Biomarker for diagnosing highly malignant gastric cancer subtype, and method for predicting prognosis using same

Abstract

The present invention relates to a biomarker capable of distinguishing PCC-NOS subtypes from other subtypes of gastric cancer at the molecular level, and a technique capable of objectively predicting the prognosis of a patient using same. The present invention provides a novel molecular diagnostic and prognostic means capable of early identification of PCC-NOS subtypes and predicting prognosis, and thus is expected to significantly contribute to personalized treatment and improved survival rates for gastric cancer patients.

Description

Biomarkers for diagnosing high-malignancy gastric cancer subtypes, and methods for predicting prognosis using the same

[0001] The present invention relates to a biomarker for diagnosing a high-malignancy gastric cancer subtype and a method for predicting prognosis using the same.

[0002] Gastric cancer is one of the cancers with a high incidence rate worldwide, and South Korea, in particular, is reported to have traditionally been one of the countries with a high incidence of gastric cancer. Despite national early screening programs and advancements in endoscopic technology, gastric cancer remains one of the most common cancers among Koreans and ranks among the leading causes of cancer death. Although gastric cancer is a critical disease for public health and socio-economic aspects, the cure rate remains limited, and the disease has not yet been fundamentally conquered.

[0003] Gastric cancer is classified into various subtypes based on histological and molecular biological characteristics. Since these subtypes exhibit distinct differences in clinical progression, drug responsiveness, and prognosis, classification by subtype is paramount for establishing personalized treatment strategies. However, current clinical classifications of gastric cancer rely primarily on histological observations, which presents a limitation in that there are insufficient objective and molecularly reproducible diagnostic methods.

[0004] In particular, the Poorly Cohesive Carcinoma-Not Otherwise Specified (PCC-NOS) subtype is known as a refractory gastric cancer with the most clinically poor prognosis. The PCC-NOS subtype exhibits low responsiveness to conventional chemotherapy and radiotherapy, and anticancer drug penetration is limited due to the specific structure of the tumor microenvironment and the composition of the extracellular matrix; consequently, recurrence and distant metastasis are frequent. Furthermore, even within the same category of undifferentiated gastric cancer, unlike signet-ring cell carcinoma, PCC-NOS is difficult to distinguish based solely on histological characteristics, and distinct molecular markers for predicting patient prognosis have not been established. Therefore, there is an urgent need for molecular-based technologies capable of accurately diagnosing the PCC-NOS subtype in advance and predicting its poor prognosis early. Beyond simply distinguishing subtypes, this technology can play a key role in selecting patient-specific treatments and identifying targets for new drug development.

[0005] Accordingly, the present invention has been devised to solve the above-mentioned problems and relates to a biomarker capable of distinguishing the PCC-NOS subtype from other subtypes at the molecular level, and a technology capable of objectively predicting a patient's prognosis using the same. By providing a novel molecular diagnostic and prognostic means capable of early identification of the PCC-NOS subtype and prediction of prognosis, the present invention is expected to significantly contribute to personalized treatment and improved survival rates for gastric cancer patients.

[0006] One objective of the present invention is to provide a composition or kit for diagnosing gastric cancer, particularly PCC-NOS subtype gastric cancer.

[0007] Another objective of the present invention is to provide information for diagnosing gastric cancer, particularly PCC-NOS subtype gastric cancer.

[0008] Another objective of the present invention is to provide a method for treating gastric cancer, particularly PCC-NOS subtype gastric cancer.

[0009] Another objective of the present invention is to provide a method for screening candidate substances for the treatment of gastric cancer, particularly PCC-NOS subtype gastric cancer.

[0010] Another objective of the present invention is to provide a composition or kit for predicting the prognosis of gastric cancer, particularly PCC-NOS subtype gastric cancer.

[0011] Another objective of the present invention is to provide information for predicting the prognosis of gastric cancer, particularly PCC-NOS subtype gastric cancer.

[0012] However, the technical problems that the present invention aims to solve are not limited to those mentioned above, and other unmentioned problems will be clearly understood by those skilled in the art from the description below.

[0013] Various embodiments described herein are described with reference to the drawings. In the following description, for a complete understanding of the invention, various specific details, such as specific forms, compositions, and processes, are described. However, specific embodiments may be practiced without one or more of these specific details, or in combination with other known methods and forms. In other examples, known processes and manufacturing techniques are not described as specific details so as not to unnecessarily obscure the invention. Reference throughout this specification to "one embodiment" or "an embodiment" means that the particular features, forms, compositions, or characteristics described in association with the embodiment are included in one or more embodiments of the invention. Accordingly, the context of "in one embodiment" or "an embodiment" expressed at various places throughout this specification does not necessarily represent the same embodiment of the invention. Additionally, particular features, forms, compositions, or characteristics may be combined in any suitable way in one or more embodiments.

[0014] Unless otherwise specifically defined in the specification, all scientific and technical terms used herein have the same meaning as commonly understood by those skilled in the art to which the present invention pertains.

[0015] Throughout the specification, when a part is described as "including" a certain component, this means that, unless specifically stated otherwise, it does not exclude other components but may include additional components.

[0016] In this specification, the term “cancer” below refers to a condition characterized by uncontrolled cell growth, in which a mass of cells called a tumor is formed due to such abnormal cell growth, infiltrates surrounding tissues, and, in severe cases, metastasizes to other organs of the body. Academically, it is also referred to as a neoplasm. Cancer is an intractable chronic disease that, even with treatment through surgery, radiation, and chemotherapy, often fails to achieve a fundamental cure, causes suffering to the patient, and ultimately leads to death. While there are various factors contributing to the development of cancer, they are classified into internal and external factors. Although the exact mechanism by which normal cells transform into cancer cells has not been precisely elucidated, it is known that a significant number of cancers arise under the influence of external factors, such as environmental factors. Internal factors include genetic factors and immunological factors, while external factors include chemical substances, radiation, and viruses. Genes involved in the development of cancer include oncogenes and tumor suppressor genes, and cancer occurs when the balance between them is disrupted by the internal or external factors described above. According to the South Korean Cancer Registry statistics released in 2021, the total number of cancer cases in the country as of 2019 was 254,718, of which gastric cancer accounted for 29,493 cases, ranking third (11.6%). In particular, gastric cancer was reported to rank second among men with 19,761 cases (14.7%) and fourth among women with 9,732 cases (8.1%).

[0017] Traditionally, gastric cancer was classified into intestinal type, diffuse type, and mixed type according to the Lauren classification (1965, widely used prior to the WHO). However, as there were many differences in the clinical progression, drug responsiveness, and prognosis of gastric cancer, the World Health Organization (WHO) further subdivided the gastric cancer subtypes as shown in Table 1 below through its 5th revision in 2019.

[0018] - Tubular adenocarcinoma - Papillary adenocarcinoma - Mucinous adenocarcinoma - Poorly cohesive carcinoma (PCC) - Signet-ring cell carcinoma (SRC) - PCC-NOS (Not Otherwise Specified) - Mixed carcinoma - Other rare types

[0019] Among these, undifferentiated carcinoma (PCC) is a representative subtype characterized by weak intercellular binding and a scattered tissue structure, resulting in a high metastatic potential and poor prognosis. Undifferentiated carcinoma is classified into SRC and PCC-NOS based on the presence of mucus production or cell morphology; PCC-NOS, unlike SRC, is a subtype composed of cells lacking a distinct signet-ring shape and is classified as a subtype with a particularly poor prognosis. The present invention relates to a biomarker capable of distinguishing the above-mentioned PCC-NOS subtype from other subtypes at the molecular level, and a technology capable of objectively predicting a patient's prognosis using this biomarker.

[0020] In this specification, the term "diagnosis" below refers to confirming the existence or characteristics of a pathological condition. For the purposes of the present invention, the diagnosis confirms the occurrence or potential for occurrence of gastric cancer, particularly PCC-NOS subtype gastric cancer, thereby enabling early prediction of the occurrence of gastric cancer, particularly PCC-NOS subtype gastric cancer.

[0021] The genes described as biomarkers below in this specification are genes of human (Homo sapiens) origin, and information regarding these genes can be easily searched in public databases that are obvious to a person skilled in the art to which this invention belongs, such as the National Center for Biotechnology Information (NCBI).

[0022] According to one embodiment of the present invention, the present invention relates to a biomarker for the diagnosis or prognosis prediction of gastric cancer, particularly PCC-NOS subtype gastric cancer.

[0023] The above biomarker is selected from the group consisting of ABCA8 (ATP Binding Cassette Subfamily A Member 8), LUM (Lumican), BGN (Biglycan), PRELP (Proline- and Arginine-Rich End Leucine-Rich Repeat Protein), ASPN (Asporin), COL14A1 (Collagen Type XIV Alpha 1 Chain), COL15A1 (Collagen Type XV Alpha 1 Chain), COL18A1 (Collagen Type XVIII Alpha 1 Chain), ANXA1 (Annexin A1), ANXA2 (Annexin A2), ANXA4 (Annexin A4), ANXA5 (Annexin A5), ANXA6 (Annexin A6), ANXA11 (Annexin A11), ABI3BP (ABI Family Member 3 Binding Protein), MFAP4 (Microfibril-Associated Protein 4), and DPT (Dermatopontin). It may be one or more genes; or a protein encoded by them. One or more genes selected above; or a protein encoded by them may have an increased expression level compared to a normal control group in gastric cancer, particularly PCC-NOS subtype gastric cancer.

[0024] Alternatively, the biomarker may be one or more genes selected from the group consisting of FBLN5 (Fibulin-5) and CCL11 (CC Motif Chemokine Ligand 11); or a protein encoded by the same. The one or more genes selected above; or the protein encoded by the same may have a reduced expression level compared to a normal control group in gastric cancer, particularly PCC-NOS subtype gastric cancer.

[0025] According to another embodiment of the present invention, the present invention relates to a composition for diagnosing or predicting the prognosis of gastric cancer, particularly PCC-NOS subtype gastric cancer.

[0026] The diagnostic composition includes ABCA8 (ATP Binding Cassette Subfamily A Member 8), LUM (Lumican), BGN (Biglycan), PRELP (Proline- and Arginine-Rich End Leucine-Rich Repeat Protein), ASPN (Asporin), COL14A1 (Collagen Type XIV Alpha 1 Chain), COL15A1 (Collagen Type (Collagen Type (Microfibril-Associated Protein 4), DPT (Dermatopontin), FBLN5 It may include any one protein selected from the group consisting of (Fibulin-5), and CCL11 (CC Motif Chemokine Ligand 11); or a preparation capable of measuring the expression level of the gene encoding it.

[0027] Alternatively, the diagnostic composition may include ABCA8 (ATP Binding Cassette Subfamily A Member 8), LUM (Lumican), BGN (Biglycan), PRELP (Proline- and Arginine-Rich End Leucine-Rich Repeat Protein), ASPN (Asporin), COL14A1 (Collagen Type XIV Alpha 1 Chain), COL15A1 (Collagen Type XV Alpha 1 Chain), COL18A1 (Collagen Type (Microfibril-Associated Protein 4), DPT (Dermatopontin), FBLN5 It may comprise any one or more proteins selected from the group consisting of (Fibulin-5), and CCL11 (CC Motif Chemokine Ligand 11); or a preparation capable of measuring the expression level of a gene encoding the same.

[0028] In the present invention, the agent for measuring the expression level of the protein is not particularly limited, but may include, for example, one or more selected from the group consisting of antibodies, oligopeptides, ligands, PNA (peptide nucleic acid), and aptamers that specifically bind to the protein.

[0029] In the present invention, the term "antibody" refers to a substance that specifically binds to an antigen and causes an antigen-antibody reaction. For the purposes of the present invention, an antibody refers to an antibody that specifically binds to the biomarker protein. The antibodies of the present invention include polyclonal antibodies, monoclonal antibodies, and recombinant antibodies. The antibodies can be easily manufactured using techniques widely known in the art. For example, polyclonal antibodies can be produced by a method widely known in the art that includes the process of injecting an antigen of the biomarker protein into an animal and collecting blood from the animal to obtain serum containing antibodies. Such polyclonal antibodies can be produced from any animal, such as goats, rabbits, sheep, monkeys, horses, pigs, cattle, dogs, etc. Additionally, monoclonal antibodies can be produced using a hybridoma method or phage antibody library technology widely known in the art. Antibodies produced by the above methods can be separated and purified using methods such as gel electrophoresis, dialysis, salt precipitation, ion exchange chromatography, and affinity chromatography. In addition, the antibody of the present invention comprises not only a complete form having two full-length light chains and two full-length heavy chains, but also functional fragments of the antibody molecule. A functional fragment of the antibody molecule means a fragment that possesses at least an antigen-binding function, and includes Fab, F(ab'), F(ab')2, and Fv.

[0030] In the present invention, the "PNA (Peptide Nucleic Acid)" comprises an artificially synthesized polymer similar to DNA or RNA. While DNA has a phosphate-ribose sugar backbone, PNA has a repeating N-(2-aminoethyl)-glycine backbone connected by peptide bonds, which significantly increases its binding affinity and stability to DNA or RNA, and is therefore used in molecular biology, diagnostic analysis, and antisense therapy.

[0031] In the present invention, the "aptamer" is an oligonucleotide or peptide molecule, and the aptamer can be manufactured in various ways that are obvious to those skilled in the art to which the present invention belongs.

[0032] In the present invention, a preparation for measuring the expression level of a gene encoding the protein may include one or more selected from the group consisting of a primer, a probe, and an antisense nucleotide that specifically binds to the gene encoding the protein.

[0033] In the present invention, the "primer" is a fragment that recognizes a target gene sequence and includes a forward and reverse primer pair, but preferably is a primer pair that provides analysis results having specificity and sensitivity. High specificity can be conferred when the nucleic acid sequence of the primer is a sequence that is inconsistent with the non-target sequence present in the sample, so that it amplifies only the target gene sequence containing the complementary primer binding site and does not induce non-specific amplification.

[0034] In the present invention, the term "probe" refers to a substance capable of specifically binding to a target substance to be detected within a sample, and means a substance capable of specifically confirming the presence of the target substance within the sample through said binding. The type of probe is not limited to substances commonly used in the art, but preferably may be PNA (peptide nucleic acid), LNA (locked nucleic acid), peptide, polypeptide, protein, RNA, or DNA, and most preferably PNA. More specifically, the probe may be a biomaterial derived from an organism or similar, or manufactured in vitro, and may be, for example, enzymes, proteins, antibodies, microorganisms, animal and plant cells and organs, nerve cells, DNA, and RNA; DNA may include cDNA, genomic DNA, and oligonucleotides; RNA may include genomic RNA, mRNA, and oligonucleotides; and examples of proteins may include antibodies, antigens, enzymes, peptides, etc.

[0035] In the present invention, "LNA (Locked nucleic acids)" refers to nucleic acid analogs containing a 2'-O, 4'-C methylene bridge. LNA nucleosides contain common nucleic acid bases of DNA and RNA and can form base pairs according to the Watson-Crick base pairing rule. However, due to the 'locking' of the molecule caused by the methylene bridge, LNAs are unable to form an ideal shape in Watson-Crick bonding. When LNAs are included in DNA or RNA oligonucleotides, LNAs can pair more quickly with complementary nucleotide chains, thereby increasing the stability of the double helix.

[0036] In the present invention, the term "antisense" refers to an oligomer having a backbone between nucleotide base sequences and subunits, wherein the antisense oligomer hybridizes with a target sequence within RNA by Watson-Crick base pairing, thereby allowing the formation of a mRNA and RNA:oligomer heterodimer within the target sequence. The oligomer may have exact sequence complementarity or approximate complementarity with respect to the target sequence.

[0037] Since the information of the biomarker protein according to the present invention or the gene encoding it is known, a person skilled in the art can easily design a primer, probe, or antisense nucleotide that specifically binds to the gene encoding the protein based on this.

[0038] In the diagnostic composition of the present invention, the expression level of the diagnostic protein or gene for gastric cancer, particularly PCC-NOS subtype gastric cancer, may be measured in decellularized tissue, and specifically, may be measured in the decellularized extracellular matrix.

[0039] In the diagnostic composition of the present invention, selected from the group consisting of ABCA8 (ATP Binding Cassette Subfamily A Member 8), LUM (Lumican), BGN (Biglycan), PRELP (Proline- and Arginine-Rich End Leucine-Rich Repeat Protein), ASPN (Asporin), COL14A1 (Collagen Type XIV Alpha 1 Chain), COL15A1 (Collagen Type XV Alpha 1 Chain), COL18A1 (Collagen Type XVIII Alpha 1 Chain), ANXA1 (Annexin A1), ANXA2 (Annexin A2), ANXA4 (Annexin A4), ANXA5 (Annexin A5), ANXA6 (Annexin A6), ANXA11 (Annexin A11), ABI3BP (ABI Family Member 3 Binding Protein), MFAP4 (Microfibril-Associated Protein 4), and DPT (Dermatopontin). If the expression level of any single protein or the gene encoding it is increased compared to a normal control group, it may be diagnosed as having a high probability of developing gastric cancer, particularly PCC-NOS subtype gastric cancer. Alternatively, it may be judged as a cancer with a poor prognosis.

[0040] In the diagnostic composition of the present invention, if the expression level of one or more proteins selected from the group consisting of FBLN5 (Fibulin-5) and CCL11 (CC Motif Chemokine Ligand 11); or the gene encoding them is reduced compared to a normal control group, it can be diagnosed that there is a high probability of developing gastric cancer, particularly PCC-NOS subtype gastric cancer.

[0041] The composition for diagnosing gastric cancer, particularly PCC-NOS subtype gastric cancer according to the present invention comprises ABCA8 (ATP Binding Cassette Subfamily A Member 8), LUM (Lumican), BGN (Biglycan), PRELP (Proline- and Arginine-Rich End Leucine-Rich Repeat Protein), ASPN (Asporin), COL14A1 (Collagen Type XIV Alpha 1 Chain), COL15A1 (Collagen Type XV Alpha 1 Chain), COL18A1 (Collagen Type XVIII Alpha 1 Chain), ANXA1 (Annexin A1), ANXA2 (Annexin A2), ANXA4 (Annexin A4), ANXA5 (Annexin A5), ANXA6 (Annexin A6), ANXA11 (Annexin A11), ABI3BP (ABI Family Member 3 Binding Protein), MFAP4 (Microfibril-Associated Protein 4), and DPT It may additionally include one or more proteins selected from the group consisting of (Dermatopontin), FBLN5 (Fibulin-5), and CCL11 (CC Motif Chemokine Ligand 11); or a preparation capable of measuring the expression level of the gene encoding therefrom. In this case, the diagnostic accuracy of gastric cancer, particularly PCC-NOS subtype gastric cancer, may be improved.

[0042] According to another embodiment of the present invention, the present invention relates to a gastric cancer diagnostic kit comprising a composition for diagnosing gastric cancer, particularly PCC-NOS subtype gastric cancer.

[0043] The kit of the present invention comprises the composition for diagnosing gastric cancer, particularly PCC-NOS subtype gastric cancer, described above. The limitations of each part constituting the composition for diagnosing gastric cancer, particularly PCC-NOS subtype gastric cancer, are redundant with those described in the composition for diagnosing gastric cancer, particularly PCC-NOS subtype gastric cancer, and are therefore omitted below to avoid excessive complexity in this specification.

[0044] In the present invention, the kit may be an RT-PCR kit, a DNA chip kit, an ELISA kit, a protein chip kit, a rapid kit, or an MRM (Multiple reaction monitoring) kit, but is not limited thereto.

[0045] The diagnostic kit of the present invention may further include one or more other component compositions, solutions, or devices suitable for the analysis method. For example, the diagnostic kit of the present invention may further include essential elements necessary to perform a reverse transcription polymerase chain reaction. The reverse transcription polymerase chain reaction kit includes a primer pair specific to a gene encoding a marker protein. The primer is a nucleotide having a sequence specific to the nucleic acid sequence of the said gene and may have a length of about 7 bp to 50 bp, more preferably about 10 bp to 30 bp. It may also include a primer specific to the nucleic acid sequence of a control gene. Furthermore, the reverse transcription polymerase chain reaction kit may include a test tube or other suitable container, a reaction buffer (with varying pH and magnesium concentration), deoxyribonucleotides (dNTPs), enzymes such as Taq-polymerase and reverse transcriptase, DNase, RNase inhibitor DEPC-water, sterile water, etc. In addition, the diagnostic kit of the present invention may include essential elements necessary for performing DNA chip. The DNA chip kit may include a substrate to which cDNA or oligonucleotides corresponding to a gene or a fragment thereof are attached, and reagents, preparations, enzymes, etc., for producing fluorescently labeled probes. Additionally, the substrate may include cDNA or oligonucleotides corresponding to a control gene or a fragment thereof. In addition, the diagnostic kit of the present invention may include essential elements necessary for performing ELISA. The ELISA kit includes an antibody specific to the protein. The antibody is an antibody having high specificity and affinity for the marker protein and little cross-reactivity with other proteins, and is a monoclonal antibody, a polyclonal antibody, or a recombinant antibody. In addition, the ELISA kit may include an antibody specific to the control protein.Other ELISA kits may include reagents capable of detecting conjugated antibodies, such as labeled secondary antibodies, chromophores, enzymes (e.g., conjugated with antibodies) and their substrates or other substances capable of binding to antibodies.

[0046] According to another embodiment of the present invention, the present invention relates to a method for providing information for diagnosing gastric cancer, particularly PCC-NOS subtype gastric cancer, comprising the step of measuring the expression level of one or more proteins selected from the biomarkers of the present invention described above or a gene encoding such proteins in a biological sample isolated from a target individual.

[0047] In the present invention, the term "target individual" refers to an individual in which the presence of gastric cancer is uncertain, or in which, even if gastric cancer has been diagnosed, the PCC-NOS subtype of the gastric cancer is unclear.

[0048] In the present invention, the "biological sample" refers to any substance, biological fluid, tissue, or cell obtained from or derived from an individual, and preferably, it is gastric tissue, as this can increase the accuracy in diagnosing gastric cancer.

[0049] The present invention may include a step of measuring the expression level of the biomarker proteins listed above or the genes encoding them in biological samples separated as described above. The step of measuring the expression level of the selected proteins or genes, or the preparation capable of measuring the expression level of the selected proteins or genes, overlaps with what is described in the composition for diagnosing gastric cancer, particularly PCC-NOS subtype gastric cancer, and is therefore omitted below to avoid excessive complexity in this specification.

[0050] According to another embodiment of the present invention, there is a method for treating PCC-NOS subtype gastric cancer, comprising the step of administering an appropriate therapeutic agent after diagnosing PCC-NOS subtype gastric cancer in particular by the method of the present invention described above.

[0051] The above therapeutic agent may be radiation therapy, surgical treatment, or administration of an anticancer drug, and the above anticancer drug may be a therapeutic agent used in conventional clinical practice or a candidate substance for the treatment of PCC-NOS subtype gastric cancer newly derived by the screening method of the present invention below.

[0052] According to another embodiment of the present invention, the present invention relates to a method for screening candidate substances for the treatment of gastric cancer, particularly PCC-NOS subtype gastric cancer.

[0053] This specifically involves the step of treating a biological sample isolated from a target individual with a candidate substance for gastric cancer treatment; and in biological samples treated with the above candidate substance ABCA8 (ATP Binding Cassette Subfamily A Member 8), LUM (Lumican), BGN (Biglycan), PRELP (Proline- and Arginine-Rich End Leucine-Rich Repeat Protein), ASPN (Asporin), COL14A1 (Collagen Type XIV Alpha 1 Chain), COL15A1 (Collagen Type XV Alpha 1 Chain), COL18A1 (Collagen Type XVIII Alpha 1 Chain), ANXA1 (Annexin A1), ANXA2 (Annexin A2), ANXA4 (Annexin A4), ANXA5 (Annexin A5), ANXA6 (Annexin A6), ANXA11 (Annexin A11), ABI3BP (ABI Family Member 3 Binding Protein), MFAP4 (Microfibril-Associated Protein 4), DPT (Dermatopontin), FBLN5 Any one protein selected from the group consisting of (Fibulin-5), and CCL11 (CC Motif Chemokine Ligand 11);Alternatively, it may include a step of measuring the expression level of the gene encoding it, and ABCA8 (ATP Binding Cassette Subfamily A Member 8), LUM (Lumican), BGN (Biglycan), PRELP (Proline- and Arginine-Rich End Leucine-Rich Repeat Protein), ASPN (Asporin), COL14A1 (Collagen Type XIV Alpha 1 Chain), COL15A1 (Collagen Type XV Alpha 1 Chain), COL18A1 (Collagen Type XVIII Alpha 1 Chain), ANXA1 (Annexin A1), ANXA2 (Annexin A2), ANXA4 (Annexin A4), ANXA5 (Annexin A5), ANXA6 (Annexin A6), ANXA11 (Annexin A11), ABI3BP (ABI Family Member 3 Binding Protein), MFAP4 (Microfibril-Associated Protein 4), DPT (Dermatopontin), One or more proteins selected from the group consisting of FBLN5 (Fibulin-5) and CCL11 (CC Motif Chemokine Ligand 11);Alternatively, it may include an additional step of measuring the expression level of the gene encoding it. In this case, ABCA8 (ATP Binding Cassette Subfamily A Member 8), LUM (Lumican), BGN (Biglycan), PRELP (Proline- and Arginine-Rich End Leucine-Rich Repeat Protein), ASPN (Asporin), COL14A1 (Collagen Type XIV Alpha 1 Chain), COL15A1 (Collagen Type XV Alpha 1 Chain), COL18A1 (Collagen Type XVIII Alpha 1 Chain), ANXA1 (Annexin A1), ANXA2 (Annexin A2), ANXA4 (Annexin A4), ANXA5 (Annexin A5), ANXA6 (Annexin A6), ANXA11 (Annexin A11), ABI3BP (ABI Family Member 3 Binding Protein), MFAP4 (Microfibril-Associated Protein 4), or DPT (Dermatopontin) are underexpressed compared to normal, or If FBLN5 (Fibulin-5) or CCL11 (CC Motif Chemokine Ligand 11) is overexpressed compared to normal levels, it may be determined that the above candidate substance has a therapeutic effect on PCC-NOS type gastric cancer.

[0054] The description regarding the preparation for measuring the expression level and the method for measuring the expression level in the screening method of the present invention overlaps with the description in the method for providing information for diagnosis of the present invention, and therefore, to prevent excessive complexity of the specification, such description is omitted below.

[0055] Since the present invention is highly effective in diagnosing the most difficult-to-treat and poor-prognostic type of gastric cancer with high accuracy, it is expected to be widely used in the medical and healthcare fields.

[0056] Figure 1 shows the results of the analysis of the characteristics of patient-derived decellularized ECM. Figure 1a illustrates the overall experimental flow of the study, showing the process of determining the proteomic profile of patient-derived tissue ECM (pdECM) using TMT mass spectrometry. Figure 1b presents clinical data of the patient and sample, including the histological type, stage, anatomical location, and depth of invasion of the tumor. Figure 1c shows representative images of the original tissue and pdECM, with a scale bar of 100 μm. Figure 1d shows the DNA quantification results of the original tissue and pdECM. Figure 1e displays the average intensity of proteins detected in identically mixed reference samples, and the number of proteins belonging to each category is specified in parentheses. Figure 1f shows the results of analyzing the term Gene Ontology: Cellular Component for the top 100 proteins with the highest intensity, where the bar graph represents the number of proteins and the dots represent statistical significance, showing that ECM-related proteins are strongly concentrated.

[0057] Figure 2 shows the results of comparing stromal-centered proteomic profiles between normal and tumor ECMs. Figure 2a presents the stromal composition of each pdECM as a hierarchical cluster heatmap and bar graph, demonstrating a clear distinction between normal and tumor tissues. Figure 2b shows the results of sorting proteins according to relative percentage composition (RPC), presenting the number of proteins accounting for 90% of the total intensity for each group. Figure 2c shows the 20 most abundant stromal proteins in each group, with the bar graph representing the average RPC and the dots representing the values ​​for each sample. Figure 2d shows the results of principal component analysis (PCA) based on the stromal protein intensity of all samples, suggesting a clear distinction between normal and tumor ECMs.

[0058] Figure 3 shows the results of the analysis of differentially expressed stromal proteins (DEPs) between NAT and tumor ECM. Figure 3a is a volcanic plot representing DEPs between NAT and tumor ECM, with the dotted line indicating the criteria of |log2FC| > 0.5 and p < 0.05. Figure 3b shows representative immunohistochemical (IHC) images of SERPINH1 abundant in tumors and HAPLN1 abundant in NAT. Figure 3c shows the results of the paired analysis of DEPs, where gray dots represent the log2FC of each patient and red dots represent the median, and the red box indicates that some samples showed an opposite trend.

[0059] Figure 4 shows the results of the specific ECM profile of the PCC-NOS subtype. Figure 4a is a PCA plot of the tumor ECM, showing that the PCC-NOS subtype samples are distinguishable from other subtypes. Figure 4b is a box plot showing the difference in proteoglycan expression between the PCC-NOS subtype and the non-PCC-NOS subtype. Figure 4c is a volcanic plot showing DEP between the two groups, with the criteria of log2FC > 0.5 and p < 0.05 indicated by a dotted line. Figure 4d is a heatmap of DEP between the two groups, defining PCC-NOS-rich stromal proteins (PEMs). Figure 4e shows the results of the paired analysis of selected PEMs, illustrating changes in log2 intensity values ​​by histological type and highlighting the specificity of the PCC-NOS subtype.

[0060] Figure 5 presents the results of identifying the origin of cells that primarily express PEM. Figure 5a presents the results of defining the origin of PEM by evaluating whether the expression of genes encoding PEM is cell type-specific based on single-cell RNA sequencing data, and considering the cell type with the highest average expression as the origin. Figure 5b is a Lollipop plot of the top 20 genes with the highest correlation to the PEM score, showing the Pearson correlation coefficient and the positive rate in fibroblasts. Figure 5c shows the expression of these genes at the single-cell level as a dot plot. Figure 5d presents the results of analyzing the expression of the same genes in isolated fibroblasts. Figure 5e presents the expression of PEM-associated genes and ABCA8 in pan-cancer single-cell analysis data, showing prominent expression in the adipogenic CAF region. Figure 5f is a scatter plot showing the correlation between the CAFadi score and the PEM score, using TCGA STAD data. Figures 5g and 5h show the results of a survival analysis performed by dividing patients into high-intensity and low-intensity groups based on PEM scores and 30 CAFadi markers, showing the poor prognosis of the high-intensity group.

[0061] Figure 6 presents the results of an analysis showing that ABCA8-positive fibroblasts are specific to the PCC-NOS subtype and are associated with a poor prognosis. Figure 6a shows a bar graph of the correlation between histological subtypes and ABCA8 expression grades in tumor microarray IHC analysis. Figure 6b presents representative images of ABCA8-positive fibroblasts in the PCC-NOS subtype and ABCA8-negative tissue in tubular adenocarcinoma, with a scale bar of 100 μm. Figure 6c shows the results of a survival analysis performed by dividing patients into Diffuse-ABCA8 positive, Diffuse-ABCA8 negative, and other subtype groups, revealing a lower survival rate in the positive group. Figure 6d shows a survival analysis based on ABCA8 gene expression, indicating that the high-expression group exhibits a poorer prognosis compared to the low-expression group.

[0062] 114 gastric cancer patients were classified into diffuse and non-diffuse groups according to Lauren's Classification, and a survival analysis was subsequently performed based on ABCA8 expression levels in the stromal region. As a result, although not statistically significant, the survival rate of the Diffuse-ABCA8 positive group decreased sharply at approximately 10 months of follow-up, a pattern that was distinctly different from the Diffuse-ABCA8 negative group and other subtype groups. Furthermore, patients with high expression of the ABCA8 gene exhibited an overall worse prognosis. This trend suggests that ABCA8 expression may serve as a potential biomarker for gastric cancer subtypes with poor prognosis.

[0063] The present invention will be explained in detail below through the following examples. However, the following examples are merely illustrative of the present invention, and the scope of the present invention is not limited by the following examples.

[0064] Examples

[0065] [Experimental Method]

[0066] 1. Experimental Design and Statistical Basis

[0067] A total of 45 tissue samples were collected from 23 individuals, including gastric cancer (GC) tissue, normal adjacent to tumor (NAT) tissue, normal (NN) tissue obtained from non-cancerous patients, and lymph node metastasis tissue. After extracellular matrix (ECM) enrichment through a detergent-based decellularization process, quantitative proteomic analysis of the decellularized gastric cancer tissue was performed using 11-fold tandem mass tag (TMT) spectroscopy.

[0068] TMT-based proteomic data were used for Hierarchical Clustering, Principal Component Analysis (PCA), Differentially Expressed Matrisome Proteins (DEP), and Pairing Analysis. Normalized intensity values ​​were scaled and clustered with the matrisome protein data based on Euclidean distance in Perseus software (www.maxquant.org / perseus). Only the normalized intensity values ​​of matrisome proteins were used for PCA. DEP was determined between two conditions (tumor vs. NAT, and Poorly Cohesive Carcinoma-Not Otherwise Specified (PCC-NOS) vs. Non-PCC-NOS) using Welch's t-test. A DEP satisfying Fold Change (FC) > √2 and p < 0.05 was selected, and PCC-NOS Enriched Matrisome Proteins (PEM) were defined as the DEP between PCC-NOS tumor samples and non-PCC-NOS tumor samples.

[0069] ECM proteomic results were integrated with single-cell RNA sequencing (scRNA-seq) analysis to identify cancer-associated fibroblasts (CAFs) that provide ECM heterogeneity. The cellular origin of DEPs was confirmed through the average expression levels of cell types, and cell type-specific genes were defined using the FindAllMarkers function of the Seurat package, with an adjusted p < 0.01 cutoff. PCC-NOS-specific CAFs were identified by calculating the PEM expression score of each single cell using the Single-cell Gene Set Variation Analysis (scGSVA) ​​R package with common scRNA-seq data. PCC-NOS-specific CAFs were validated through gene expression analysis of isolated primary CAFs and Tumor Microarray (TMA) immunohistochemistry (IHC) of 114 consecutive gastric cancer cases.

[0070] 2. Collection of Patient and Tissue Samples

[0071] Postoperative specimens were collected from a total of 23 patients who underwent gastrectomy at a tertiary hospital in South Korea between January and August 2021. Of these, 19 were patients with gastric cancer, and 4 were non-cancer patients who underwent sleeve gastrectomy due to morbid obesity. During surgery, tumor tissue and normal tissue (each 2×2 cm or smaller) were collected by the surgeon from the resected specimens. The collected tissues were transported to the laboratory on ice in Dulbecco's Modified Eagle's Medium (DMEM; GIBCO Life Technologies) in a solution containing 5% antibiotics (Penicillin / Streptomycin) and amphotericin B. Subsequently, the samples were washed with phosphate-buffered saline (PBS; containing 5% Penicillin / Streptomycin), cut into appropriate sizes, frozen, and stored at -195 °C. This process was carried out under the approval of the Institutional Review Board (IRB No. 2006-052-1131) of Seoul National University Hospital.

[0072] A total of 45 tissue samples were obtained, consisting of 19 tumor tissues, 17 normal tissues (gastric cancer patients), 5 lymph node tissues (gastric cancer patients), and 4 normal tissues (non-cancerous patients). This study was conducted in accordance with the Declaration of Helsinki, and the study protocol was carried out under the approval of the Institutional Review Board of Seoul National University Hospital (IRB No. 2304-144-1427).

[0073] 3. Tissue Decellularization Process

[0074] The collected tissues were decellularized using a detergent-based method. The solution used was a mixture of 1% Triton X-100 (Sigma-Aldrich) and 0.1% ammonium hydroxide (Sigma-Aldrich) in distilled water.

[0075] Tissue samples were cut into 3×3×3 mm pieces and processed for at least 3 hours, with the solution replaced every 30 minutes or whenever it became opaque. Once the tissue became colorless, patient-derived ECM (pdECM) samples were washed with Dulbecco's Phosphate-Buffered Saline (DPBS, Welgene) for 2 days, with the solution replaced every hour. Subsequently, DPBS residues were removed by washing with distilled water four times for 10 minutes each. This process was performed at room temperature using an orbital shaker at 70 rpm. Finally, the pdECM samples were lyophilized for 1 day and stored at -20 °C.

[0076] 4. pdECM Characterization

[0077] For Hematoxylin & Eosin (H&E) staining, the original tissue and decellularized tissue were fixed in 4% paraformaldehyde (PFA) for 1 day and embedded in Paraplast, and sections were prepared to a thickness of 10 μm. The sections were stained according to the standard H&E staining protocol. The DNA content of pdECM was quantified using a DNA extraction kit (Bioneer), and the concentration was measured using a DS-11 spectrophotometer (DeNovix).

[0078] 5. S-Trap Protein Digestion

[0079] Protein digestion was performed using S-Trap Mini (ProtiFi). Specifically, approximately 5 mg of decellularized gastric tissue was mixed with 1× SDS buffer (5% Sodium Dodecyl Sulfate, SDS; 50 mM Triethylammonium Bicarbonate, TEAB; pH 8.5) and sonicated using a sonicator (VCX 130; Sonics) according to the manufacturer's instructions. Each sonicated sample was centrifuged at 13,000 g for 10 minutes, and the supernatant was collected in a 1.5 mL tube. Dithiothreitol (DTT) was added to achieve a final concentration of 20 mM, and the mixture was heated at 95 °C for 10 minutes. Subsequently, the solution was cooled to room temperature, and the alkylation reaction was carried out by treating with 40 mM Iodoacetamide (IAA) under dark conditions for 30 minutes to achieve a final concentration. Subsequently, 12% water-soluble phosphoric acid (1:10 dilution; final concentration 1.2%) and 7 times the volume of binding buffer (90% water-soluble methanol with a final concentration of 100 mM TEAB, pH 7.1) were added to the Sodium Dodecyl Sulfate (SDS) solution. After light mixing, the protein solution was loaded onto an S-Trap filter and centrifuged at 3,000 g for 1 minute to recover the flow-through, which was then reloaded onto the filter. This step was repeated twice. The filter was then washed three times with 400 μL of binding buffer. Finally, 10 μg of Trypsin (Promega) and 125 μL of digestion buffer (50 mM TEAB) were added to the filter at a ratio of 1:25 (w / w), and digestion was performed at 37 °C for 16 hours. To recover digested peptides, 50 mM TEAB, 0.2% formic acid (FA) diluted in water, and 50% acetonitrile / 0.Three buffer solutions of 2% formic acid (80 μL each, repeated twice) were applied. The digested peptide solutions were combined, lyophilized, and then desalted according to the protocol of a Pierce Peptide Desalting Spin Column (Thermo Fisher Scientific).

[0080] 6. TMT 11-Flex Labeling

[0081] Multiplexing was applied to compare data between samples, and a total of five sets of TMT 11-Plex were used. These included 19 tumor tissues, 17 normal tissues, and 5 lymph node tissues obtained from gastric cancer (GC) patients, as well as 4 normal tissues from non-cancerous patients. A pooled common control was constructed as a reference for integrating the datasets, consisting of a mixture of equal amounts of total peptides from each sample used in the experiment. Each TMT set included two equal amounts of reference samples for quality verification, labeled with the 131N and 131C tags, respectively. The remaining nine channels contained individual tissue samples.

[0082] 100 μg of peptide from each sample was quantified using the Pierce Quantitative Fluorometric Peptide Assay kit (Thermo Fisher Scientific). Subsequently, the desalted and dried peptides were redissolved in 100 mM Triethylammonium Bicarbonate (TEAB, 100 μL) and labeled with TMT 11-Plex reagent (Thermo Fisher Scientific) according to the manufacturer's instructions. 0.8 mg of TMT reagent (41 μL) was added to each sample and reacted at room temperature for 1 hour. The reaction was terminated by adding 5% Hydroxylamine (8 μL) and treating at room temperature for 15 minutes.

[0083] Labeled samples (25-100 μg) were mixed and dried, and desalted using Pierce Peptide Desalting Spin Columns (Thermo Fisher Scientific). The eluent was then dried and stored at -80 °C.

[0084] 7. High pH reversed-phase fraction

[0085] TMT-labeled peptides were fractionated using a Shimadzu HPLC system. The system consisted of a binary pump, an autosampler, a degasser, a variable wave detector, and a fraction collector.

[0086] Fractionation was performed using a Waters XBridge BEH C18 column (4.6 × 150 mm, 2.5 μm), with mobile phase A being 5 mM Ammonium Formate in 100% water and mobile phase B being 5 mM Ammonium Formate in 95% Acetonitrile.

[0087] Sample separation was performed under the following linear gradient conditions.

[0088] 5% B, 15 mins; 5% → 15% B, 5 mins; 15% → 40% B, 30 mins; 40% B, 5 mins; 40% → 95% B, 4 mins; 95% B, 4 mins; 95% → 5% B, 1 min; and 5% B, an additional 9 mins.

[0089] Fractions were collected for 21 to 61 minutes to obtain a total of 40 fractions, each with a volume of approximately 1 mL. A variable wavelength detector was monitored at 214 nm.

[0090] The 40 collected fractions were paired and mixed (e.g., 1+21, 2+22, etc.), and finally 20 fractions were produced. Each fraction was dissolved in 200 μL of water / formic acid (99.9:0.1, v / v) and used for subsequent liquid chromatography-tandem mass spectrometry (LC-MS / MS) analysis.

[0091] 8. Nano LC-Electrospray Ionization-MS / MS Analysis

[0092] The UltiMate 3000 RSLC nano System (Thermo Fisher Scientific) and Orbitrap Eclipse Tribrid Mass Spectrometer (Thermo Fisher Scientific) were used for proteomic analysis. Fractionated peptides were injected and separated into an EASY-Spray PepMap RSLC C18 Column (2 μm, 100 Å, 75 μm × 50 cm, Thermo Fisher Scientific) and operated at 45 °C.

[0093] Mobile phase A was water / formic acid (99.9:0.1, v / v), and mobile phase B was acetonitrile / formic acid (99.9:0.1, v / v). A 5%–95% B gradient was applied for 140 minutes, and the flow rate was 250 nL / min. The electrospray ionization voltage was 1800–1900 V, and the ion transfer tube temperature was 275 °C.

[0094] Data was collected in Data-dependent Top-speed Mode and included as many MS2 scans as possible within 3 seconds per cycle. MS1 scans were performed using an Orbitrap analyzer, resolution 120K, mass range 400-2000 m / z, Automatic Gain Control (AGC) standard mode, maximum injection time (auto), charge states 2-6, and dynamic exclusion 30 seconds, while MS2 scans were performed using Higher-energy C-trap Dissociation (HCD) collision mode, Orbitrap resolution 30K, fixed collision energy 37% (for isotransitive labeled peptides), injection time auto, separation window 0.7, AGC standard mode, first mass 110, and Turbo TMT mode.

[0095] 9. Data Processing

[0096] For proteomics analysis, raw files were converted to MS (.ms1) and MS2 (.ms2) files using RawConverter (Scripps Research Institute), and protein discovery and database generation were performed using IP2 (Integrated Platform for Mass Spectrometry Data Analysis, Bruker). ProLuCID, DTASelect2, and Census were used for the analysis. The database was generated based on the UniProt Human Proteome (20,645 entries, updated January 1, 2020), and the IP2 parameters were as specified in Table 2.

[0097] - Precursor tolerance: 10 ppm- Fragment tolerance: 200 ppm- Enzyme: Trypsin- Allowable miscleavage: ≤ 2- Static Modification: Cysteine ​​(+57.0215 Da), Lysine and N-terminus (+229.1629 Da)- Differential Modification: Methionine (+15.9949 Da)- Minimum peptide number: 2

[0098] Spectral files generated from 20 fractions were compared with forward and reverse databases using the same parameters, and the False Discovery Rate (FDR) at the spectral level was set to less than 0.01 for peptide validation. TMT reporter ion analysis was performed using Census software, and the reporter ion mass tolerance was set to 20 ppm.10. Protein Abundance Normalization

[0099] Due to differences in sample processing and experimental environments, systematic and sample-specific biases were present in the quantification of protein abundance. To eliminate this, a correction was performed by calculating the median after log2 transformation of peptide abundance, subtracting the median from each column value to center the common median to zero, calculating the mean of the medians and adding it back to the zero-centered values, and then applying the y = 2^x transformation.

[0100] In addition, for intensity normalization between samples, the relative intensity value of each protein was calculated by dividing the protein intensity of the corresponding sample by the internal reference (R2) value and multiplying it by the average normalized intensity value in the R2 column. The final value was then subjected to the y = 2^x transformation. This normalized abundance value was used for subsequent proteomic analysis.

[0101] 11. Immunohistochemical staining

[0102] Immunostaining was performed on the markers in Table 3.

[0103] Serpin Family H Member 1 (SERPINH1; HSP47, Abcam)Hyaluronan and Proteoglycan Link Protein 1 (HAPLN1, Bio-Techne)ATP Binding Cassette Subfamily A Member 8 (ABCA8, Sigma-Aldrich)

[0104] ABCA8 immunostaining was performed on a tissue microarray (rabbit polyclonal antibody, HPA044914, Sigma-Aldrich, 1:200) consisting of 114 consecutive gastric cancer cases. Staining intensity of the peritumoral stroma was classified as 0 (no staining), 1 (weak staining), 2 (medium staining), and 3 (strong staining), and ABCA8 expression levels were classified as ABCA8- (negative; intensity 0 or 1) or ABCA+ (positive, intensity 2 or 3).12. Bioinformatics Data Analysis

[0105] Single-cell RNA sequencing (scRNA-seq) analysis of gastric cancer (GC) tissues utilized data published in a prior study (Kumar, V. et al, Cancer Discov. 12, 670-691). Briefly, single-cell gastric cancer anaphoresis from the Singapore cohort was collected, and a barcoded sequencing library was constructed according to the manufacturer's instructions. The analysis included only patients classified as intestinal or diffuse according to the Lauren classification. Specific parameters, reagent kits, and pipelines for sequencing followed those previously described.

[0106] Five major cell types were annotated using previously described markers: epithelial cells, stromal cells, T cells, macrophages, and B cells. Within stromal cells, fibroblasts and endothelial cells were distinguished by Plasmalemma Vesicle-Associated Protein (PLVAP) expression. Subsequently, the cellular origin of Differentially Expressed Matrisome Proteins (DEPs) was identified using the average expression levels of each cell type. Cell type-specific genes were defined using the FindAllMarkers function of the Seurat package (https: / satijalab.org / seurat / ), and a adjusted p < 0.01 threshold was applied to determine whether specific gene expression was cell type-specific. The average expression levels by cell type were calculated using the AverageExpression function of the Seurat package, and the cell type with the highest average expression level was considered the cellular origin of the corresponding gene.

[0107] Subsequently, PEM (PCC-NOS Enriched Matrisome Proteins) scores were calculated using only cell populations annotated as fibroblast types. The gene expression patterns of fibroblasts were normalized and clustered according to the following procedure:

[0108] (1) Perform linear dimensional reduction using the RunPCA function by considering all substrate genes as features,

[0109] (2) Use the FindNeighbors function to search for nearest neighbors with the parameter dims = 1:20,

[0110] (3) Clustering using the FindClusters function with the resolution = 0.5 parameter,

[0111] (4) Visualize the fibroblast in dimensional space using the RunUMAP function with the parameter dims = 1:20.

[0112] Single-cell Gene Set Variation Analysis (scGSVA) ​​was performed using the PEM gene list from the R package (https: / github.com / guokai8 / scGSVA), where the PEM score for each cell was calculated, and the Pearson correlation coefficient for each gene was calculated along with the PEM score and the Transcripts Per Million (TPM) value.

[0113] The expression of PEM markers in the Cancer-Associated Fibroblast Atlas was confirmed using publicly available single-cell RNA-seq data. Cell types were already annotated, and markers of adipogenic CAF (CAFadi; cluster 4) were sorted according to log2 fold change values. The top 30 CAFadi markers were selected and used to calculate CAFadi scores in The Cancer Genome Atlas-Stomach Adenocarcinoma (TCGA-STAD) bulk RNA-seq data.

[0114] The TCGA-STAD gene expression dataset and clinical dataset were collected from the TCGAbiolinks package (https: / bioconductor.org / packages / release / bioc / html / TCGAbiolinks.html). Raw counts downloaded from the Illumina platform were converted into normalized TPM values, and a total of 375 tumor samples were analyzed.

[0115] The expression patterns of specific gene sets in each TCGA sample were evaluated using Single-sample Gene Set Enrichment Analysis (ssGSEA). The ssGSEA scores for PEM and CAFadi were calculated using the ssGSEAprojection module of the GenePattern web-based tool (https: / www.genepattern.org / modules / docs / ssGSEAProjection / 4#gsc.tab=0).

[0116] Overall survival (OS) was visualized using the KM plotter database, and the association between each marker gene set and the prognosis of the ABCA8 gene was evaluated.

[0117] 13. Primary Fibroblast Isolation and Transcriptome Analysis

[0118] Gastric mucosal tissue samples obtained as described in patient and tissue sample collection were washed with phosphate-buffered saline (PBS) and 5% penicillin / streptomycin, and then cut into square pieces 2–3 mm in size. The cut tissues were washed three additional times with PBS. The washed tissue pieces were placed at regular intervals in a 6-well plate, covered with a 22 mm cover glass (#HSU-0101060), and cultured in Dulbecco's Modified Eagle's Medium (DMEM; GIBCO Life Technologies) supplemented with 10% fetal bovine serum (FBS) and 1% penicillin / streptomycin.

[0119] Fibroblasts were isolated from gastric cancer (GC) tissue via the outgrowth method. The isolated fibroblasts were validated by immunochemical staining for alpha smooth muscle actin (α-SMA) and fibroblast activation protein (FAP), which are representative markers of fibroblasts. All experiments were performed using fibroblasts maintained at nine or fewer passages.

[0120] RNA was extracted using an RNA prep kit (PureLink RNA Mini Kit; Thermo Fisher Scientific), and reverse transcription was subsequently performed using the PrimeScript 1st Strand Complementary DNA (cDNA) Synthesis Kit (Takara Bio Inc.).

[0121] RNA sequencing (RNA-seq) was performed on an Illumina HiSeq2500 Sequencer (Theragen Etex Bio Institute). Raw Illumina sequence data was converted into FASTQ files after demultiplexing.

[0122] Messenger RNA (mRNA) sequencing reads were mapped to the Genome Reference Consortium’s Homo sapiens genome assembly GRCh38.p13 and aligned using Spliced ​​Transcripts Alignment to a Reference (STAR, version 2.7.10; https: / / github.com / alexdobin / STAR). The aligned reads were assembled with known genes and quantified using RSEM (RNA-Seq by Expectation-Maximization, version 1.3.3; https: / / github.com / deweylab / RSEM). Quantification was expressed in read counts and standardized values ​​(e.g., FPKM, TPM).

[0123] 14. Survival Analysis

[0124] Patients were followed from the date of surgery until death or the final follow-up. Overall survival (OS) was calculated using the Kaplan-Meier curve and the Log-rank test, with p < 0.05 defined as statistical significance.

[0125] [Experimental Results]

[0126] 1. Quantitative proteomic analysis of the extracellular matrix of patient-derived tissues

[0127] To comprehensively analyze the ECM proteomic profiles of normal gastric tissues and histologically annotated gastric cancer (GC) tissues, the ECM was enriched using decellularization technology, and Tandem Mass Tag (TMT)-based quantitative proteomics was performed. In addition to acellular components, transcriptomic analysis of cellular components within the stroma was incorporated. To this end, a multimodal analysis approach combining single-cell RNA sequencing (scRNA-seq) and ECM proteomics was adopted (Fig. 1a). A total of 45 tissue samples were collected, consisting of gastric cancer tissues (GC tissues), normal adjacent to tumor (NAT) tissues, normal (NN) tissues from non-cancerous patients, and lymph node metastasis tissues, obtained from a total of 23 individuals. It was confirmed that the collected samples showed no specific bias in tumor stage, anatomical region, depth of tumor invasion, or histology, suggesting that the samples were composed of a diverse patient population (Fig. 1b). The efficiency of the decellularization process for the removal of cellular components was confirmed through H&E staining (Fig. 1c) and DNA quantification (Fig. 1d). As a result, cell nuclei were effectively removed in pdECM, and the content of DNA corresponding to the nucleus was significantly reduced.

[0128] A total of 4,838 proteins were detected through TMT-based quantitative proteomic analysis. Based on the Matrisome database, 376 of these proteins were identified as matrisome proteins, including 37 collagens (COLs), 102 ECM glycoproteins (GLY), 18 proteoglycans (PRO), 99 ECM regulators (REG), 55 affiliated proteins (AFF), and 63 secreted factors (SEC). Analysis of a representative pdECM sample prepared by mixing all samples in equal proportions revealed that the 376 matrisome proteins accounted for 62.6% of the total intensity, while 4,464 non-matrisome proteins accounted for the remaining small proportion (Fig. 1e). The top 100 high-intensity proteins were annotated according to the Cellular Component (CC) category of Gene Ontology (GO) and analyzed via DAVID (https: / david.ncifcrf.gov / ). The category with the highest statistical significance was identified as being related to the extracellular region and ECM, suggesting a strong enrichment of ECM-related proteins (Fig. 1f).

[0129] These results demonstrate that the ECM enrichment process enables large-scale quantitative profiling of tissue ECM components.

[0130] 2. The pdECM profile shows differences between normal and tumor tissues and tumor heterogeneity.

[0131] Hierarchical clustering based on the intensity of matrisome proteins formed clusters that clearly separated normal tissues (Normal, NN, and NAT) from tumor tissues, revealing differential profiles between normal and tumor ECMs. A comparison of the relative percentage composition (RPC) of major matrisome categories between normal and tumor tissues showed that collagen (COL) content was significantly increased in normal tissues, and proteoglycan (PRO) content was also relatively high. In contrast, tumor tissues exhibited higher levels of ECM glycoproteins (GLY) (Fig. 2a).

[0132] The average RPCs of all proteins detected under each condition were sorted in descending order (Fig. 2b). The number of proteins accounting for 90% of the total RPCs was 231 in NN tissue and 292 in NAT tissue, but was significantly higher in tumor tissue at 605 compared to normal tissue. Furthermore, a comparison of the 20 most abundant stromal proteins by condition revealed that the top six proteins were identical across NN, NAT, and tumor tissues, consisting of the collagen COL1 family, COL6 family, and fibrillin-1 (FBN1). However, there were differences in their detailed composition (Fig. 2c). For example, in tumor tissue, fibrinogen (FGA / FGB / FGG), mucin-2 (MUC2), and TGF-β-induced protein (Transforming Growth Factor Beta Induced, TGFBI) were located higher, whereas in normal tissue, heparan sulfate proteoglycan-2 (HSPG2), decorin (DCN), and laminin subunit gamma-1 (LAMC1) were found to be relatively high.

[0133] The ECM profiles obtained through dimensional reduction techniques more clearly demonstrated the differences between normal and tumor tissues. In normal tissues, NN and NAT samples exhibited similar ECM profiles and were clearly distinguishable from tumor tissues in most cases. However, tumor tissues were not only distinctly different from normal tissues but also showed significant variation within the tumors (Fig. 2d). An examination of the distribution of the top 20 stromal proteins revealed that tumor tissues showed a much wider distribution than normal tissues, highlighting the heterogeneity of the tumor ECM (Fig. 2c).

[0134] These results demonstrate that tumor tissue possesses a unique ECM profile distinct from normal tissue, while simultaneously exhibiting distinct heterogeneity within the tumor ECM.

[0135] 3. Differentially expressed stromal proteins between tumor-adjacent normal (NAT) tissue and tumor tissue

[0136] Differentially expressed stromal proteins (DEPs) between NAT tissue and tumor tissue were identified. For each protein, the FC value (average intensity of tumor / average intensity of NAT) and p-value were calculated, and proteins satisfying the statistical criteria (|log2FC| > 0.5, p < 0.05) were defined as DEPs (Fig. 3a).

[0137] As a result, 20 tumor-enriched DEPs and 77 NAT-enriched DEPs were identified. The tumor-enriched DEPs included four ECM glycoproteins (GLYs), namely Serpin Family H Member 1 (SERPINH1), Annexin Family (ANXA3 / 4 / 5 / 13), S100A Family (S100A6 / 8 / 9), Matrix Metalloproteinase 14 (MMP14), and other matrix-associated proteins. Increased expression of the COL6 family, COL14A1, COL15A1, and COL28A1 was observed in collagen (COL) proteins, and in GLY proteins, Multimerin-1 / 2 (MMRN1 / 2), Sushi Repeat-Containing Protein X-Linked (SRPX), Laminin Subunit Alpha-5 (LAMA5), Laminin Subunit Beta-2 (LAMB2), and Microfibril-Associated Protein 5 (MFAP5) were significantly upregulated in tumor tissue. On the other hand, specific proteoglycans were more abundant in NAT tissue, including Heparan Sulfate Proteoglycan 2 (HSPG2), Decorin (DCN), Hyaluronan and Proteoglycan Link Protein 1 (HAPLN1), Proline- and Arginine-Rich End Leucine-Rich Repeat Protein (PRELP), and Lumican (LUM).

[0138] At the actual tissue level (in vivo), SERPINH1 and HAPLN1 were verified as tumor-rich and NAT-rich proteins, respectively, via immunohistochemistry (IHC) (Fig. 3b). Pairing analysis for DEP showed the same trend as FC-based analysis. The median DEP FC of all patients was consistent with the overall trend of DEP (Fig. 3c). However, due to tumor heterogeneity, some samples did not follow this trend (e.g., red box in Fig. 3c). These findings highlight the importance of further research on tumor heterogeneity within the ECM, even though identifying DEP between normal and tumor tissues provides useful insights.

[0139] 4. The undifferentiated gastric cancer-other unspecified (PCC-NOS) subtype has a unique ECM profile distinct from other histological types.

[0140] To investigate the heterogeneity of the tumor ECM, dimensional reduction of the ECM profile was performed using only tumor samples, and histological information was overlaid on it. Interestingly, PCC-NOS (Poorly Cohesive Carcinoma-Not Otherwise Specified) subtype samples formed clusters in close proximity to other types (Fig. 4a).

[0141] First, based on the matrisome category, the PCC-NOS subtype had a significantly higher proteoglycan (PRO) content compared to the non-PCC-NOS subtype (Fig. 4b). Specifically, the average PRO content of the PCC-NOS subtype was 6.8%, while that of the non-PCC-NOS subtype was 3.3%. Interestingly, the value of the PCC-NOS subtype (6.8%) was higher than the average PRO content of the NAT tissue (4.7%), which indicates that the abundant accumulation of PRO is a characteristic of the PCC-NOS subtype.

[0142] DEPs between PCC-NOS and non-PCC-NOS subtypes were identified based on FC values ​​and p-values ​​(Fig. 4c). As a result, a total of 27 substrate proteins were abundantly expressed in the PCC-NOS subtype, whereas 16 substrate proteins were abundantly expressed in the non-PCC-NOS subtype (Fig. 4d).

[0143] In particular, the PCC-NOS subtype was remarkably rich in proteoglycans, specifically small leucine-rich repeat proteins (SLRPs). In addition, various Annexin family proteins (ANXA1 / 2 / 4 / 5 / 6 / 11), ABI3BP (ABI Family Member 3 Binding Protein), Microfibril-Associated Protein 4 (MFAP4), Dermatopontin (DPT), and COL14 / 15 / 18A1 type collagen were significantly upregulated in the PCC-NOS subtype.

[0144] Furthermore, a pairing analysis including PEMs (PCC-NOS Enriched Matrisome Proteins) revealed unique cancer-related ECM characteristics specific to the PCC-NOS subtype that were not observed in other subtypes. In most PCC-NOS subtype tissues, PEM expression was elevated under tumor conditions relative to NAT in the same patient. Conversely, in other subtypes, PEM expression showed a noticeable decreasing trend (Fig. 4e).

[0145] These findings suggest that the characteristics of the ECM vary depending on the histological subtype, and that the PCC-NOS subtype, in particular, possesses an abundant set of unique ECM proteins.

[0146] 5. PCC-NOS-specific ECM proteins are primarily expressed in adipogenesis cancer-associated fibroblasts (CAFadi).

[0147] Fibroblasts are cells that primarily produce ECM proteins and can influence the unique ECM profile of the PCC-NOS subtype. To confirm this, we analyzed publicly available single-cell RNA sequencing (scRNA-seq) data to investigate the single-cell level expression of PEM (PCC-NOS Enriched Matrisome Proteins). The single cells were categorized into seven major cell types, including fibroblasts, epithelial cells, macrophages, endothelial cells, T cells, B cells, and mast cells. The analysis confirmed that more than 70% of PEM is primarily expressed in fibroblasts (Fig. 5a).

[0148] Based on the expression of genes encoding PEM, the degree of PEM expression in each fibroblast was calculated as a score using the Gene Set Variation Analysis (GSVA) ​​package. Cells with higher PEM scores were more closely associated with PEM-associated fibroblasts. The correlation between the expression of each gene and the PEM score was evaluated using Pearson's Correlation Coefficient (Fig. 5b). In this analysis, Lumican (LUM) was the gene showing the highest correlation with PEM-associated fibroblasts. Genes with such high correlation can be used as markers to distinguish PCC-NOS subtypes. On the other hand, genes with low expression rates in positive fibroblasts, such as CCL11 and FBLN5, can be usefully employed to distinguish PCC-NOS subtypes based on their expression levels.

[0149] To verify the fibroblast specificity of PEM-correlated genes, we compared the average expression levels and positive cell ratios across seven major cell types. As a result, it was confirmed that most PEM-correlated genes are expressed in fibroblasts (Fig. 5c). Subsequently, to further verify PCC-NOS specificity, we isolated primary fibroblasts from samples representative of each histological subtype (PCC-NOS / PCC-signet-ring cell phenotype / Tubular Adenocarcinoma) and performed bulk RNA sequencing. The analysis confirmed that PEM-correlated genes were expressed at much higher levels in PCC-NOS subtype fibroblasts than in other subtypes (Fig. 5d).

[0150] In addition, we explored the relationship between PEM-associated fibroblasts and well-known fibroblast subtypes within tumors. We used publicly available single-cell RNA-seq data as reference data, and most PEM-associated genes were prominently expressed in a specific cluster (cluster 4), which was a population classified as adipogenic cancer-associated fibroblasts (CAFadi) (Fig. 5e). When comparing the molecular similarities between PEM-associated fibroblasts and CAFadi using bulk RNA-seq data from Stomach Adenocarcinoma (STAD) from The Cancer Genome Atlas (TCGA), the top 30 CAFadi markers with the highest PEM scores and log2 FC values ​​showed a positive correlation (Fig. 5f). These two molecular panels (PEMs and CAFadi markers) were confirmed to have clinical significance. When both PEM and CAFadi marker expression were high, the prognosis of gastric cancer patients was found to be poor (Fig. 5g, and Fig. 5h).

[0151] Importantly, this correlation between PEM and CAFadi markers suggests that PEM-associated fibroblasts may represent a subset of CAFadi, demonstrating that higher expression is associated with a poor prognosis in gastric cancer patients. This highlights the clinical relevance of the findings of this study.

[0152] 6. ABCA8-positive fibroblasts are specific to the PCC-NOS subtype and are associated with a poor prognosis.

[0153] To selectively detect CAFadi (Adipogenic Cancer-Associated Fibroblasts) cells in tumor sections, we focused on cell-surface markers of CAFadi instead of matrisome markers. Among the top 30 CAFadi markers (based on the highest log2 fold change value), the ATP Binding Cassette Subfamily A Member 8 (ABCA8) protein was selected as a CAFadi marker because the expression of the ABCA8 gene is specific in fibroblasts and not in other cell types.

[0154] The level of ABCA8 protein expression in tumor tissues from 114 gastric cancer (GC) patients was evaluated through immunohistochemistry (IHC) analysis of tissue microarrays (TMA). Surprisingly, only the patient group of the PCC-NOS subtype showed high levels of ABCA8 expression predominantly, while ABCA8 protein expression was rarely observed in most patients of other histological subtypes (Fig. 6a). Furthermore, ABCA8 expression was observed mainly in the stromal region rather than in cancer cells, which further supports the fact that ABCA8-positive fibroblasts are specific to the PCC-NOS subtype (Fig. 6b).

[0155] 114 gastric cancer patients were classified into diffuse and non-diffuse groups according to Lauren's Classification, and a survival analysis was subsequently performed based on ABCA8 expression levels in the stromal region (Fig. 6c). As a result, although not statistically significant, the survival rate of the Diffuse-ABCA8 positive group decreased sharply at approximately 10 months of follow-up, a pattern that was distinctly different from the Diffuse-ABCA8 negative group and other subtype groups. Furthermore, patients with high expression of the ABCA8 gene showed an overall worse prognosis (Fig. 6d).

[0156] This trend suggests that ABCA8 expression may serve as a potential biomarker for gastric cancer subtypes with a poor prognosis.

[0157] Foregoing, specific parts of the present invention have been described in detail. It is evident to those skilled in the art that such specific descriptions are merely preferred embodiments and do not limit the scope of the invention. Accordingly, the actual scope of the invention is defined by the appended claims and their equivalents.

[0158] The present invention is expected to significantly contribute to personalized treatment and improved survival rates for gastric cancer patients by providing a novel molecular diagnostic and prognostic prediction means capable of early identification of PCC-NOS subtypes and prediction of prognosis.

Claims

1. ABCA8 (ATP Binding Cassette Subfamily A Member 8), LUM (Lumican), BGN (Biglycan), PRELP (Proline- and Arginine-Rich End Leucine-Rich Repeat Protein), ASPN (Asporin), COL14A1 (Collagen Type XIV Alpha 1 Chain), COL15A1 (Collagen Type XV Alpha 1 Chain), COL18A1 (Collagen Type (Microfibril-Associated Protein 4), DPT (Dermatopontin), FBLN5 (Fibulin-5), A composition for diagnosing gastric cancer comprising any one protein selected from the group consisting of and CCL11 (CC Motif Chemokine Ligand 11); or a preparation capable of measuring the expression level of a gene encoding the same.

2. In Paragraph 1, The composition includes ABCA8 (ATP Binding Cassette Subfamily A Member 8), LUM (Lumican), BGN (Biglycan), PRELP (Proline- and Arginine-Rich End Leucine-Rich Repeat Protein), ASPN (Asporin), COL14A1 (Collagen Type XIV Alpha 1 Chain), COL15A1 (Collagen Type (Collagen Type (Microfibril-Associated Protein 4), DPT (Dermatopontin), FBLN5 (Fibulin-5), A composition further comprising one or more proteins selected from the group consisting of and CCL11 (CC Motif Chemokine Ligand 11); or a preparation capable of measuring the expression level of a gene encoding the same.

3. In Paragraph 1, A composition in which the above gastric cancer is a PCC-NOS (Poorly cohesive carcinoma, Not Otherwise Specified) type gastric cancer.

4. In Paragraph 1, A composition in which the expression level of the above protein or gene is measured in decellularized tissue.

5. In Paragraph 4, A composition in which the expression level of the above protein or gene is measured in a decellularized extracellular matrix.

6. A gastric cancer diagnostic kit comprising a diagnostic composition according to any one of claims 1 to 5.

7. In Paragraph 6, Kit, which states that the above gastric cancer is a PCC-NOS (Poorly cohesive carcinoma, Not Otherwise Specified) type gastric cancer.

8. ABCA8 (ATP Binding Cassette Subfamily A Member 8), LUM (Lumican), BGN (Biglycan), PRELP (Proline- and Arginine-Rich End Leucine-Rich Repeat Protein), ASPN (Asporin), COL14A1 (Collagen Type XIV Alpha 1 Chain), COL15A1 (Collagen Type XV Alpha 1 Chain), COL18A1 (Collagen Type XVIII Alpha 1 Chain), ANXA1 (Annexin A1), ANXA2 (Annexin A2), ANXA4 (Annexin A4), ANXA5 (Annexin A5), ANXA6 (Annexin A6), ANXA11 (Annexin A11), ABI3BP (ABI Family Member 3 Binding Protein), MFAP4 (Microfibril-Associated Protein 4), DPT (Dermatopontin), A method for providing information for diagnosing gastric cancer, comprising the step of measuring the expression level of any one protein selected from the group consisting of FBLN5 (Fibulin-5) and CCL11 (CC Motif Chemokine Ligand 11); or a gene encoding the same.

9. In Paragraph 8, The method includes ABCA8 (ATP Binding Cassette Subfamily A Member 8), LUM (Lumican), BGN (Biglycan), PRELP (Proline- and Arginine-Rich End Leucine-Rich Repeat Protein), ASPN (Asporin), COL14A1 (Collagen Type XIV Alpha 1 Chain), COL15A1 (Collagen Type (Collagen Type (Microfibril-Associated Protein 4), DPT (Dermatopontin), FBLN5 (Fibulin-5), and A method further comprising the step of measuring the expression level of one or more proteins selected from the group consisting of CCL11 (CC Motif Chemokine Ligand 11); or a gene encoding the same.

10. In Paragraph 8, A method in which the above gastric cancer is a PCC-NOS (Poorly cohesive carcinoma, Not Otherwise Specified) type gastric cancer.

11. In Paragraph 10, The above method is, ABCA8 (ATP Binding Cassette Subfamily A Member 8), LUM (Lumican), BGN (Biglycan), PRELP (Proline- and Arginine-Rich End Leucine-Rich Repeat Protein), ASPN (Asporin), COL14A1 (Collagen Type XIV Alpha 1 Chain), COL15A1 (Collagen Type XVIII Alpha 1 Chain), ANXA1 (Annexin A1), ANXA2 (Annexin A2), ANXA4 (Annexin A4), ANXA5 (Annexin A5), ANXA6 (Annexin A6), ANXA11 (Annexin A11), ABI3BP (ABI Family Member 3 Binding Protein), MFAP4 (Microfibril-Associated Protein) 4), or DPT (Dermatopontin) is overexpressed compared to normal, or A method comprising an additional step of determining that the gastric cancer is of the PCC-NOS type if FBLN5 (Fibulin-5) or CCL11 (CC Motif Chemokine Ligand 11) is underexpressed compared to normal.

12. In Paragraph 8, A method in which the expression level of the above protein or gene is measured in decellularized tissue.

13. In Paragraph 12, A method in which the expression level of the above protein or gene is measured in a decellularized extracellular matrix.

14. A step of treating a biological sample isolated from a target individual with a candidate substance for gastric cancer treatment; and ABCA8 (ATP Binding Cassette Subfamily A Member 8), LUM (Lumican), BGN (Biglycan), PRELP (Proline- and Arginine-Rich End Leucine-Rich Repeat Protein), ASPN (Asporin), COL14A1 (Collagen Type XIV Alpha 1 Chain), COL15A1 (Collagen Type XV Alpha 1 Chain), COL18A1 (Collagen Type XVIII Alpha 1 Chain), ANXA1 (Annexin A1), ANXA2 (Annexin A2), ANXA4 (Annexin A4), ANXA5 (Annexin A5), ANXA6 (Annexin A6), ANXA11 (Annexin A11), ABI3BP (ABI Family Member 3 Binding Protein), MFAP4 (Microfibril-Associated Protein 4), DPT (Dermatopontin), FBLN5 in biological samples treated with the above candidate substance A screening method for drugs for treating gastric cancer, comprising the step of measuring the expression level of any one protein selected from the group consisting of (Fibulin-5) and CCL11 (CC Motif Chemokine Ligand 11); or a gene encoding the same.

15. In Paragraph 14, The method includes ABCA8 (ATP Binding Cassette Subfamily A Member 8), LUM (Lumican), BGN (Biglycan), PRELP (Proline- and Arginine-Rich End Leucine-Rich Repeat Protein), ASPN (Asporin), COL14A1 (Collagen Type XIV Alpha 1 Chain), COL15A1 (Collagen Type (Collagen Type (Microfibril-Associated Protein 4), DPT (Dermatopontin), FBLN5 (Fibulin-5), and A method further comprising the step of measuring the expression level of one or more proteins selected from the group consisting of CCL11 (CC Motif Chemokine Ligand 11); or a gene encoding the same.

16. In Paragraph 14, A method in which the above gastric cancer is a PCC-NOS (Poorly cohesive carcinoma, Not Otherwise Specified) type gastric cancer.

17. In Paragraph 16, The above method is, ABCA8 (ATP Binding Cassette Subfamily A Member 8), LUM (Lumican), BGN (Biglycan), PRELP (Proline- and Arginine-Rich End Leucine-Rich Repeat Protein), ASPN (Asporin), COL14A1 (Collagen Type XIV Alpha 1 Chain), COL15A1 (Collagen Type XVIII Alpha 1 Chain), ANXA1 (Annexin A1), ANXA2 (Annexin A2), ANXA4 (Annexin A4), ANXA5 (Annexin A5), ANXA6 (Annexin A6), ANXA11 (Annexin A11), ABI3BP (ABI Family Member 3 Binding Protein), MFAP4 (Microfibril-Associated Protein) 4), or DPT (Dermatopontin) is expressed lower than normal, or A method comprising an additional step of determining that the candidate substance has a therapeutic effect on PCC-NOS type gastric cancer if FBLN5 (Fibulin-5) or CCL11 (CC Motif Chemokine Ligand 11) is overexpressed compared to normal.

18. In Paragraph 14, A method in which the above biological sample is a decellularized tissue.

19. In Paragraph 18, A method in which the above biological sample is a decellularized extracellular matrix.

20. ABCA8 (ATP Binding Cassette Subfamily A Member 8), LUM (Lumican), BGN (Biglycan), PRELP (Proline- and Arginine-Rich End Leucine-Rich Repeat Protein), ASPN (Asporin), COL14A1 (Collagen Type XIV Alpha 1 Chain), COL15A1 (Collagen Type XV Alpha 1 Chain), COL18A1 (Collagen Type XVIII Alpha 1 Chain), ANXA1 (Annexin A1), ANXA2 (Annexin A2), ANXA4 (Annexin A4), ANXA5 (Annexin A5), ANXA6 (Annexin A6), ANXA11 (Annexin A11), ABI3BP (ABI Family Member 3 Binding Protein), MFAP4 (Microfibril-Associated Protein 4), DPT (Dermatopontin), A method for providing information to predict the prognosis of gastric cancer, comprising the step of measuring the expression level of any one protein selected from the group consisting of FBLN5 (Fibulin-5) and CCL11 (CC Motif Chemokine Ligand 11); or a gene encoding the same.

21. In Paragraph 20, The method includes ABCA8 (ATP Binding Cassette Subfamily A Member 8), LUM (Lumican), BGN (Biglycan), PRELP (Proline- and Arginine-Rich End Leucine-Rich Repeat Protein), ASPN (Asporin), COL14A1 (Collagen Type XIV Alpha 1 Chain), COL15A1 (Collagen Type (Collagen Type (Microfibril-Associated Protein 4), DPT (Dermatopontin), FBLN5 (Fibulin-5), and A method further comprising the step of measuring the expression level of one or more proteins selected from the group consisting of CCL11 (CC Motif Chemokine Ligand 11); or a gene encoding the same.

22. In Paragraph 20, The above method is, ABCA8 (ATP Binding Cassette Subfamily A Member 8), LUM (Lumican), BGN (Biglycan), PRELP (Proline- and Arginine-Rich End Leucine-Rich Repeat Protein), ASPN (Asporin), COL14A1 (Collagen Type XIV Alpha 1 Chain), COL15A1 (Collagen Type XVIII Alpha 1 Chain), ANXA1 (Annexin A1), ANXA2 (Annexin A2), ANXA4 (Annexin A4), ANXA5 (Annexin A5), ANXA6 (Annexin A6), ANXA11 (Annexin A11), ABI3BP (ABI Family Member 3 Binding Protein), MFAP4 (Microfibril-Associated Protein) 4), or DPT (Dermatopontin) is overexpressed compared to normal, or A method for predicting a poor prognosis for gastric cancer when FBLN5 (Fibulin-5) or CCL11 (CC Motif Chemokine Ligand 11) is underexpressed compared to normal.

23. In Paragraph 20, A method in which the expression level of the above protein or gene is measured in decellularized tissue.

24. In Paragraph 23, A method in which the expression level of the above protein or gene is measured in a decellularized extracellular matrix.

25. ABCA8 (ATP Binding Cassette Subfamily A Member 8), LUM (Lumican), BGN (Biglycan), PRELP (Proline- and Arginine-Rich End Leucine-Rich Repeat Protein), ASPN (Asporin), COL14A1 (Collagen Type XIV Alpha 1 Chain), COL15A1 (Collagen Type XV Alpha 1 Chain), COL18A1 (Collagen Type XVIII Alpha 1 Chain), ANXA1 (Annexin A1), ANXA2 (Annexin A2), ANXA4 (Annexin A4), ANXA5 (Annexin A5), ANXA6 (Annexin A6), ANXA11 (Annexin A11), ABI3BP (ABI Family Member 3 Binding Protein), MFAP4 (Microfibril-Associated Protein 4), DPT (Dermatopontin), A step of measuring the expression level of any one protein selected from the group consisting of FBLN5 (Fibulin-5) and CCL11 (CC Motif Chemokine Ligand 11); or the gene encoding it; and, A method for treating gastric cancer comprising the step of administering an anticancer agent when the above individual is determined to have gastric cancer of the PCC-NOS (Poorly cohesive carcinoma, Not Otherwise Specified) type.

Images