Prediction method for response to immune checkpoint inhibitor

Abstract

This invention provides a method for predicting the response of cancer patients to immune checkpoint inhibitors, comprising: (1) The step of determining the presence or absence of one or more values ​​or SNPs selected from the group consisting of (i) to (v), which are as follows: (i) The amount of microorganisms present in the feces or intestinal contents of the aforementioned subjects. (ii) Scoring of genomic pathways in the feces or intestinal contents of the aforementioned subjects. (iii) The amount of metabolites present in the blood, serum or plasma of the aforementioned subjects. (iv) The expression levels of genes in the blood, serum, or plasma of the aforementioned subjects, and (v) Single nucleotide polymorphisms (SNPs) selected from one or more genes in the group consisting of IL6R and NLRC5 from the aforementioned subjects; and (2) Using the value obtained in step (1) above as an indicator, the step of predicting the response of the aforementioned object to immune checkpoint inhibitors.

Description

[Technical Field]

[0001] This invention relates to a method for predicting responses to immune checkpoint inhibitors. [Previous Technology]

[0002] Various immune checkpoint molecules exist in the microenvironment of cancer cells or cancer cells, which hinder the immune response to cancer. Immune checkpoint inhibitors can relieve immunosuppression mechanisms and are a novel treatment for activating the immune response to cancer. In terms of immune checkpoint inhibitors, ipilimumab, an anti-CTLA-4 (cytotoxic T-lymphocyte-associated antigen-4) antibody, and nivolumab and pembrolizumab, an anti-PD-1 (planned cell death protein-1) antibody, have been approved in Japan and abroad for use in cancer treatment.

[0003] Immune checkpoint inhibitors have revolutionized cancer treatment, bringing hope to cancer patients who have previously struggled to achieve a complete cure. However, immune checkpoint inhibitors do not have a consistent effect on all cancer patients. Data on progression-free survival shows that in clinical trials using anti-PD-1 antibodies, one type of immune checkpoint inhibitor, a "non-responsive group" was observed where disease progression occurred within three months. On the other hand, the group that responded to anti-PD-1 antibodies for more than one year showed almost no further disease progression, achieving a near-cure state. This suggests that in clinical efficacy, there exist three distinct subgroups: the "non-responsive group," the "responsive group," and the "intermediate group."

[0004] It is problematic not only from a medical perspective, but also from a medical economic perspective, to administer immune checkpoint inhibitors, which are expected to be the standard treatment for almost all cancers, to patients in the ineffective group.

[0005] In order to solve the above problems, methods for predicting the response to immune checkpoint inhibitors have been explored, and methods using intestinal bacteria as biomarkers have been proposed (Patent Documents 1 to 5).

[0005] [Previous Technical Documents]

[0005] [Patent Documents]

[0006] [Patent Document 1] International Publication No. 2018 / 064165

[0006] [Patent Document 2] International Publication No. 2019 / 191309

[0006] [Patent Document 3] International Publication No. 2019 / 129753

[0006] [Patent Document 4] International Publication No. 2018 / 226690

[0006] [Patent Document 5] Chinese Patent Application Publication No. 1096800085 Specification

[0007] The subject of this invention is to predict the response of cancer patients to immune checkpoint inhibitors.

[0008] The inventors, through their dedicated investigation in order to solve the aforementioned problems, surprisingly discovered that by determining the presence of a specific bacterial species previously unknown or the score of a specific genome pathway in the feces or intestinal contents of the subject, the presence of a specific metabolite in the subject's blood, serum, or plasma, the expression level of a specific gene in the subject's blood, serum, or plasma, or the presence or absence of a specific single nucleotide polymorphism (SNP) in the subject, the response of the subject to immune checkpoint inhibitors can be predicted, thus completing the present invention.

[0009] That is, the present invention may include the following states.

[0010] [1] A method for predicting the response of cancer patients to immune checkpoint inhibitors, comprising:

[0010] (1) The step of determining the presence or absence of one or more values ​​or SNPs selected from the group consisting of (i) to (v), wherein (i) to (v) are as follows:

[0010] (i) The amount of microorganisms present in the feces or intestinal contents of the aforementioned objects.

[0010] (ii) Genomic pathway scores in the feces or intestinal contents of the aforementioned subjects.

[0010] (iii) The amount of metabolites present in the blood, serum or plasma of the aforementioned object,

[0010] (iv) The expression levels of genes in the blood, serum, or plasma of the aforementioned subjects, and

[0010] (v) Single nucleotide polymorphisms (SNPs) of one or more genes selected from the group consisting of IL6R and NLRC5 in the aforementioned objects;

[0010] (2) Using the value obtained in step (1) above or the presence or absence of SNP as an indicator, the step of predicting the response of the immune checkpoint inhibitor to the aforementioned object;

[0010] The aforementioned microorganisms are selected from one or more species of microorganisms in the group consisting of the genera *Geobacillus*, *Gordonibacter*, *Odoribacter*, *Veillonella*, *Corynebacterium*, *Porphyromonas*, and *Arthrobacter*.

[0010] The aforementioned genomic pathways are selected from one or more of the following genomic pathways: bacterial invasion of epithelial cells, fatty acid degradation, flagellar assembly, fatty acid biosynthesis, PPAR signaling pathway, peptidoglycan biosynthesis, nucleotide metabolism, purine metabolism, phenylalanine metabolism, and fatty acid metabolism.

[0010] The aforementioned metabolites are one or more metabolites selected from the group consisting of lactic acid, pyruvic acid, glucose, 2-sideoxybutyric acid, glyceric acid, caprylic acid, citrulline, 2-hydroxybutyric acid and 2-piperidinecarboxylic acid.

[0010] The aforementioned genes are genes related to the MAPK pathway, the Type I IFN receptor complex pathway, and the TCR signaling pathway, as well as one or more genes selected from the group consisting of BCL11B, ERCC3, ERCC6, FCRL1, FCRL3, MS4A1 and TCF7.

[0010] The aforementioned SNP is one or more SNPs selected from the group consisting of the following rs numbers: rs2228145.1, rs7190199, and rs7185320.

[0010] [2] The method described in item 1, wherein the aforementioned step (1) is a step of determining (i) the amount of microorganisms present in the feces or intestinal contents of the aforementioned object.

[0010] [3] The method as described in item 1 or 2, wherein the presence of the aforementioned *Geobacillus* genus, the aforementioned *Gordonibacter* genus, and / or the aforementioned *Veillonella* genus microorganisms is higher than a set threshold, and / or

[0010] When the presence of the aforementioned Odoribacter genus microorganisms is lower than the set threshold, it is predicted that the aforementioned subjects will respond well to immune checkpoint inhibitors.

[0010] [4] The method as described in item 1 or 2, wherein the presence of the aforementioned Veillonella microorganisms is higher than a set threshold, and / or

[0010] When the presence of the aforementioned Odoribacter genus microorganisms is lower than the set threshold, it is predicted that the aforementioned subjects will respond well to immune checkpoint inhibitors.

[0010] [5] The method as described in item 1 or 2, wherein the presence of the aforementioned *Geobacillus* genus, the aforementioned *Gordonibacter* genus, and / or the aforementioned *Veillonella* genus microorganisms is lower than a set threshold, and / or

[0010] When the presence of the aforementioned Odoribacter genus microorganisms is higher than the set threshold, it is predicted that the aforementioned subjects will have a poor response to immune checkpoint inhibitors.

[0010] [6] The method as described in item 1 or 2, wherein the presence of the aforementioned Veillonella microorganisms is lower than a set threshold, and / or

[0010] When the presence of the aforementioned Odoribacter genus microorganisms is higher than the set threshold, it is predicted that the aforementioned subjects will have a poor response to immune checkpoint inhibitors.

[0010] [7] The method described in item 1 or 2, wherein when the presence of the aforementioned Arthrobacter microorganisms is higher than a set threshold, it is predicted that harmful skin phenomena will be induced in the aforementioned subjects due to immune checkpoint inhibitors.

[0010] [8] The method described in any of items 1 to 6, wherein the quantity of the aforementioned microorganisms is determined by metagenomic analysis.

[0010] [9] The method described in item 1, wherein the aforementioned step (1) is (ii) the step of determining the genomic pathway score in the feces or intestinal contents of the aforementioned subject.

[0010]

[10] The method as described in item 1 or 9, wherein the scores for the aforementioned bacterial invasion of epithelial cells, the aforementioned fatty acid degradation, the aforementioned flagellar assembly, the aforementioned PPAR signaling pathway and / or the aforementioned phenylalanine metabolism are lower than a set threshold, and / or

[0010] If the scores for the aforementioned fatty acid biosynthesis, the aforementioned nucleotide metabolism, and / or the aforementioned peptidoglycan biosynthesis are higher than the set threshold, it is predicted that the aforementioned subjects will respond well to immune checkpoint inhibitors.

[0010]

[11] The method as described in item 1 or 9, wherein the scores for bacterial invasion of epithelial cells and / or phenylalanine metabolism are lower than a set threshold, and / or

[0010] If the scores for the aforementioned nucleotide metabolism and / or the aforementioned peptidoglycan biosynthesis are higher than the set threshold, it is predicted that the aforementioned subjects will respond well to immune checkpoint inhibitors.

[0010]

[12] The method as described in item 1 or 9, wherein the scores for the aforementioned bacterial invasion of epithelial cells, the aforementioned fatty acid degradation, the aforementioned flagellar assembly, the aforementioned PPAR signaling pathway and / or the aforementioned phenylalanine metabolism are higher than a set threshold, and / or

[0010] If the scores for the aforementioned fatty acid biosynthesis, nucleotide metabolism, and / or peptidoglycan biosynthesis are lower than the set thresholds, it is predicted that the aforementioned subjects will have a poor response to immune checkpoint inhibitors.

[0010]

[13] The method as described in item 1 or 9, wherein the scores for bacterial invasion of epithelial cells and / or phenylalanine metabolism are higher than a set threshold, and / or

[0010] If the scores for the aforementioned nucleotide metabolism and / or the aforementioned peptidoglycan biosynthesis are lower than the set threshold, it is predicted that the aforementioned subjects will have a poor response to immune checkpoint inhibitors.

[0010]

[14] The method described in item 1 or 9, wherein if the score of the aforementioned fatty acid metabolism is higher than a set threshold, it is predicted that harmful skin phenomena will be induced in the aforementioned subjects due to immune checkpoint inhibitors.

[0010]

[15] The method described in any of items 1 and 9 to 14, wherein the scores of the aforementioned genomic pathways are determined by metagenomic analysis.

[0010]

[16] The method as described in item 1, wherein the aforementioned step (1) is (iii) the step of determining the amount of metabolites present in the blood, serum or plasma of the aforementioned object.

[0010]

[17] The method as described in item 1 or 16, wherein the amounts of the aforementioned lactic acid, the aforementioned pyruvate, the aforementioned glucose, the aforementioned glyceric acid, the aforementioned caprylic acid, the aforementioned 2-hydroxybutyric acid and / or the aforementioned 2-sideoxybutyric acid are lower than a set threshold, and / or

[0010] When the amount of the aforementioned 2-piperidinecarboxylic acid and / or the aforementioned citrulline is higher than the set threshold, it is predicted that the aforementioned subjects will respond well to immune checkpoint inhibitors.

[0010]

[18] The method as described in item 1 or 16, wherein the amounts of the aforementioned lactic acid, the aforementioned pyruvate, the aforementioned glucose, the aforementioned glyceric acid, the aforementioned caprylic acid, the aforementioned 2-hydroxybutyric acid and / or the aforementioned 2-sideoxybutyric acid are higher than a set threshold, and / or

[0010] If the amount of the aforementioned 2-piperidinecarboxylic acid and / or the aforementioned citrulline is lower than the set threshold, it is predicted that the aforementioned subjects will have a poor response to immune checkpoint inhibitors.

[0010]

[19] The method described in any of items 1 and 16 to 18, wherein the amount of the aforementioned metabolite is determined by metabolome analysis.

[0010]

[20] The method described in item 1, wherein the aforementioned step (1) is (iv) the step of determining the expression level of genes in the blood, serum or plasma of the aforementioned subject.

[0010]

[21] The method as described in item 1 or 20, wherein the expression level of the gene related to one or more of the pathways selected from the group consisting of the genes related to the aforementioned MAPK pathway, the aforementioned Type I IFN receptor complex pathway, and the aforementioned TCR signaling pathway is higher than a set threshold, or

[0010] If the expression levels of one or more genes selected from the group consisting of the aforementioned BCL11B, the aforementioned ERCC3, the aforementioned ERCC6, the aforementioned FCRL1, the aforementioned FCRL3, the aforementioned MS4A1 and the aforementioned TCF7 are higher than a set threshold, it is predicted that the aforementioned subjects will respond well to immune checkpoint inhibitors.

[0010]

[22] The method as described in item 1 or 20, wherein the expression level of the gene related to one or more of the pathways selected from the group consisting of the genes related to the aforementioned MAPK pathway, the aforementioned Type I IFN receptor complex pathway, and the aforementioned TCR signaling pathway is lower than a set threshold, or

[0010] When the expression levels of one or more genes selected from the group consisting of BCL11B, ERCC3, ERCC6, FCRL1, FCRL3, MS4A1 and TCF7 are lower than a set threshold, it is predicted that the subject will have a poor response to immune checkpoint inhibitors.

[0010]

[23] The method described in any of items 20 to 22, wherein the expression level of the aforementioned gene is determined by transcriptome analysis.

[0010]

[24] The method described in item 1, wherein the aforementioned step (1) is (v) a step of determining the presence or absence of a single nucleotide polymorphism (SNP) in one or more genes selected from the group consisting of the aforementioned IL6R and the aforementioned NLRC5 in the aforementioned object.

[0010]

[25] The method described in item 1 or 24, wherein in the presence of the aforementioned SNP, it is predicted that harmful skin phenomena will be induced in the aforementioned subjects due to immune checkpoint inhibitors.

[0010]

[26] The method described in any of items 1 to 25, wherein the aforementioned cancer is gastric cancer.

[0010]

[27] The method described in any of items 1 to 26, wherein the aforementioned immune checkpoint inhibitor is an inhibitor of an immune checkpoint molecule selected from the group consisting of CTLA-4, PD-1, PD-L1, PD-L2, LAG-3, TIM3, BTLA, B7H3, B7H4, 2B4, CD160, A2aR, KIR, VISTA, IDO1, arginase I, TIGIT, and CD115.

[0010]

[28] The method described in any of items 1 to 26, wherein the aforementioned immune checkpoint inhibitor is an anti-PD-1 antibody.

[0011]

[29] A cancer treatment agent containing an immune checkpoint inhibitor, characterized in that it is administered to a cancer patient who is determined (predicted) to be responsive to an immune checkpoint inhibitor by any of the methods described in any of the preceding items 1 to 28.

[0011]

[30] A cancer therapeutic agent containing an immune checkpoint inhibitor, characterized in that:

[0011] (i) Subjects to cancer where the presence of microorganisms of the genera *Geobacillus*, *Gordonibacter*, and / or *Veillonella* is higher than a set threshold, and / or the presence of microorganisms of the genera *Odoribacter* is lower than a set threshold (preferably subjects to cancer where the presence of microorganisms of the genera *Veillonella* is higher than a set threshold, and / or the presence of microorganisms of the genera *Odoribacter* is lower than a set threshold).

[0011] (ii) Subjects to cancer cells whose scores for bacterial invasion of epithelial cells, fatty acid degradation, flagellar assembly, PPAR signaling pathway and / or phenylalanine metabolism are lower than a set threshold, and / or whose scores for fatty acid biosynthesis, nucleotide metabolism and / or peptidoglycan biosynthesis are higher than a set threshold (preferably subjects to cancer cells whose scores for bacterial invasion of epithelial cells are lower than a set threshold);

[0011] (iii) Subjected to cancer patients in whom the presence of lactic acid, pyruvate, glucose, glyceric acid, caprylic acid, 2-hydroxybutyric acid and / or 2-sideoxybutyric acid is lower than a set threshold, and / or the presence of 2-piperidinic acid and / or citrulline is higher than a set threshold; and / or

[0011] (iv) Subjected to: cancer patients whose expression levels of one or more genes from the group consisting of genes related to the MAPK pathway, the Type I IFN receptor complex pathway, and the TCR signaling pathway are higher than a set threshold, and / or whose expression levels of one or more genes from the group consisting of BCL11B, ERCC3, ERCC6, FCRL1, FCRL3, MS4A1, and TCF7 are higher than a set threshold.

[0011]

[31] A cancer therapeutic agent containing an immune checkpoint inhibitor, characterized in that it is administered to a cancer patient in whom the presence of microorganisms of the genera *Geobacillus*, *Gordonibacter*, and / or *Veillonella* is higher than a set threshold, and / or the presence of microorganisms of the genera *Odoribacter* is lower than a set threshold (preferably administered to a cancer patient in whom the presence of microorganisms of the genera *Veillonella* is higher than a set threshold, and / or the presence of microorganisms of the genera *Odoribacter* is lower than a set threshold).

[0011]

[32] A cancer therapeutic agent containing an immune checkpoint inhibitor, characterized in that the scores for bacterial invasion of epithelial cells, the aforementioned fatty acid degradation, the aforementioned flagellar assembly, the PPAR signaling pathway and / or phenylalanine metabolism are lower than a set threshold, and / or the scores for fatty acid biosynthesis, nucleotide metabolism and / or peptidoglycan biosynthesis are higher than a set threshold (preferably cancer subjects administered to epithelial cells with a score for bacterial invasion of epithelial cells lower than a set threshold).

[0011]

[33] A cancer therapeutic agent containing an immune checkpoint inhibitor, characterized in that it is administered to a cancer patient in which the presence of lactic acid, pyruvate, glucose, glyceric acid, caprylic acid, 2-hydroxybutyric acid and / or 2-sideoxybutyric acid is lower than a set threshold, and / or the presence of 2-piperidinecarboxylic acid and / or citrulline is higher than a set threshold.

[0012]

[34] A method for treating a cancer patient, comprising:

[0012] (1) The steps to determine the presence or absence of one or more values ​​or SNPs selected from the group consisting of the following (i) to (v):

[0012] (i) The amount of microorganisms present in the feces or intestinal contents of the aforementioned object;

[0012] (ii) Scoring of genomic pathways in the feces or intestinal contents of the aforementioned subjects;

[0012] (iii) The amount of metabolites present in the blood, serum or plasma of the aforementioned object;

[0012] (iv) The expression levels of genes in the blood, serum, or plasma of the aforementioned subjects; and

[0012] (v) Single nucleotide polymorphisms (SNPs) of one or more genes selected from the group consisting of IL6R and NLRC5 in the aforementioned objects;

[0012] (2) Using the value obtained in step (1) above or the presence or absence of SNP as an indicator, the step of predicting the response of the above object to immune checkpoint inhibitor;

[0012] (3) The step of administering the aforementioned immune checkpoint inhibitor to the aforementioned subject who is determined to have a good response to the immune checkpoint inhibitor and / or not induce harmful skin phenomena.

[0012] of which

[0012] The aforementioned microorganisms are one or more species selected from the group consisting of the genera *Geobacillus*, *Gordonibacter*, *Odoribacter*, *Veillonella*, *Corynebacterium*, *Porphyromonas*, and *Arthrobacter*.

[0012] The aforementioned genomic pathways are one or more genomic pathways selected from the group consisting of bacterial invasion of epithelial cells, fatty acid degradation, flagellar assembly, fatty acid biosynthesis, PPAR signaling pathway, peptidoglycan biosynthesis, nucleotide metabolism, purine metabolism, phenylalanine metabolism, and fatty acid metabolism.

[0012] The aforementioned metabolites are one or more metabolites selected from the group consisting of lactic acid, pyruvic acid, glucose, 2-sideoxybutyric acid, glyceric acid, caprylic acid, citrulline, 2-hydroxybutyric acid and 2-piperidinecarboxylic acid.

[0012] The aforementioned genes are selected from genes related to the MAPK pathway, the Type I IFN receptor complex pathway, and the TCR signaling pathway, as well as one or more genes from the group consisting of BCL11B, ERCC3, ERCC6, FCRL1, FCRL3, MS4A1, and TCF7.

[0012] The aforementioned SNP is one or more SNPs selected from the group consisting of the following rs numbers: rs2228145.1; rs7190199; and rs7185320.

[0013]

[35] A system for determining the responsiveness of a cancerous subject to an immune checkpoint inhibitor, comprising a memory unit, an input unit, a data processing unit, and an output unit, characterized in that:

[0013] The memory system remembers one or more of the following (1-i) to (1-v) cutoff values ​​used to determine the responsiveness of immune checkpoint inhibitors, as follows:

[0013] (1-i) The amount of microorganisms present in feces or intestinal contents;

[0013] (1-ii) Scoring of genomic pathways in feces or intestinal contents;

[0013] (1-iii) The amount of metabolites present in blood, serum or plasma;

[0013] (1-iv) The level of gene expression in blood, serum, or plasma; and

[0013] (1-v) Single nucleotide polymorphisms (SNPs) selected from one or more genes in the group consisting of IL6R and NLRC5;

[0013] The memory unit receives one or more values ​​or the presence or absence of SNPs selected from the group consisting of the following (2-i) to (2-v) from the input unit, and stores them in the memory unit. The following are (2-i) to (2-v):

[0013] (2-i) The amount of microorganisms present in the feces or intestinal contents of the aforementioned object;

[0013] (2-ii) Scoring of genomic pathways in the feces or intestinal contents of the aforementioned subjects;

[0013] (2-iii) The amount of metabolites present in the blood, serum or plasma of the aforementioned object;

[0013] (2-iv) The expression levels of genes in the blood, serum, or plasma of the aforementioned subjects; and

[0013] (2-v) Single nucleotide polymorphisms (SNPs) of one or more genes selected from the group consisting of IL6R and NLRC5 in the aforementioned objects;

[0013] The data processing unit compares the previously stored values ​​or the presence or absence of SNPs with the aforementioned cutoff value to determine the responsiveness of the aforementioned object to immune checkpoint inhibitors;

[0013] The output unit outputs the results of determining the responsiveness of the aforementioned object to immune checkpoint inhibitors for cancer, wherein...

[0013] The aforementioned microorganisms are one or more species selected from the group consisting of the genera *Geobacillus*, *Gordonibacter*, *Odoribacter*, *Veillonella*, *Corynebacterium*, *Porphyromonas*, and *Arthrobacter*.

[0013] The aforementioned genomic pathway is one or more genomic pathways selected from the group consisting of bacterial invasion of epithelial cells, fatty acid degradation, flagellar assembly, fatty acid biosynthesis, PPAR signaling pathway, peptidoglycan biosynthesis, nucleotide metabolism, purine metabolism, phenylalanine metabolism, and fatty acid metabolism.

[0013] The aforementioned metabolites are one or more metabolites selected from the group consisting of lactic acid, pyruvic acid, glucose, 2-sideoxybutyric acid, glyceric acid, caprylic acid, citrulline, 2-hydroxybutyric acid and 2-piperidinecarboxylic acid.

[0013] The aforementioned genes are selected from genes related to the MAPK pathway, the Type I IFN receptor complex pathway, and the TCR signaling pathway, as well as one or more genes from the group consisting of BCL11B, ERCC3, ERCC6, FCRL1, FCRL3, MS4A1, and TCF7.

[0013] The aforementioned SNP is one or more SNPs selected from the group consisting of the following rs numbers: rs2228145.1; rs7190199; and rs7185320.

[0013]

[36] The system as described in item 35 further includes an analytical measuring unit, wherein

[0013] The aforementioned analytical assay unit determines the quantity and / or genomic pathway score of microorganisms present in the feces or intestinal contents of the aforementioned subject.

[0013] The aforementioned analytical measuring system determines the amount of metabolites present in the blood, serum, or plasma of the aforementioned subject.

[0013] The aforementioned analytical assay system determines the expression level of genes in the blood, serum, or plasma of the aforementioned subject, and / or

[0013] The aforementioned analytical assay determines the presence or absence of single nucleotide polymorphisms (SNPs) in one or more genes selected from the group consisting of IL6R and NLRC5 in the aforementioned subjects.

[0013] Instead of the aforementioned input unit, or via the aforementioned input unit, input the quantity of the aforementioned microorganism, the score of the aforementioned genomic pathway, the quantity of the aforementioned metabolites, the expression level of the aforementioned genes, and / or the presence or absence of the aforementioned SNPs as determined by the aforementioned analysis and determination unit.

[0014] By employing this invention, the response to immune checkpoint inhibitors can be predicted in cancer patients. As a result, for patients who respond well to immune checkpoint inhibitors, appropriate use of immune checkpoint inhibitors can be considered, with the expectation of improved treatment efficacy. [Simplified Explanation of the Diagram]

[0015] Figure 1 shows the relationship between the diversity of fecal microbiota and the effectiveness of anti-PD-1 antibodies. (A) The first half (training cohort) represents the data, and (B) The second half (validation cohort) represents the data. Furthermore, Ace and Chao1 in the figure represent diversity indicators.

[0015] Figure 2 shows the results of the overall metagenomic analysis of feces from PD (progressive disease: tumor enlargement) and non-PD (non-tumor enlargement) groups, and a list of bacterial species (genus) showing significant differences.

[0015] Figure 3 shows the results of the overall metagenomic analysis of feces from PD and non-PD groups, identifying a list of KEGG pathways with significant differences.

[0015] Figure 4 shows the results of analysis of the bacterial species that were identified as significantly different in the overall metagenomic analysis of the feces of the first half of the patient group, and the results of analysis in the second half of the patient group (PD group and non-PD group).

[0015] Figure 5 shows the results of analysis of the bacterial species (Geobacillus and Gordonibacter) that were identified as significantly different in the overall metagenomic analysis of the feces of the first half of the patient group in the second half of the patient group (PD group and non-PD group).

[0015] Figure 6 shows the results of the analysis of the KEGG pathway, which showed significant differences in the overall metagenomics analysis of the feces of the first half of the patient group, and the analysis results of the second half of the patient group (PD group and non-PD group).

[0015] Figure 7 shows the results of the analysis of the KEGG pathway (fatty acid biosynthesis), which showed significant differences in the overall metagenomic analysis of feces in the first half of the patient group, and the results in the second half of the patient group (PD group and non-PD group). (A) Comparison of scores of fatty acid biosynthesis pathway. (B) Comparison of scores of fatty acid breakdown pathway. (C) Schematic diagram of fatty acid biosynthesis pathway.

[0015] Figure 8 shows the results of the analysis of the KEGG pathway (PPAR signaling pathway) in the feces of the first half of the patient group, which showed significant differences, and the results of the analysis in the second half of the patient group (PD group and non-PD group). (A) Comparison of PPAR signaling pathway scores in the first half of the patient group. (B) Comparison of PPAR signaling pathway scores in the second half of the patient group. (C) Schematic diagram of PPAR signaling pathway.

[0015] Figure 9 shows the overall metagenomic analysis of feces from the first and second halves of the patient population (PD group and non-PD group), narrowing down the scope to the most frequently occurring bacterial species (above 0.01%) and analyzing the results. (A) First half of the patient population, (B) Second half of the patient population.

[0015] Figure 10 shows the scores of Odoribacter (A) and Veillonella (B) in the stool of patients in the first and second halves of the patient population (PD group and non-PD group).

[0015] Figure 11 shows the profile of the entire bacterial species obtained by the overall metagenomics analysis of the feces of the first half of the patient group.

[0015] Figure 12 shows the profile of the entire bacterial species obtained by the overall metagenomics analysis of the feces of the second half of the patient group.

[0015] Figure 13 shows the results of plasma metabolome analysis for patient groups (PD group and non-PD group).

[0015] Figure 14-1 shows the results of plasma metabolome analysis for the first and second halves of the patient population (PD group and non-PD group).

[0015] Figure 14-2 shows the results of plasma metabolome analysis for the first and second halves of the patient population (PD group and non-PD group).

[0015] Figure 15 shows the results of the analysis of fecal metagenomics and plasma metabolome. The analysis of fecal metagenomics for the non-PD and PD groups shows the downregulated KEGG pathway (thick blue line) and upregulated KEGG pathway (thick dotted red line) in the non-PD group. Furthermore, in the plasma metabolome, among the metabolites showing significant differences, downregulated metabolites are indicated by "black circles," and upregulated metabolites are indicated by "slanted circles."

[0015] Figure 16 shows the results of whole blood gene expression analysis (RNA-Seq analysis) in patient groups (PD group and non-PD group).

[0015] Figure 17-1 shows the results of whole blood gene expression analysis (RNA-Seq analysis) in patient populations (PD population and non-PD population). List of genes with significant differences in expression levels between the two populations.

[0015] Figures 17-2 and 17-3 show the results of whole blood gene expression analysis (RNA-Seq analysis) (MAPK signaling pathway) for the first and second halves of the patient population (ratio of PD population to non-PD population). In the effective population (non-PD population), this pathway is overactive.

[0015] Figure 17-3 is the same as above.

[0015] Figures 17-4 and 17-5 show the results of whole blood gene expression analysis (RNA-Seq analysis) (TCR signaling pathway) for the first and second halves of the patient population (ratio of PD population to non-PD population). In the effective population (non-PD population), this pathway is hyperactive.

[0015] Figure 17-5 is the same as above.

[0015] Figures 17-6 and 17-7 show the results of whole blood gene expression analysis (RNA-Seq analysis) (type I IFN receptor complex) for the first and second halves of the patient population (ratio of PD group to non-PD group). In the effective group (non-PD group), this pathway is hyperactive.

[0015] Figure 17-7 is the same as above.

[0015] Figure 18 shows the results of drug efficacy prediction for patient groups (PD group and non-PD group) using a combination of multiple markers (Odoribacter spp. and Veliionella spp.) as indicators.

[0015] Figure 19 shows the results of drug efficacy prediction for patient groups (PD group and non-PD group) using a combination of multiple markers (Gordonibacter, Geobacillus, Odoribacter and Veliionella) as indicators.

[0015] Figure 20-1 shows the results of drug efficacy prediction for the first half of the patient group (PD group and non-PD group) using a combination of multiple markers (Pyruvic acid, Pipecolinic acid, Glyceric acid) as indicators.

[0015] Figure 20-2 shows the results of drug efficacy prediction for the second half of the patient group (PD group and non-PD group) using a combination of multiple markers (pyruvic acid, 2-piperidinecarboxylic acid, glyceric acid) as indicators.

[0015] Figure 20-3 shows the results of drug efficacy prediction for the first half of the patient group (PD group and non-PD group) using a combination of multiple markers (pyruvic acid, glyceric acid, lactic acid) as indicators.

[0015] Figure 20-4 shows the results of drug efficacy prediction for the second half of the patient group (PD group and non-PD group) using a combination of multiple markers (pyruvic acid, glyceric acid, lactic acid) as indicators.

[0015] Figure 21-1 shows the results of drug efficacy prediction for the first half of the patient population (PD group and non-PD group) using a combination of multiple markers (18 genes) as indicators.

[0015] Figure 21-2 shows the results of drug efficacy prediction for the second half of the patient population (PD group and non-PD group) using a combination of multiple markers (18 genes) as indicators.

[0015] Figure 22-1 shows the results of drug efficacy prediction for the first half of the patient population (PD group and non-PD group) using a combination of multiple markers (15 factors) as indicators.

[0015] Figure 22-2 shows the results of drug efficacy prediction for the second half of the patient population (PD group and non-PD group) using a combination of multiple markers (15 factors) as indicators.

[0015] Figure 23 shows the results of drug efficacy prediction for the first half of the patient population (PD group and non-PD group) using a combination of multiple markers (12 factors) as indicators.

[0015] Figure 24 shows the results of drug efficacy prediction for the first half of the patient population (PD group and non-PD group) using a combination of multiple labels (7 factors) as indicators.

[0015] Figure 25 shows the results of drug efficacy prediction for the first half of the patient population (PD group and non-PD group) using a combination of multiple labels (4 factors) as indicators.

[0015] Figure 26 shows a structural diagram of the system of the present invention.

[0015] Figure 27 illustrates the significant differences in bacterial species and KEGG pathways identified by analyzing the overall survival (OS) data from the feces of patients in the first and second halves of the patient population. *: Items with significant differences identified by the Bonferroni method.

[0015] Figure 28 illustrates the significantly different bacterial species and KEGG pathways identified by analyzing the progression-free survival (PFS) using metagenomic data from feces of patients in the first and second halves of the patient population. *: Items identified as significantly different using the Bonferroni method.

[0015] Figure 29 shows the OS analysis of the patient population using overall metagenomics data from the feces of the first and second halves of the patient population, which identified significant differences in nucleotide metabolic pathways and was scored as a comparison between the surviving and deceased patient populations during the observation period.

[0015] Figure 30 shows the survival curves of patients (first and second halves) whose expression (score) of nucleotide metabolic pathways was divided into 3.4 or higher or less than 3.4, as revealed by the overall metagenomics analysis of feces.

[0015] Figure 31 shows the PFS analysis based on the overall metagenomics data in the feces of patients from the first and second halves of the patient population. Significant differences in nucleotide metabolic pathways can be seen, and patients are scored in groups with / without cancer progression during the observation period.

[0015] Figure 32 shows the progression-free survival (PFS) curves for patients with a nucleotide metabolic pathway performance score of 3.4 or higher or less than 3.4 as shown by fecal metagenomics analysis.

[0015] Figure 33 shows the score of bacterial diversity (Chao1) as shown by the overall metagenomics analysis of feces from patients who survived or died during the observation period (first half and second half).

[0015] Figure 34 shows the survival curves of patients (first and second halves) with a Chao1 score of bacterial diversity (more than 606 or less than 606, as shown by the metagenomic analysis of feces.

[0015] Figure 35 shows the score of bacterial diversity (Chao1) in the stool samples of patients (first half and second half) who were divided into groups with cancer progression / non-progression during the observation period.

[0015] Figure 36 shows the progression-free survival (PFS) curves for patients with a score of 606 or higher or less than 606 (first and second halves) based on the score of the bacterial diversity (Chao1) shown by the overall metagenomics analysis of feces.

[0015] Figure 37 shows the OS analysis of the peptidoglycan biosynthesis pathways that identified significant differences by using overall metagenomics data from the feces of the first and second halves of the patient population, and the scores for comparison between the surviving and deceased patient populations during the observation period.

[0015] Figure 38 shows the survival curves of patients (first and second halves) whose expression level (score) of peptidoglycan biosynthesis pathway was divided into 1.32 or higher or less than 1.32, as shown by the metagenomic analysis of feces.

[0015] Figure 39 shows the metabolome analysis score of the amount of 2-hydroxybutyric acid in the plasma (normalized by internal standard (2-isopropylmalic acid)) of patients who survived or died during the observation period (first half and second half).

[0015] Figure 40 shows the survival curves of patient groups (first and second halves) classified by metabolome analysis scores of plasma 2-hydroxybutyric acid (normalized by internal standard (2-isopropylmalic acid)) with scores above 0.76 or below 0.76.

[0015] Figure 41 shows the metabolome analysis score of the amount of 2-side-oxybutyric acid in the plasma of patients (first half and second half) who survived or died during the observation period (normalized by internal standard (2-isopropylmalic acid)).

[0015] Figure 42 shows the survival curves of patient groups (first and second halves) classified as having a median score of 2-side-oxybutyric acid (normalized by internal standard (2-isopropylmalic acid)) in plasma, categorized as either 0 or above or below 0.

[0015] Figure 43 shows the inter-group comparison results of whole blood gene expression analysis (RNA-Seq analysis) data obtained from patient groups that survived or died during the observation period, and a table of reproducible markers in the first and second halves of the patient groups.

[0015] Figure 44 shows the sum of RNA-Seq analysis scores of the 7 genes recorded in Figure 43 in patients who survived or died during the observation period (first half and second half).

[0015] Figure 45 shows the survival curves of the patient groups (first and second halves) classified by the sum of RNA-Seq analysis performance scores of the 7 genes recorded in Figure 43, with scores of 0.5 or higher or lower than 0.5.

[0015] Figure 46 shows the scores of patients whose cancer progressed during the observation period based on the significant differences in phenylalanine metabolic pathways identified in the PFS analysis using overall metagenomics data from the feces of patients in the first and second halves of the patient population, categorized as those with and without cancer progression.

[0015] Figure 47 shows the progression-free survival (PFS) curves for patients classified as having a performance level (score) of phenylalanine metabolic pathway as shown by fecal metagenomics analysis, categorized as 0.16 or above or below 0.16.

[0015] Figure 48 shows the metabolome analysis score of plasma triglyceride levels (normalized by internal standard (2-isopropylmalic acid)) in patients (first half and second half) whose cancer had worsened / not worsened during the observation period.

[0015] Figure 49 shows the progression-free survival (PFS) curves for patients (first and second halves) classified as patients with plasma glyceric acid levels (normalized by internal standard (2-isopropylmalic acid)) metabolome analysis scores, categorized as above or below -0.00033.

[0015] Figure 50 shows the scores of extracted markers (Arthrobacter genus, fatty acid metabolism pathway) by inter-group comparative analysis of overall metagenomics data in the feces of patients with and without harmful phenomena in the skin.

[0015] Figure 51 shows the SNPs markers of immune-related genes extracted by inter-population comparative analysis of whole blood genomic data from patients with / without harmful phenomena in the skin.

Implementation Method

[0016] The present invention will now be described in detail. Unless otherwise defined, the terms (technical and scientific terms) used in this specification have the same meaning as those commonly understood by those skilled in the art. Furthermore, all references cited in this specification are also incorporated herein by reference, and their entire contents are included herein.

[0017] The present invention discovers that by comparing tumor enlargement groups (PD groups) with non-tumor enlargement groups (non-PD groups), survival groups with death groups, or groups with / without cancer progression in cancer patients who are treated with immune checkpoint inhibitors, the results of exploring biomarkers can determine the amount of a specific bacterial species or the score of a specific genomic pathway that was previously unknown, or the amount of a specific metabolite in the blood, serum or plasma of the subject, or the expression level of a specific gene in the blood, serum or plasma of the subject, and thus predict the subject's response to immune checkpoint inhibitors.

[0018] Furthermore, it has been found that by comparing the presence / absence of harmful phenomena (e.g., harmful skin phenomena) among cancer patients treated with immune checkpoint inhibitors, the results of exploring novel biomarkers can detect the presence of previously unknown specific bacterial species, scores of specific genomic pathways, or specific single nucleotide polymorphisms (SNPs) in the subjects, thereby predicting the risk of adverse phenomena of immune checkpoint inhibitors in the subjects.

[0019] That is, in one embodiment, the present invention provides a method for predicting the response of a cancer patient to an immune checkpoint inhibitor, comprising:

[0019] (1) The step of determining the presence or absence of one or more values ​​or SNPs selected from the group consisting of (i) to (v), wherein (i) to (v) are as follows:

[0019] (i) The amount of microorganisms present in the feces or intestinal contents of the aforementioned object,

[0019] (ii) Scoring of genomic pathways in the feces or intestinal contents of the aforementioned subjects,

[0019] (iii) The amount of metabolites present in the blood, serum or plasma of the aforementioned object.

[0019] (iv) The expression levels of genes in the blood, serum, or plasma of the aforementioned subjects, and

[0019] (v) Single nucleotide polymorphisms (SNPs) selected from one or more genes in the group consisting of IL6R and NLRC5 in the aforementioned objects;

[0019] (2) Using the value obtained in step (1) above or the presence or absence of SNP as an indicator, the step of predicting the response of the above object to immune checkpoint inhibitor;

[0019] Among them,

[0019] The aforementioned microorganisms are selected from one or more species of microorganisms in the group consisting of the genera *Geobacillus*, *Gordonibacter*, *Odoribacter*, *Veillonella*, *Corynebacterium*, *Porphyromonas*, and *Arthrobacter*.

[0019] The aforementioned genomic pathway is one or more genomic pathways selected from the group consisting of bacterial invasion of epithelial cells, fatty acid degradation, flagellar assembly, fatty acid biosynthesis, PPAR signaling pathway, peptidoglycan biosynthesis, nucleotide metabolism, purine metabolism, phenylalanine metabolism, and fatty acid metabolism.

[0019] The aforementioned metabolites are one or more metabolites selected from the group consisting of lactic acid, pyruvic acid, glucose, 2-sideoxybutyric acid, glyceric acid, caprylic acid, citrulline, 2-hydroxybutyric acid and 2-piperidinecarboxylic acid.

[0019] The aforementioned genes are selected from genes related to the MAPK pathway, the Type I IFN receptor complex pathway, and the TCR signaling pathway, as well as one or more genes from the group consisting of BCL11B, ERCC3, ERCC6, FCRL1, FCRL3, MS4A1, and TCF7.

[0019] The aforementioned SNP is one or more SNPs selected from the group consisting of the following rs numbers: rs2228145.1, rs7190199, and rs7185320.

[0020] In this specification, the term "cancer" is not specifically limited. For example, examples include leukemia (e.g., acute myeloid leukemia, chronic myeloid leukemia, acute lymphoblastic leukemia, chronic lymphoblastic leukemia), malignant lymphoma (Hodgkin's lymphoma, non-Hodgkin's lymphoma (e.g., adult T-cell leukemia, follicular lymphoma, diffuse large cell B-cell lymphoma)), multiple myeloma, myelomecosis syndrome, head and neck cancer, esophageal cancer, esophageal adenocarcinoma, gastric cancer, colorectal cancer, colon cancer, rectal cancer, liver cancer (e.g., hepatocellular carcinoma), gallbladder and bile duct cancer, bile duct cancer, pancreatic cancer, thyroid cancer, lung cancer, etc. Cancers include (e.g., non-small cell lung cancer (e.g., squamous non-small cell lung cancer, non-squamous non-small cell lung cancer), small cell lung cancer), breast cancer, ovarian cancer (e.g., serous ovarian cancer), cervical cancer, uterine cancer, endometrial cancer, vaginal cancer, vulvar cancer, kidney cancer (e.g., renal cell carcinoma), urothelial carcinoma (e.g., bladder cancer, upper urinary tract cancer), prostate cancer, testicular tumors (e.g., germ cell tumors), bone / soft tissue sarcoma, skin cancer (e.g., uveal malignant melanoma, malignant melanoma, Merkel cell carcinoma), glioma, brain tumors (e.g., glioblastoma), pleural mesothelioma, and cancers of unknown primary origin). In this invention, for cancers that can be predicted to respond to immune checkpoint inhibitors, gastric cancer is preferred. More preferably, unresectable progressive gastric cancer.

[0021] In this specification, "cancer (malignant tumor) treatment" includes, for example, treatments intended to (i) reduce cancer proliferation, (ii) reduce symptoms caused by cancer, (iii) improve the quality of life of cancer patients, (iv) reduce the dosage of other anticancer drugs or adjuvant cancer treatments already administered, and / or (v) prolong the survival of cancer patients. Furthermore, treatment includes recurrence suppression. "Recurrence suppression" means the preventive suppression of cancer recurrence in patients whose cancer has been completely or substantially eliminated or removed through cancer treatment or surgical resection.

[0022] Among them, "other anticancer drugs" may include, for example, alkylating agents, platinum preparations, metabolic antagonists (e.g., folic acid metabolism antagonists, pyridine metabolism inhibitors, purine metabolism inhibitors), ribonucleotide reductase inhibitors, nucleotide analogs, topoisomerase inhibitors, microtubule polymerization inhibitors, microtubule depolymerization inhibitors, antitumor antibiotics, cytokine preparations, antihormonal drugs, molecularly targeted drugs, and cancer immunotherapy drugs.

[0023] The therapeutic agent of the present invention, (1) for enhancing the therapeutic effect on cancer, (2) for reducing the dosage of other drugs used in combination, and / or (3) for reducing the side effects of other drugs used in combination, may be used in combination with one or more other drugs (mainly anticancer drugs) used for the purpose of treating cancer as described above. In the present invention, the dosage form of the prescription formulated in combination with other drugs may be a type of admixture in which two components are mixed in one preparation, or it may be a dosage form that forms individual preparations. When the therapeutic agent of the present invention is given separately from other drugs, the therapeutic agent of the present invention may be given first, and then other drugs may be given after the treatment agent is given, or other drugs may be given first, and then the therapeutic agent of the present invention may be given. In the above dosage, there may be a period during which two drugs are given simultaneously. Furthermore, the dosage methods of various drugs may be the same or different. Depending on the properties of the pharmaceutical agent, it can also be provided as a package containing the therapeutic agent of the present invention and other pharmaceutical agents. The dosage of the other pharmaceutical agents can be appropriately selected based on the dosage used clinically. Furthermore, any two or more other pharmaceutical agents can be combined and administered in a suitable ratio. Moreover, the aforementioned other pharmaceutical agents include not only those currently existing but also those that will emerge in the future.

[0024] In this specification, "immune checkpoint inhibitor" means a drug that exerts immunosuppressive function by inhibiting the transmission of inhibitory co-signals by immune checkpoint molecules. Regarding immune checkpoint molecules, known ones include CTLA-4, PD-1, PD-L1 (programmed cell death-ligand 1), PD-L2 (programmed cell death-ligand 2), LAG-3 (lymphocyte activation gene 3), TIM3 (T cell immunoglobulin and mucin-3), BTLA (B and T lymphocyte attenuator), B7H3, B7H4, 2B4, CD160, A2aR (adenosine A2a receptor), KIR (killer inhibitory receptor), and VISTA (V-domain Ig-containing suppressor of T cell activation). activation), IDO1 (indoleamine 2,3-dioxygenase), arginase I, TIGIT (T cell immunoglobulin and ITIM domain), CD115, etc. (see Nature Reviews Cancer, 12, 252-264, 2012; Cancer Cell, 27, 450-461, 2015), however, there are no particular limitations if the molecules have a consistent function.

[0025] In this invention, the applicable immune checkpoint inhibitor is not particularly limited to any substance that can inhibit the function (signal) of immune checkpoint molecules.

[0026] In terms of immune checkpoint inhibitors, it is preferable to be an inhibitor of human immune checkpoint molecules, and more preferably a neutralizing antibody against human immune checkpoint molecules.

[0027] Regarding immune checkpoint inhibitors, for example, inhibitors of immune checkpoint molecules selected from the group consisting of CTLA-4, PD-1, PD-L1, PD-L2, LAG-3, TIM3, BTLA, B7H3, B7H4, 2B4, CD160, A2aR, KIR, VISTA, and TIGIT can be listed. Examples of immune checkpoint inhibitors can be listed below, but immune checkpoint inhibitors are not limited to these.

[0028] Regarding immune checkpoint inhibitors, for example, anti-CTLA-4 antibodies (e.g., ipilimumab (YERVOY (registered trademark)), tremelimumab, AGEN-1884), anti-PD-1 antibodies (e.g., nivolumab (Opdivo (registered trademark)), REGN-2810, pembrolizumab (KEYTRUDA (registered trademark)), PDR-001, BGB-A317, AMP-514 (MEDI0680), BCD-100, IBI-308, JS-001, PF-06801591, TSR-042), and anti-PD-L1 antibodies (e.g., arteriosclerosis). Atezolizumab (RG7446, MPDL3280A), Avelumab (PF-06834635, MSB0010718C), Durvalumab (MEDI4736), BMS-936559, CA-170, LY-3300054), anti-PD-L2 antibodies (e.g., rHIgM12B7), PD-L1 fusion proteins, PD-L2 fusion proteins (e.g., AMP-224), anti-Tim-3 antibodies (e.g., MBG453), anti-LAG-3 antibodies (e.g., BMS-986016, LAG525), anti-KIR antibodies (e.g., lirilumab), etc. Furthermore, antibodies containing the heavy and light chain complementarity-determining regions (CDRs) or variable regions (VRs) of the aforementioned known antibodies are also a type of immune checkpoint inhibitor. For example, as a further example of anti-PD-1 antibodies, antibodies containing the heavy and light chain complementarity-determining regions (CDRs) or variable regions (VRs) of nivolumab can be cited.

[0029] Among the immune checkpoint inhibitors applicable to this invention, anti-CTLA-4 antibodies, anti-PD-1 antibodies, anti-PD-L1 antibodies, anti-PD-L2 antibodies, PD-L1 fusion proteins, and PD-L2 fusion proteins are preferred. More preferably, anti-PD-1 antibodies, anti-PD-L1 antibodies, anti-PD-L2 antibodies, PD-L1 fusion proteins, and PD-L2 fusion proteins are also preferred. Particularly preferred are anti-PD-1 antibodies. Regarding anti-PD-1 antibodies, antibodies containing complementarity-determining regions (CDRs) or variable regions (VR) of the heavy and light chains of nivolumab (including nivolumab) are preferred, and nivolumab is even more preferred.

[0030] Any one or any plurality of antibodies or fusion proteins among these immune checkpoint inhibitors may be used in this invention.

[0031] In this invention, the dosage of immune checkpoint inhibitors administered to subjects who respond well to immune checkpoint inhibitors varies with age, weight, symptoms, treatment effect, administration method, treatment time, etc., but can be adjusted to achieve the best desired effect.

[0032] For example, when using anti-PD-1 antibodies, the dosage is typically 0.1 to 20 mg / kg body weight. Also, when using antibodies containing the heavy and light chain complementarity-determining regions (CDRs) or variable regions (VR) of nivolumab (e.g., nivolumab), the dosage is typically 0.3 to 10 mg / kg body weight. Preferably, for adults, nivolumab can be administered by intravenous infusion at the following intervals: (1) 1 mg / kg (body weight) once every 3 weeks, (2) 3 mg / kg (body weight) once every 2 weeks, (3) 2 mg / kg (body weight) once every 3 weeks, (4) 80 mg once every 3 weeks, (5) 240 mg once every 2 weeks, (6) 360 mg once every 3 weeks, or (7) 480 mg once every 4 weeks.

[0033] In this specification, "feces" means that which is expelled from the intestines to the outside, and "intestinal contents" means the contents before they are expelled from the intestines. In this invention, by determining the amount of microorganisms present in the feces or intestinal contents of a cancer patient, the patient's responsiveness to immune checkpoint inhibitors can be predicted from the amount of microorganisms present.

[0034] In this specification, the "step of determining the value of the quantity of microorganisms" can determine whether the desired microorganisms exist, the absolute quantity of the desired microorganisms, or the relative quantity of the desired microorganisms.

[0035] The step of determining the quantity of microorganisms can be done using well-known methods for identifying microorganisms. For example, the quantity of microorganisms can be determined by metagenomic analysis. Metagenomic analysis refers to the analysis of the overall genome of a microbial colony without a culture process. In this invention, metagenomic analysis can be performed using well-known methods, such as next-generation sequencer methods and analysis. The identification of microorganisms by metagenomic analysis can be, for example, by comparing the base sequences of the 16S rRNA gene in the ribosomal subunits for microbial classification, or by identifying the species by the sequences of other genomic regions of the microorganism; there is no particular limitation. Furthermore, the quantity of the microorganism can also be determined by performing metagenomic analysis.

[0036] In one embodiment, the inventors first discovered that the presence of one or more microorganisms selected from the group consisting of *Geobacillus*, *Gordonibacter*, *Odoribacter*, *Veillonella*, *Corynebacterium*, *Porphyromonas*, and *Arthrobacter* in the feces or intestinal contents of the subject can be used as an indicator to predict the subject's response to immune checkpoint inhibitors. For example, the inventors found that in cancer patients who showed a good response to immune checkpoint inhibitors, the presence of microorganisms of the genus *Geobacillus*, *Gordonibacter*, or *Veillonella* was significantly higher than in cancer patients who did not show a good response to immune checkpoint inhibitors. In other words, in an embodiment, if the presence of *Geobacillus*, *Gordonibacter*, or *Veillonella* microorganisms in the feces or intestinal contents of a subject is higher than a set threshold, it can predict that the subject will respond well to immune checkpoint inhibitors. Using the presence of *Veillonella* microorganisms as an indicator is even better.

[0037] Furthermore, in other samples, if the amount of *Geobacillus*, *Gordonibacter*, or *Veillonella* microorganisms present in the feces or intestinal contents of the subject is lower than a set threshold, it can predict that the subject will not respond well to immune checkpoint inhibitors. However, using the amount of *Veillonella* microorganisms as an indicator is better.

[0038] In one embodiment, the inventors first discovered that the presence of *Odoribacter* microorganisms in the feces or intestinal contents of a subject can predict the subject's response to immune checkpoint inhibitors. For example, the inventors found that the presence of *Odoribacter* microorganisms in cancer patients who showed a good response to immune checkpoint inhibitors was significantly lower than that in cancer patients who did not show a good response to immune checkpoint inhibitors. That is, in one embodiment, when the presence of *Odoribacter* microorganisms in the feces or intestinal contents of a subject is lower than a set threshold, a good response to immune checkpoint inhibitors can be predicted for that subject.

[0039] Furthermore, in other samples, if the presence of Odoribacter microorganisms in the feces or intestinal contents of the subject is higher than a set threshold, it can be predicted that the subject will respond poorly to immune checkpoint inhibitors.

[0040] Furthermore, in other samples, if the amount of Arthrobacter microorganisms present in the feces or intestinal contents of the subject is higher than a set threshold, it can be predicted that harmful skin phenomena will be induced in the aforementioned subject due to immune checkpoint inhibitors.

[0041] In this invention, the specific value of the "set threshold" or "cutoff value" is not limited as it varies depending on the analysis method or measurement conditions. For example, it can be determined by clinical studies comparing cancer patients who show a good response to immune checkpoint inhibitors (non-PD group) and cancer patients who do not show a good response to immune checkpoint inhibitors (PD group); analysis of the overall survival (OS) and progression-free survival (PFS) of cancer patients given immune checkpoint inhibitors; analysis of the presence or absence of harmful phenomena in cancer patients given immune checkpoint inhibitors, etc.

[0042] In this specification, "genomic pathway" means a pathway (a network of intermolecular interactions in a metabolic or signaling system) that can be identified by the base sequence obtained from the feces or intestinal contents of a subject, for example, a pathway registered in the KEGG (Kyoto Encyclopedia of Genes and Genomes) PATHWAY database (https: / / www.genome.jp / kegg / pathway.html). Genomic pathways can be identified from the feces or intestinal contents of a subject using a metagenomic dataset obtained by the same method as the metagenomic analysis described above, and then using the aforementioned KEGG PATHWAY database. By analyzing a total metagenomics dataset obtained from the feces or intestinal contents of subjects, the inventors discovered that the scores of one or more genomic pathways selected from the group consisting of bacterial invasion of epithelial cells, fatty acid degradation, flagellar assembly, fatty acid biosynthesis, and the PPAR signaling pathway showed significant differences between cancer patients who showed good response to immune checkpoint inhibitors (non-PD group) and cancer patients who did not show good response to immune checkpoint inhibitors (PD group). Furthermore, it was found that scores for one or more genomic pathways selected from the group consisting of peptidoglycan biosynthesis, nucleotide metabolism, purine metabolism, phenylalanine metabolism, and fatty acid metabolism showed significant differences between cancer patients who showed good response to immune checkpoint inhibitors (survivors or those without progression) and cancer patients who did not show good response to immune checkpoint inhibitors (deceased or those with progression).

[0043] That is, in one embodiment, if the scores of the genomic pathways in the feces or intestinal contents of the subject for bacterial invasion of epithelial cells, fatty acid degradation, flagellar assembly, PPAR signaling pathway, and / or phenylalanine metabolism are lower than a set threshold, it can be predicted that the subject will respond well to the immune checkpoint inhibitor. Furthermore, in other embodiments, if the scores of the genomic pathways in the feces or intestinal contents of the subject for bacterial invasion of epithelial cells, fatty acid degradation, flagellar assembly, PPAR signaling pathway, and / or phenylalanine metabolism are higher than a set threshold, it can be predicted that the subject will respond poorly to the immune checkpoint inhibitor.

[0044] Furthermore, in one embodiment, if the scores for fatty acid biosynthesis, nucleotide metabolism, and / or peptidoglycan biosynthesis in the genomic pathways of the subject's feces or intestinal contents are higher than a set threshold, it can predict that the subject will respond well to the immune checkpoint inhibitor. Furthermore, in other embodiments, if the scores for fatty acid biosynthesis, nucleotide metabolism, and / or peptidoglycan biosynthesis in the genomic pathways of the subject's feces or intestinal contents are lower than a set threshold, it can predict that the subject will respond poorly to the immune checkpoint inhibitor.

[0045] Furthermore, in other samples, if the score of fatty acid metabolism in the genomic pathway of the subject's feces or intestinal contents is higher than a set threshold, it can be predicted that immune checkpoint inhibitors will induce harmful skin phenomena in the aforementioned subjects.

[0046] The genomic pathway score used in this invention can be determined by performing a metagenomics analysis on the feces or intestinal contents of the subject and inputting the obtained metagenomics dataset into the KEGG PATHWAY database. For example, it can be implemented by referring to the reference: Abubucker et al. PLoS Computational Biology 2012, (8) 6 e1002358.

[0047] In one embodiment, the present invention can predict a subject's response to an immune checkpoint inhibitor by determining the amount of metabolites present in the subject's blood, serum, or plasma. The inventors have discovered that the amount of one or more metabolites, particularly selected from the group consisting of lactic acid, pyruvate, glucose, 2-hydroxybutyric acid, glyceric acid, caprylic acid, citrulline, 2-hydroxybutyric acid, and 2-piperidinic acid, present in the subject's blood, serum, or plasma can be used as an indicator to predict a subject's response to an immune checkpoint inhibitor. That is, in one embodiment, if the amount of lactic acid, pyruvate, glucose, glyceric acid, caprylic acid, 2-hydroxybutyric acid, or 2-hydroxybutyric acid present in the subject's blood, serum, or plasma is lower than a set threshold, a good response to an immune checkpoint inhibitor can be predicted. Furthermore, in other implementations, if the levels of lactic acid, pyruvate, glucose, glyceric acid, caprylic acid, 2-hydroxybutyric acid, or 2-sideoxybutyric acid in the metabolites present in the subject's blood, serum, or plasma are higher than a set threshold, it can predict that the subject will not respond well to immune checkpoint inhibitors.

[0048] Furthermore, in other embodiments, if the amount of 2-piperidinecarboxylic acid and / or citrulline present in the metabolites of the subject's blood, serum, or plasma is higher than a set threshold, it can be predicted that the subject has a good response to the immune checkpoint inhibitor. Furthermore, in other embodiments, if the amount of 2-piperidinecarboxylic acid and / or citrulline present in the metabolites of the subject's blood, serum, or plasma is lower than a set threshold, it can be predicted that the subject has a poor response to the immune checkpoint inhibitor.

[0049] The method for determining the amount of metabolites present in blood, serum, or plasma is not particularly limited, as long as a well-known metabolome analysis method is used, such as mass analysis or nuclear magnetic resonance. Mass analysis refers to a method in which blood, serum, or plasma samples are ionized using an ion source to form gaseous ions (ionization), and then the ionized blood, serum, or plasma samples are separated according to their mass-charge ratio using electromagnetic force or time-of-flight difference, and measured using a detectable mass analyzer. Regarding ionization methods using ion sources, suitable options include electron ionization (EI), chemical ionization (CI), electric field deionization (FD), high-speed atomic impact (FAB), matrix-supported laser deionization (MALDI), and electrospray ionization (ESI). In the analytical section, suitable methods for separating ionized blood, serum, or plasma samples include magnetic field deflection, quadrupole, ion trap, time-of-flight (TOF), and Fourier transform ion cyclotron resonance methods. Furthermore, tandem mass analysis (MS / MS) combining two or more mass analysis methods can be used. Additionally, gas chromatography (GC), liquid chromatography (LC), or high-performance liquid chromatography (HPLC) can be used to separate and purify metabolic groups from impurities for analysis.

[0050] In one embodiment, step (1) may also be (iv) the step of determining the expression level of genes in the blood, serum, or plasma of the subject. In this case, the genes determined are preferably genes related to one or more pathways selected from the group consisting of the MAPK pathway, the Type I IFN receptor complex pathway, and the TCR signaling pathway. The inventors found that the expression of genes related to one or more pathways selected from the group consisting of the MAPK pathway, the Type I IFN receptor complex pathway, and the TCR signaling pathway differed significantly between cancer patients who showed a good response to immune checkpoint inhibitors (non-PD group) and cancer patients who did not show a good response to immune checkpoint inhibitors (PD group). Furthermore, the inventors, through analyzing the expression levels of genes in the blood, serum, or plasma of the subjects, discovered significant differences in the expression of the following genes: BCL11B, ERCC3, ERCC6, FCRL1, FCRL3, MS4A1, and TCF7, between cancer patients who showed a good response to immune checkpoint inhibitors (survivor group) and cancer patients who did not show a good response to immune checkpoint inhibitors (deceased group).

[0051] In this specification, "MAPK pathway (also known as MAPK signaling pathway)" refers to the signaling pathway involving MAP kinase (mitogen-activated protein kinase). MAPK is a serine / threonine kinase involved in various cellular functions such as metabolism, proliferation, division, movement, and apoptosis. In this invention, genes related to the MAPK pathway that can be used to predict responses to immune checkpoint inhibitors may be listed, for example, one or more genes selected from MAPKAPK5, MAPKAPK5-AS1, MAP3K14, MAPK3K7, MAP3K1, MAPK9, MAP3K5, and MAPK14.

[0052] In this specification, genes related to the "Type I IFN receptor complex" pathway may include PIK3R1, PTPN11, STAT1, FYN, EIF4B, RAC1, MAP3K1, RPS6KB1, PDCD4, REL, MAPK14, RPS6KA5, STAT2, CRK, MAP2K6, RAPGEF1, etc.

[0053] In this specification, genes related to the "TCR signaling pathway" may include MAP3K14, CD4, FYN, DLG1, PDK1, NFKB1, MALT1, LAT, SOS1, MAP3K7, IL10, GRAP2, CD40LG, PIK3CA, PRKCQ, NFATC2, CALM1, etc.

[0054] In one embodiment, a high expression level (e.g., a performance score calculated by any method) of genes related to one or more pathways selected from the group consisting of the MAPK pathway, the Type I IFN receptor complex pathway, and the TCR signaling pathway compared to a set threshold can predict a good response to immune checkpoint inhibitors. Conversely, in other embodiments, a low expression level (e.g., a performance score calculated by any method) of genes related to one or more pathways selected from the group consisting of the MAPK pathway, the Type I IFN receptor complex pathway, and the TCR signaling pathway compared to a set threshold can predict a poor response to immune checkpoint inhibitors.

[0055] In one embodiment, a high expression level (e.g., expression score calculated by any method) of one or more genes selected from the group consisting of BCL11B, ERCC3, ERCC6, FCRL1, FCRL3, MS4A1, and TCF7 compared to a predetermined threshold indicates a good response to an immune checkpoint inhibitor. Conversely, in other embodiments, a low expression level (e.g., expression score calculated by any method) of one or more genes selected from the group consisting of BCL11B, ERCC3, ERCC6, FCRL1, FCRL3, MS4A1, and TCF7 compared to a predetermined threshold indicates a poor response to an immune checkpoint inhibitor.

[0056] In one embodiment, the step of determining the expression level of genes in the blood, serum or plasma of the subject can also be determined by well-known transcriptome analysis.

[0057] In one embodiment, step (1) can also be (v) determining the presence or absence of single nucleotide polymorphisms (SNPs) in one or more genes selected from the group consisting of IL6R and NLRC5 in the aforementioned subjects, wherein the SNP can be one or more SNPs selected from the group consisting of the following rs numbers: rs2228145.1; rs7190199; and rs7185320. The inventors analyzed the whole blood genomic SNP arrays of cancer patients (those with / without harmful skin phenomena) administered immune checkpoint inhibitors and found a correlation between specific SNPs and the occurrence of harmful skin phenomena. That is, in cases where the subject has one or more SNPs selected from the group consisting of rs2228145.1; rs7190199; and rs7185320, it can be predicted that immune checkpoint inhibitors will induce harmful skin phenomena.

[0058] According to the method of predicting the response of cancer patients to immune checkpoint inhibitors according to the present invention, the response of patients to immune checkpoint inhibitors can also be predicted by combining one or more values ​​or the presence or absence of SNPs selected from the group consisting of (i) to (v) below, which are as follows:

[0058] (i) The amount of microorganisms present in the feces or intestinal contents of the aforementioned object;

[0058] (ii) Scoring of genomic pathways in the feces or intestinal contents of the aforementioned subjects;

[0058] (iii) The amount of metabolites present in the blood, serum or plasma of the aforementioned object;

[0058] (iv) The expression levels of genes in the blood, serum, or plasma of the aforementioned subjects; and

[0058] (v) Single nucleotide polymorphisms (SNPs) of one or more genes selected from the group consisting of IL6R and NLRC5 in the aforementioned objects.

[0058] By combining the above indicators, the accuracy of diagnosis becomes higher.

[0059] By means of the method of the present invention, by administering immune checkpoint inhibitors to subjects who are determined to have a good response to immune checkpoint inhibitors, not only can improved treatment efficacy be expected, but it can also contribute to the medical economy. Therefore, in one embodiment, the present invention can provide a method for treating cancer patients, which includes...

[0059] (1) The step of determining the presence or absence of one or more values ​​or SNPs selected from the group consisting of the following (i) to (v):

[0059] (i) The amount of microorganisms present in the feces or intestinal contents of the aforementioned object;

[0059] (ii) Scoring of genomic pathways in the feces or intestinal contents of the aforementioned subjects;

[0059] (iii) The amount of metabolites present in the blood, serum or plasma of the aforementioned object;

[0059] (iv) The expression levels of genes in the blood, serum, or plasma of the aforementioned subjects; and

[0059] (v) Single nucleotide polymorphisms (SNPs) of one or more genes selected from the group consisting of IL6R and NLRC5 in the aforementioned objects;

[0059] (2) Using the value obtained in step (1) above or the presence or absence of SNP as an indicator, the step of predicting the response of the aforementioned object to immune checkpoint inhibitors;

[0059] (3) The step of administering the aforementioned immune checkpoint inhibitor to the aforementioned subject who is determined to have a good response to the immune checkpoint inhibitor;

[0059] The aforementioned microorganisms are selected from one or more species of microorganisms in the group consisting of the genera *Geobacillus*, *Gordonibacter*, *Odoribacter*, *Veillonella*, *Corynebacterium*, *Porphyromonas*, and *Arthrobacter*.

[0059] The aforementioned genomic pathway is one or more genomic pathways selected from the group consisting of bacterial invasion of epithelial cells, fatty acid degradation, flagellar assembly, fatty acid biosynthesis, PPAR signaling pathway, peptidoglycan biosynthesis, nucleotide metabolism, purine metabolism, phenylalanine metabolism, and fatty acid metabolism.

[0059] The aforementioned metabolites are one or more metabolites selected from the group consisting of lactic acid, pyruvic acid, glucose, 2-sideoxybutyric acid, glyceric acid, caprylic acid, citrulline, 2-hydroxybutyric acid and 2-piperidinecarboxylic acid.

[0059] The aforementioned genes are selected from genes related to the MAPK pathway, the Type I IFN receptor complex pathway, and the TCR signaling pathway, as well as one or more genes from the group consisting of BCL11B, ERCC3, ERCC6, FCRL1, FCRL3, MS4A1, and TCF7.

[0059] The aforementioned SNP is one or more SNPs selected from the group consisting of the following rs numbers: rs2228145.1; rs7190199; and rs7185320.

[0060] Furthermore, in one embodiment, the present invention may be a system 10 (FIG. 26), which is a system 10 for determining the responsiveness of a cancer subject to an immune checkpoint inhibitor, comprising a memory unit 11, an input unit 12, a data processing unit 14, and an output unit 15, characterized in that:

[0060] The memory unit 11 stores one or more cutoff values ​​selected from the group consisting of (1-i) to (1-v) for determining responsiveness to immune checkpoint inhibitors, as follows:

[0060] (1-i) The amount of microorganisms present in feces or intestinal contents;

[0060] (1-ii) Scoring of genomic pathways in feces or intestinal contents;

[0060] (1-iii) The amount of metabolites present in blood, serum or plasma;

[0060] (1-iv) The level of gene expression in blood, serum, or plasma; and

[0060] (1-v) Single nucleotide polymorphisms (SNPs) selected from one or more genes in the group consisting of IL6R and NLRC5;

[0060] The memory unit 11 inputs one or more values ​​or the presence or absence of SNPs selected from the group consisting of (2-i) to (2-v) from the input unit 12, and stores them in the memory unit 11, wherein (2-i) to (2-v) are as follows:

[0060] (2-i) The amount of microorganisms present in the feces or intestinal contents of the aforementioned object;

[0060] (2-ii) Scoring of genomic pathways in the feces or intestinal contents of the aforementioned subjects;

[0060] (2-iii) The amount of metabolites present in the blood, serum or plasma of the aforementioned object;

[0060] (2-iv) The expression levels of genes in the blood, serum, or plasma of the aforementioned subjects; and

[0060] (2-v) Single nucleotide polymorphisms (SNPs) of one or more genes selected from the group consisting of IL6R and NLRC5 in the aforementioned objects;

[0060] The data processing unit 14 compares the previously memorized value or the presence or absence of SNP with the aforementioned cutoff value and determines the responsiveness of the aforementioned object to the immune checkpoint inhibitor.

[0060] Output unit 15 outputs the results of the determination of the responsiveness of the aforementioned object to the immune checkpoint inhibitor of cancer.

[0061] The system 10 may further include an analytical measuring unit 13, which determines the amount of microorganisms present or a genomic pathway score in the feces or intestinal contents of the aforementioned object.

[0061] The aforementioned analytical measuring unit 13 determines the amount of metabolites present in the blood, serum, or plasma of the aforementioned subject.

[0061] The aforementioned analytical measuring unit 13 determines the expression level of genes in the blood, serum, or plasma of the aforementioned subject, and / or

[0061] The aforementioned analytical assay unit 13 determines the single nucleotide polymorphism (SNP) of one or more genes selected from the group consisting of IL6R and NLRC5.

[0061] Alternatively, the aforementioned input unit 12 may be replaced, or the aforementioned input unit 12 may be used to input the aforementioned microbial quantity, the aforementioned genomic pathway score, the aforementioned metabolite quantity, the aforementioned gene expression level and / or the aforementioned SNP quantity determined by the aforementioned analysis and determination unit 13 (Fig. 26).

[0062] The memory unit 11 may include memory devices such as RAM, ROM, and flash memory; fixed magnetic disk devices such as hard disk drives; or removable memory devices such as floppy disks and optical disks. In addition to data measured by the analysis and measurement unit 13, data and instructions input from the input unit 12, and calculation results performed by the data processing unit 14, the memory unit 11 may also store computer programs and databases used in various processes of the data processing device. Computer programs may be stored via computer-readable recording media such as CD-ROMs and DVD-ROMs, or via the internet. Computer programs may be stored in the memory unit 11 containing known installation programs.

[0063] The input unit 12 is an interface, including an operating unit such as a keyboard and mouse. Herein, the input unit 12 can input data measured by the analysis and measurement unit 13, instructions for calculations performed by the data processing unit 14, etc. Furthermore, if the input unit 12, such as the analysis and measurement unit 13, is external, in addition to the operating unit, it may include an interface for inputting the measured data via a network or memory medium.

[0064] The analytical measurement unit 13 performs the aforementioned steps of measuring the presence of microorganisms, scoring of genomic pathways, presence of metabolites, gene expression levels, and / or SNPs. Therefore, the analytical measurement unit 13 is configured to measure the presence of microorganisms, scoring of genomic pathways, presence of metabolites, gene expression levels, and / or SNPs. The analytical measurement unit 13 is not intended to be limited to the following, but can be used alone or in combination with known machines such as next-generation sequencers capable of performing overall metagenomic analysis, mass analyzers capable of determining the presence of metabolites, or nuclear magnetic resonance devices. The analytical measurement unit 13 can be configured separately from the system 10, or the measured data can be input via the input unit 12 using a network or memory medium.

[0065] The data processing unit 14 can determine the responsiveness of a subject to an immune checkpoint inhibitor by using values ​​input from the input unit 12 or the analysis and measurement unit 13, and a cutoff value used to determine the responsiveness to the immune checkpoint inhibitor. The data processing unit 14 performs various calculations on the data stored in the memory unit 11, measured by the analysis and measurement unit 13, according to a program stored in the memory unit 11. These calculations are executed by the CPU included in the data processing unit 14. The CPU includes a functional module that controls the analysis and measurement unit 13, the input unit 12, the memory unit 11, and the output unit 15, and can perform various adjustments. These components can be constructed using separate integrated circuits, microprocessors, firmware, etc.

[0066] The output unit 15 is configured to output the results of the calculations performed by the data processing unit. The output unit 15 may be an output means such as a display device such as a liquid crystal display or a printer that directly displays the results of the calculations, or it may be an interface for outputting to an external memory device or via a network.

[0067] [Inspection and Testing Kit]

[0067] The present invention also includes an invention kit for testing or measuring the evaluation items related to the present invention ((1-i) the amount of microorganisms present in feces or intestinal contents; (1-ii) the score of genomic pathways in feces or intestinal contents; (1-iii) the amount of metabolites present in blood, serum or plasma; (1-iv) the expression level of genes in blood, serum or plasma; or (1-v) the single nucleotide polymorphism (SNP) of one or more genes in the group consisting of IL6R and NLRC5).

[0068] All patent and non-patent documents or references that are clearly cited in this specification may be incorporated in their entirety as part of this specification.

[0069] Hereinafter, although the present invention has been described in detail with reference to embodiments, the present invention is not limited to these embodiments. Those skilled in the art can easily modify and alter the present invention based on the description in this specification, and such modifications and alterations are also included within the technical scope of the present invention.

[0069] [Example]

[0070] The following analysis was performed on specimens (before drug administration) obtained from the observational study (DELIVER trial) (JACCRO GC-08) in patients with unresectable progressive gastric cancer using biomarkers containing nivolumab. Details of the DELIVER trial can be found in the UMIN-CTR clinical trial registry (UMIN trial ID: UMIN000030850).

[0071] Furthermore, in the embodiments and diagrams, "first half" refers to the training cohort and "second half" refers to the validation cohort.

[0072] [Example 1]

[0072] <Materials and Analysis Methods>

[0072] 1. Analysis of fecal overall metagenomics

[0072] 1-1. Collection of Specimens

[0072] From 478 cases (196 samples from the first half + 282 samples from the second half), stool samples approximately the size of a small bean were collected using a special medicine spoon and immediately placed into a container containing guanidine solution. The container was shaken 5 to 6 times to mix the contents, and then submitted to the medical authority within one week for storage at -20°C or -80°C.

[0073] 1-2. Modulation of the specimen

[0073] Zirconium oxide beads were added to the sample, and the bacterial cells were broken up (5 m / s, 2 min). The cells were then purified using magnetic beads, dissolved, and the total metagenomic DNA was extracted. (Reference: Hosomi et al. Scientific Reports 2017, (7) 4339.)

[0074] The quality was confirmed by agarose gel electrophoresis. The concentration was confirmed using a NanoPhotometer spectrophotometer (IMPLEN, CA, USA) and a Qubit 2.0 Flurometer (Life Technologies, CA, USA).

[0075] 1-3. Sequencing

[0075] The sequence library was prepared using 1 μg of genomic DNA, modulated using the NEBNext Ultra DNA Library Prep Kit for Illumina (NEB, USA). For library purification, the AMPure XP system was used, and the quality of the sequence library was confirmed using an Agilent 2100 Bioanalyzer (Agilent Technologies, Santa Clara, CA). The sequences in the sequence library were sequenced using a NovaSeq 6000 (Illumina, Inc., San Diego, CA) with 150 bp paired end sequencing.

[0076] 2. Analysis of RNA Expression in Blood Cells (RNA-Seq)

[0076] 2-1. Collection of Specimens

[0076] The blood used for RNA extraction was collected from 439 cases (174 samples from the first half + 265 samples from the second half). EDTA-2K was added to the blood collection tubes, and the blood was immediately inverted and mixed more than 10 times. The blood was then stored at -20°C or -80°C.

[0077] 2-2. Preparation of the specimen

[0077] Total RNA was obtained by dissolving whole blood in TRIzol (trademark) LS Reagent (Thermo Fisher) and then extracting it using the miRNeasy Micro Kit (QIAGEN, Valencia, CA, USA). The quality of Total RNA was confirmed by electrophoresis using an Agilent 2100 Bioanalyzer RNA 6000 Pico Kit (Agilent Technologies, Santa Clara, CA).

[0078] 2-3. Sequencing

[0078] The sequence library was constructed using Total RNA, modulated using the NEBNext Ultra II Directional RNA Library Prep Kit for Illumina (New England Biolabs, MA) and the NEBNext rRNA Depletion Kit (Human / Mouse / Rat) (New England Biolabs, MA). The quality of the sequence library was confirmed using an Agilent 2200 TapeStation High Sensitivity D1000 (Agilent Technologies, Santa Clara, CA), and the concentration was determined by qPCR using KOD SYBR qPCR Mix (TOYOBO). The sequences in the library were sequenced using a NovaSeq6000 (Illumina, Inc., San Diego, CA) with 50bp paired end sequencing.

[0079] 3. Metabolome Analysis

[0079] 3-1. Collection of Specimens

[0079] Blood from plasma collected from 489 cases (195 samples from the first half + 294 samples from the second half) was mixed with EDTA-2K. After collection from the blood collection tube, the mixture was immediately inverted and mixed more than 10 times. The mixture was centrifuged at 1200xg for 10 minutes. The supernatant (plasma) was separated and dispensed into dispensing tubes. The dispensing tubes were stored at -20°C or -80°C.

[0080] 3-2. Preparation and determination of the sample

[0080] After extracting metabolites from 50 μl of each plasma sample, two derivative reactions, namely methyl oxime and trimethylsilylation, were performed. The derivative samples were analyzed by metabolome analysis using a gas chromatography-mass spectrometry (GC-QP2010 Ultra) to quantify 105 metabolites, including organic acids, sugars, and amino acids. The specific method was in accordance with the method in the following reference (Reference: Rodriguez-Martinez et al. Anal. Chem., 2017, 89(21), pp 11405-11412).

[0081] 4. Comparison between PD (tumor enlargement) group and non-PD (non-tumor enlargement) group

[0081] PD group and non-PD group are defined as follows in this embodiment.

[0082] ‧PD group: A group of cases in which tumor enlargement is visible on the initial imaging evaluation at the start of treatment.

[0082] ‧non-PD group: a group of cases where the tumor size has been maintained or shrunk upon initiation of treatment and initial imaging evaluation.

[0083] Based on the above definition, the results of fecal metagenomics analysis, blood cell RNA expression analysis, and metabolome analysis were compared and reviewed in the PD group and non-PD group.

[0084] <Result>

[0084] 1. Analysis of fecal overall metagenomics

[0084] 1-1. Complexity of species

[0084] Regarding indicators of diversity, the scoring results using ACE and CHAO1 show high diversity in the PR group (Figure 1). Furthermore, the calculation methods for ACE and CHAO1 are based on the following reference (Chao et al. "Stopping rules and estimation for recapture debugging with unequal failure rates". Biometrika, 1993, 80, 193-201.). From these results, the species diversity in feces can serve as a predictive biomarker for the efficacy of immune checkpoint inhibitors.

[0085] 1-2. Correlation analysis of PD (tumor enlargement) group to non-PD group

[0085] The correlation coefficients were calculated with the overall fecal metagenomics data of 180 samples from the first half of the 196 samples for which clinical data (data on the presence / absence of drug efficacy) were obtained. Based on these values, the PD group and non-PD group were identified as significantly different bacterial species and functional KEGG pathways. (Figures 2 and 3)

[0086] Regarding bacterial species, at the genus level, *Geobacillus*, *Gordonibacter*, and *Sebaldella* were extracted as markers. Regarding functional KEGG pathways, bacterial invasion of epithelial cells, flagellar assembly, and fatty acid degradation were extracted.

[0087] Subsequently, in order to verify the markers obtained in the first half, the fecal metagenomics data of 257 samples obtained from the clinical data (data on the presence or absence of drug effects) in the second half of the 282 samples were analyzed in the same way.

[0088] The results showed that, regarding the functional KEGG pathway, bacterial invasion of epithelial cells reappeared, representing a pathway through which the PD group differed significantly from the non-PD group. (Figures 4 to 6)

[0089] Furthermore, when investigating the functional KEGG pathway in detail, differences were observed in the PPAR signaling pathway and fatty acid-related pathways between the PD group and the non-PD group. (Figures 7 and 8)

[0090] Furthermore, the analysis was performed on the most frequently occurring bacterial species (0.01% or higher).

[0091] The results showed that, in the first half, significant differences were observed between the PD group and the non-PD group in the genera *Eubacterium*, *Actinomyces*, *Odoribacter*, *Gordonibacter*, *Eggerthlla*, and *Veillonella*. In the second half, significant differences were observed between the PD group and the non-PD group in the genera *Odoribacter*, *Veillonella*, *Parabacteroides*, and *Streptococcus*.

[0092] Regarding the common bacterial species in the first and second halves, two species can be identified: Odoribacter and Veillonella. (Figures 9 and 10)

[0093] On the other hand, when analyzing the overall profile of the bacterial species, no differences were observed between the PD group and the non-PD group. (Fig. 11, Fig. 12)

[0094] 2. Analysis of plasma metabolites

[0094] Using GC-MS analysis data from 181 samples out of the first 195 samples, for which clinical data (data on the presence / absence of drug efficacy) were obtained, the correlation coefficients between the PD group and the non-PD group were calculated and analyzed. The results yielded metabolites such as lactic acid, pyruvate, glucose, 2-butyric acid, glyceric acid, and 2-piperidinic acid, as shown in Figure 13, as markers of differential efficacy.

[0095] Notably, various short-chain fatty acids (lactic acid, butyric acid) were identified as markers for distinguishing PD groups from non-PD groups. (Figure 14-1)

[0096] Furthermore, in both the first and second half of the sample groups, the results of the analysis of the metabolite markers of the identifiable PD group against the non-PD group showed that lactic acid, pyruvic acid, glucose, glyceric acid, caprylic acid, and citrulline were identified (Figure 14-2, Figure 14-3).

[0097] The identified metabolites that differ between the PD group and the non-PD group, and the results of the aforementioned fecal overall metagenomics analysis (the pathways that differ between the PD group and the non-PD group), are represented by a KEGG pathway diagram. If their relationship is investigated, it can be seen that there is a correlation between some pathways and metabolites. (Figure 15)

[0098] This allows us to assess the functional metabolic pathway data obtained in the intestine, which reflects the metabolites in the blood.

[0099] 3. Whole blood gene expression analysis (RNA-Seq)

[0099] Of the first 174 samples, for 161 samples for which clinical data (data on the presence / absence of drug efficacy) was obtained, the gene expression profile in whole blood was measured, the correlation coefficient between the PD group and the non-PD group was calculated, and the results were analyzed. The genes shown in Figure 16 were identified as markers.

[0100] Furthermore, by examining gene expression data and using functional annotation (GO-Term) data, we can investigate whether gene groups showing differences between the PD and non-PD groups exhibit bias towards specific GO-Terms. This reveals that pathways such as the MAPK pathway, TCR signaling pathway, NF-κB, and cytokine signaling (e.g., the Type I IFN receptor complex) show differences in expression levels between the PD and non-PD groups. (Figures 17-1 to 17-7)

[0101] 4. Predicting drug effects using marker combinations

[0101] Regarding the several effect prediction biomarkers identified to date, we examine whether combining multiple markers can improve the accuracy of effect prediction.

[0102] First, by combining species, using two genera, *Odoribacter* and *Veliionella*, the frequency data (expression values) were normalized and scored separately. The results of comparing the PD group with the non-PD group were obtained, with AUC ranging from 0.571 to 0.584 and p-values ​​ranging from 0.020 to 0.036. (Figure 18)

[0103] Furthermore, by scoring using combinations of four genera (Gordonibacter, Geobacillus, Odoribacter, and Veillonella), the AUC for predicting drug efficacy ranged from 0.571 to 0.634, with p-values ​​ranging from 0.001 to 0.036. The maximum correct diagnostic rate for predicting efficacy was 60%, while the maximum correct diagnostic rate for predicting no efficacy was 70%. (Figure 19)

[0104] Subsequently, in the combination of metabolites in the first half of the sample, the results of scoring using three metabolites—pyruvic acid, pipecolinic acid, and glyceric acid—showed an AUC of 0.708. The correct diagnostic rate for predicting efficacy was 51%, while the correct diagnostic rate for predicting no efficacy was 84%. (Figure 20-1)

[0105] Even in the combination of metabolites in the latter half of the sample, the validation results showed that the prediction of efficacy was 53% with a correct diagnostic rate of 53%; the prediction of inefficiency was 62% with a correct diagnostic rate of 62%. (Figure 20-2)

[0106] Scoring was performed using three metabolites: pyruvic acid, glyceric acid, and lactic acid. In the first half of the sample, the correct diagnostic rate for predicting efficacy was 51%, and for predicting no efficacy, it was 73%. In the second half of the sample, the correct diagnostic rate for predicting efficacy was 60%, and for predicting no efficacy, it was 60%. (Figures 20-3 and 20-4)

[0107] In the combination of blood gene expression markers (first half of the sample), the expression levels of 18 genes (including genes from the TCR signaling pathway, MAPK signaling pathway, and IF type I signaling pathway: CD247, STAT1, CCR5, CBLB, FYB, PIK3R1, RIPK1, STK38, FYN, GATA3, CD3G, CD4, AEBP2, EIF4B, MAPKAPK5, PTPN11, MAP3K14, BID) were used for scoring. The results showed an AUC of 0.598. The correct diagnostic rate for those with predictive power was 43%, and the correct diagnostic rate for those without predictive power was 73%. (Figure 21-1)

[0108] Similarly, for the second half of the sample, the results of scoring using the expression levels of the aforementioned 18 genes showed that the correct diagnostic rate for effective predictions was 53%, while the correct diagnostic rate for ineffective predictions was 59%. (Figure 21-2)

[0109] Furthermore, the species, pathway, and plasma metabolites of feces are combined to form several combinations, which are then scored to predict the effects.

[0110] The biomarkers used in the combination are the following 15.

[0110] Bacterial invasion of epithelial cells, fatty acid degradation, flagellar assembly, fatty acid biosynthesis, PPAR signaling pathway, Geobacillus, Gordonibacter, Odoribacter, Veillonella, lactic acid, pyruvate, glucose, 2-sideoxybutyric acid, glyceric acid, 2-piperidinecarboxylic acid.

[0111] The results of scoring combinations of 4 to 15 markers were as follows: AUC ranged from 0.631 to 0.682, p-values ​​ranged from 0.00003 to 0.0025, with the highest correct diagnostic rate of 67% for predictive effectiveness and 74% for predictive ineffectiveness. (Figures 22 to 25)

[0112] Thus, it is believed that combining multiple biomarkers to predict effects can provide a more stable and robust prediction.

[0113] [Example 2]

[0113] Analysis of overall survival (OS) and progression-free survival (PFS) using fecal total genomics analysis, blood cell RNA expression analysis (RNA-Seq), and metabolome analysis.

[0114] Using the same sample as in Example 1, fecal metagenomics analysis, blood cell RNA expression analysis (RNA-Seq) and metabolome analysis were performed to obtain data for overall survival (OS) and progression-free survival (PFS) analysis.

[0115] In the OS analysis of this embodiment, the survivor group and the dead group are defined as follows.

[0116] ‧Survivor Group: The group confirmed to be alive during the observation period.

[0116] ‧Death Cluster: Clusters of confirmed deaths during the observation period.

[0117] Furthermore, in the PFS analysis of this embodiment, the deterioration group and the non-deterioration group refer to the following definitions.

[0118] ‧Deteriorating Group: A group that was confirmed to have deteriorated during the observation period.

[0118] ‧No deterioration group: The group that was confirmed to have no deterioration during the observation period.

[0119] Based on the above definition, the results of fecal overall metagenomics analysis, blood cell RNA expression analysis, and metabolome analysis were compared and reviewed by the surviving population and the dead population, as well as the population with and without deterioration.

[0120] The markers showing significant differences in expression levels between the two groups are shown in Figures 27 and 28, based on OS analysis of the overall metagenomics of the surviving and dead populations and PFS analysis of the deteriorating and non-deteriorating populations.

[0121] In both OS and PFS analyses, even within the functional KEGG pathway, those with significantly higher performance scores in the nucleotide metabolism pathway showed longer survival and no cancer progression, indicating higher efficacy of immune checkpoint inhibitor therapy. (Figures 29 to 32)

[0122] Furthermore, using the CHAO1 index, which displays diversity, the results of analyzing the bacterial diversity in feces of the survivors and the dead (OS analysis) and the worsening and non-worsening (PFS analysis) groups were clearly shown. It is evident that those with high species diversity have longer survival periods and no cancer progression, that is, they have higher therapeutic effects from immune checkpoint inhibitors. (Figures 33 to 36)

[0123] Furthermore, a comparison of the results of the survivor and death groups (OS analysis) clearly shows that even among the functional KEGG pathways, those with higher performance scores in the peptidoglycan biosynthesis pathway have longer survival periods, meaning that the treatment effect of immune checkpoint inhibitors is high. (Figures 37 to 38)

[0124] Furthermore, the results of comparing the levels of metabolites in the plasma of the survivors and the deceased (OS analysis) clearly show that those with higher performance scores for 2-hydroxybutyric acid and 2-oxobutyric acid in plasma have longer survival periods, that is, they are more effective in treating with immune checkpoint inhibitors. (Figures 39 to 42)

[0125] After normalizing the RNA expression analysis (RNA-Seq) data of blood cells, inter-group comparisons were performed on the data of the survival and death populations to extract candidate marker genes that reappeared in the first and second halves of the sample. The results of exploring marker genes with a median difference greater than 0.5 and a t-test p-value less than 0.01 were obtained, resulting in the extraction of 7 genes (BCL11B, ERCC3, ERCC6, FCRL1, FCRL3, MS4A1, and TCF7) as shown in Figure 43. The sum of the expression scores of these 7 genes, compared with the survival and death populations, showed a significantly higher expression in the survival population. (Figures 44 and 45)

[0126] The PFS analysis comparing the worsening and non-worsening groups clearly shows that even in the functional KEGG pathway, those with low performance scores in the phenylalanine metabolism pathway do not experience cancer progression, meaning that treatment with immune checkpoint inhibitors is highly effective. (Figures 46 to 47)

[0127] The analysis of PFS between the worsening and non-worsening groups clearly shows that those with lower plasma glyceric acid levels do not experience cancer progression, meaning that treatment with immune checkpoint inhibitors is more effective. (Figures 48-49)

[0128] [Example 3]

[0128] Analysis of harmful phenomena through fecal metagenomics

[0128] Using the same samples as in Example 1, fecal metagenomics analysis was performed. Inter-population comparisons were conducted between the 56 samples exhibiting skin lesions and the remaining samples, and markers recurring in both the first and second halves of the sample were extracted. Markers with a median difference greater than 0.01 and a t-test p-value less than 0.05 were explored.

[0129] In the analysis of harmful phenomena in this embodiment, the group with harmful phenomena and the group without harmful phenomena are defined as follows.

[0130] ‧Group with harmful phenomena: Groups that exhibit any of the following harmful phenomena during the observation period: acne-like rash, dry skin, or pruritus.

[0130] ‧Group without harmful phenomena: Groups in which no harmful phenomena such as acne-like rashes, dry skin, or pruritus were observed during the observation period.

[0131] The results showed that, in terms of bacterial species, *Arthrobacter* genus, in terms of the functional KEGG pathway, fatty acid metabolism can serve as a marker for predicting the occurrence of harmful skin conditions. (Figure 50)

[0132] [Example 4]

[0132] Analysis of harmful phenomena through fecal metagenomics

[0132] Using the same samples as in Example 1, genomic single nucleotide polymorphism (SNP) array analysis was performed on whole blood samples. Genomic DNA was extracted from the first 175 samples and the second 265 samples, and tagged. For these, SNP analysis was performed using a microarray (Illumina: Infinium Asian Screening Array BeadChip (up to 600,000 tags)). Genome Studio Software was used for data analysis.

[0133] An inter-group comparison was conducted between the group with harmful phenomena (50 samples) and the group without harmful phenomena (390 samples) classified in the same way as in Example 3 above, and SNPs targeting immune-related genes were extracted. (Fig. 51) Among these markers, SNPs present in the exon portion of coated proteins were identified as being associated with rs2228145.1 on the IL6R gene, and rs7190199 and rs7185320 on the NLRC5 gene.

Claims

1. A method for predicting the response of a subject with gastric cancer to an immune checkpoint inhibitor, comprising: (1) determining the values ​​of (i) to (iii) as follows: (i) the amount of microorganisms present in the subject's feces or intestinal contents; (ii) a score of genomic pathways in the subject's feces or intestinal contents; and (iii) the amount of metabolites present in the subject's blood, serum, or plasma; (2) using the value obtained in step (1) as an indicator to predict the subject's response to an immune checkpoint inhibitor; wherein the microorganisms are selected from one or more species of microorganisms in the group consisting of the genera *Geobacillus*, *Gordonibacter*, *Odoribacter*, *Veillonella*, and *Arthrobacter*. When the presence of *Geobacillus*, *Gordonibacter*, and / or *Veillonella* species is higher than a set threshold, and / or the presence of *Odoribacter* species is lower than a set threshold, the subject is predicted to respond well to immune checkpoint inhibitors. When the presence of *Geobacillus*, *Gordonibacter*, and / or *Veillonella* species is lower than a set threshold, and / or the presence of *Odoribacter* species is higher than a set threshold, the subject is predicted to respond poorly to immune checkpoint inhibitors. When the presence of *Arthrobacter* species is higher than a set threshold, it is predicted that the subject will experience harmful skin reactions induced by immune checkpoint inhibitors.This genomic pathway is selected from one or more of the following genomic pathways within the group comprising bacterial invasion of epithelial cells, fatty acid degradation, flagellar assembly, fatty acid biosynthesis, PPAR signaling pathway, peptidoglycan biosynthesis, nucleotide metabolism, phenylalanine metabolism, and fatty acid metabolism; the pathway is defined as follows: when the scores for bacterial invasion of epithelial cells, fatty acid degradation, flagellar assembly, PPAR signaling pathway, and / or phenylalanine metabolism are lower than a set threshold; and / or when the scores for fatty acid biosynthesis, nucleotide metabolism, etc., are lower than a set threshold. If the scores for bacterial invasion of epithelial cells, fatty acid degradation, flagellar assembly, PPAR signaling pathway, and / or phenylalanine metabolism are higher than the set thresholds, and / or if the scores for fatty acid biosynthesis, nucleotide metabolism, and / or peptidoglycan biosynthesis are lower than the set thresholds, the subject is predicted to have a poor response to immune checkpoint inhibitors.If the fatty acid metabolism score is higher than a set threshold, it is predicted that the subject will experience harmful skin reactions induced by immune checkpoint inhibitors; the metabolite is selected from one or more metabolites of the group consisting of lactic acid, pyruvate, glucose, 2-butyric acid, glyceric acid, caprylic acid, citrulline, 2-hydroxybutyric acid, and 2-piperidinecarboxylic acid; if the presence of lactic acid, pyruvate, glucose, glyceric acid, caprylic acid, 2-hydroxybutyric acid, and / or 2-butyric acid is lower than a set threshold, and / or the presence of 2-piperidinecarboxylic acid and / or citrulline is higher than a set threshold, it is predicted that the subject will respond well to immune checkpoint inhibitors; If the levels of lactic acid, pyruvate, glucose, glyceric acid, caprylic acid, 2-hydroxybutyric acid, and / or 2-sideoxybutyric acid are higher than a set threshold, and / or if the levels of 2-piperidinecarboxylic acid and / or citrulline are lower than a set threshold, the subject is predicted to have a poor response to the immune checkpoint inhibitor; the immune checkpoint inhibitor is an anti-PD-1 antibody.

2. The method as claimed in claim 1, wherein step (1) further comprises determining (iv) the expression level of a gene in the blood, serum, or plasma of the subject; and / or (v) the presence or absence of a single nucleotide polymorphism (SNP) of one or more genes selected from the group consisting of IL6R and NLRC5 in the blood, serum, or plasma of the subject; the gene being selected from genes associated with the MAPK pathway, the Type I IFN receptor complex pathway, and the TCR signaling pathway, and one or more genes selected from the group consisting of BCL11B, ERCC3, ERCC6, FCRL1, FCRL3, MS4A1, and TCF7; the MAPK pathway-related gene being selected from one or more genes selected from MAPKAPK5, MAPKAPK5-AS1, MAP3K14, MAPK3K7, MAP3K1, MAPK9, MAP3K5, and MAPK14; The genes associated with the type I IFN receptor complex pathway are selected from one or more of the following genes: PIK3R1, PTPN11, STAT1, FYN, EIF4B, RAC1, MAP3K1, RPS6KB1, PDCD4, REL, MAPK14, RPS6KA5, STAT2, CRK, MAP2K6, and RAPGEF1; the genes associated with the TCR signaling pathway are selected from one or more of the following genes: MAP3K14, CD4, FYN, DLG1, PDK1, NFKB1, MALT1, LAT, SOS1, MAP3K7, IL10, GRAP2, CD40LG, PIK3CA, PRKCQ, NFATC2, and CALM1; the SNP is selected from the following rs number: rs2228145.1; one or more SNPs in the group consisting of rs7190199; and rs7185320; If the expression level of a gene related to one or more pathways selected from the group consisting of genes related to the MAPK pathway, the Type I IFN receptor complex pathway, and the TCR signaling pathway is higher than a set threshold, or if the expression level of one or more genes selected from the group consisting of BCL11B, ERCC3, ERCC6, FCRL1, FCRL3, MS4A1, and TCF7 is higher than a set threshold, it is predicted that the subject responds well to immune checkpoint inhibitors; If the expression level of a gene selected from the group consisting of genes related to the MAPK pathway, the Type I IFN receptor complex pathway, and the TCR signaling pathway is higher than a set threshold, it is predicted that the subject responds well to immune checkpoint inhibitors; If the expression levels of one or more genes related to the complex pathway and the TCR signaling pathway are lower than a set threshold, or if the expression levels of one or more genes selected from the group consisting of BCL11B, ERCC3, ERCC6, FCRL1, FCRL3, MS4A1, and TCF7 are lower than a set threshold, it is predicted that the subject will have a poor response to immune checkpoint inhibitors. In the presence of this SNP, it is predicted that the subject will experience harmful skin phenomena induced by immune checkpoint inhibitors.

3. The method as described in claim 1, wherein the presence of a *Veillonella* species is higher than a set threshold, and / or the presence of an *Odoribacter* species is lower than a set threshold, is predicted to indicate that the subject is responding well to an immune checkpoint inhibitor.

4. The method as described in claim 1, wherein a low level of *Veillonella* microorganisms and / or a high level of *Odoribacter* microorganisms are predicted to indicate poor response to immune checkpoint inhibitors in the subject.

5. The method as described in any one of claims 1 to 4, wherein the quantity of the microorganism is determined by metagenomic analysis.

6. The method as described in claim 1, wherein if the scores for bacterial invasion of epithelial cells and / or phenylalanine metabolism are lower than a set threshold, and / or if the scores for nucleotide metabolism and / or peptidoglycan biosynthesis are higher than a set threshold, the subject is predicted to respond well to an immune checkpoint inhibitor.

7. The method as described in claim 1, wherein if the scores for bacterial invasion of epithelial cells and / or phenylalanine metabolism are higher than a set threshold, and / or if the scores for nucleotide metabolism and / or peptidoglycan biosynthesis are lower than a set threshold, the subject is predicted to have a poor response to an immune checkpoint inhibitor.

8. The method as described in any of claims 1, 2, 6 and 7, wherein the score of the genomic pathway is determined by metagenomic analysis.

9. The method as described in claim 1, wherein the amount of the metabolite is determined by metabolome analysis.

10. The method as described in claim 2, wherein the expression level of the gene is determined by transcriptome analysis.

Images