (CD5 antigen-like), CH3L1 (Chitinase-3 -like protein 1), CO7 (Complement component C7), FA11 (Coagulation factor XI), FBLN1 (Fibulin-1), FBLN3 (EGF-containing fibulin-like extracellular matrix protein 1), FCG3A (Low affinity immunoglobulin gamma Fc region receptor III-A), FSTL1 (Follistatin-related protein 1), GPX3 (Glutathione peroxidase 3), IGHG1 (Ig gamma-1 chain C region), IGHG3 (Ig gamma-3 chain C region), IGJ (Immunoglobulin J chain), Immunoglobulin superfamily containing leucine-rich repeat protein (ISLR), LG3BP (Galectin-3-binding protein), LUM (Lumican), NRP1 (Neuropilin-1), QSOX1 (Sulfhydryl oxidase 1), SHBG (Sex hormone-binding globulin), One or more biomarkers selected from the group consisting of extracellular superoxide dismutase [Cu-Zn] (SODE) and thyroxine-binding globulin (THBG) Of containing the detection reagent, seashell penip reaction composition for the prediction.
2. The method of claim 1,

   The marker is selected from the group consisting of CD5L, IGHG1, IGHG3, IGJ, LG3BP and QSOX1, composition for predicting sorafenib response.
3. The method of claim 1,

   The marker is any one selected from the group consisting of CD5L, IGHG1, IGHG3, IGJ, LG3BP and QSOX1 and C163A, C1QB, CIQC, CATB, CH3L1, CO7, FA11, FBLN1, FBLN3, FCG3A, FSTL1, GPX3, IGHG1, A composition for predicting sorafenib response, which is a combination of one or more markers selected from the group consisting of ISLR, LUM, NRP1, SHBG, SODE, and THBG.
4. The method of claim 1,

   Said one or more markers include CD5L, IGHG1, IGHG3, IGJ, LG3BP and QSOX1; FBLN1, LG3BP, CO7 and CD5L, or LG3BP, IGHG1, IGHG3, CD5L and IGJ, composition for predicting sorafenib response.
5. The method of claim 1,

   The detection reagent is a reagent that can detect the marker at the protein or nucleic acid level, composition for predicting sorafenib reaction.
6. The method of claim 5, wherein

   The reagent for detecting the protein level is Western blot, ELISA, radioimmunoassay, immunodiffusion, immunoelectrophoresis, tissue immunostaining, immunoprecipitation assay, complement fixation assay, FACS, mass spectrometry, or a reagent for protein microarray,

   The detection reagent of the nucleic acid level is nucleic acid amplification reaction, polymerase chain reaction, reverse transcription polymerase chain reaction, competitive polymerase chain reaction, Nuclease protection assay (RNase, S1 nuclease assay), in situ hybridization, nucleic acid microarray or northern A composition for predicting sorafenib reaction, which is a reagent for lot.
7. The method of claim 6,

   The protein level detection reagent is an antibody, antibody fragment, aptamer, avider or peptidomimetics, receptor, ligand that specifically recognizes the full length or fragment thereof of the marker. Including,

   The nucleic acid level detecting reagent includes a nucleic acid sequence of the marker, a nucleic acid sequence complementary to the nucleic acid sequence, a primer or a probe or a primer and a probe specifically recognizing the nucleic acid sequence and fragments of the complementary sequence. , Composition for predicting sorafenib reaction.
8. The method of claim 1,

   The composition is for ELISA analysis, dip stick rapid kit (dip stick rapid kit) analysis, MRM analysis, microarray, nucleic acid amplification, or immunoassay, composition for predicting sorafenib response.
9. The method of claim 8,

   The composition is for MRM analysis, the peptide of each marker used in the MRM analysis is a composition for predicting sorafenib response, which are described in Table 3-1 and Table 3-2.
10. C163A, C1QB, CIQC, CATB, CD5L, CH3L1, CO7, FA11, FBLN1, FBLN3, FCG3A, FSTL1, GPX3, IGHG1, IGHG3, IGJ, ISLR, LG3BP, LUM, NRP1 from biological samples of subjects in need of sorafenib Detecting the concentration of nucleic acid and / or protein of at least one biomarker selected from the group consisting of QSOX1, SHBG, SODE and THBG;

    Comparing the result of detecting the concentration of the nucleic acid or protein with the result of the corresponding marker of the control sample; And

    Comparing with the control sample, if there is a change in the nucleic acid or protein concentration of the subject sample, comprising the step of determining the subject to the sorafenib compliance group, how to determine the response to sorafenib.
11. The method of claim 10,

    The subject is a liver cancer patient,

    The biological sample is whole blood, serum or plasma.
12. The method of claim 10 or 11,

    Wherein said biomarker is selected from the group consisting of CD5L, IGHG1, IGHG3, IGJ, LG3BP, and QSOX1.
13. The method according to any one of claims 10 to 12,

    The marker is any one selected from the group consisting of CD5L, IGHG1, IGHG3, IGJ, LG3BP and QSOX1 and C163A, C1QB, CIQC, CATB, CH3L1, CO7, FA11, FBLN1, FBLN3, FCG3A, FSTL1, GPX3, IGHG1, And a combination of one or more markers selected from the group consisting of ISLR, LUM, NRP1, SHBG, SODE, and THBG.
14. The method according to any one of claims 10 to 13,

    Said one or more markers include CD5L, IGHG1, IGHG3, IGJ, LG3BP and QSOX1; FBLN1, LG3BP, CO7 and CD5L, or LG3BP, IGHG1, IGHG3, CD5L and IGJ.
15. The method according to any one of claims 10 to 14,

    The detecting step is performed by protein or nucleic acid microarray analysis, nucleic acid amplification, antigen-antibody reaction, or mass spectrometry.
16. The method according to any one of claims 10 to 15,

    The detecting step is performed by MRM analysis, wherein the peptides of each marker used in the MRM analysis are those described in Tables 3-1 and 3-2.

Images