WO2021249355A1 - 用于选择化疗响应患者组的生物标志物和方法及其用途 - Google Patents

用于选择化疗响应患者组的生物标志物和方法及其用途 Download PDF

Info

Publication number
WO2021249355A1
WO2021249355A1 PCT/CN2021/098716 CN2021098716W WO2021249355A1 WO 2021249355 A1 WO2021249355 A1 WO 2021249355A1 CN 2021098716 W CN2021098716 W CN 2021098716W WO 2021249355 A1 WO2021249355 A1 WO 2021249355A1
Authority
WO
WIPO (PCT)
Prior art keywords
folfox
marker
responders
markers
combination
Prior art date
Application number
PCT/CN2021/098716
Other languages
English (en)
French (fr)
Inventor
孙恬
Original Assignee
碳逻辑生物科技(香港)有限公司
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 碳逻辑生物科技(香港)有限公司 filed Critical 碳逻辑生物科技(香港)有限公司
Publication of WO2021249355A1 publication Critical patent/WO2021249355A1/zh

Links

Images

Classifications

    • C—CHEMISTRY; METALLURGY
    • C12—BIOCHEMISTRY; BEER; SPIRITS; WINE; VINEGAR; MICROBIOLOGY; ENZYMOLOGY; MUTATION OR GENETIC ENGINEERING
    • C12Q—MEASURING OR TESTING PROCESSES INVOLVING ENZYMES, NUCLEIC ACIDS OR MICROORGANISMS; COMPOSITIONS OR TEST PAPERS THEREFOR; PROCESSES OF PREPARING SUCH COMPOSITIONS; CONDITION-RESPONSIVE CONTROL IN MICROBIOLOGICAL OR ENZYMOLOGICAL PROCESSES
    • C12Q1/00—Measuring or testing processes involving enzymes, nucleic acids or microorganisms; Compositions therefor; Processes of preparing such compositions
    • C12Q1/68—Measuring or testing processes involving enzymes, nucleic acids or microorganisms; Compositions therefor; Processes of preparing such compositions involving nucleic acids
    • C12Q1/6876—Nucleic acid products used in the analysis of nucleic acids, e.g. primers or probes
    • C12Q1/6883—Nucleic acid products used in the analysis of nucleic acids, e.g. primers or probes for diseases caused by alterations of genetic material
    • C12Q1/6886—Nucleic acid products used in the analysis of nucleic acids, e.g. primers or probes for diseases caused by alterations of genetic material for cancer
    • C—CHEMISTRY; METALLURGY
    • C12—BIOCHEMISTRY; BEER; SPIRITS; WINE; VINEGAR; MICROBIOLOGY; ENZYMOLOGY; MUTATION OR GENETIC ENGINEERING
    • C12Q—MEASURING OR TESTING PROCESSES INVOLVING ENZYMES, NUCLEIC ACIDS OR MICROORGANISMS; COMPOSITIONS OR TEST PAPERS THEREFOR; PROCESSES OF PREPARING SUCH COMPOSITIONS; CONDITION-RESPONSIVE CONTROL IN MICROBIOLOGICAL OR ENZYMOLOGICAL PROCESSES
    • C12Q1/00—Measuring or testing processes involving enzymes, nucleic acids or microorganisms; Compositions therefor; Processes of preparing such compositions
    • C12Q1/68—Measuring or testing processes involving enzymes, nucleic acids or microorganisms; Compositions therefor; Processes of preparing such compositions involving nucleic acids
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
    • G01N33/00—Investigating or analysing materials by specific methods not covered by groups G01N1/00 - G01N31/00
    • G01N33/48—Biological material, e.g. blood, urine; Haemocytometers
    • G01N33/50—Chemical analysis of biological material, e.g. blood, urine; Testing involving biospecific ligand binding methods; Immunological testing
    • G01N33/53—Immunoassay; Biospecific binding assay; Materials therefor
    • G01N33/574—Immunoassay; Biospecific binding assay; Materials therefor for cancer
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
    • G01N33/00—Investigating or analysing materials by specific methods not covered by groups G01N1/00 - G01N31/00
    • G01N33/48—Biological material, e.g. blood, urine; Haemocytometers
    • G01N33/50—Chemical analysis of biological material, e.g. blood, urine; Testing involving biospecific ligand binding methods; Immunological testing
    • G01N33/53—Immunoassay; Biospecific binding assay; Materials therefor
    • G01N33/574—Immunoassay; Biospecific binding assay; Materials therefor for cancer
    • G01N33/57407—Specifically defined cancers
    • G01N33/57419—Specifically defined cancers of colon
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
    • G01N33/00—Investigating or analysing materials by specific methods not covered by groups G01N1/00 - G01N31/00
    • G01N33/48—Biological material, e.g. blood, urine; Haemocytometers
    • G01N33/50—Chemical analysis of biological material, e.g. blood, urine; Testing involving biospecific ligand binding methods; Immunological testing
    • G01N33/53—Immunoassay; Biospecific binding assay; Materials therefor
    • G01N33/574—Immunoassay; Biospecific binding assay; Materials therefor for cancer
    • G01N33/57484—Immunoassay; Biospecific binding assay; Materials therefor for cancer involving compounds serving as markers for tumor, cancer, neoplasia, e.g. cellular determinants, receptors, heat shock/stress proteins, A-protein, oligosaccharides, metabolites
    • G—PHYSICS
    • G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
    • G16B—BIOINFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR GENETIC OR PROTEIN-RELATED DATA PROCESSING IN COMPUTATIONAL MOLECULAR BIOLOGY
    • G16B20/00—ICT specially adapted for functional genomics or proteomics, e.g. genotype-phenotype associations
    • G—PHYSICS
    • G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
    • G16B—BIOINFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR GENETIC OR PROTEIN-RELATED DATA PROCESSING IN COMPUTATIONAL MOLECULAR BIOLOGY
    • G16B40/00—ICT specially adapted for biostatistics; ICT specially adapted for bioinformatics-related machine learning or data mining, e.g. knowledge discovery or pattern finding
    • C—CHEMISTRY; METALLURGY
    • C12—BIOCHEMISTRY; BEER; SPIRITS; WINE; VINEGAR; MICROBIOLOGY; ENZYMOLOGY; MUTATION OR GENETIC ENGINEERING
    • C12Q—MEASURING OR TESTING PROCESSES INVOLVING ENZYMES, NUCLEIC ACIDS OR MICROORGANISMS; COMPOSITIONS OR TEST PAPERS THEREFOR; PROCESSES OF PREPARING SUCH COMPOSITIONS; CONDITION-RESPONSIVE CONTROL IN MICROBIOLOGICAL OR ENZYMOLOGICAL PROCESSES
    • C12Q2600/00—Oligonucleotides characterized by their use
    • C12Q2600/106—Pharmacogenomics, i.e. genetic variability in individual responses to drugs and drug metabolism
    • C—CHEMISTRY; METALLURGY
    • C12—BIOCHEMISTRY; BEER; SPIRITS; WINE; VINEGAR; MICROBIOLOGY; ENZYMOLOGY; MUTATION OR GENETIC ENGINEERING
    • C12Q—MEASURING OR TESTING PROCESSES INVOLVING ENZYMES, NUCLEIC ACIDS OR MICROORGANISMS; COMPOSITIONS OR TEST PAPERS THEREFOR; PROCESSES OF PREPARING SUCH COMPOSITIONS; CONDITION-RESPONSIVE CONTROL IN MICROBIOLOGICAL OR ENZYMOLOGICAL PROCESSES
    • C12Q2600/00—Oligonucleotides characterized by their use
    • C12Q2600/158—Expression markers

Definitions

  • the present invention relates to the field of medicine and diagnosis, and in particular to biomarkers and methods for selecting groups of patients responding to chemotherapy and their uses.
  • FOLFOX is a combination of chemotherapeutic drugs containing leucovorin, 5-FU and oxaliplatin, and FOLFOX is widely used to treat colorectal cancer (de Gramont et al., 2000).
  • the objective response to FOLFOX observed in metastatic colorectal cancer is about 30-50% (Tsuji et al., 2012).
  • Selecting colorectal cancer patients who may benefit from or resist FOLFOX treatment will allow the most effective treatment to be selected at the beginning and avoid unnecessary side effects to the patient.
  • the further classification of non-responders to FOLFOX can prompt the selection of appropriate drugs for different tumor characteristics of patients who do not respond to FOLFOX. Therefore, it is important to develop a method for stratifying FOLFOX responders and FOLFOX non-responders.
  • the purpose of the present invention is to provide a specific marker and detection method for early judgment of FOLFOX responsiveness with high sensitivity and high specificity.
  • a use of a gene, mRNA, cDNA, protein, or detection reagent of a FOLFOX responsive marker for preparing a diagnostic reagent or kit said diagnostic reagent or kit Used to (a) judge the responsiveness of a subject to FOLFOX therapy, and/or (b) evaluate the therapeutic effect of a subject using FOLFOX to treat colorectal cancer; and/or classify non-responders to FOLFOX chemotherapy as IML1 or IML2 type;
  • the FOLFOX responsive marker is selected from the following group:
  • A Any marker selected from A1 to A74, or a combination thereof: (A1) LEPR; (A2) NAIP; (A3) ZBTB37; (A4) GPATCH2L; (A5) ZNF224; (A6) CLASP2; (A7 ) U2SURP; (A8) CTDSPL2; (A9) ARID4A; (A10) TM2D1; (A11) MYSM1; (A12) AGO3; (A13) TRPM7; (A14) CDK16; (A15) ALS2; (A16) YIPF2; (A17 ) SPAG9; (A18) DDX5; (A19) PRR7; (A20) IFT80; (A21) PEX12; (A22) MLF2; (A23) RUNDC1; (A24) RNF111; (A25) PIKFYVE; (A26) CCNT2; (A27 ) FAM149B1; (A28) C2orf49
  • B Any marker selected from B1 to B74, or a combination thereof: (B1) MXRA5; (B2) GAS1; (B3) SFRP2; (B4) PRRX1; (B5) CCL8; (B6) TWIST1; (B7 ) INHBA; (B8) COL15A1; (B9) F13A1; (B10) CTHRC1; (B11) SULF1; (B12) COLEC12; (B13) ZNF423; (B14) PLAU; (B15) ZCCHC24; (B16) NDN; (B17 ) COL6A2; (B18) ANTXR1; (B19) RAB31; (B20) DDR2; (B21) MLLT11; (B22) NOX4; (B23) TGFB1I1; (B24) MEIS1; (B25) FLNA; (B26) EFEMP2; (B27 )LAYN; (B28) MRC2; (B
  • the FOLFOX responsive marker includes:
  • the A1 to A74 markers are selected from Table A:
  • the B1 to B74 markers are selected from Table B:
  • the detection reagent includes:
  • a primer or primer pair, probe or chip (such as a nucleic acid chip or a protein chip) that specifically amplifies the mRNA or cDNA of the FOLFOX responsive marker.
  • the diagnosis includes early diagnosis, auxiliary diagnosis, or a combination thereof.
  • the gene, mRNA, cDNA, or protein of any one of the markers shown in Table A and/or Table B of the FOLFOX responsive marker is derived from humans.
  • the subject is a human.
  • the subject is a tumor patient.
  • the tumor patient includes a colon cancer patient, especially a colon cancer patient.
  • the patients include colon cancer patients at stage I, stage II, stage III, and stage IV.
  • the gene, mRNA, cDNA, or protein of the FOLFOX responsive marker is derived from human.
  • the detection is a detection of an isolated sample.
  • the in vitro sample includes: a blood sample, a serum sample, a tissue sample, a body fluid sample, or a combination thereof.
  • the sample is a mononuclear cell sample isolated from peripheral blood.
  • the detection is to detect the expression level of any gene shown in Table A and/or Table B in peripheral blood mononuclear cells.
  • the detection reagent is coupled or has a detectable label.
  • the detectable label is selected from the following group: chromophore, chemiluminescent group, fluorophore, isotope or enzyme.
  • the antibody is a monoclonal antibody or a polyclonal antibody.
  • the diagnostic reagents include antibodies, primers, probes, sequencing libraries, nucleic acid chips (such as DNA chips) or protein chips.
  • the nucleic acid chip includes a substrate and a specific oligonucleotide probe spotted on the substrate, and the specific oligonucleotide probe includes any of the FOLFOX A probe that specifically binds to the polynucleotide (mRNA or cDNA) of the responsive marker.
  • the protein chip includes a substrate and a specific antibody spotted on the substrate, and the specific antibody includes a specific antibody against the FOLFOX responsive marker.
  • the antibody is a monoclonal antibody or a polyclonal antibody.
  • the use includes predicting the effect (prognosis) of FOLFOX treatment in patients with bowel cancer.
  • the colorectal cancer patient is suitable for FOLFOX treatment, and/or the effect of FOLFOX treatment is good:
  • the marker includes any marker from A1 to A20 or a combination thereof.
  • the marker includes a combination of at least 2 or at least 3 markers from A1 to A20.
  • the markers also include any markers from A21 to A48 or a combination thereof.
  • the markers also include any marker from A49 to A65 or a combination thereof.
  • the markers also include any marker from A66 to A74 or a combination thereof.
  • the marker includes a combination of n markers from A1 to A65, where n is any positive integer from 3 to 65 (ie 3, 4, 5, 6, 7 , 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32 , 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56, 57 , 58, 59, 60, 61, 62, 63, 64, 65).
  • n is any positive integer from 3 to 65 (ie 3, 4, 5, 6, 7 , 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32 , 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56, 57 , 58, 59, 60, 61, 62,
  • the marker includes a combination of n markers from A1 to A74, where n is any positive integer from 3 to 74 (ie 3, 4, 5, 6, 7 , 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32 , 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56, 57 , 58, 59, 60, 61, 62, 63, 64, 65, 66, 67, 68, 69, 70, 71, 72, 73, 74).
  • n is any positive integer from 3 to 74 (ie 3, 4, 5, 6, 7 , 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32 , 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52,
  • the markers include any one of B1 to B6 or a combination thereof.
  • the marker includes a combination of at least two or at least three markers from B1 to B6.
  • the markers also include any marker from B7 to B19 or a combination thereof.
  • the markers also include any marker from B20 to B61 or a combination thereof.
  • the markers also include any one of B62 to B74 or a combination thereof.
  • the markers also include a combination of n markers from B1 to B61, where n is any positive integer from 3 to 61 (ie 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60, 61).
  • n is any positive integer from 3 to 61 (ie 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60, 61).
  • the marker includes a combination of n markers from B1 to B74, wherein n is any positive integer from 3 to 74 (ie 3, 4, 5, 6, 7 , 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32 , 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56, 57 , 58, 59, 60, 61, 62, 63, 64, 65, 66, 67, 68, 69, 70, 71, 72, 73, 74).
  • n is any positive integer from 3 to 74 (ie 3, 4, 5, 6, 7 , 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32 , 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52
  • the reagents are PCR primer pairs.
  • a kit which contains a detection reagent for detecting genes, mRNA, cDNA, protein, or a combination of FOLFOX responsive markers,
  • the FOLFOX responsive marker is selected from the following group:
  • the FOLFOX responsive marker includes:
  • the kit contains genes, mRNA, cDNA and/or protein of FOLFOX responsive markers as reference substances or quality control substances.
  • the kit further includes a label or instructions, and the label or instructions indicate that the kit is used for (a) judging the responsiveness of a subject to FOLFOX therapy, and/or (b ) To evaluate the therapeutic effect of a subject using FOLFOX to treat colorectal cancer.
  • the reagents are PCR primer pairs.
  • test result of the FOLFOX responsive marker of the test subject meets the following conditions, it indicates that the colorectal cancer patient is suitable for FOLFOX treatment, and/or the effect of FOLFOX treatment is good:
  • a detection method which includes the steps:
  • test sample being a blood sample or a tissue sample (such as a bowel cancer tissue excised during an operation);
  • the FOLFOX responsive marker is selected from the following group:
  • test result of the FOLFOX responsive marker of the test subject meets the following conditions, it indicates that the colorectal cancer patient is suitable for FOLFOX treatment, and/or the effect of FOLFOX treatment is good:
  • the detection method further includes: if the detection result indicates that the colorectal cancer patient is not suitable for FOLFOX treatment, and/or the effect of FOLFOX treatment is not good, then further classifying the colorectal cancer patient It is IML1 type or IML2 type.
  • the FOLFOX responsive marker includes:
  • the tissue sample is a tissue sample of intestinal cancer excised during the operation of the patient.
  • the expression amount of the marker gene is the amount of mRNA.
  • the method further includes: performing mRNA extraction on the colorectal cancer tissue sample to obtain the extracted mRNA.
  • the sample is from a test object.
  • the detection object is a human.
  • control reference value C0 is the concentration of the FOLFOX responsive marker in the same sample in the FOLFOX non-responsive normal population.
  • a method for typing FOLFOX responsiveness which includes the steps:
  • test sample from the subject to be tested, and the test sample is selected from the following group: blood sample or tissue sample (such as colon cancer tissue excised during surgery);
  • the bowel cancer patient is a FOLFOX responder suitable for FOLFOX treatment:
  • the FOLFOX responsive marker is selected from the following group:
  • the FOLFOX responsive marker includes:
  • step (c) it further includes: if the test result indicates that the bowel cancer patient is not suitable for FOLFOX treatment, and/or the effect of FOLFOX treatment is not good, then further treating the bowel cancer
  • the patient is classified as IML1 or IML2;
  • the colorectal cancer patient is determined to be a non-responsive IML1 type of FOLFOX chemotherapy:
  • the colorectal cancer patient is determined to be a non-responsive IML2 type of FOLFOX chemotherapy:
  • a gene, mRNA, cDNA, or protein of a FOLFOX responsive marker which is used as a marker for judging FOLFOX responsiveness and/or evaluating the therapeutic effect of FOLFOX, and/ Or used as a marker to classify non-responders to FOLFOX chemotherapy;
  • the FOLFOX responsive marker is selected from the following group:
  • the FOLFOX responsive marker includes:
  • a marker set for typing FOLFOX responsiveness and/or further typing of non-responders to FOLFOX chemotherapy includes:
  • (C) A combination of one or more markers from A1 to A74 and one or more markers from B1 to B74, and the combination includes at least 3 markers.
  • the FOLFOX responsive marker includes:
  • the set of markers includes any marker from A1 to A20 or a combination thereof.
  • the marker set includes a combination of at least 2 or at least 3 markers from A1 to A20.
  • the set of markers further includes any marker from A21 to A48 or a combination thereof.
  • the set of markers further includes any marker from A49 to A65 or a combination thereof.
  • the marker set includes a combination of n markers from A1 to A74, where n is any positive integer from 3 to 74.
  • the set of markers includes any marker from B1 to B6 or a combination thereof.
  • the marker set includes a combination of at least 2 or at least 3 markers from B1 to B6.
  • the set of markers further includes any marker from B7 to B19 or a combination thereof.
  • the set of markers further includes any marker from B20 to B61 or a combination thereof.
  • the marker set includes a combination of n markers from B1 to B74, where n is any positive integer from 3 to 74
  • a method for typing human colorectal cancer comprising:
  • the chemotherapy includes: FOLFOX, 5-FU treatment, oxaliplatin treatment, or chemotherapy targeting DNA replication.
  • the method further includes: using chemotherapy to treat patients who are classified as responders to chemotherapy for colorectal cancer.
  • the at least three markers include any one of A1 to A20 or a combination thereof.
  • the at least three markers include a combination of at least two or at least three markers from A1 to A20.
  • the at least three markers further include any one of A21 to A48 or a combination thereof.
  • the at least three markers also include any marker from A49 to A65 or a combination thereof.
  • the at least three markers include any one of B1 to B6 or a combination thereof.
  • the at least three markers include a combination of at least two or at least three markers from B1 to B6.
  • the at least three markers also include any one of B7 to B19 or a combination thereof.
  • the at least three markers also include any one of B20 to B61 or a combination thereof.
  • a method for typing human colorectal cancer comprising:
  • the chemotherapy includes: FOLFOX, 5-FU treatment, oxaliplatin treatment, or chemotherapy targeting DNA replication.
  • the method further includes: using chemotherapy to treat patients who are classified as responders to chemotherapy for colorectal cancer.
  • the at least three markers are as defined above.
  • the mRNA gene expression level is obtained by a technique selected from the group consisting of microarray, RNAseq, RT-PCR, or a combination thereof.
  • the method further includes performing normalization processing by a method selected from the group consisting of fRMA, RMA, RNAseq CPM, RNAseq FPKM.
  • step (3) the similarity is calculated by a method selected from the following group:
  • a device for typing colorectal cancer including:
  • a data processing unit that processes the input data of mRNA gene expression level, and the data processing unit includes a normalization processing subunit, a similarity calculation subunit, and a similarity difference calculation subunit unit;
  • the normalization processing subunit is used to normalize the gene expression values of at least three marker genes in Table A or Table B;
  • the similarity calculation subunit is used to calculate the first similarity between the at least three marker genes in Table A or Table B and the average gene expression values of these genes in FOLFOX responders; and to calculate Table A or Table The second degree of similarity between the at least three marker genes described in B and the average gene expression value of these genes in FOLFOX non-responders;
  • the similarity difference calculation subunit is used to calculate the difference between the first similarity and the second similarity of each marker gene
  • (P3) Typing unit the typing unit classifies the subject based on the difference between the marker genes, and the typing classifies the test subject as responders or non-responders to chemotherapy , So as to obtain typing results;
  • the classification unit is also used to further classify non-responders to FOLFOX chemotherapy into IML1 type or IML2 type.
  • the subject is a patient with bowel cancer.
  • the chemotherapy includes: FOLFOX, 5-FU therapy, oxaliplatin therapy, or chemotherapy targeting DNA replication.
  • the at least three markers are as defined above.
  • Figure 1 shows the workflow of the iterative machine learning (IML) method. Finding the convergence point of the machine learning process, divides FOLFOX non-responders into subgroups, and then analyzes the FOLFOX resistance mechanism of the FOLFOX non-responders subgroup.
  • IML iterative machine learning
  • the recurrence events of patients in this data set are recorded, and the follow-up time of patients in this data set is estimated to be a uniformly distributed time series within 60 months.
  • Figure 4A shows that any combination of the 3 genes in Table 1A has the predictive value of FOLFOX response.
  • the X axis is the number of genes used, starting with 3 and ending with 50.
  • the Y axis is the average performance of 200 rounds of random combinations using a selected number of genes.
  • the upper solid line the sensitivity of nonresponder; the middle long dotted line: the average overall performance; the lower dotted line: the specificity of nonresponder.
  • Figure 4B shows that any combination of the 3 genes in Table 1B has the predictive value of FOLFOX response.
  • the X axis is the number of genes used, starting with 3 and ending with 50.
  • the Y axis is the average performance of 200 rounds of random combinations using a selected number of genes.
  • the upper solid line the sensitivity of non-responder; the middle long dashed line: the average overall performance; the lower dotted line: the specificity of non-responder.
  • Figure 5 shows a box plot showing the correlation between the expression levels of known markers of FOLFOX resistance (5A.ERCC1, 5B.DPYD, 5C.BRCA1, 5D.SRBC, 5E.TYMP) and IML prediction results sex.
  • the Y axis is the expression value of this marker
  • the green box is the FOLFOX responder predicted by IML
  • the red box is the FOLFOX non-responder predicted by IML gene signature feature 1
  • the yellow box is the predicted FOLFOX non-responder of IML gene signature feature 2.
  • the potential main FOLFOX resistance mechanisms of different resistance subgroups are significantly different.
  • Figure 6 shows the gene-protein interaction network in IML gene signature feature 1, which shows the highly correlated network of cell cycle and mitosis.
  • the present invention provides biomarkers and methods for classifying colorectal cancer patients into chemotherapy responders and non-responders, wherein colorectal cancer patients classified as responders have significantly better survival benefits than non-responders .
  • the present invention has been completed on this basis.
  • FOLFOX is not a single targeted therapy with a known single protein target.
  • the main cytotoxic components of FOLFOX 5-Fu and oxaliplatin have different mechanisms of action.
  • 5-FU is designed to inhibit thymine synthase (which is a key enzyme in DNA synthesis), and the metabolites of 5-FU are similar in structure to nucleotides, which can be incorporated into DNA and cause cell death (Longley et al., 2003).
  • Oxaliplatin forms intra-strand connections between DNA and disrupts DNA replication (Hector et al., 2001).
  • colorectal cancer is a heterogeneous disease, and there are at least four common molecular subtypes (Guinney et al., 2015). Different molecular subtypes have different dependence on DNA replication and cell cycle of colorectal tumor cells.
  • tumors may resist FOLFOX treatment in different ways (Hammond et al., 2016). Therefore, drug resistance may be caused by a combination of many factors, and a typical simple method of sequencing a drug target gene is unlikely to work (Hammond et al, 2016).
  • the inventors developed a novel iterative supervised machine learning (IML) method by using genomic data of patients with stage IV colorectal cancer and objective FOLFOX responsiveness data to solve these complexities.
  • IML iterative supervised machine learning
  • the IML method finds statistical convergence points in the machine learning process, and identifies a subgroup of colorectal cancer patients with the same basic mechanism of FOLFOX resistance, and selects patients who benefit from FOLFOX treatment.
  • the inventors validated this method in two independent validation data sets: one data set is stage IV colorectal cancer patients treated with FOLFOX; the other data set is stage III colorectal cancer patients treated with FOLFOX.
  • sample refers to a material specifically associated with a subject, from which specific information related to the subject can be determined, calculated, or inferred.
  • the sample may be composed wholly or partly of biological material from the subject.
  • the term "expression” includes the production of mRNA from genes or gene parts, and includes the production of proteins encoded by RNA or genes or gene parts, as well as the appearance of detection substances related to expression.
  • cDNA binding of a binding ligand (such as an antibody) to genes or other oligonucleotides, proteins or protein fragments, and the chromogenic part of the binding ligand are all included within the scope of the term "expression”. Therefore, the increase in half-dot density on immunoblotting such as Western blotting is also within the scope of the term "expression” based on biological molecules.
  • the term "reference value” or “control reference value” refers to a value that is statistically related to a specific result when compared to an analysis result.
  • the reference value is determined based on the comparison of mRNA expression and/or protein expression of FOLFOX responsive markers and statistical analysis. Some of these studies are shown in the example section of this document. However, research from the literature and user experience of the methods disclosed herein can also be used to produce or adjust reference values. The reference value can also be determined by considering the conditions and results that are particularly relevant to the patient's medical history, genetics, age, and other factors.
  • FOLFOX responsive marker of the present invention refers to one or more of the markers shown in Table A and/or Table B.
  • FOLFOX responsive marker protein of the present invention refers to having any one or more of the FOLFOX responsive markers of the present invention.
  • FOLFOX responsive marker gene and “FOLFOX responsive marker polynucleotide” are used interchangeably, and both refer to any FOLFOX responsive marker shown in Table A and/or Table B The nucleotide sequence of the thing.
  • a nucleic acid sequence encoding it can be constructed based on it, and specific probes can be designed based on the nucleotide sequence.
  • the full-length nucleotide sequence or its fragments can usually be obtained by PCR amplification, recombination or artificial synthesis.
  • primers can be designed according to the nucleotide sequence of the FOLFOX responsive marker disclosed in the present invention, especially the open reading frame sequence, and use a commercially available cDNA library or according to the routine known to those skilled in the art.
  • the cDNA library prepared by the method was used as a template, and the relevant sequences were obtained by amplification. When the sequence is long, it is often necessary to perform two or more PCR amplifications, and then splice the amplified fragments together in the correct order.
  • the recombination method can be used to obtain the relevant sequence in large quantities. This is usually done by cloning it into a vector, then transferring it into a cell, and then isolating the relevant sequence from the proliferated host cell by conventional methods.
  • artificial synthesis methods can also be used to synthesize related sequences, especially when the fragment length is short. Usually, by first synthesizing multiple small fragments, and then ligating to obtain fragments with very long sequences.
  • the DNA sequence encoding the protein (or fragment or derivative thereof) of the present invention can be obtained completely through chemical synthesis.
  • the DNA sequence can then be introduced into various existing DNA molecules (such as vectors) and cells known in the art.
  • polynucleotide sequence of the present invention can be used to express or produce recombinant FOLFOX responsive markers.
  • antibody of the present invention and “specific antibody against a FOLFOX responsive marker” are used interchangeably.
  • the present invention also includes polyclonal antibodies and monoclonal antibodies, especially monoclonal antibodies, which have specificity to FOLFOX responsive markers (Table A and/or Table B).
  • the present invention not only includes complete monoclonal or polyclonal antibodies, but also includes immunologically active antibody fragments, such as Fab' or (Fab) 2 fragments; antibody heavy chains; antibody light chains; genetically engineered single-chain Fv molecules ( Ladner et al., U.S. Patent No. 4,946,778); or chimeric antibodies, such as antibodies that have murine antibody binding specificity but still retain human-derived antibody portions.
  • immunologically active antibody fragments such as Fab' or (Fab) 2 fragments
  • antibody heavy chains such as antibody heavy chains; antibody light chains; genetically engineered single-chain Fv molecules ( Ladner et al., U.S. Patent No. 4,946,778); or chimeric antibodies, such as antibodies that have murine antibody binding specificity but still retain human-derived antibody portions.
  • the antibody of the present invention can be prepared by various techniques known to those skilled in the art. For example, purified gene products of human FOLFOX responsive markers or antigenic fragments thereof can be administered to animals to induce the production of polyclonal antibodies. Similarly, cells expressing human FOLFOX responsive marker protein or antigenic fragments can be used to immunize animals to produce antibodies.
  • the antibody of the present invention may also be a monoclonal antibody. Such monoclonal antibodies can be prepared using hybridoma technology
  • Antibodies against human FOLFOX responsive marker proteins can be used in immunohistochemistry techniques to detect human FOLFOX responsive marker proteins in specimens (especially tissue samples or blood samples). Since the FOLFOX responsive marker protein exists in blood samples or tissue samples (intestinal cancer tissue excised during surgery), its expression level can be the target of detection.
  • the present invention Based on the differential expression of FOLFOX responsive markers in tissue samples or blood samples, the present invention also provides a corresponding method for judging FOLFOX responsiveness.
  • the invention relates to a diagnostic test method for quantitatively and locally detecting the protein level or mRNA level of human FOLFOX responsive markers. These tests are well known in the art.
  • the human FOLFOX responsive marker protein level or mRNA level detected in the test can be used to judge (including auxiliary judgment) whether it is suitable for FOLFOX treatment or whether it has FOLFOX responsiveness.
  • a preferred method is to perform quantitative detection of mRNA or cDNA by PCR.
  • a preferred method is to quantitatively detect mRNA or cDNA by sequencing.
  • the polynucleotide of FOLFOX responsiveness marker can be used for the diagnosis of FOLFOX responsiveness.
  • a part or all of the polynucleotide of the present invention can be used as probes to be fixed on a microarray or a DNA chip, and used for differential expression analysis and genetic diagnosis of genes in the analysis.
  • Antibodies against FOLFOX responsive markers can be immobilized on a protein chip to detect FOLFOX responsive proteins in samples.
  • FOLFOX responsive markers can be used as markers for determining FOLFOX responsiveness.
  • the present invention also provides a kit for judging FOLFOX responsiveness.
  • the kit contains a detection reagent for detecting genes, mRNA, cDNA, protein, or a combination of FOLFOX responsive markers.
  • the kit contains the antibody or immunoconjugate of the anti-FOLFOX responsive marker of the present invention, or an active fragment thereof; or contains a primer or primer pair that specifically amplifies the mRNA or cDNA of the FOLFOX responsive marker , Probe or chip.
  • the kit further includes a label or instructions.
  • the inventors developed a novel iterative supervised machine learning (IML) method using genomic data of stage III and IV colorectal cancer patients with objective FOLFOX response data.
  • the IML method finds statistical convergence points in the process of machine learning, and identifies subgroups of colorectal cancer patients with the same potential biological mechanism of FOLFOX resistance, and selects patients who benefit from FOLFOX treatment.
  • stage IV colorectal cancer patients treated with FOLFOX showed a sensitivity of 97.6% and a specificity of 100%.
  • the prediction scores show that there is a correlation with the expression levels of known single gene markers including ERCC1, DPYD, BRCA1, CAVIN3 and TYMP.
  • FOLFOX resistance has two different dominant modes, each of which depends on the up-regulation of different types of DNA damage repair proteins.
  • the main mode of FOLFOX resistance may be the synergistic effect of tumor cells' ability to resist apoptosis and change the cell cycle.
  • Some tumor samples from patients with CMS4 mesenchymal subtype also show these patterns and can still benefit from FOLFOX treatment.
  • the predicted score of the IML model reflects the underlying mechanism of FOLFOX resistance, which will help design FOLFOX combination therapies.
  • the gene expression data was analyzed using R/Bioconductor software (Gentleman et al., 2004).
  • the published gene expression data and objective FOLFOX response data of 83 patients with stage IV colorectal cancer (GSE28702) were used as the training set (Tsuji et al., 2012).
  • the gene expression data and survival data of 32 patients with stage IV colorectal cancer (GSE72970) treated with FOLFOX were used as an independent validation set (Del Rio et al., 2017).
  • the gene expression data and survival data of 55 FOLFOX-treated stage III colorectal cancer patients (GSE81653) were used as an independent validation set (Lin et al., 2017).
  • vRMA vRMA
  • fRMA vRMA
  • GSE28702 and GSE72970 Affymetrix Human Genome U133 Plus 2.0 array data
  • This normalization method is designed for clinical diagnostic settings.
  • Each diagnostic sample is Need to be treated separately (McCal et al., 2010).
  • Use ComBat to delete the batch effect of samples in GSE28702 and GSE72970 Johnson et al., 2007.
  • the standardized Affymetrix human gene 2.0ST array data (GSE81653) is downloaded from the Gene Expression Omnibus database.
  • the model was trained using six iterations. In each round, a selected subset of samples in the non-responder group is compared with all tumor samples in the responder group. The first and fourth rounds of learning are used to pre-select non-responders tumors with the same characteristics in the subset, and t-test the non-responders and responder groups used in this round.
  • Two statistical criteria are used to select genes in the features of these two learning rounds: (1) genes with p value ⁇ 0.01 and (2) genes with the absolute value of the difference between the mean of the non-responder group and the responder group need to be high Change of 1.2 times.
  • the second, third, fifth and sixth learning rounds were used to strengthen the previous learning rounds, and 200 rounds of 10-fold cross-validation were performed, and in each cross-validation round, t-test and p-value were performed To be ranked.
  • a statistical criterion is used to select the genes in the characteristics of these four learning rounds: in the 200 rounds of verification, at least 90% of the genes have p-values in the top 250 genes.
  • the gene signatures selected in the first, second, fourth and fifth rounds are pre-screened gene signature features, and they are not used in the final scoring function.
  • the scores of these four pre-screened gene signature features are calculated by using a simple closest centroid model.
  • the machine learning method produces statistical convergence points in the third and sixth iterations of supervised learning rounds. In this case, the average area under the curve for 200 rounds of 10-folder cross-validation is always higher than 0.8.
  • the gene signature features selected in the third and sixth rounds are the final gene signature features.
  • all 42 FOLFOX responder samples and 30 FOLFOX non-responder samples are used for statistical analysis, and 74 genetic signature features IMLSig.1 ⁇ are trained (Table 1A).
  • In the sixth supervised iterative learning round all 42 FOLFOX responder samples and a subset of 13 FOLFOX non-responder samples were used for statistical analysis, and 74 genetic signature features IMLSig 2 ⁇ (Table 1B ).
  • IMLSig 1 ⁇ and IMLSig 2 ⁇ are used to construct two k nearest neighbor regression scoring functions S 1 and S 2 .
  • the two scores are combined into one score S.
  • DAVID Human et al., 2009
  • STRING Szklarczyk et al., 2019
  • the predicted median overall survival time (13.4 months) of the FOLFOX non-responder group was significantly shorter than that of the predicted FOLFOX responder group (36.6 months) ( Figure 2).
  • the inventors used 166 stage III colorectal cancer patients who received FOLFOX as an adjuvant therapy.
  • the inventors only used the consistent molecular subtypes of the CMS4 subtype.
  • Test the response to FOLFOX treatment Record the recurrence of patients in this data set.
  • the follow-up time of patients in this data set is not available and is estimated to be a uniformly distributed time series within 60 months.
  • the IML method can still identify patient subgroups among the CMS4 subtype patients who benefit from FOLFOX treatment.
  • the IML method was specifically developed to predict the FOLFOX response, and it provides additional predictive value.
  • the potential main FOLFOX resistance mechanisms of the two different resistance subgroups are different.
  • the clustering of GO biological process terms for gene enrichment function analysis in the signature feature indicates the process of apoptosis (enrichment score 0.805).
  • the score of FOFOX non-responders predicted by Feature 1 (IML1 type) shows a trend of high ERCC1 and high DPYD, which indicates that most tumor cells in FOLFOX non-responders may have relatively high 5-FU catabolism rates, and It has effective ERCC1 nucleotide excision repair ability, thereby overcoming the apoptosis induced by oxaliplatin (red box, Figure 5A, 5B).
  • the scores of FOLFOX non-responders predicted by Signature Feature 2 are associated with high levels of BRCA1 and low levels of the BRCA1 inactivator CAVIN3, which indicates that tumor cells in genetic signature Feature Type 2 FOLFOX non-responders may be dependent on BRCA1 repair Oxaliplatin induced double-strand break (yellow box, Figure 5C, 5D).
  • the pattern of activated BRCA1 is consistent with the absence of a mesenchymal phenotype, because BRCA1 is known to inhibit epithelial-mesenchymal transition and stem cell dedifferentiation (Bai et al., 2014).
  • the FOLFOX non-responders predicted by feature 2 showed a mild tendency to TYMP (yellow box, Figure 5E), which indicates that the tumor cells of the FOLFOX non-responders of feature 2 were converted from 5-FU to its active metabolite 5- The efficiency of FdUMP may be lower.
  • IML1 and IML2 FOLFOX-resistant patients suggest that these two subtypes may have different sensitivity to other drugs.
  • IML1 and IML2 have different degrees of dependence on BRCA1, and their sensitivity to PARP inhibitor drugs may also be different.
  • Table 1A Using 30 non-responder samples and 42 responder samples, 74 genes in Table 1A were selected (iteration round 3, Figure 1). 200 10-fold cross-validation was performed, and t-test was performed in each cross-validation round, and the p-values were ranked. Use a statistical criterion to select the genes in the features of these four learning rounds: in the 200 rounds of validation, at least 90% of the genes have p-values in the top 250 genes.
  • Table 1A List of genes in IML signature feature 1
  • Table 1B The 74 genes in Table 1B were selected by using 13 samples of non-responders and 42 samples of responders (Iteration 6th round, Figure 1). 200 10-fold cross-validation was performed, and t-test was performed in each cross-validation round, and the p-values were ranked. Use a statistical criterion to select the genes in the features of these four learning rounds: in the 200 rounds of validation, at least 90% of the genes have p-values in the top 250 genes.
  • FOLFOX is a combination of chemotherapy drugs that is used to treat colorectal cancer. About half of the patients benefit from FOLFOX, and half of the patients show resistance.
  • IML iterative machine learning method
  • the inventors provided an IML model that can predict the response to FOLFOX treatment, and validated the model in FOLFOX-treated stage IV patients and FOLFOX-treated stage III patients.
  • the main advantage of the IML model is that the underlying training workflow assumes that there may be multiple causes of FOLFOX resistance in colorectal cancer patients.
  • the inventor's results indicate that there are at least two different main mechanisms of FOLFOX resistance in colorectal cancer. Each mechanism depends on the up-regulation of different types of DNA damage repair proteins.
  • the main mechanism of FOLFOX drug resistance is the synergistic effect of tumor cell resistance to apoptosis and altered cell cycle, accounting for about 75% of non-responders.
  • the second mechanism of FOLFOX resistance is the activation of BRCA1, which accounts for approximately 25% of non-responders.
  • the gene signatures of the IML model show that they are related to known single gene markers (such as ERCC1, DPYD, BRCA1, CAVIN3 and TYMP). However, it should be pointed out that a single gene cannot fully explain FOLFOX resistance, which is caused by multiple factors (Hammond et al, 2016).
  • the epithelial-mesenchymal subtype is thought to tend to resist chemotherapy.
  • the IML method identifies a subgroup of patients who can still benefit from FOLFOX treatment. These results not only indicate that the IML method provides specific additional predictive value for FOLFOX responsiveness, but also that epithelial-mesenchymal transition alone is not sufficient to induce resistance to FOLFOX treatment. It is also necessary to obtain a synergistic effect against apoptosis and up-regulate different types of DNA damage repair proteins. This process may be independent of epithelial-mesenchymal transition, because functional DNA damage repair proteins (such as BRCA1) may inhibit the mesenchymal phenotype.
  • the inventors developed an IML method that can layer FOLFOX responders and non-responders.
  • the IML method has been validated in both stage III and IV colorectal cancer patient groups.
  • the main advantage of the inventor's method is that IML will not treat all patients who are resistant to FOLFOX with the same treatment, but will fairly layer the non-responsive subgroups, and these subgroups have the same resistance synergistic mechanism.
  • the prediction score of the IML method reflects the potential different molecular mechanisms of FOLFOX resistance. Therefore, the prediction score of the IML model also provides an indication of the molecular mechanism of FOLFOX resistance.
  • the tumor characteristics of patients with IML1 and IML2 FOLFOX resistance suggest that these two subtypes may have different sensitivity to other drugs.
  • IML1 and IML2 have different degrees of dependence on BRCA1, and their sensitivity to PARP inhibitor drugs may also be different. According to the classification of the tumor characteristics of patients with FOLFOX resistance, FOLFOX can be used in combination with other drugs to combat different FOLFOX tumor resistance.

Landscapes

  • Health & Medical Sciences (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Engineering & Computer Science (AREA)
  • Chemical & Material Sciences (AREA)
  • Immunology (AREA)
  • Physics & Mathematics (AREA)
  • Molecular Biology (AREA)
  • Proteomics, Peptides & Aminoacids (AREA)
  • Analytical Chemistry (AREA)
  • Biotechnology (AREA)
  • General Health & Medical Sciences (AREA)
  • Organic Chemistry (AREA)
  • Bioinformatics & Cheminformatics (AREA)
  • Biophysics (AREA)
  • Genetics & Genomics (AREA)
  • Pathology (AREA)
  • Microbiology (AREA)
  • Wood Science & Technology (AREA)
  • Zoology (AREA)
  • Biochemistry (AREA)
  • Biomedical Technology (AREA)
  • Hematology (AREA)
  • Urology & Nephrology (AREA)
  • Medical Informatics (AREA)
  • Oncology (AREA)
  • Hospice & Palliative Care (AREA)
  • Evolutionary Biology (AREA)
  • Theoretical Computer Science (AREA)
  • Spectroscopy & Molecular Physics (AREA)
  • Cell Biology (AREA)
  • Bioinformatics & Computational Biology (AREA)
  • General Physics & Mathematics (AREA)
  • Medicinal Chemistry (AREA)
  • Food Science & Technology (AREA)
  • General Engineering & Computer Science (AREA)
  • Artificial Intelligence (AREA)
  • Software Systems (AREA)
  • Public Health (AREA)
  • Evolutionary Computation (AREA)
  • Epidemiology (AREA)

Abstract

本发明提供了用于选择化疗响应患者组的生物标志物和方法及其用途。具体地，本发明提供了FOLFOX响应性标志物(表A和/或表B中所示的任一标志物)的基因、mRNA、cDNA、蛋白或其检测试剂的用途，用于制备判断FOLFOX响应性的诊断试剂或试剂盒。研究表明，FOLFOX响应性标志物可作为判断FOLFOX响应性和评价FOLFOX治疗效果的标记物，具有高灵敏度和特异性。此外，本发明的FOLFOX响应性标志物还可将FOLFOX化疗的无响应者划分为IML1型和IML2型2种亚型，便于对无响应者的用药进行指导。

Description

用于选择化疗响应患者组的生物标志物和方法及其用途 技术领域
本发明涉及医学和诊断领域，具体地涉及用于选择化疗响应患者组的生物标志物和方法及其用途。
背景技术
FOLFOX是包含亚叶酸，5-FU和奥沙利铂的化疗药物的组合，并且FOLFOX被广泛用于治疗结直肠癌(de Gramont et al.,2000)。在转移性结直肠癌中观察到的对FOLFOX的客观响应约为30-50％(Tsuji et al.,2012)。选择可能将受益于或抵抗FOLFOX治疗的结直肠癌患者将允许在开始时选择最有效的治疗方法，并避免对患者产生不必要的副作用。此外，对FOLFOX无响应者进一步分型，可以针对不同FOLFOX无响应患者的不同肿瘤特征提示选用适用药物。因此，开发对FOLFOX响应者和FOLFOX无响应者进行分层的方法很重要。
然而，本领域尚缺乏令人满意的对FOLFOX响应性进行早期的和准确的区分方法，更缺乏对FOLFOX无响应者进一步分型的方法，因此，本领域迫切需要开发可用于高灵敏和高特异性地早期对FOLFOX响应性进行评判的特异标志物和检测方法。
发明内容
本发明的目的就是提供一种高灵敏和高特异性地早期对FOLFOX响应性进行评判的特异标志物和检测方法。
在本发明的第一方面，提供了一种FOLFOX响应性标志物的基因、mRNA、cDNA、蛋白、或其检测试剂的用途，用于制备一诊断试剂或试剂盒，所述诊断试剂或试剂盒用于(a)判断某一对象采用FOLFOX疗法的响应性，和/或(b)评价某一对象采用FOLFOX治疗肠癌的治疗效果；和/或将FOLFOX化疗的无响应者分为IML1型或IML2型；
其中，所述的FOLFOX响应性标志物选自下组：
(A)选自A1至A74的任一标志物、或其组合：(A1)LEPR；(A2)NAIP；(A3)ZBTB37；(A4)GPATCH2L；(A5)ZNF224；(A6)CLASP2；(A7)U2SURP；(A8)CTDSPL2；(A9)ARID4A；(A10)TM2D1；(A11)MYSM1；(A12)AGO3；(A13)TRPM7；(A14)CDK16；(A15)ALS2；(A16)YIPF2；(A17)SPAG9；(A18)DDX5；(A19)PRR7；(A20)IFT80；(A21)PEX12；(A22)MLF2；(A23)RUNDC1；(A24)RNF111；(A25)PIKFYVE；(A26) CCNT2；(A27)FAM149B1；(A28)C2orf49；(A29)CDC27；(A30)NF1；(A31)H2AFX；(A32)FKBP8；(A33)EPS15；(A34)COL28A1；(A35)NFKBIB；(A36)TRIM14；(A37)STX10；(A38)IMPDH1；(A39)GGNBP2；(A40)SCAMP4；(A41)TAOK1；(A42)ADAM15；(A43)FOXN3；(A44)BCLAF1；(A45)NLGN2；(A46)LSM4；(A47)JMJD8；(A48)NHP2；(A49)HMCN2；(A50)CTDSPL；(A51)SRF；(A52)ZC3H7B；(A53)HTATIP2；(A54)PPFIA3；(A55)RNF212；(A56)ITPR3；(A57)KCNC4；(A58)JMJD4；(A59)CNOT9；(A60)PDZD7；(A61)MAP3K10；(A62)ADPRHL1；(A63)SCYL1；(A64)UPP2；(A65)H2BFM；(A66)MAP3K2；(A67)SMURF2；(A68)APPBP2；(A69)NPEPPS；(A70)HELZ；(A71)MBTD1；(A72)AMZ2P1；(A73)CEP290；(A74)WASHC4；
(B)选自B1至B74的任一标志物、或其组合：(B1)MXRA5；(B2)GAS1；(B3)SFRP2；(B4)PRRX1；(B5)CCL8；(B6)TWIST1；(B7)INHBA；(B8)COL15A1；(B9)F13A1；(B10)CTHRC1；(B11)SULF1；(B12)COLEC12；(B13)ZNF423；(B14)PLAU；(B15)ZCCHC24；(B16)NDN；(B17)COL6A2；(B18)ANTXR1；(B19)RAB31；(B20)DDR2；(B21)MLLT11；(B22)NOX4；(B23)TGFB1I1；(B24)MEIS1；(B25)FLNA；(B26)EFEMP2；(B27)LAYN；(B28)MRC2；(B29)EOGT；(B30)SERPINH1；(B31)NAP1L3；(B32)KIAA1462；(B33)CERCAM；(B34)GLI3；(B35)HS3ST3A1；(B36)C3orf80；(B37)CAVIN1；(B38)PCDHB2；(B39)CMTM3；(B40)TSHZ3；(B41)EVC；(B42)MFGE8；(B43)MSRB3；(B44)EIF4B；(B45)RASA3；(B46)EHD2；(B47)TWIST2；(B48)OMA1；(B49)PCDH7；(B50)FBXL7；(B51)CHSY3；(B52)HLX；(B53)ATP10A；(B54)PRR16；(B55)DPH5；(B56)VASN；(B57)ZBTB46；(B58)PDGFRA；(B59)RPLP0；(B60)RPS3A；(B61)RPL9；(B62)SPOCK1；(B63)COL5A2；(B64)COL6A3；(B65)AEBP1；(B66)COL5A1；(B67)COL6A1；(B68)PCOLCE；(B69)BNC2；(B70)WISP1；(B71)CLEC11A；(B72)THY1；(B73)IGFBP5；(B74)PDGFRB；
(C)A1至A74中一个或多个标志物与B1至B74中一个或多个标志物所构成的组合。
在另一优选例中，所述的FOLFOX响应性标志物包括：
(A)选自A1至A65的任一标志物、或其组合；
(B)选自B1至B61的任一标志物、或其组合；
(C)A1至A65中一个或多个标志物与B1至B61中一个或多个标志物所构成的组合。
在另一优选例中，所述的A1至A74标志物选自表A：
表A
Figure PCTCN2021098716-appb-000001
Figure PCTCN2021098716-appb-000002
在另一优选例中，所述的B1至B74标志物选自表B：
表B
Figure PCTCN2021098716-appb-000003
Figure PCTCN2021098716-appb-000004
Figure PCTCN2021098716-appb-000005
在另一优选例中，所述检测试剂包括：
(a)针对所述FOLFOX响应性标志物的特异性抗体、特异性结合分子；和/或
(b)特异性扩增所述FOLFOX响应性标志物的mRNA或cDNA的引物或引物对、探针或芯片(如核酸芯片或蛋白质芯片)。
在另一优选例中，所述的诊断包括早期诊断、辅助性诊断、或其组合。
在另一优选例中，所述FOLFOX响应性标志物表A和/或表B中所示的任一标志物的基因、mRNA、cDNA、或蛋白来源于人。
在另一优选例中，所述的对象为人。
在另一优选例中，所述的对象为肿瘤患者。
在另一优选例中，所述的肿瘤患者包括肠癌患者，尤其是大肠癌患者。
在另一优选例中，所述的患者包括处于I期、II期、III期和IV期的肠癌患者。
在另一优选例中，所述FOLFOX响应性标志物的基因、mRNA、cDNA、或蛋白来源于人。
在另一优选例中，所述检测是针对离体样本的检测。
在另一优选例中，所述的离体样本包括：血液样本、血清样本、组织样本、体液样本或其组合。
在另一优选例中，所述的样本为分离自外周血的单个核细胞样本。
在另一优选例中，所述检测是检测外周血的单个核细胞中表A和/或表B中所示的任一基因的表达量。
在另一优选例中，所述的检测试剂偶联有或带有可检测标记。
在另一优选例中，所述可检测标记选自下组：生色团、化学发光基团、荧光团、同位素或酶。
在另一优选例中，所述的抗体是单克隆抗体或多克隆抗体。
在另一优选例中，所述诊断试剂包括抗体、引物、探针、测序文库、核酸芯片(如DNA芯片)或蛋白质芯片。
在另一优选例中，所述的核酸芯片包括基片和点样在基片上的特异性寡核苷酸探针，所述的特异性寡核苷酸探针包括与任一所述的FOLFOX响应性标志物的多核苷酸(mRNA或cDNA)特异性结合的探针。
在另一优选例中，所述的蛋白质芯片包括基片和点样在基片上的特异性抗体，所述的特异性抗体包括抗所述FOLFOX响应性标志物的特异性抗体。
在另一优选例中，所述的抗体是单克隆抗体或多克隆抗体。
在另一优选例中，所述用途包括用于预测肠癌患者采用FOLFOX治疗的效果(预后)。
在另一优选例中，当满足以下条件时，则提示所述肠癌患者适合采用FOLFOX治疗，和/或采用FOLFOX治疗的效果好：
(i)当某一标志物在FOLFOX响应者中是上调的标志物，且所述标志物的表达水平高于参考值或标准值时；和/或
(ii)当某一标志物在FOLFOX响应者中是下调的标志物，且所述标志物的表达水平低于参考值或标准值时。
在另一优选例中，所述的标志物包括了A1～A20中任一标志物或其组合。
在另一优选例中，所述的标志物包括了A1～A20中至少2个或至少3个标志物所构成的组合。
在另一优选例中，所述的标志物还包括了A21～A48中任一标志物或其组合。
在另一优选例中，所述的标志物还包括了A49～A65中任一标志物或其组合。
在另一优选例中，所述的标志物还包括了A66～A74中任一标志物或其组合。
在另一优选例中，所述的标志物包括了A1～A65中n个标志物所构成的组合，其中，n为3-65的任一正整数(即3、4、5、6、7、8、9、10、11、12、13、14、15、16、17、18、19、20、21、22、23、24、25、26、27、28、29、30、31、32、33、34、35、36、37、38、39、40、41、42、43、44、45、46、47、48、49、50、51、52、53、54、55、56、57、58、59、60、61、62、63、64、65)。
在另一优选例中，所述的标志物包括了A1～A74中n个标志物所构成的组合，其中，n为3-74的任一正整数(即3、4、5、6、7、8、9、10、11、12、13、14、15、16、17、18、19、20、21、22、23、24、25、26、27、28、29、30、31、32、33、34、35、36、37、38、39、40、41、42、43、44、45、46、47、48、49、50、51、52、53、54、55、56、57、58、59、60、61、62、63、64、65、66、67、68、69、 70、71、72、73、74)。
在另一优选例中，所述的标志物包括了B1～B6中任一标志物或其组合。
在另一优选例中，所述的标志物包括了B1～B6中至少2个或至少3个标志物所构成的组合。
在另一优选例中，所述的标志物还包括了B7～B19中任一标志物或其组合。
在另一优选例中，所述的标志物还包括了B20～B61中任一标志物或其组合。
在另一优选例中，所述的标志物还包括了B62～B74中任一标志物或其组合。
在另一优选例中，所述的标志物还包括了B1～B61中n个标志物所构成的组合，其中，n为3-61的任一正整数(即3、4、5、6、7、8、9、10、11、12、13、14、15、16、17、18、19、20、21、22、23、24、25、26、27、28、29、30、31、32、33、34、35、36、37、38、39、40、41、42、43、44、45、46、47、48、49、50、51、52、53、54、55、56、57、58、59、60、61)。
在另一优选例中，所述的标志物包括了B1～B74中n个标志物所构成的组合，其中，n为3-74的任一正整数(即3、4、5、6、7、8、9、10、11、12、13、14、15、16、17、18、19、20、21、22、23、24、25、26、27、28、29、30、31、32、33、34、35、36、37、38、39、40、41、42、43、44、45、46、47、48、49、50、51、52、53、54、55、56、57、58、59、60、61、62、63、64、65、66、67、68、69、70、71、72、73、74)。
在另一优选例中，所述的试剂为PCR引物对。
在本发明的第二方面，提供了一种试剂盒，所述的试剂盒含有一检测试剂，所述检测试剂用于检测FOLFOX响应性标志物的基因、mRNA、cDNA、蛋白、或其组合，
其中，所述的FOLFOX响应性标志物选自下组：
(A)选自A1至A74的任一标志物、或其组合；
(B)选自B1至B74的任一标志物、或其组合；
(C)A1至A74中一个或多个标志物与B1至B74中一个或多个标志物所构成的组合。
在另一优选例中，所述的FOLFOX响应性标志物包括：
(A)选自A1至A65的任一标志物、或其组合；
(B)选自B1至B61的任一标志物、或其组合；
(C)A1至A65中一个或多个标志物与B1至B61中一个或多个标志物所构成的组合。
在另一优选例中，所述的试剂盒含有FOLFOX响应性标志物的基因、mRNA、cDNA和/或蛋白作为对照品或质控品。
在另一优选例中，所述的试剂盒还包括标签或说明书，所述标签或说明书注 明所述试剂盒用于(a)判断某一对象采用FOLFOX疗法的响应性，和/或(b)评价某一对象采用FOLFOX治疗肠癌的治疗效果。
在另一优选例中，所述的试剂为PCR引物对。
在另一优选例中，所述的标签或说明书中注明以下内容：
如果检测对象的FOLFOX响应性标志物的检测结果满足以下条件时，则提示所述肠癌患者适合采用FOLFOX治疗，和/或采用FOLFOX治疗的效果好：
(i)当某一标志物在FOLFOX响应者中是上调的标志物，且所述标志物的表达水平高于参考值或标准值时；和/或
(ii)当某一标志物在FOLFOX响应者中是下调的标志物，且所述标志物的表达水平低于参考值或标准值时。
在本发明第三方面，提供了一种检测方法，包括步骤：
(a)提供一检测样本，所述检测样本为血液样本或组织样本(如手术中切下的肠癌组织)；
(b)检测所述检测样本中FOLFOX响应性标志物基因的表达量，记为C1；和
(c)将所述FOLFOX响应性标志物的浓度C1与对照参比值C0进行比较，
其中，所述的FOLFOX响应性标志物选自下组：
(A)选自A1至A74的任一标志物、或其组合；
(B)选自B1至B74的任一标志物、或其组合；
(C)A1至A74中一个或多个标志物与B1至B74中一个或多个标志物所构成的组合；
如果检测对象的FOLFOX响应性标志物的检测结果满足以下条件时，则提示所述肠癌患者适合采用FOLFOX治疗，和/或采用FOLFOX治疗的效果好：
(i)当某一标志物在FOLFOX响应者中是上调的标志物，且所述标志物的表达水平高于参考值或标准值C0时；和/或
(ii)当某一标志物在FOLFOX响应者中是下调的标志物，且所述标志物的表达水平低于参考值或标准值C0时。
在另一优选例中，所述检测方法还包括：如果检测结果提示所述肠癌患者不适合采用FOLFOX治疗，和/或采用FOLFOX治疗的效果不好，则进一步将所述肠癌患者分型为IML1型或IML2型。
在另一优选例中，所述的FOLFOX响应性标志物包括：
(A)选自A1至A65的任一标志物、或其组合；
(B)选自B1至B61的任一标志物、或其组合；
(C)A1至A65中一个或多个标志物与B1至B61中一个或多个标志物所构成的组合。
在另一优选例中，所述的组织样本是病人手术中切下的肠癌组织样本。
在另一优选例中，所述的标志物基因的表达量是mRNA量。
在另一优选例中，所述方法还包括：对所述肠癌组织样本进行mRNA抽提，从而获得经抽提的mRNA。
在另一优选例中，所述的样本来自一检测对象。
在另一优选例中，所述的检测对象为人。
在另一优选例中，所述的对照参比值C0为FOLFOX无响应正常人群中相同样本中的FOLFOX响应性标志物的浓度。
在本发明第四方面，提供了一种对FOLFOX响应性进行分型的方法，包括步骤：
a)提供一来自待测对象的测试样本，所述的测试样本选自下组：血液样本或组织样本(如手术中切下的肠癌组织)；
b)检测所述测试样本中FOLFOX响应性标志物的mRNA和/或蛋白的浓度，记为C1；和
c)将所述的FOLFOX响应性标志物的mRNA和/或蛋白的浓度C1与对照参比值C0进行比较，
如果检测对象的FOLFOX响应性标志物的检测结果满足以下条件时，则提示所述肠癌患者为适合采用FOLFOX治疗的FOLFOX响应者：
(i)当某一标志物在FOLFOX响应者中是上调的标志物，且所述标志物的表达水平高于参考值或标准值C0时；和/或
(ii)当某一标志物在FOLFOX响应者中是下调的标志物，且所述标志物的表达水平低于参考值或标准值C0时；
其中，所述的FOLFOX响应性标志物选自下组：
(A)选自A1至A74的任一标志物、或其组合；
(B)选自B1至B74的任一标志物、或其组合；
(C)A1至A74中一个或多个标志物与B1至B74中一个或多个标志物所构成的组合。
在另一优选例中，所述的FOLFOX响应性标志物包括：
(A)选自A1至A65的任一标志物、或其组合；
(B)选自B1至B61的任一标志物、或其组合；
(C)A1至A65中一个或多个标志物与B1至B61中一个或多个标志物所构成的组合。
在另一优选例中，在步骤(c)中，还包括：如果检测结果提示所述肠癌患者不适合采用FOLFOX治疗，和/或采用FOLFOX治疗的效果不好，则进一步将所述肠癌患者分型为IML1型或IML2型；
其中，当满足以下条件(Z1)和/或(Z2)时，则将所述肠癌患者确定为FOLFOX化疗的无响应IML1型：
(Z1)当表A某一标志物在FOLFOX响应者中是上调的标志物，且所述标志物的表达水平低于参考值或标准值C0时；和/或
(Z2)当表A某一标志物在FOLFOX响应者中是下调的标志物，且所述标志物的表达水平高于参考值或标准值C0时；
当满足以下条件(Z3)和/或(Z4)时，则将所述肠癌患者确定为FOLFOX化疗的无响应IML2型：
(Z3)当表B某一标志物在FOLFOX响应者中是上调的标志物，且所述标志物的表达水平低于参考值或标准值C0时；和/或
(Z4)当表B某一标志物在FOLFOX响应者中是下调的标志物，且所述标志物的表达水平高于参考值或标准值C0时。
在本发明的第五方面，提供了一种FOLFOX响应性标志物的基因、mRNA、cDNA、或蛋白的用途，它们被用作判断FOLFOX响应性和/或评价FOLFOX治疗效果的标志物，和/或用作将FOLFOX化疗的无响应者进行分型的标志物；
其中，所述的FOLFOX响应性标志物选自下组：
(A)选自A1至A74的任一标志物、或其组合；
(B)选自B1至B74的任一标志物、或其组合；
(C)A1至A74中一个或多个标志物与B1至B74中一个或多个标志物所构成的组合。
在另一优选例中，所述的FOLFOX响应性标志物包括：
(A)选自A1至A65的任一标志物、或其组合；
(B)选自B1至B61的任一标志物、或其组合；
(C)A1至A65中一个或多个标志物与B1至B61中一个或多个标志物所构成的组合。
在本发明的第六方面，提供了一种用于对FOLFOX响应性进行分型和/或对 FOLFOX化疗的无响应者进一步分型的标志物集合(set)，所述的集合包括：
(A)选自A1至A74的3-74种标志物；
(B)选自B1至B74的3-74种标志物；
(C)A1至A74中一个或多个标志物与B1至B74中一个或多个标志物所构成的组合，所述组合至少包含3种标志物。
在另一优选例中，所述的FOLFOX响应性标志物包括：
(A)选自A1至A65的2-65种标志物；
(B)选自B1至B61的2-61种标志物；
(C)A1至A65中一个或多个标志物与B1至B61中一个或多个标志物所构成的组合。
在另一优选例中，所述的标志物集合包括了A1～A20中任一标志物或其组合。
在另一优选例中，所述的标志物集合包括了A1～A20中至少2个或至少3个标志物所构成的组合。
在另一优选例中，所述的标志物集合还包括了A21～A48中任一标志物或其组合。
在另一优选例中，所述的标志物集合还包括了A49～A65中任一标志物或其组合。
在另一优选例中，所述的标志物集合包括了A1～A74中n个标志物所构成的组合，其中，n为3-74的任一正整数。
在另一优选例中，所述的标志物集合包括了B1～B6中任一标志物或其组合。
在另一优选例中，所述的标志物集合包括了B1～B6中至少2个或至少3个标志物所构成的组合。
在另一优选例中，所述的标志物集合还包括了B7～B19中任一标志物或其组合。
在另一优选例中，所述的标志物集合还包括了B20～B61中任一标志物或其组合。
在另一优选例中，所述的标志物集合包括了B1～B74中n个标志物所构成的组合，其中，n为3-74的任一正整数
在本发明的第七方面，提供了一种对人大肠癌进行分型的方法，所述方法包括：
(1)获得某一检测对象的表A中至少三个标志物基因在所述样品中的mRNA基因表达水平；
(2)归一化表A中至少三个标志物基因的基因表达值；
(3)计算表A中所述至少三个标志物基因与这些基因在FOLFOX应答者中的平均基因表达值的第一相似度；以及计算表A中所述至少三个标志物基因与这些基因在FOLFOX无应答者中的平均基因表达值的第二相似度；
(4)分别针对各标志物基因，计算第一相似度与第二相似度的差值；
(5)基于所述差值，对大肠癌患者进行分型，所述分型将检测对象划分为针对化疗的响应者或无响应者(即IML1型)；
其中，所述的化疗包括：FOLFOX、5-FU治疗、奥沙利铂治疗、或靶向DNA复制的化学治疗。
在另一优选例中，所述方法还包括：用化疗对分型为化疗响应者进行结直肠癌治疗。
在另一优选例中，所述的至少三个标志物包括了A1～A20中任一标志物或其组合。
在另一优选例中，所述的至少三个标志物包括了A1～A20中至少2个或至少3个标志物所构成的组合。
在另一优选例中，所述的至少三个标志物还包括了A21～A48中任一标志物或其组合。
在另一优选例中，所述的至少三个标志物还包括了A49～A65中任一标志物或其组合。
在另一优选例中，所述的至少三个标志物包括了B1～B6中任一标志物或其组合。
在另一优选例中，所述的至少三个标志物包括了B1～B6中至少2个或至少3个标志物所构成的组合。
在另一优选例中，所述的至少三个标志物还包括了B7～B19中任一标志物或其组合。
在另一优选例中，所述的至少三个标志物还包括了B20～B61中任一标志物或其组合。
在本发明的第八方面，提供了一种对人大肠癌进行分型的方法，所述方法包括：
(1)获得某一检测对象的表B中至少三个标志物基因在所述样品中的mRNA基因表达水平；
(2)归一化表B中至少三个标志物基因的基因表达值；
(3)计算表B中所述至少三个标志物基因与这些基因在FOLFOX应答者中的平均基因表达值的第一相似度；以及计算表B中所述至少三个标志物基因与这些基因在FOLFOX无应答者中的平均基因表达值的第二相似度；
(4)分别针对各标志物基因，计算第一相似度与第二相似度的差值；
(5)基于所述差值，对大肠癌患者进行分型，所述分型将检测对象划分为针对化疗的响应者或无响应者(即IML2型)；
其中，所述的化疗包括：FOLFOX、5-FU治疗、奥沙利铂治疗、或靶向DNA复制的化学治疗。
在另一优选例中，所述方法还包括：用化疗对分型为化疗响应者进行结直肠癌治疗。
在另一优选例中，所述的至少三个标志物如上定义。
在另一优选例中，所述的mRNA基因表达水平用选自下组的技术获得：microarray,RNAseq,RT-PCR、或其组合。
在另一优选例中，所述的方法还包括用选自下组的方法进行归一化处理：fRMA,RMA,RNAseq CPM，RNAseq FPKM。
在另一优选例中，在步骤(3)中，用选自下组的方法计算相似性：
欧氏距离(euclidean distance)，曼哈顿距离(manhattan distance)，明可夫斯基距离(minkowski distance)，切比雪夫距离(chebyshev distance)，雅卡德距离(chebyshev distance)，皮尔逊相关性(pearson correlation)，余弦相关性(consine correlation)或回归值(regression values)。
在本发明的第九方面，提供了一种用于对大肠癌进行分型的设备，包括：
(P1)输入单元，所述输入单元用于输入某一对象的表A或表B中至少三个标志物基因在所述样品中的mRNA基因表达水平的数据；
(P2)数据处理单元，所述数据处理单元对输入的mRNA基因表达水平的数据进行处理，并且所述数据处理单元包括归一化处理子单元、相似度计算子单元和相似度差值计算子单元；
其中，所述的归一化处理子单元用于对表A或表B中至少三个标志物基因的基因表达值进行归一化处理；
所述的相似度计算子单元用于计算表A或表B中所述至少三个标志物基因与这些基因在FOLFOX应答者中的平均基因表达值的第一相似度；以及计算表A或表B中所述至少三个标志物基因与这些基因在FOLFOX无应答者中的平均基因表达值的第二相似度；
所述的相似度差值计算子单元用于计算各标志物基因的第一相似度与第二相似度的差值；
(P3)分型单元，所述的分型单元基于所述各标志物基因的差值，对所述对象进行分型，所述分型将检测对象划分为针对化疗的响应者或无响应者，从而获得分型结果；和
(P4)输出设备，所述的输出设备用于输出所述的分型结果。
在另一优选例中，所述分型单元还用于将FOLFOX化疗的无响应者进一步分为IML1型或IML2型。
在另一优选例中，所述的对象为肠癌患者。
在另一优选例中，所述的化疗包括：FOLFOX、5-FU治疗、奥沙利铂治疗、或靶向DNA复制的化学治疗。
在另一优选例中，所述的至少三个标志物如上定义。
应理解，在本发明范围内中，本发明的上述各技术特征和在下文(如实施例)中具体描述的各技术特征之间都可以互相组合，从而构成新的或优选的技术方案。限于篇幅，在此不再一一累述。
附图说明
图1显示了迭代式机器学习(IML)方法的工作流找到了机器学习过程的收敛点，将FOLFOX无响应者划分为亚组，然后分析FOLFOX无响应者亚组的FOLFOX抵抗机制。
图2显示了验证组中FOLFOX治疗的4期大肠癌患者(n＝32)的生存曲线，其表明IML预测的FOLFOX响应者组的总体生存率明显高于IML预测的FOLFOX无响应者组，HR＝2.6(p值＝0.02)。
图3显示了在佐剂(adjuvant)设置中，验证集中FOLFOX治疗的3期大肠癌患者(n＝55)的生存曲线，其表明，IML预测的FOLFOX响应者组比IML预测的FOLFOX无响应者组具有更好的无复发生存率。HR＝2.36(p值＝0.02)。在该数据集中患者的复发事件被记录，并该数据集的患者的随访时间被估计为60个月内均匀分布的时间序列。
图4A显示了在表1A中3个基因的任何组合都具有FOLFOX响应的预测价值。X轴是使用的基因数量，从3个开始到50个结束。Y轴是使用选定数量的基因进行200轮随机组合获得的平均性能。上面那条实线：无响应者的灵敏度(sensitivity of  nonresponder)；中间长虚线：平均总体性能(overall performance)；下面点虚线：无响应者的特异性(specificity of nonresponder)。
图4B显示了表1B中3个基因的任何组合都具有FOLFOX响应的预测价值。X轴是使用的基因数量，从3个开始到50个结束。Y轴是使用选定数量的基因进行200轮随机组合获得的平均性能。上面那条实线：无响应者的灵敏度(sensitivity of nonresponder)；中间长虚线：平均总体性能(overall performance)；下面点虚线：无响应者的特异性(specificity of nonresponder)。
图5显示了箱线图，其显示了FOLFOX抗性的已知标记(5A.ERCC1、5B.DPYD，5C.BRCA1、5D.SRBC，5E.TYMP)的表达水平与IML预测结果之间的关联性。Y轴是此标记的表达值，绿色框是IML预测的FOLFOX响应者，红色框是IML基因签名特征1预测的FOLFOX无响应者，黄色框是IML基因签名特征2的预测FOLFOX无响应者。不同抗性亚组的潜在主要FOLFOX抗性机制明显不同。
图6显示了IML基因签名特征1中的基因蛋白质相互作用网络，其显示了高度相关联的细胞周期与有丝分裂相关的网络。
具体实施方式
本发明人通过广泛而深入的研究，首次开发了可用于选择化疗响应患者的生物标志物和方法。具体地，本发明提供了将结肠直肠癌患者分为化疗应答者和非应答者的生物标记物和方法，其中被分类为应答者的结肠直肠癌患者比非应答者具有明显更好的生存益处。在此基础上完成了本发明。
研究已表明，预测FOLFOX响应不是一件容易的事，其复杂性至少表现在三个层面上。首先，FOLFOX不是具有已知单一蛋白质靶标的单一靶向疗法。FOLFOX的主要细胞毒性成分5-Fu和奥沙利铂具有不同的作用机理。5-FU旨在抑制胸腺嘧啶合酶(这是DNA合成中的关键酶)，并且5-FU的代谢产物与核苷酸的结构相似，它可被掺入DNA并导致细胞死亡(Longley等，2003)。奥沙利铂在DNA之间形成链内连接并破坏DNA复制(Hector等，2001)。其次，大肠癌是一种异质性疾病，至少存在四种共有分子亚型(Guinney等人，2015)。不同分子亚型对结直肠肿瘤细胞DNA复制和细胞周期的依赖性不同。第三，肿瘤可能以不同方式抵抗FOLFOX治疗(Hammond等，2016)。因此，耐药性可能是由多种因素共同引起的，典型的简单测序一种药物靶标基因方法不太可能奏效(Hammond et al，2016)。在本发明中，本发明人通过使用IV期结直肠癌患者的基因组数据和客观的FOLFOX响应性数据，开发一种新颖的迭代监督机 器学习(IML)方法，解决了这些复杂性。IML方法在机器学习过程中找到统计收敛点，并识别出具有相同的FOLFOX耐药基本机制的结直肠癌患者亚组，并选择受益于FOLFOX治疗的患者。本发明人在两个独立的验证数据集中验证此方法：一个数据集是FOLFOX治疗的IV期结直肠癌患者；另一个数据集是FOLFOX治疗的III期结直肠癌患者。
术语
本文中使用的术语“样品”或“样本”是指与受试者特异地相关联的材料，从其中可以确定、计算或推断出与受试者有关的特定信息。样本可以全部或部分由来自受试者的生物材料构成。
如本文所用，术语“表达”包括mRNA从基因或基因部分的产生，并且包括由RNA或基因或基因部分所编码的蛋白质的产生，还包括与表达相关的检测物质的出现。例如，cDNA，结合配体(如抗体)与基因或其它寡核苷酸、蛋白质或蛋白质片段的结合以及结合配体的显色部分都包括在术语“表达”的范围内。因此，在免疫印迹如Western印迹上半点密度的增加也处于以生物学分子为基础的术语“表达”的范围内。
如本文所用，术语“参比值”或“对照参比值”是指当与分析结果相比时与特定结果统计学相关的值。在优选的实施方案中，参比值是根据对比较FOLFOX响应性标志物的mRNA表达和/或蛋白的表达，并进行统计学分析来确定的。在本文的实施例部分中显示了一些这样的研究。但是，来自文献的研究和本文公开的方法的用户经验也可用于生产或调整参比值。参比值也可以通过考虑与患者的医疗史、遗传学、年龄和其它因素特别相关的情况和结果来确定。
FOLFOX响应性标志物
如本文所用，术语“本发明的FOLFOX响应性标志物”指表A和/或表B中所示的一种或多种标志物。
在本发明中，术语“本发明的FOLFOX响应性标志物蛋白”、“本发明蛋白”、“本发明的多肽”、或“表A和/或表B中所示的标志物”可互换使用，都指具有本发明的FOLFOX响应性标志物中任何一种或多种。
在本发明中，术语“FOLFOX响应性标志物基因”、“FOLFOX响应性标志物的多核苷酸”可互换使用，都指表A和/或表B中所示的任一FOLFOX响应性标志物的核苷酸序列。
需理解的是，当编码相同的氨基酸时，密码子中核苷酸的取代是可接受的。另外需理解的是，由核苷酸取代而产生保守的氨基酸取代时，核苷酸的变换也是可被接受的。
在得到了FOLFOX响应性标志物的信息的情况下，可根据其构建出编码它的核酸序列，并且根据核苷酸序列来设计特异性探针。核苷酸全长序列或其片段通常可以用PCR扩增法、重组法或人工合成的方法获得。对于PCR扩增法，可根据本发明所公开的FOLFOX响应性标志物的核苷酸序列，尤其是开放阅读框序列来设计引物，并用市售的cDNA库或按本领域技术人员已知的常规方法所制备的cDNA库作为模板，扩增而得有关序列。当序列较长时，常常需要进行两次或多次PCR扩增，然后再将各次扩增出的片段按正确次序拼接在一起。
一旦获得了有关的序列，就可以用重组法来大批量地获得有关序列。这通常是将其克隆入载体，再转入细胞，然后通过常规方法从增殖后的宿主细胞中分离得到有关序列。
此外，还可用人工合成的方法来合成有关序列，尤其是片段长度较短时。通常，通过先合成多个小片段，然后再进行连接可获得序列很长的片段。
目前，已经可以完全通过化学合成来得到编码本发明蛋白(或其片段，衍生物)的DNA序列。然后可将该DNA序列引入本领域中已知的各种现有的DNA分子(如载体)和细胞中。
通过常规的重组DNA技术，可利用本发明的多核苷酸序列可用来表达或生产重组的FOLFOX响应性标志物。
特异性抗体
在本发明中，术语“本发明抗体”和“抗FOLFOX响应性标志物的特异性抗体”可互换使用。
本发明还包括对FOLFOX响应性标志物(表A和/或表B)具有特异性的多克隆抗体和单克隆抗体，尤其是单克隆抗体。
本发明不仅包括完整的单克隆或多克隆抗体，而且还包括具有免疫活性的抗体片段，如Fab'或(Fab)2片段；抗体重链；抗体轻链；遗传工程改造的单链Fv分子(Ladner等人，美国专利No.4,946,778)；或嵌合抗体，如具有鼠抗体结合特异性但仍保留来自人的抗体部分的抗体。
本发明的抗体可以通过本领域内技术人员已知的各种技术进行制备。例如，纯化的人FOLFOX响应性标志物的基因产物或者其具有抗原性的片段，可被施用于 动物以诱导多克隆抗体的产生。与之相似的，表达人FOLFOX响应性标志物蛋白或其具有抗原性的片段的细胞可用来免疫动物来生产抗体。本发明的抗体也可以是单克隆抗体。此类单克隆抗体可以利用杂交瘤技术来制备
抗人FOLFOX响应性标志物蛋白的抗体可用于免疫组织化学技术中，检测标本(尤其是组织样本或血液样本)中的人FOLFOX响应性标志物蛋白。由于FOLFOX响应性标志物蛋白存在于血液样本或组织样本(手术中切下的肠癌组织)中，因此其表达量可成为检测对象。
检测方法
基于FOLFOX响应性标志物在组织样本或血液样本中差异表达，本发明还提供了相应的判断FOLFOX响应性的方法。
本发明涉及定量和定位检测人FOLFOX响应性标志物的蛋白水平或mRNA水平的诊断试验方法。这些试验是本领域所熟知的。试验中所检测的人FOLFOX响应性标志物蛋白水平或mRNA水平，可以用于判断(包括辅助判断)是否适合进行FOLFOX治疗或是否具有FOLFOX响应性。
一种优选的方法是对mRNA或cDNA，进行PCR进行定量检测。
一种优选的方法是对mRNA或cDNA，测序进行定量检测。
FOLFOX响应性标志物的多核苷酸可用于FOLFOX响应性的诊断。本发明的多核苷酸的一部分或全部可作为探针固定在微阵列或DNA芯片上，用于分析中基因的差异表达分析和基因诊断。抗FOLFOX响应性标志物的抗体可以固定在蛋白质芯片上，用于检测样本中的FOLFOX响应性蛋白。
检测试剂盒
基于FOLFOX响应性标志物与FOLFOX响应性的相关性，因此FOLFOX响应性标志物可以作为FOLFOX响应性的判断标志物。
本发明还提供了一种判断FOLFOX响应性的试剂盒，所述的试剂盒含有一检测试剂，所述检测试剂用于检测FOLFOX响应性标志物的基因、mRNA、cDNA、蛋白、或其组合。优选地，所述试剂盒含有本发明的抗FOLFOX响应性标志物的抗体或免疫偶联物，或其活性片段；或者含有特异性扩增FOLFOX响应性标志物的mRNA或cDNA的引物或引物对、探针或芯片。
在另一优选例中，所述的试剂盒还包括标签或说明书。
在本发明中，本发明人使用具有客观FOLFOX响应数据的III期和IV期大肠癌患者的基因组数据，开发了一种新颖的迭代监督机器学习(IML)方法。IML方法在机器学习过程中找到统计收敛点，并识别出具有相同的FOLFOX耐药潜在生物学机制的结直肠癌患者亚组，并选择受益于FOLFOX治疗的患者。
本发明的主要优点包括：
(a)FOLFOX治疗的IV期大肠癌患者训练集的表现为97.6％的敏感性和100％的特异性。IML方法在两个独立的验证集中进行了验证，其中一个验证集是FOLFOX治疗的IV期大肠癌患者(HR＝2.6，p值＝0.02，预测的响应者组和预测的无响应者组3年总生存率18)0.8％)，另一个验证集是FOLFOX治疗的III期大肠癌患者(估计HR＝2.36，p值＝0.02)。预测分数显示：与包括ERCC1，DPYD，BRCA1，CAVIN3和TYMP在内的已知单基因标记的表达水平存在相关性。
(b)功能分析和蛋白质-蛋白质网络首次显示，FOLFOX抗药性有两种不同的优势模式，每种模式都依赖于不同类型的DNA损伤修复蛋白的上调。FOLFOX耐药的主要模式可能是肿瘤细胞自身抗凋亡和改变细胞周期的内在能力的协同效应。CMS4间充质亚型患者的部分肿瘤样品也显示出这些模式，并且仍可从FOLFOX治疗中受益。IML模型的预测分数反映了FOLFOX耐药性的潜在机制，这将有助于设计FOLFOX的联合疗法。
下面结合具体实施例，进一步阐述本发明。应理解，这些实施例仅用于说明本发明而不用于限制本发明的范围。下列实施例中未注明具体条件的实验方法，通常按照常规条件，例如Sambrook等人，分子克隆：实验室手册(New York:Cold Spring Harbor Laboratory Press,1989)中所述的条件，或按照制造厂商所建议的条件。除非另外说明，否则百分比和份数是重量百分比和重量份数。
实施例1
1.1数据
使用R/Bioconductor软件(Gentleman等，2004)分析基因表达数据。将83例IV期结直肠癌患者(GSE28702)的公开基因表达数据和客观FOLFOX响应数据用作训练集(Tsuji等，2012)。将32例经FOLFOX治疗的IV期大肠癌患者(GSE72970)的基因表达数据和生存数据用作独立验证集(Del Rio等人，2017)。55例FOLFOX治疗的III期结直肠癌患者(GSE81653)的基因表达数据和生存数据被用作独立验证集(Lin等人，2017)。使用frma软件包中的vRMA(fRMA)方法对Affymetrix人类基因组U133  Plus 2.0阵列数据(GSE28702和GSE72970)进行归一化，并且该归一化方法专为临床诊断设置而设计的，每个诊断样品均需单独处理(McCal等人，2010年)。使用ComBat删除了GSE28702中的样本和GSE72970中的样本的批处理效果(Johnson等人，2007)。标准化的Affymetrix人类基因2.0ST阵列数据(GSE81653)从Gene Expression Omnibus数据库下载。
1.2迭代机器学习(IML)方法
总共使用了83个IV期CRC肿瘤样本(FOLFOX响应者n＝42，FOLFOX无响应者n＝41)来建立模型。如图1(图1)中IML机器学习流程的总体设计所示，该模型使用六个迭代轮进行了训练。在每一轮中，将无响应者组中选定的样品子集与响应者组中的所有肿瘤样品进行比较。第一轮和第四轮学习用于预选子集中具有相同特征的无响应者肿瘤，并对该轮中使用的无响应者组和响应者组进行t-检验。
使用两个统计标准来选择这两个学习回合的特征中的基因：(1)p值<0.01的基因和(2)无响应者组与响应者组的均值之差的绝对值的基因需要高于1.2倍的变化。第二、第三、第五和第六次学习轮次用于加强先前的学习轮次，并进行了200轮10倍交叉验证，并且在每个交叉验证轮中，进行了t检验和p值被排名。一个统计标准用于选择这四个学习回合的特征中的基因：在200个回合验证中，至少90％的基因的p值位于排名前250位的基因中。在第一轮，第二轮，第四轮和第五轮中选择的基因签名特征(gene signatures)是预筛选基因签名特征，它们未用于最终评分功能。通过使用简单的最接近质心模型来计算这四个预筛选基因签名特征的分数。机器学习方法在第三次和第六次迭代监督学习回合中产生统计收敛点，在此情况下，200轮10倍交叉验证(10-folder cross-validation)的曲线下平均面积始终高于0.8。第三轮和第六轮选择的基因签名特征是最终基因签名特征。在第三轮监督学习循环中，所有42个FOLFOX响应者样本和30个FOLFOX无响应者样本的子集都用于统计分析，并训练了74个基因签名特征IMLSig.1 □(表1A)。在第六次有监督的迭代学习轮中，所有42个FOLFOX响应者样本和13个FOLFOX无响应者样本的子集都用于统计分析，并训练了74个基因签名特征IMLSig 2 □(表1B)。
IMLSig 1 □和IMLSig 2 □被用来构造两个k最近邻回归得分函数S 1和S 2。通过组合规则S＝(S 1≥λ 1)∨(S 2≥λ 2)，，将两个分数合并为一个分数S。CRC肿瘤样本可以分为两组：FOLFOX响应者(S＝0)和FOLFOX无响应者(S＝1)。使用DAVID(Huang等人，2009)进行基因签名特征的富集功能分析；使用STRING(Szklarczyk等人，2019)进行蛋白质网络分析。
1.3结果
1.3.1使用IV期CRC肿瘤开发IML机器学习模型
总共在六个迭代有监督的学习轮中训练了两种基因签名特征。机器学习流程的总体设计如图1所示。机器学习方法在第三次和第六次迭代监督学习回合中产生统计收敛点，由此200轮10折交叉验证的曲线下平均面积始终高于0.8。第三和第六次迭代监督学习回合用于选择最终基因签名特征中的基因。在第三轮监督学习循环中，所有42个FOLFOX响应者样本和30个FOLFOX无响应者样本的子集都用于统计分析(请参阅详细的统计筛选方法)，并训练了74个基因签名特征(基因签名特征1，表1A)。在第六次有监督的迭代学习回合中，所有42个FOLFOX响应者样本和13个FOLFOX无响应者样本的子集都用于统计分析，并训练出了74个基因签名特征(基因签名特征2，表1B)。
1.3.2 IML机器学习模型在训练集和独立验证集中的表现
在83个FOLFOX治疗的IV期CRC肿瘤样本的训练集中(响应者n＝42，无响应者n＝41)，IML预测模型的灵敏度为97.6％，特异性为100％。
在32个FOLFOX治疗的IV期CRC样本的独立验证集中。IML模型将50％的样本(n＝16)预测为FOLFOX无响应者，IML模型将50％的样本(n＝16)预测为FOLFOX响应者。尽管样本量很小，但IML预测模型的生存分析导致显着的危险比HR＝2.6(p值＝0.02)。预测的FOLFOX响应者组的3年生存率是61.9％[95％CI，41.9％-91.4％]，而预测的FOLFOX无响应者组的3年生存率是18.8％[95％CI，6.8％-52.0％]。推断的95％置信区间大是由于样本量较小(n＝32)。预测的FOLFOX无响应者组的中位总体生存时间(13.4个月)明显短于预测的FOLFOX响应者组(36.6个月)(图2)。
为了进一步测试IML预测模型在辅助环境中的预测能力，本发明人使用了166名接受FOLFOX作为辅助治疗的III期大肠癌患者。为了消除由肿瘤内在预后特征引起的潜在偏倚，本发明人使用CMScaller将166位III期大肠癌患者分为四种CMS共有分子亚型(CMS1n＝16，CMS2n＝32，CMS3n＝19，CMS4n＝55，未分类n＝44)(Eide等人，2017；Guinney等人，2015)。因为CMS4组的预后往往最差，并且在这一III期患者队列中，所以CMS4组的患者数量也最多(n＝55)进行统计分析，本发明人仅使用了CMS4亚型的一致分子亚型测试对FOLFOX治疗的响应。记录该数据集中患者的复发事件。此数据集患者的随访时间不可用，并且估计为60个月内分布均匀的时间序列。在这55名接受FOLFOX治疗的III期患者中，IML模型具有良好的预测能力，预计无响应者组与有响应者组的危险比估计为HR＝2.36(p值＝0.02)(图3)。
尽管普遍认为所有CMS4间充质肿瘤都倾向于抗化疗，但必须注意的是，IML方法仍可以从FOLFOX治疗中受益的CMS4亚型患者中识别出患者亚组。IML方法是专门为预测FOLFOX响应而开发的，它提供了附加的预测价值。
1.3.3基因最小组合的表现
为了测试具有预测能力所需的最小数量的基因k，计算过程逐渐增加了基因k的数量，并且从k＝3开始并在k＝50结束，每次k增加1。对于每个值对于k个基因，对(表1A)或(表1B)中k个基因的200个随机组合进行了模拟，并针对k个基因的每个随机组合，测试了基因签名特征的性能。从3个基因开始，平均总体表现已超过60％(图4A，图4B)。
1.3.4 FOLFOX抗药性的分子机制和蛋白-蛋白相互作用网络
两个不同耐药性亚组的潜在主要FOLFOX耐药性机制是不同的。签名特征中的基因富集功能分析的GO生物过程术语的聚类表明了细胞凋亡过程(富集评分0.805)。由特征1(IML1型)预测的FOFOX无响应者得分显示出高ERCC1和高DPYD的趋势，这表明大部分FOLFOX无响应者中的肿瘤细胞可能具有相对较高的5-FU分解代谢率，并具有有效的ERCC1核苷酸切除修复能力，从而克服了奥沙利铂诱导的细胞凋亡(红色方框，图5A，5B)。此外，对签名特征基因的蛋白质相互作用网络分析表明，与细胞周期相关的蛋白质具有高度相互联系的网络(图6)，其中蛋白质参与相互作用的富集的GO术语主要是细胞周期和有丝分裂相关的功能术语(表3)。综上所述，这些结果表明，FOLFOX耐药性的主要机制可能是肿瘤细胞自身抗凋亡和改变细胞周期的内在能力的协同作用。
签名特征中基因的富集功能分析所提出的抗药性机制尚不清楚。总体而言，签名特征中SPOCK1、TGFB1I1、WISP1、TWIST1、TWIST2等基因的下调显示了对间充质表型的抑制。据报道，活跃的间充质表型可抵抗化疗(Guinney等人，2015；Roepman等人，2014)。但是，间充质表型的抑制与对FOLFOX的耐药性之间的正相关可能是功能性DNA损伤修复蛋白的作用。签名特征2(IML2型)预测的FOLFOX无响应者得分显示与高水平BRCA1和低水平的BRCA1失活剂CAVIN3相关联，这表明基因签名特征2型FOLFOX无响应者中的肿瘤细胞可能依赖BRCA1修复奥沙利铂所诱导的双链断裂(黄色方框，图5C，5D)。激活的BRCA1的模式与不存在间充质表型相一致，因为已知BRCA1抑制上皮间充质转化和干细胞去分化(Bai等人，2014)。此外，特征2预测的FOLFOX无响应者表现出轻度的TYMP倾向(黄色方框，图5E)，这表明特征2的 FOLFOX无响应者的肿瘤细胞从5-FU转化为其活性代谢物5-FdUMP的效率可能较低。
IML1型和IML2型的FOLFOX耐药的患者的肿瘤特性提示，这2种亚型对其它药物的敏感程度可能不同。比如，IML1型和IML2型对BRCA1的依赖程度不同，对PARP抑制剂类药物的敏感程度也可能不同。
表1A：使用30个无应答者样品和42个应答者样品选择了表1A中的74个基因(迭代第3轮，图1)。进行了200次10倍交叉验证，并在每个交叉验证回合中进行了t检验，并对p值进行了排名。使用一个统计标准来选择这四个学习回合的特征中的基因：在200个回合验证中，至少90％的基因的p值位于排名前250位的基因中。
表1A.IML签名特征1中的基因列表
Figure PCTCN2021098716-appb-000006
Figure PCTCN2021098716-appb-000007
表1B：通过使用13个无应答者样品和42个应答者样品选择出了表1B中的74个基因(迭代第6轮，图1)。进行了200次10倍交叉验证，并在每个交叉验证回合中进行了t检验，并对p值进行了排名。使用一个统计标准来选择这四个学习回合的特征中的基因：在200个回合验证中，至少90％的基因的p值位于排名前250位的基因中。
表1B.IML签名特征2中的基因列表
Figure PCTCN2021098716-appb-000008
Figure PCTCN2021098716-appb-000009
实施例2
在本实施例中，基于实施例1获得了表1A和表1B两组FOLFOX响应性签名特征， 选取任3个验证。
Table1A中的3个基因和Table1B中的3个基因的4个例子的结果如下表2所示：
表2 3个基因例子
Figure PCTCN2021098716-appb-000010
结果表明，采用Table1A中的3个基因或Table1B中的3个基因，其总体准确性(或总性能值)均在0.6以上。
讨论
FOLFOX是一种化疗药物的组合，它被用于治疗结直肠癌。大约一半的患者对FOLFOX获益，一半的患者显示出耐药性。在本发明中，本发明人开发了一种新颖的迭代机器学习方法(IML)，该方法可以找到具有相同的FOLFOX耐药机制的患者亚组的统计收敛点，并预测患者是否将从FOLFOX治疗中受益。
在本发明中，本发明人提供了可以预测对FOLFOX治疗的响应的IML模型，并在FOLFOX治疗的IV期患者和FOLFOX治疗的III期患者中验证了该模型。IML模型的主要优点是潜在的训练工作流程假设大肠癌患者中可能存在多种导致FOLFOX耐药的原因。本发明人的结果表明，大肠癌中至少有两种不同的FOLFOX耐药性主要机制。每种机制都取决于不同类型的DNA损伤修复蛋白的上调。FOLFOX耐药性的主要机制是肿瘤细胞的抗凋亡和改变的细胞周期构成的协同作用，约占无应答者的75％。FOLFOX耐药性的第二个机制是BRCA1的激活，约占非应答者的25％。
IML模型的基因特征(signatures)显示与已知的单个基因标记(如ERCC1，DPYD，BRCA1，CAVIN3和TYMP)相关。然而，应该指出的是，单个基因不能完全解释FOLFOX耐药性，而耐药性是由多种因素共同引起的(Hammond et al，2016)。在训练集GSE28702和验证集GSE72970(图5A，5B，5C，5D，5E)的样品中，这些与已知单基因标记和特征之间观察到的相关性是一致的，这表明通过IML方法所鉴定出的位于底层的耐药性生物学是有力的，而IML模型确实捕捉到了可协同作用的对细胞凋亡的抗性、细胞周期改变、药物代谢功能障碍和上调的DNA修复失调的基因表达模 式，这些均导致了FOLFOX耐药性。
上皮间质亚型被认为倾向于抵抗化疗。但是，在上皮间质亚型患者中，IML方法确定了仍可从FOLFOX治疗中受益的患者亚组。这些结果不仅表明IML方法为FOLFOX响应性提供了特定的额外预测值，而且表明仅上皮-间质转化不足以引起对FOLFOX治疗的耐药性。还需要获得抗凋亡的协同作用以及上调不同类型的DNA损伤修复蛋白。该过程可能独立于上皮-间充质转化，因为功能性DNA损伤修复蛋白(如BRCA1)可能抑制间充质表型。
总而言之，本发明人开发了可以对FOLFOX响应者和无响应者进行分层的IML方法。IML方法在III期和IV期大肠癌患者组中均得到验证。本发明人方法的主要优势在于，IML不会对所有对FOLFOX耐药的患者进行相同的治疗，而是公正地将无响应的亚组分层，这些亚组具有相同的耐药协同作用机制。IML方法的预测得分反映了FOLFOX耐药性的潜在不同分子机制。因此，IML模型的预测得分也为FOLFOX耐药性的分子机制提供了指示。IML1型和IML2型的FOLFOX耐药的患者的肿瘤特性提示，这2种亚型对其它药物的敏感程度可能不同。比如，IML1型和IML2型对BRCA1的依赖程度不同，对PARP抑制剂类药物的敏感程度也可能不同。根据FOLFOX耐药的患者的肿瘤特性的分型，通过将FOLFOX与其他药物联合使用，可以对抗不同的FOLFOX肿瘤耐药性。
在本发明提及的所有文献都在本申请中引用作为参考，就如同每一篇文献被单独引用作为参考那样。此外应理解，在阅读了本发明的上述讲授内容之后，本领域技术人员可以对本发明作各种改动或修改，这些等价形式同样落于本申请所附权利要求书所限定的范围。

Claims (12)

  1. 一种FOLFOX响应性标志物的基因、mRNA、cDNA、蛋白、或其检测试剂的用途，其特征在于，用于制备一诊断试剂或试剂盒，所述诊断试剂或试剂盒用于(a)判断某一对象采用FOLFOX疗法的响应性，和/或(b)评价某一对象采用FOLFOX治疗肠癌的治疗效果；和/或将FOLFOX化疗的无响应者分为IML1型或IML2型；
    其中，所述的FOLFOX响应性标志物选自下组：
    (A)选自A1至A74的任一标志物、或其组合：(A1)LEPR；(A2)NAIP；(A3)ZBTB37；(A4)GPATCH2L；(A5)ZNF224；(A6)CLASP2；(A7)U2SURP；(A8)CTDSPL2；(A9)ARID4A；(A10)TM2D1；(A11)MYSM1；(A12)AGO3；(A13)TRPM7；(A14)CDK16；(A15)ALS2；(A16)YIPF2；(A17)SPAG9；(A18)DDX5；(A19)PRR7；(A20)IFT80；(A21)PEX12；(A22)MLF2；(A23)RUNDC1；(A24)RNF111；(A25)PIKFYVE；(A26)CCNT2；(A27)FAM149B1；(A28)C2orf49；(A29)CDC27；(A30)NF1；(A31)H2AFX；(A32)FKBP8；(A33)EPS15；(A34)COL28A1；(A35)NFKBIB；(A36)TRIM14；(A37)STX10；(A38)IMPDH1；(A39)GGNBP2；(A40)SCAMP4；(A41)TAOK1；(A42)ADAM15；(A43)FOXN3；(A44)BCLAF1；(A45)NLGN2；(A46)LSM4；(A47)JMJD8；(A48)NHP2；(A49)HMCN2；(A50)CTDSPL；(A51)SRF；(A52)ZC3H7B；(A53)HTATIP2；(A54)PPFIA3；(A55)RNF212；(A56)ITPR3；(A57)KCNC4；(A58)JMJD4；(A59)CNOT9；(A60)PDZD7；(A61)MAP3K10；(A62)ADPRHL1；(A63)SCYL1；(A64)UPP2；(A65)H2BFM；(A66)MAP3K2；(A67)SMURF2；(A68)APPBP2；(A69)NPEPPS；(A70)HELZ；(A71)MBTD1；(A72)AMZ2P1；(A73)CEP290；(A74)WASHC4；
    (B)选自B1至B74的任一标志物、或其组合：(B1)MXRA5；(B2)GAS1；(B3)SFRP2；(B4)PRRX1；(B5)CCL8；(B6)TWIST1；(B7)INHBA；(B8)COL15A1；(B9)F13A1；(B10)CTHRC1；(B11)SULF1；(B12)COLEC12；(B13)ZNF423；(B14)PLAU；(B15)ZCCHC24；(B16)NDN；(B17)COL6A2；(B18)ANTXR1；(B19)RAB31；(B20)DDR2；(B21)MLLT11；(B22)NOX4；(B23)TGFB1I1；(B24)MEIS1；(B25)FLNA；(B26)EFEMP2；(B27)LAYN；(B28)MRC2；(B29)EOGT；(B30)SERPINH1；(B31)NAP1L3；(B32)KIAA1462；(B33)CERCAM；(B34)GLI3；(B35)HS3ST3A1；(B36)C3orf80；(B37)CAVIN1；(B38)PCDHB2；(B39)CMTM3；(B40)TSHZ3；(B41)EVC；(B42)MFGE8；(B43)MSRB3；(B44)EIF4B；(B45)RASA3；(B46)EHD2；(B47)TWIST2；(B48)OMA1；(B49)PCDH7；(B50)FBXL7；(B51)CHSY3；(B52)HLX；(B53)ATP10A；(B54)PRR16；(B55)DPH5；(B56)VASN；(B57)ZBTB46；(B58)PDGFRA；(B59)RPLP0；(B60)RPS3A； (B61)RPL9；(B62)SPOCK1；(B63)COL5A2；(B64)COL6A3；(B65)AEBP1；(B66)COL5A1；(B67)COL6A1；(B68)PCOLCE；(B69)BNC2；(B70)WISP1；(B71)CLEC11A；(B72)THY1；(B73)IGFBP5；(B74)PDGFRB；
    (C)A1至A74中一个或多个标志物与B1至B74中一个或多个标志物所构成的组合。
  2. 如权利要求1所述的用途，其特征在于，所述的FOLFOX响应性标志物包括：
    (A)选自A1至A65的任一标志物、或其组合；
    (B)选自B1至B61的任一标志物、或其组合；
    (C)A1至A65中一个或多个标志物与B1至B61中一个或多个标志物所构成的组合。
  3. 一种试剂盒，其特征在于，所述的试剂盒含有一检测试剂，所述检测试剂用于检测FOLFOX响应性标志物的基因、mRNA、cDNA、蛋白、或其组合，
    其中，所述的FOLFOX响应性标志物选自下组：
    (A)选自A1至A74的任一标志物、或其组合；
    (B)选自B1至B74的任一标志物、或其组合；
    (C)A1至A74中一个或多个标志物与B1至B74中一个或多个标志物所构成的组合。
  4. 一种检测方法，其特征在于，包括步骤：
    (a)提供一检测样本，所述检测样本为血液样本或组织样本(如手术中切下的肠癌组织)；
    (b)检测所述检测样本中FOLFOX响应性标志物基因的表达量，记为C1；和
    (c)将所述FOLFOX响应性标志物的浓度C1与对照参比值C0进行比较，
    其中，所述的FOLFOX响应性标志物选自下组：
    (A)选自A1至A74的任一标志物、或其组合；
    (B)选自B1至B74的任一标志物、或其组合；
    (C)A1至A74中一个或多个标志物与B1至B74中一个或多个标志物所构成的组合；
    如果检测对象的FOLFOX响应性标志物的检测结果满足以下条件时，则提示所述肠癌患者适合采用FOLFOX治疗，和/或采用FOLFOX治疗的效果好：
    (i)当某一标志物在FOLFOX响应者中是上调的标志物，且所述标志物的表达水平高于参考值或标准值C0时；和/或
    (ii)当某一标志物在FOLFOX响应者中是下调的标志物，且所述标志物的表达 水平低于参考值或标准值C0时。
  5. 如权利要求4所述的检测方法，其特征在于，还包括：如果检测结果提示所述肠癌患者不适合采用FOLFOX治疗，和/或采用FOLFOX治疗的效果不好，则进一步将所述肠癌患者分型为IML1型或IML2型。
  6. 一种对FOLFOX响应性进行分型的方法，其特征在于，包括步骤：
    a)提供一来自待测对象的测试样本，所述的测试样本选自下组：血液样本或组织样本(如手术中切下的肠癌组织)；
    b)检测所述测试样本中FOLFOX响应性标志物的mRNA和/或蛋白的浓度，记为C1；和
    c)将所述的FOLFOX响应性标志物的mRNA和/或蛋白的浓度C1与对照参比值C0进行比较，
    如果检测对象的FOLFOX响应性标志物的检测结果满足以下条件时，则提示所述肠癌患者为适合采用FOLFOX治疗的FOLFOX响应者：
    (i)当某一标志物在FOLFOX响应者中是上调的标志物，且所述标志物的表达水平高于参考值或标准值C0时；和/或
    (ii)当某一标志物在FOLFOX响应者中是下调的标志物，且所述标志物的表达水平低于参考值或标准值C0时；
    其中，所述的FOLFOX响应性标志物选自下组：
    (A)选自A1至A74的任一标志物、或其组合；
    (B)选自B1至B74的任一标志物、或其组合；
    (C)A1至A74中一个或多个标志物与B1至B74中一个或多个标志物所构成的组合。
  7. 如权利要求6所述的方法，其特征在于，在步骤(c)中，还包括：如果检测结果提示所述肠癌患者不适合采用FOLFOX治疗，和/或采用FOLFOX治疗的效果不好，则进一步将所述肠癌患者分型为IML1型或IML2型；
    其中，当满足以下条件(Z1)和/或(Z2)时，则将所述肠癌患者确定为FOLFOX化疗的无响应IML1型：
    (Z1)当表A某一标志物在FOLFOX响应者中是上调的标志物，且所述标志物的表达水平低于参考值或标准值C0时；和/或
    (Z2)当表A某一标志物在FOLFOX响应者中是下调的标志物，且所述标志物的表达水平高于参考值或标准值C0时；
    当满足以下条件(Z3)和/或(Z4)时，则将所述肠癌患者确定为FOLFOX化疗的无 响应IML2型：
    (Z3)当表B某一标志物在FOLFOX响应者中是上调的标志物，且所述标志物的表达水平低于参考值或标准值C0时；和/或
    (Z4)当表B某一标志物在FOLFOX响应者中是下调的标志物，且所述标志物的表达水平高于参考值或标准值C0时。
  8. 一种FOLFOX响应性标志物的基因、mRNA、cDNA、或蛋白的用途，其特征在于，它们被用作判断FOLFOX响应性和/或评价FOLFOX治疗效果的标志物，和/或用作将FOLFOX化疗的无响应者进行分型的标志物；
    其中，所述的FOLFOX响应性标志物选自下组：
    (A)选自A1至A74的任一标志物、或其组合；
    (B)选自B1至B74的任一标志物、或其组合；
    (C)A1至A74中一个或多个标志物与B1至B74中一个或多个标志物所构成的组合。
  9. 一种用于对FOLFOX响应性进行分型的标志物集合(set)，其特征在于，所述的集合包括：
    (A)选自A1至A74的3-74种标志物；
    (B)选自B1至B74的3-74种标志物；
    (C)A1至A74中一个或多个标志物与B1至B74中一个或多个标志物所构成的组合，所述组合至少包含3种标志物。
  10. 一种对人大肠癌进行分型的方法，其特征在于，所述方法包括：
    (1)获得某一检测对象的表A中至少三个标志物基因在所述样品中的mRNA基因表达水平；
    (2)归一化表A中至少三个标志物基因的基因表达值；
    (3)计算表A中所述至少三个标志物基因与这些基因在FOLFOX应答者中的平均基因表达值的第一相似度；以及计算表A中所述至少三个标志物基因与这些基因在FOLFOX无应答者中的平均基因表达值的第二相似度；
    (4)分别针对各标志物基因，计算第一相似度与第二相似度的差值；
    (5)基于所述差值，对大肠癌患者进行分型，所述分型将检测对象划分为针对化疗的响应者或无响应者(即IML1型)；
    其中，所述的化疗包括：FOLFOX、5-FU治疗、奥沙利铂治疗、或靶向DNA复制的化学治疗。
  11. 一种对人大肠癌进行分型的方法，其特征在于，所述方法包括：
    (1)获得某一检测对象的表B中至少三个标志物基因在所述样品中的mRNA基因表达水平；
    (2)归一化表B中至少三个标志物基因的基因表达值；
    (3)计算表B中所述至少三个标志物基因与这些基因在FOLFOX应答者中的平均基因表达值的第一相似度；以及计算表B中所述至少三个标志物基因与这些基因在FOLFOX无应答者中的平均基因表达值的第二相似度；
    (4)分别针对各标志物基因，计算第一相似度与第二相似度的差值；
    (5)基于所述差值，对大肠癌患者进行分型，所述分型将检测对象划分为针对化疗的响应者或无响应者(即IML2型)；
    其中，所述的化疗包括：FOLFOX、5-FU治疗、奥沙利铂治疗、或靶向DNA复制的化学治疗。
  12. 一种用于对大肠癌进行分型的设备，其特征在于，包括：
    (P1)输入单元，所述输入单元用于输入某一对象的表A或表B中至少三个标志物基因在所述样品中的mRNA基因表达水平的数据；
    (P2)数据处理单元，所述数据处理单元对输入的mRNA基因表达水平的数据进行处理，并且所述数据处理单元包括归一化处理子单元、相似度计算子单元和相似度差值计算子单元；
    其中，所述的归一化处理子单元用于对表A或表B中至少三个标志物基因的基因表达值进行归一化处理；
    所述的相似度计算子单元用于计算表A或表B中所述至少三个标志物基因与这些基因在FOLFOX应答者中的平均基因表达值的第一相似度；以及计算表A或表B中所述至少三个标志物基因与这些基因在FOLFOX无应答者中的平均基因表达值的第二相似度；
    所述的相似度差值计算子单元用于计算各标志物基因的第一相似度与第二相似度的差值；
    (P3)分型单元，所述的分型单元基于所述各标志物基因的差值，对所述对象进行分型，所述分型将检测对象划分为针对化疗的响应者或无响应者，从而获得分型结果；和
    (P4)输出设备，所述的输出设备用于输出所述的分型结果。
PCT/CN2021/098716 2020-06-09 2021-06-07 用于选择化疗响应患者组的生物标志物和方法及其用途 WO2021249355A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN202010526422.8 2020-06-09
CN202010526422.8A CN113774130A (zh) 2020-06-09 2020-06-09 用于选择化疗响应患者组的生物标志物和方法及其用途

Publications (1)

Publication Number Publication Date
WO2021249355A1 true WO2021249355A1 (zh) 2021-12-16

Family

ID=78835039

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2021/098716 WO2021249355A1 (zh) 2020-06-09 2021-06-07 用于选择化疗响应患者组的生物标志物和方法及其用途

Country Status (2)

Country Link
CN (1) CN113774130A (zh)
WO (1) WO2021249355A1 (zh)

Families Citing this family (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN114427003B (zh) * 2022-01-27 2022-08-30 中山大学附属第一医院 Nhp2在预测癌症放疗敏感性和预后中的应用

Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2016046640A2 (en) * 2014-09-26 2016-03-31 Medical Prognosis Institute A/S Methods for predicting drug responsiveness
US20190137495A1 (en) * 2017-06-04 2019-05-09 Rappaport Family Institute for Research in the Me dical Sciences Method of Predicting Personalized Response to Cancer Therapy, Method of Treating Cancer, and Kit Therefor

Family Cites Families (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
EP2202320A1 (en) * 2008-12-24 2010-06-30 Agendia B.V. Methods and means for typing a sample comprising colorectal cancer cells
EP2759605B1 (en) * 2013-01-25 2018-11-14 Signature Diagnostics AG A method for predicting a manifestation of an outcome measure of a cancer patient
CN103710451B (zh) * 2013-12-26 2015-06-24 上海锐赛生物技术有限公司 Pik3c2g在结直肠癌化疗疗效判断和检测试剂盒中的应用
CN103695560B (zh) * 2014-01-09 2016-09-14 上海交通大学医学院附属瑞金医院 Ppp1r12a基因在结直肠癌化疗疗效判断和检测试剂盒中的应用
CN104878083B (zh) * 2015-03-04 2018-05-22 上海市杨浦区中心医院 检测folfox化疗药物对于结直肠癌有效性的试剂盒
CN109988708B (zh) * 2019-02-01 2022-12-09 碳逻辑生物科技(中山)有限公司 一种用于对患有结肠直肠癌的患者进行分型的系统

Patent Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2016046640A2 (en) * 2014-09-26 2016-03-31 Medical Prognosis Institute A/S Methods for predicting drug responsiveness
US20190137495A1 (en) * 2017-06-04 2019-05-09 Rappaport Family Institute for Research in the Me dical Sciences Method of Predicting Personalized Response to Cancer Therapy, Method of Treating Cancer, and Kit Therefor

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
LU WEI, FU DONGLIANG, KONG XIANGXING, HUANG ZHIHENG, HWANG MAXWELL, ZHU YINGSHUANG, CHEN LIUBO, JIANG KAI, LI XINLIN, WU YIHUA, LI: "FOLFOX treatment response prediction in metastatic or recurrent colorectal cancer patients via machine learning algorithms", CANCER MEDICINE, JOHN WILEY & SONS LTD., GB, vol. 9, no. 4, 1 February 2020 (2020-02-01), GB , pages 1419 - 1429, XP055878491, ISSN: 2045-7634, DOI: 10.1002/cam4.2786 *

Also Published As

Publication number Publication date
CN113774130A (zh) 2021-12-10

Similar Documents

Publication Publication Date Title
AU2016281766B2 (en) Epigenetic chromosome interactions
US10260097B2 (en) Method of using a gene expression profile to determine cancer responsiveness to an anti-angiogenic agent
EP2925885B1 (en) Molecular diagnostic test for cancer
US20200327962A1 (en) Statistical ai for advanced deep learning and probabilistic programing in the biosciences
AU2012261820A1 (en) Molecular diagnostic test for cancer
US20160063179A1 (en) System for predicting prognosis of locally advanced gastric cancer
Pan et al. Transcriptome analyses reveal molecular mechanisms underlying phenotypic differences among transcriptional subtypes of glioblastoma
Su et al. Identification of the key genes and pathways in esophageal carcinoma
Lee et al. Meta-analysis of tumor stem-like breast cancer cells using gene set and network analysis
WO2021249355A1 (zh) 用于选择化疗响应患者组的生物标志物和方法及其用途
Li et al. Significance of pyroptosis-related gene in the diagnosis and classification of rheumatoid arthritis
Cheng et al. Identification of differential intestinal mucosa transcriptomic biomarkers for ulcerative colitis by bioinformatics analysis
Hao et al. Essential genes and MiRNA–mRNA network contributing to the pathogenesis of idiopathic pulmonary arterial hypertension
Yan et al. A prognostic model based on seven immune-related genes predicts the overall survival of patients with hepatocellular carcinoma
US20170088902A1 (en) Expression profiling for cancers treated with anti-angiogenic therapy
Li et al. Noninvasive Identification of Immune‐Related Biomarkers in Hepatocellular Carcinoma
US20120129711A1 (en) Biomarkers for the prognosis and high-grade glioma clinical outcome
US20220290243A1 (en) Identification of patients that will respond to chemotherapy
Feng et al. A Novel Biomarker Based on miRNA to Predict the Prognosis of Muscle‐Invasive Bladder Urothelial Carcinoma
Tseng et al. Investigating multi-cancer biomarkers and their cross-predictability in the expression profiles of multiple cancer types
US20240150453A1 (en) Methods of predicting response to anti-tnf blockade in inflammatory bowel disease
WO2015181556A1 (en) Pro-angiogenic expression signature
Zhang et al. Identification of common mechanisms and biomarkers of atrial fibrillation and heart failure based on machine learning
Theodorou Examination of pathway crosstalk and functional modules in papillary thyroid cancer dedifferentiation to anaplastic thyroid cancer
CN116426635A (zh) 用于预测ii期结直肠癌复发风险的生物标志物和方法及其用途

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 21821745

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 21821745

Country of ref document: EP

Kind code of ref document: A1