Detection Method

Abstract

This paper provides materials and methods for detecting colorectal neoplasia and colon cancer based on the expression level of stool-derived eukaryotic RNA biomarkers in the eukaryotic nucleic acid present in the stool samples from subjects (e.g., patients).This method can be used to detect high-risk adenoma and colorectal neoplasia molecular subtypes.

Description

[Technical field]

[0001] FIELD OF THEINVENTION The present invention relates to the extraction of eukaryotic nucleic acid from stool samples and the use of said nucleic acid for the diagnosis and treatment of intestinal disease. [Background technology]

[0002] background Gastrointestinal disorders, such as gastrointestinal cancer and other digestive disorders, such as ulcerative colitis, irritable bowel syndrome and Crohn's disease, are widespread. In the United States, gastrointestinal disorders are estimated to affect 60-70 million people annually. For some disorders, early screening and diagnosis have led to reduced mortality and improved quality of life for patients. However, standard diagnostic methods, such as colonoscopy, are invasive, time-consuming and relatively expensive. There is a continuing need for non-invasive methods to diagnose gastrointestinal disorders in both humans and animals. Summary of the Invention

[0003] overview As used herein, the level of expression of 2, 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, or 29 stool-derived eukaryotic RNA biomarkers selected from the biomarkers listed in Table 1 or Table 2, or a combination of Tables 1 and 2, in eukaryotic nucleic acid extracted from a stool sample from a subject is measured. comparing the measured expression levels of the 2, 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, or 29 stool-derived eukaryotic RNA biomarkers in the stool sample with the measured expression levels of the 2, 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, or 29 stool-derived eukaryotic RNA biomarkers in a control; comparing the measured expression levels of 2, 23, 24, 25, 26, 27, 28 or 29 stool-derived eukaryotic RNA biomarkers in the control to the measured expression levels of the 2, 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 or 29 stool-derived eukaryotic RNA biomarkers in the control; A difference in the measured expression levels of the 2, 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 or 29 stool-derived eukaryotic RNA biomarkers in the stool sample is indicative of the subject having colorectal neoplasia.Also provided is a method of detecting colorectal neoplasia in a subject comprising the steps of: measuring a variant allele frequency of one or more variant biomarker genes selected from the biomarker genes listed in Table 3 in eukaryotic nucleic acid extracted from a stool sample derived from the subject; comparing the measured variant allele frequency of the one or more variant biomarker genes in the stool sample with a measured variant allele frequency of the one or more variant biomarker genes in a control, wherein a difference in the variant allele frequency of the one or more variant biomarker genes relative to the variant allele frequency of the one or more variant biomarker genes in the control indicates that the subject has or is at risk for colorectal cancer. Also provided is a method of detecting a molecular subtype of colorectal cancer in a subject comprising the steps of: measuring a level of expression in eukaryotic nucleic acid extracted from a stool sample from a subject of two or more biomarker genes selected from any of the colorectal neoplasia molecular subtype biomarker genes listed in Table 4; comparing the measured expression levels of the two or more colorectal neoplasia molecular subtype biomarker genes in the biological sample with the measured expression levels of the two or more colorectal neoplasia molecular subtype biomarker genes in a control, wherein a difference in the measured expression levels of the two or more colorectal neoplasia molecular subtype biomarker genes in the biological sample from the measured expression levels of the two or more colorectal neoplasia molecular subtype biomarker genes in the control is indicative of the molecular subtype of colorectal cancer. [The present invention 1001] a) measuring the level of expression in eukaryotic nucleic acid extracted from a stool sample from the subject of 2, 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 or 29 stool-derived eukaryotic RNA biomarkers selected from the biomarkers listed in Table 1 or Table 2, or a combination of Tables 1 and 2; b) comparing the measured expression levels of the 2, 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, or 29 stool-derived eukaryotic RNA biomarkers in the stool sample with the measured expression levels of the 2, 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, or 29 stool-derived eukaryotic RNA biomarkers in a control; a difference in the measured expression levels of the 2, 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, or 29 stool-derived eukaryotic RNA biomarkers in the stool sample relative to the measured expression levels of the 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, or 29 stool-derived eukaryotic RNA biomarkers in the stool sample is indicative of the subject having colorectal neoplasia, 16. A method for detecting colorectal neoplasia in a subject, comprising: [The present invention 1002] The method of claim 1001, wherein the subject is a human. [The present invention 1003] The method of claim 1001, wherein the colorectal neoplasia is selected from the group consisting of colorectal cancer, high risk adenoma, intermediate risk adenoma and low risk adenoma. [The present invention 1004] The method of claim 1001, wherein the stool-derived eukaryotic RNA biomarkers are selected from the group consisting of ACY1, TNFRSF10B, DST, EGLN2, PER3, CTNNB1, ACHE, SMAD4, EDN1, ERBB2 and GAPDH. [The present invention 1005] 1001. The method of claim 1001, wherein said nucleic acid comprises RNA, total RNA, mRNA, seRNA, tRNA, rRNA, ncRNA, smRNA or snoRNA, or any combination of RNA, total RNA, mRNA, seRNA, tRNA, rRNA, ncRNA, smRNA or snoRNA. [The present invention 1006] The method of claim 1001, wherein said expression level is measured by nucleic acid sequencing, microarray sequencing, molecular barcoding, amplicon sequencing, probe capture, polymerase chain reaction (PCR), ddPCR, dPCR, RT-PCR or RT-qPCR. [The present invention 1007] The method of claim 1001, further comprising determining demographic information of the subject. [The present invention 1008] The method of claim 1001, further comprising the step of administering a fecal immunochemical test (FIT) to the subject. [The present invention 1009] 1. A method for selecting a clinical regimen for a subject having or at risk for colorectal neoplasia, comprising: a) measuring the level of expression in eukaryotic nucleic acid present in a stool sample from the subject of 2, 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 or 29 stool-derived eukaryotic RNA biomarkers selected from the biomarkers listed in Table 1 or Table 2, or a combination of Tables 1 and 2; b) comparing the measured expression levels of the 2, 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, or 29 stool-derived eukaryotic RNA biomarkers in the stool sample with the measured expression levels of the 2, 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, or 29 stool-derived eukaryotic RNA biomarkers in a control; a difference in the measured expression levels of the 2, 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, or 29 stool-derived eukaryotic RNA biomarkers in the stool sample relative to the measured expression levels of 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, or 29 stool-derived eukaryotic RNA biomarkers indicates that the subject has or is at risk for colorectal neoplasia; c) performing a diagnostic procedure or treatment or a combination of a diagnostic procedure and treatment The method comprising: [The present invention 1010] The method of claim 1009, wherein the subject is a human. [The present invention 1011] The method of claim 1009, wherein the colorectal neoplasia is selected from the group consisting of colorectal cancer, high risk adenoma, intermediate risk adenoma and low risk adenoma. [The present invention 1012] 1009. The method of claim 1009, wherein the stool-derived eukaryotic RNA biomarkers are selected from the group consisting of ACY1, TNFRSF10B, DST, EGLN2, PER3, CTNNB1, ACHE, SMAD4, EDN1, ERBB2 and GAPDH. [The present invention 1013] 1009. The method of claim 1009, wherein said nucleic acid comprises RNA, total RNA, mRNA, seRNA, tRNA, rRNA, ncRNA, smRNA or snoRNA, or any combination of RNA, total RNA, mRNA, seRNA, tRNA, rRNA, ncRNA, smRNA or snoRNA. [The present invention 1014] The method of claim 1009, wherein said expression level is measured by nucleic acid sequencing, microarray sequencing, molecular barcoding, amplicon sequencing, probe capture, polymerase chain reaction (PCR), ddPCR, dPCR, RT-PCR or RT-qPCR. [The present invention 1015] The method of any one of claims 10 to 12, further comprising determining demographic information of the subject. [The present invention 1016] The method of claim 1009, further comprising the step of administering a fecal immunochemical test (FIT) to the subject. [The present invention 1017] The method of any one of claims 10 to 12, wherein the clinical protocol comprises a diagnostic procedure or treatment. [The present invention 1018] The method of claim 1009, wherein the diagnostic procedure comprises a colonoscopy. [The present invention 1019] The method of claim 1009, wherein the treatment comprises surgery, chemotherapy, radiation therapy, targeted therapy or immunotherapy. [The present invention 1020] The method of claim 1019, wherein the chemotherapy comprises administration of 5-fluorouracil, leucovorin, capecitabine, oxaliplatin, irinotecan, or a combination thereof. [The present invention 1021] The method of the present invention 1019, wherein the targeted therapy comprises administration of bevacizumab (anti-VEGF), ramucilamab (anti-VEGFR2), aflibercept, regorafenib, cetuximab (anti-EGFR), panitumumab, tripfluridine-tipiracil, or a combination thereof. [The present invention 1022] a) generating a random forest model configured to determine the presence of colorectal neoplasia based on a predetermined set of features comprising expression levels of 2, 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 or 29 stool-derived eukaryotic RNA biomarkers selected from the biomarkers listed in Table 1 or Table 2, or a combination of Tables 1 and 2; b) measuring in the subject the expression level of 2, 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, or 29 stool-derived eukaryotic RNA biomarkers selected from the biomarkers listed in Table 1 or Table 2, or a combination of Tables 1 and 2; and c) determining the presence of colorectal neoplasia using the random forest model based on the expression levels in the subject of 2, 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, or 29 stool-derived eukaryotic RNA biomarkers selected from the biomarkers listed in Table 1 or Table 2, or a combination of Tables 1 and 2. 16. A method for detecting colorectal neoplasia in a subject, comprising: [The present invention 1023] The method of the present invention 1022, wherein the predetermined set of features includes population statistical information. [The present invention 1024] The method of the present invention 1022, comprising applying a positive immunochemical fecal test (FIT) forcing model by a random forest model. [The present invention 1025] The method of the present invention 1022, wherein the 2, 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 or 29 stool-derived eukaryotic RNA biomarkers are selected from the group consisting of ACY1, TNFRSF10B, DST, EGLN2, PER3, CTNNB1, ACHE, SMAD4, EDN1, ERBB2 and GAPDH. [The present invention 1026] a) determining that a stool sample from a patient expresses increased levels of 2, 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, or 29 stool-derived eukaryotic RNA biomarkers selected from the biomarkers listed in Table 1 or Table 2, or a combination of Tables 1 and 2, in eukaryotic nucleic acids present in the stool sample from the subject; and b) administering to the subject a colonoscopy, surgery, chemotherapy, radiation therapy, targeted therapy, or immunotherapy. 23. A method of treating colorectal neoplasia in a subject, comprising: [The present invention 1027] a) measuring variant allele frequencies of one or more variant biomarker genes selected from the biomarker genes listed in Table 3 in eukaryotic nucleic acid extracted from a stool sample from the subject; b) comparing the measured variant allele frequency of the one or more variant biomarker genes in the stool sample to the measured variant allele frequency of the one or more variant biomarker genes in a control, wherein a difference in the variant allele frequency of the one or more variant biomarker genes relative to the variant allele frequency of the one or more variant biomarker genes in the control is indicative of the subject having or at risk of colorectal cancer. 16. A method for detecting colorectal neoplasia in a subject, comprising: [The present invention 1028] a) measuring variant allele frequencies of one or more variant biomarker genes selected from the biomarker genes listed in Table 3 in eukaryotic nucleic acid extracted from a stool sample from the subject; b) comparing the measured variant allele frequency of the one or more variant biomarker genes in the stool sample to the measured variant allele frequency of the one or more variant biomarker genes in a control, wherein a difference in the variant allele frequency of the one or more variant biomarker genes relative to the variant allele frequency of the one or more variant biomarker genes in the control is indicative of the subject having or at risk of colorectal cancer. 4. A method for determining whether a subject has or is at risk for colorectal cancer, comprising: [The present invention 1029] 1. A method for selecting a clinical regimen for a subject having or at risk for colorectal cancer, comprising: a) detecting, in a eukaryotic nucleic acid in a biological sample from the subject, one or more variant alleles of a biomarker gene selected from the biomarker genes listed in Table 3, wherein the variants are associated with colorectal cancer tumorigenesis; b) performing a diagnostic procedure or treatment or a combination of a diagnostic procedure and treatment The method comprising: [The present invention 1030] a) detecting, in eukaryotic nucleic acid present in a stool sample from a subject, one or more variant alleles of a biomarker gene selected from the biomarker genes listed in Table 3, wherein the variants are associated with colorectal cancer tumorigenesis; and b) administering to the subject a colonoscopy, surgery, chemotherapy, radiation therapy, targeted therapy, or immunotherapy. 23. A method of treating colorectal cancer in a subject, comprising: [The present invention 1031] The method of any of claims 1027 to 1030, wherein the variant allele comprises a silent mutation, a missense mutation, an insertion, a deletion, a frameshift mutation and / or a nonsense mutation. [The present invention 1032] The method according to any one of claims 1027 to 1031, wherein the biological sample is a stool sample. [The present invention 1033] The method according to any one of claims 1027 to 1032, wherein the stool sample is a human stool sample. [The present invention 1034] The method of any of claims 1027 to 1033, wherein the nucleic acid comprises cDNA, RNA, total RNA, mRNA, seRNA, tRNA, rRNA, ncRNA, smRNA or sno RNA, or any combination of cDNA, RNA, total RNA, mRNA, seRNA, tRNA, rRNA, ncRNA, smRNA or sno RNA. [The present invention 1035] The method of any one of claims 1027 to 1034, wherein the expression level is measured by nucleic acid sequencing, microarray sequencing, molecular barcoding, probe capture, polymerase chain reaction (PCR), ddPCR, RT-PCR or RT-qPCR. [The present invention 1036] The method of any of claims 1027 to 1035, wherein said clinical regimen comprises one or more of surgery, chemotherapy, radiation therapy, immunotherapy and targeted therapy. [The present invention 1037] a) measuring the level of expression in eukaryotic nucleic acid extracted from a stool sample from the subject of two or more biomarker genes selected from any of the colorectal neoplasia molecular subtype biomarker genes listed in Table 4; b) comparing the measured expression levels of the two or more colorectal neoplasia molecular subtype biomarker genes in a biological sample with the measured expression levels of the two or more colorectal neoplasia molecular subtype biomarker genes in a control, wherein a difference in the measured expression levels of the two or more colorectal neoplasia molecular subtype biomarker genes in the biological sample relative to the measured expression levels of the two or more colorectal neoplasia molecular subtype biomarker genes in the control is indicative of the molecular subtype of colorectal cancer. 1. A method for detecting a molecular subtype of colorectal cancer in a subject, comprising: [The present invention 1038] a) measuring the level of expression in eukaryotic nucleic acid extracted from a stool sample from the subject of two or more biomarker genes selected from any of the colorectal neoplasia molecular subtype biomarker genes listed in Table 4; b) comparing the measured expression levels of the two or more colorectal neoplasia molecular subtype biomarker genes in a biological sample with the measured expression levels of the two or more colorectal neoplasia molecular subtype biomarker genes in a control, wherein a difference in the measured expression levels of the two or more colorectal neoplasia molecular subtype biomarker genes in the biological sample from the measured expression levels of the two or more colorectal neoplasia molecular subtype biomarker genes in the control indicates that the subject has or is at risk for a molecular subtype of colorectal cancer. 23. A method for determining whether a subject has or is at risk for a molecular subtype of colorectal cancer, comprising: [The present invention 1039] a) measuring the level of expression in eukaryotic nucleic acid in a stool sample from a patient of two or more biomarker genes selected from any of the colorectal neoplasia molecular subtype biomarker genes listed in Table 4 that are associated with genomic instability, microsatellite instability, or immune infiltration; b) comparing the measured expression levels of the two or more colorectal neoplasia molecular subtype biomarker genes in a biological sample with the measured expression levels of the two or more colorectal neoplasia molecular subtype biomarker genes in a control, wherein an altered level of the colorectal neoplasia molecular subtype biomarker genes associated with microsatellite instability or immune infiltration relative to the measured levels of the two or more genes in a control indicates that the patient is a candidate for immunotherapy; and c) administering immunotherapy 23. A method of treating colorectal cancer in a patient, comprising: [The present invention 1040] The method of claim 1039, wherein the immunotherapy is an inhibitory checkpoint molecule or an immune checkpoint inhibitor. [The present invention 1041] 1. A method for selecting a clinical regimen for a subject having or at risk for colorectal cancer, comprising: a) measuring the level of expression in eukaryotic nucleic acid in a biological sample from the subject of two or more biomarker genes selected from any of the colorectal neoplasia molecular subtype biomarker genes listed in Table 4; b) comparing the measured expression levels of the two or more colorectal neoplasia molecular subtype biomarker genes in the biological sample with the measured expression levels of the two or more colorectal neoplasia molecular subtype biomarker genes in a control, thereby identifying the molecular subtype of colorectal cancer; c) performing a diagnostic procedure or treatment or a combination of a diagnostic procedure and treatment The method comprising: [The present invention 1042] a) determining that a stool sample from a patient expresses an increased level of any of the colorectal neoplasia molecular subtype biomarker genes listed in Table 4 in eukaryotic nucleic acids present in a stool sample from the subject; and b) administering to the patient a colonoscopy, surgery, chemotherapy, radiation therapy, targeted therapy or immunotherapy. 23. A method for treating colorectal neoplasia comprising: [The present invention 1043] The method of any of claims 1037 to 1042, wherein the colorectal neoplasia molecular subtype is selected from the group consisting of CMS1, CMS2, CMS3 and CMS4. [The present invention 1044] The method according to any one of claims 1037 to 1043, wherein the biological sample is a stool sample. [The present invention 1045] The method according to any one of claims 1037 to 1044, wherein the stool sample is a human stool sample. [The present invention 1046] The method of any of claims 1037 to 1045, wherein the nucleic acid comprises cDNA, RNA, total RNA, mRNA, seRNA, tRNA, rRNA, ncRNA, smRNA or snoRNA, or any combination of cDNA, RNA, total RNA, mRNA, seRNA, tRNA, rRNA, ncRNA, smRNA or snoRNA. [The present invention 1047] The method of any one of claims 1037 to 1046, wherein the expression level is measured by nucleic acid sequencing, microarray sequencing, molecular barcoding, probe capture, polymerase chain reaction (PCR), ddPCR, RT-PCR or RT-qPCR. [The present invention 1048] The method of any of claims 1037 to 1047, wherein said clinical regimen comprises one or more of surgery, chemotherapy, radiation therapy, immunotherapy and targeted therapy. [The present invention 1049] a) determining that a stool sample from a patient contains, relative to a control, an increased level of two or more stool-derived eukaryotic biomarkers selected from the biomarkers listed in Table 1 or Table 2, or a combination of Tables 1 and 2, or Table 3 or Table 4, in eukaryotic nucleic acid present in a stool sample from the subject; b) determining, based on said determining, that said patient is at risk for colorectal neoplasia; and c) performing a colonoscopy on the patient. The method includes: [The present invention 1050] The method of any one of claims 1001 to 1049, further comprising the step of preparing a stool sample. [Brief description of the drawings]

[0004] These and other features and advantages of the present invention will be more fully disclosed, or will become apparent, in the following detailed description of the preferred embodiments of the invention, which should be considered in conjunction with the accompanying drawings, in which like numerals indicate like parts and in which:

[0005] [Figure 1] Figure 1A shows a separation of electrophoretic files. Electrophoretic analysis was used to check the quality of RNA extracted based on the method described in the literature. Figure 1B shows a separation of electrophoretic files. Electrophoretic analysis was used to check the quality of RNA extracted based on the method described herein. [Diagram 2]Figure 2A shows the separation of electrophoretic files. Electrophoretic analysis was used to check the quality of seRNA in samples extracted immediately without incubation in stabilization buffer. Figure 2B shows the separation of electrophoretic files. Electrophoretic analysis was used to check the quality of seRNA in samples incubated in stabilization buffer and stored at room temperature for 24 hours before extraction. Figure 2C shows the separation of electrophoretic files. Electrophoretic analysis was used to check the quality of seRNA in samples incubated in stabilization buffer and stored at room temperature for 48 hours before extraction. [Diagram 3] Figure 3A shows the ROC analysis in different patient populations obtained during internal validation of the SVM, and Figure 3B shows the sensitivity of prediction with the SVM used in an independent test set. [Figure 4] Figure 4A is a table listing 274 colorectal neoplasm molecular subtype biomarker genes used in the Colorectal Cancer Subtyping Consortium classifier. Figure 4B is a table listing 25 exemplary colorectal neoplasm molecular subtype biomarker genes useful for identifying colorectal cancer subtype CMS1. [Diagram 5] Figure 5 is a heatmap summarizing the stratification of patients by CMS (consensus molecular subtype) of colorectal cancer using the Colorectal Cancer Subtyping Consortium classifier. [Figure 6] Figure 6 shows the correlation of four biological replicate pairs comparing transcript expression levels of 398 genes as measured by the Affymetrix Human Transcriptome Array 2.0 and the Illumina Targeted RNA Custom Panel. [Figure 7] FIG. 7 is a principal component analysis graph showing hierarchical clustering of 13 patients with colorectal cancer, adenoma and no neoplastic findings. [Figure 8] FIG. 8 shows six putative somatic variants identified in stool samples from human subjects diagnosed with adenoma and colorectal cancer. [Figure 9] FIG. 9 is a table listing biomarkers associated with cancer, colorectal neoplasia, and / or gastrointestinal health in which putative somatic variants may be identified. [Figure 10] FIG. 10 is a table summarizing patient demographics and processing metrics associated with the prospective training set, the prospective holdout test set, the retrospective holdout test set, and the total study cohort. [Figure 11A] FIG. 11A is a flowchart of candidate feature selection using bootstrapping of the test set. [Figure 11B] FIG. 11B is a graph of the selected candidate features. [Figure 12] FIG. 12 is a graph of raw GAPDH levels in patients with no colonoscopy findings, benign polyps, low risk adenomas, intermediate risk adenomas, high risk adenomas, and colorectal cancer. [Figure 13] FIG. 13 is a graph showing the performance of the model for detection of high-risk adenomas (HRA) based on internal cross-validation (n=154 patients). [Figure 14] Figure 14 is a table showing the features with their Gini importance ranking. [Figure 15] Figure 15A is a graph showing the performance of a model for detection of HRA based on an independent hold-out test set (n=110 patients) without and with FIT features. Figure 15B is a graph showing the performance of a model for detection of HRA based on an independent hold-out test set (n=110 patients) with FIT features. [Figure 16] Figure 16A shows model predictions sorted by disease severity without and with the FIT feature. Figure 16B shows model predictions sorted by disease severity with and without the FIT feature. [Figure 17]Figure 17A is a graph showing the results of incremental downsampling analysis without the Fecal Immunochemical Testing (FIT) feature, and Figure 17B is a graph showing the results of incremental downsampling analysis with the Fecal Immunochemical Testing (FIT) feature. [Figure 18] FIG. 18 is a graph showing the performance of the model on all samples in the holdout test set, which contained 11 additional colorectal cancer (CRC) samples. [Figure 19] FIG. 19 is a graph showing the performance of the model on all samples in the hold-out test set, which includes 11 additional colorectal cancer (CRC) samples, extrapolated to the generalized screening population. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0006] Detailed Description This description of the preferred embodiment is intended to be read in conjunction with the accompanying drawings, which are to be considered as part of the entire written description of the invention. The drawing figures are not necessarily to scale, and certain features of the invention may be shown in exaggerated scale or somewhat schematic form for clarity and conciseness. In this description, relative terms such as "horizontal", "vertical", "upper", "lower", "upper" and derivatives thereof (e.g., "horizontally", "downwards", "upwardly", etc.) should be interpreted as indicating the exact orientation at the time of describing the particular drawing figure under discussion or as shown in the particular drawing figure under discussion. Such relative terms are for convenience of description and are not usually intended to require a particular orientation. For example, terms such as "inwardly" versus "outwardly", "longitudinal" versus "transversely" should be interpreted relative to each other or relative to a longitudinal axis or axis or center of rotation, as appropriate. Terms such as coupling in connection, e.g., "connected" and "interconnected," refer to both a relationship in which structures are fixed or connected to one another either directly or indirectly through an intervening structure, as well as a movable or fixed connection or relationship, unless expressly stated otherwise. The term "operably connected" refers to a connection, coupling, or linkage that enables the directly related structures to function as intended by virtue of the relationship. Where only one machine is illustrated, the term "machine" should be construed to encompass any machine assembly that alone or in concert executes a set (or sets) of instructions for carrying out any one or more of the methodologies discussed herein. In the claims, means-plus-function clauses, when used, are intended to encompass structures for performing the functions described, suggested, or obvious in the written description or drawings, e.g., structural equivalents as well as equivalent structures.

[0007] The present invention is based in part on the inventors' development of a method for separating eukaryotic cells from bacterial cells in a stool sample, such as a stool sample taken from a mammal. In the colon, there are approximately 1×10 13 There are approximately 1000 bacterial cells in the colon. This colonic microbiota can include 300-1000 species. Stool or fecal samples are complex macromolecular mixtures that contain not only eukaryotic cells shed from the intestinal lumen of the gastrointestinal tract, but also microorganisms, e.g. bacteria and any gastrointestinal parasites, indigestible unabsorbed food residues, secretions from enterocytes and excretions, e.g. mucus and pigments. Normal stool is composed of approximately 75% water and 25% solids. Bacteria make up approximately 60% of the total dry mass of feces. High bacterial loads can contribute to a poor signal-to-noise ratio for the detection of eukaryotic biomarkers from stool samples. Furthermore, eukaryotic signals can be severely degraded. Such extraction and processing of eukaryotic nucleic acids can promote or accelerate degradation, which severely limits further analysis.

[0008] The extraction method allows for the isolation of high-quality eukaryotic RNA from stool samples.The method is described in International Application WO2018 / 081580, the entirety of which is incorporated herein by reference.The inventors may refer to stool-derived eukaryotic RNA (seRNA) to identify the eukaryotic RNA that is preserved during stool formation, and which is subsequently extracted from stool samples by the method disclosed in International Application WO2018 / 081580.

[0009] Thus, the inventors have developed materials and methods for non-invasively evaluating the transcriptome of human colorectal cancer and colorectal neoplasia.The materials and methods disclosed herein provide efficient and sensitive detection of eukaryotic nucleic acid in human stool samples.The inventors have found that colorectal neoplasia can be detected based on the expression levels and variants of stool-derived eukaryotic RNA biomarkers in eukaryotic nucleic acid present in stool samples from subjects.The detection method can be configured in a manner that is useful for detecting various forms and subtypes of colorectal cancer or colorectal neoplasia.

[0010] More specifically, the materials and methods disclosed herein can be used to detect high-risk adenoma (HRA) based on the expression level of stool-derived eukaryotic RNA biomarkers in eukaryotic nucleic acid present in stool samples from subjects.Disclosed herein is a model-based approach for predicting or identifying colorectal neoplasms, particularly high-risk adenomas.In some embodiments, the model can be based on the expression level of two or more stool-derived eukaryotic RNA biomarkers listed in Table 1 and Table 2 in eukaryotic nucleic acid present in stool samples from subjects. In some embodiments, the model may be based on the expression levels of two or more stool-derived eukaryotic RNA biomarkers, such as 2, 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, or 29, selected from the stool-derived eukaryotic RNA biomarkers listed in Table 1 or Table 2 or a combination of Tables 1 and 2. In some embodiments, the model may be based on the expression levels of two or more stool-derived eukaryotic RNA biomarkers, such as 2, 3, 4, 5, 6, 7, 8, 9, 10, or 11, selected from the stool-derived eukaryotic RNA biomarkers listed in Table 1. The model may also include demographic features, such as the age and smoking status of the subject. In some embodiments, the model may also include the results of a stool immunochemical test (FIT) administered to a stool sample from the subject. In some aspects, the materials and methods disclosed herein may be used to identify medium risk adenomas (MRA), low risk adenomas (LRA), or benign polyps.

[0011] Also provided are materials and methods for detecting colorectal cancer based on detection of variant biomarkers in eukaryotic nucleic acid in stool samples from subjects.In some embodiments, variant biomarkers can be associated with colorectal cancer tumorigenesis.Variants can be variants of any biomarkers listed in Table 3.Variants can be variants of colorectal cancer driver genes, such as TP53, KRAS, PIK3CA, BRAF, APC, BMP3, NDRG4, SMAD4, MLH1, CTNNB1, EGFR, BRCA1, CDKN2A, CDH1, PTEN, VEGFA, MAPK3 or NRAS.

[0012] The inventors have found that gene expression signatures associated with consensus molecular subtypes (CMS) as defined by the Colorectal Cancer Subtyping Consortium (CRCSC) can be effectively detected in stool-derived eukaryotic RNA. More specifically, the materials and methods disclosed herein could be used to isolate seRNA from stool samples that could indicate the presence of a particular subtype of colorectal cancer (e.g., CMS1) as defined by CRCSC. Approximately 14% of individuals diagnosed with colorectal cancer have a CMS1 classification. CMS1 tumors are characterized by microsatellite instability (MSI-H), hypermutation, and increased immune infiltrates. Such features are consistent with tumors in which the immune system plays an active role in detecting and monitoring tumor sites. Patients with such tumors may benefit from targeted immunotherapy, such as immune checkpoint blockade therapy. For example, Keytruda™ (pembrolizumab) and Opdivo™ (nivolumab) are both FDA-approved for the treatment of adult and pediatric patients with unresectable or metastatic solid tumors that are MSI-H and who have not benefited from first-line chemotherapy.

[0013] Therefore, herein, materials and methods are provided for determining whether a human subject with colorectal cancer has gene expression signature associated with CMS1.In the context of disease monitoring, the method can selectively identify this patient group non-invasively and provide treatment guidance using seRNA.The method can be effectively performed non-invasively using stool samples rather than blood samples or biopsy samples.The method is useful in the development of clinical planning and treatment methods for subjects with or at risk of colorectal cancer. In some embodiments, the two or more biomarkers may comprise 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 30, 35, 40, 45, 50, 60, 70, 80, 90, 100, 120, 140, 160, 180 or more combinations of the markers in Figure 4 or Table 4. In some embodiments, the markers may be included in differentially expressed transcript clusters and / or common pathways associated with colorectal cancer. Exemplary pathways include microsatellite instability (MSI), chromosomal instability (CIN) and CpG island methylator phenotype (CIMP). In some embodiments, the pathways may be cellular component pathways, cellular response to stress, stress and RNA binding pathways.

[0014] In the context of disease monitoring, the method can selectively identify patient population non-invasively and provide treatment guidance.The method can be effectively performed non-invasively using stool samples rather than blood samples or biopsy samples.The method is useful in developing clinical planning and treatment methods for subjects with colorectal neoplasia or colorectal cancer or at risk of colorectal neoplasia or colorectal cancer.

[0015] The methods and materials disclosed herein include methods for isolating eukaryotic nucleic acids from stool samples. Such eukaryotic nucleic acids can be evaluated for levels of specific biomarkers that may indicate gastrointestinal disorders or diseases, such as colorectal neoplasia or colorectal cancer, in eukaryotic organisms, such as mammals. The mammals can be human or non-human animals, such as humans, dogs, cats, non-human primates, ruminants, ursids, equines, pigs, sheep, goats, camels, buffalo, deer, elk, moose, weasels, rabbits, guinea pigs, hamsters, rats, mice, pachyderms, rhinos, or chinchillas.

[0016] The inventors have found that eukaryotic cells can be effectively separated from bacterial cells in eukaryotic stool samples. The inventors have also found that eukaryotic biomarkers can be detected in RNA isolated from such eukaryotic cells. Such biomarkers can be useful for detecting gastrointestinal disorders, such as colorectal cancer, celiac disease, Crohn's disease, ulcerative colitis, gastritis, gastroenteritis, gastric cancer, gastric ulcer, necrotizing enterocolitis, gastrointestinal stromal tumors, gastrointestinal lymphoma, gastrointestinal neoplasia, lymphosarcoma, adenoma, hyperplastic changes, adenocarcinoma, inflammatory bowel disease, irritable bowel syndrome, intrapancreatic neoplasia, intrahepatic neoplasia, cholangiocarcinoma, and colitis. Provided herein are materials and methods for determining whether a subject, such as a human, dog, or cat, is at risk for gastrointestinal disease, such as colorectal neoplasia, such as high-risk adenoma or colorectal cancer. Also provided are materials and methods for diagnosing disease and methods for identifying the health status of a subject.

[0017] The methods and compositions disclosed herein are widely and variously useful for the detection, diagnosis, classification and treatment of gastrointestinal disorders, such as colorectal neoplasms or colorectal cancer.The detection method can include measuring the expression level of one, two or more biomarkers in a stool sample from a sample from a subject, such as a patient, who has or is suspected of having a gastrointestinal disorder, and comparing the measured expression level with the measured expression level of the one, two or more biomarkers in a control.The difference in the measured expression level of one, two or more biomarkers in the sample of the subject relative to the measured expression level of the one, two or more biomarkers in the control is an indication that the subject has a gastrointestinal disorder.In some embodiments, the difference in the measured expression level of one, two or more biomarkers in the sample of the subject relative to the measured expression level of the one, two or more biomarkers in the control is an indication that the subject, such as a patient, is at risk of having a gastrointestinal disorder.

[0018] In some embodiments, the detection method comprises detecting 2, 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 or more stool-derived eukaryotic RNA biomarkers in a sample from a subject, e.g., a patient, having or suspected of having a gastrointestinal disorder, e.g., a colorectal neoplasm. The method may include measuring expression levels of the 2, 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 or more stool-derived eukaryotic RNA biomarkers in a control. A difference in the measured expression levels of the 2, 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 or more stool-derived eukaryotic RNA biomarkers in a subject sample relative to the measured expression levels of the 2, 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 or more stool-derived eukaryotic RNA biomarkers in a control is an indication that the subject has a gastrointestinal disorder, e.g., a colorectal neoplasm. In some embodiments, a difference in the measured expression levels of the 2, 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 or more stool-derived eukaryotic RNA biomarkers in a subject sample relative to the measured expression levels of the 2, 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 or more stool-derived eukaryotic RNA biomarkers in a control is an indication that the subject, e.g., patient, is at risk for a gastrointestinal disorder, e.g., a colorectal neoplasm.In some embodiments, a difference in the measured expression levels of the 2, 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 or more stool-derived eukaryotic RNA biomarkers in a subject sample relative to the measured expression levels of the 2, 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 or more stool-derived eukaryotic RNA biomarkers in a control is an indication that the subject, e.g., patient, is at risk for a particular type of colorectal neoplasia, e.g., adenoma, more particularly high risk adenoma. In any of the above embodiments, the stool-derived eukaryotic RNA biomarkers may be selected from the stool-derived eukaryotic RNA biomarkers listed in Table 1 or Table 2, or a combination of Tables 1 and 2. Detection methods may also include analysis of variants of specific biomarkers.

[0019] In another embodiment, a method of detecting a disease may comprise determining the relative proportions, e.g., relative ratios, of expression levels of 2, 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 or more stool-derived eukaryotic RNA biomarkers in a stool sample from a subject, and comparing such relative proportions of stool-derived eukaryotic RNA biomarkers to the relative proportions of expression levels of the 2, 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 or more stool-derived eukaryotic RNA biomarkers in a control. A difference in the relative percentage of measured expression levels of the 2, 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 or more stool-derived eukaryotic RNA biomarkers in a subject sample relative to a control is an indication that the subject has a gastrointestinal disease, e.g., a colorectal neoplasm. In some embodiments, a difference in the ratio of measured expression levels of the 2, 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 or more stool-derived eukaryotic RNA biomarkers in a subject sample relative to the ratio of measured expression levels of the 2, 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 or more stool-derived eukaryotic RNA biomarkers in a control is an indication that the subject is at risk for a gastrointestinal disorder, e.g., a colorectal neoplasm.In some embodiments, a difference in the ratio of measured expression levels of the 2, 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 or more stool-derived eukaryotic RNA biomarkers in a subject sample relative to the ratio of measured expression levels of the 2, 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 or more stool-derived eukaryotic RNA biomarkers in a control is an indication that the subject is at risk for a particular type of colorectal neoplasia, e.g., adenoma, more particularly high risk adenoma. In any of the above embodiments, the stool-derived eukaryotic RNA biomarkers may be selected from the stool-derived eukaryotic RNA biomarkers listed in Table 1 or Table 2, or a combination of Tables 1 and 2. Detection methods may also include analysis of variants of specific biomarkers.

[0020] The method comprises determining whether levels of 2, 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 or more stool-derived eukaryotic RNA biomarkers are greater than or equal to the same 2, 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 or more stool-derived eukaryotic RNA biomarkers in a control. The method may include determining whether the level of expression of 2, 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 or more stool-derived eukaryotic RNA biomarkers in eukaryotic RNA isolated from a stool sample collected from a subject is different from the level of more stool-derived eukaryotic RNA biomarkers. Exemplary stool-derived eukaryotic RNA biomarkers are listed in Tables 1 and 2. Exemplary stool-derived eukaryotic RNA biomarkers may include ACY1, TNFRSFlOB, DST, EGLN2, PER3, CTNNB1, ACHE, SMAD4, EDN1, ERBB2, GAPDH.ABCB1, MAPK3, VEZF1, KRAS, PTEN, CREBBP, SUZ12, CDHR5, CABLES1 AREG, SPATA2, PPARGC1A, DBP, CDH1, PDGFA, OGG1, CGN, and TCF7L2.

[0021] Table 1. Fecal-derived eukaryotic RNA biomarkers TIFF0007650663000001.tif61128

[0022] Table 2. Fecal-derived eukaryotic RNA biomarkers TIFF0007650663000002.tif97128

[0023] In some embodiments, the stool-derived eukaryotic RNA biomarkers may also include a subset of the stool-derived eukaryotic RNA biomarkers listed in Table 1 and Table 2. Some or all of the stool-derived eukaryotic RNA biomarkers listed in Table 1 or Table 2 or a combination of Table 1 and Table 2 may constitute a panel. For example, some or all of the stool-derived eukaryotic RNA biomarkers in Table 1 may constitute a panel (Panel A). For example, Panel A may include some or all of the stool-derived eukaryotic RNA biomarkers ACY1, TNFRSFlOB, DST, EGLN2, PER3, CTNNB1, ACHE, SMAD4, EDN1, ERBB2, GAPDH. The composition may include a gene array and a probe set configured for specific detection of the marker panel disclosed herein. The composition may also include a kit with a gene array and a probe set configured for specific detection of the marker panel disclosed herein. The method can include identifying the presence of a variant in the nucleic acid sequence of a stool-derived eukaryotic RNA biomarker, such as a stool-derived eukaryotic RNA biomarker listed in Tables 1 and 2.

[0024] Also provided is a method for detecting variants in the nucleic acid sequence of a biomarker in eukaryotic nucleic acid (e.g., seRNA). The variant can be any mutation that contributes to tumor survival, tumor progression, or tumor metastasis. We may refer to such mutations as "driver mutations" or "progressor mutations". Such mutations may include silent mutations, missense mutations, insertions, deletions, frameshift mutations, or nonsense mutations. We may also refer to the expression of any particular variant as "variant allele frequency" (VAF). Such variants may include variants of any of the biomarkers listed in Figures 8 or 9. The variant may be a variant of a colorectal cancer driver gene, such as TP53, KRAS, PIK3CA, BRAF, APC, BMP3, NDRG4, SMAD4, MLH1, CTNNB1, EGFR, BRCA1, CDKN2A, CDH1, PTEN, VEGFA, MAKP3, or NRAS. Exemplary stool-derived eukaryotic RNA variant biomarkers are listed in Table 3.

[0025] Table 3. Stool-derived eukaryotic RNA biomarkers and variants TIFF0007650663000003.tif148166TIFF0007650663000004.tif140166

[0026] The method may include identifying the presence of a variant in the nucleic acid sequence of a biomarker, for example, a biomarker listed in Table 3. Some or all of the colorectal neoplasia biomarker genes listed in Table 3 may comprise a panel (Panel B). In some embodiments, the colorectal neoplasia biomarker genes listed in Table 3 may also comprise a subset of colorectal neoplasia subtype biomarkers. The composition may include a gene array and a probe set configured for specific detection of a marker panel disclosed herein. The composition may also include a kit with a gene array and a probe set configured for specific detection of a marker panel disclosed herein.

[0027] In another embodiment, a method for detecting disease may include measuring the relative variant allele frequency, e.g., relative ratio, of variants in one, two or more of any biomarker genes listed in Table 3 in a stool sample of a subject, and comparing the relative variant allele frequency of such biomarkers with the variant allele frequency of the one, two or more biomarkers in a control.The difference in the measured variant allele frequency of the one, two or more biomarkers in a sample of a subject relative to the measured variant allele frequency of the one, two or more biomarkers in a control is an indication that the subject has a gastrointestinal disease.In some embodiments, the difference in the measured variant allele frequency of the one, two or more biomarkers in a sample of a subject relative to the measured variant allele frequency of the one, two or more biomarkers in a control is an indication that the subject is at risk of gastrointestinal disorder.

[0028] Also provided is a method for detecting colorectal neoplasia molecular subtype biomarkers. Colorectal cancer can be classified into four distinct molecular subtypes based on the expression of certain markers. The four consensus molecular subtypes (CMS1-4) are predicted based on the expression of 274 genes (based on unique HUGO gene name identifiers) shown in Figure 4A. In the random forest classifier shown by CRCSC, the expression of the 274 genes is used as features to accurately identify the molecular subtype classification. The four CMS subtypes include CMS1-4. CMS1 is associated with hypermutation and microsatellite instability. CMS1 tumors typically have immune infiltrates. CMS1 tumors tend to have high histopathological grade at diagnosis and are associated with poor survival. CMS2, also referred to as the "canonical" subtype, is an epithelial tumor characterized by significant WNT and MYC signaling activation and increased copy number variants, and tends to be associated with long-term survival. CMS3 is an epithelial tumor characterized by pronounced metabolic dysregulation and mutations in the KRAS, receptor tyrosine kinase and MAPK pathways. CMS4 tumors are mesenchymal tumors characterized by transforming growth factor-β activation, stromal invasion and angiogenesis. CMS4 tumors tend to be diagnosed at advanced stages (stages III and IV) and correlate with poor overall and recurrence-free survival. The 25 genes that are particularly potent in predicting CMS1 (based on unique HUGO gene name identifiers) are shown in Figure 4B and Table 4.

[0029] Table 4. Stool-derived eukaryotic RNA biomarkers for CMS1 colorectal cancer subtype TIFF0007650663000005.tif132128

[0030] The method may include determining the level of expression of two or more colorectal neoplasia subtype biomarker genes in human RNA isolated from a stool sample collected from a subject by determining whether the level of two or more colorectal neoplasia subtype biomarker genes in a stool sample from the subject is different from the level of the same two or more colorectal neoplasia subtype biomarker genes in a control. Exemplary colorectal neoplasia subtype biomarker genes are shown in Table 4. Some or all of the colorectal neoplasia biomarker genes listed in Table 4 may constitute a panel (Panel C). In some embodiments, the colorectal neoplasia biomarker genes listed in Table 4 may also comprise a subset of colorectal neoplasia subtype biomarkers. The composition may include a gene array and a probe set configured for specific detection of a marker panel disclosed herein. The composition may also include a kit with a gene array and a probe set configured for specific detection of a marker panel disclosed herein.

[0031] In another embodiment, a method for detecting a disease may include measuring the relative proportions, e.g., relative ratios, of expression levels of one, two or more colorectal neoplasia subtype biomarkers in a stool sample of a subject, and comparing the relative proportions of such biomarkers with the relative proportions of expression levels of the one, two or more biomarkers in a control. The difference in the relative proportions of the measured expression levels of the one, two or more biomarkers in a sample of a subject relative to the control may indicate a molecular subtype of colorectal cancer. In some embodiments, the difference in the proportions of the measured expression levels of the one, two or more biomarkers in a sample of a subject relative to the proportions of the measured expression levels of the one, two or more biomarkers in a control is an indication that the subject may develop a particular subtype of colorectal cancer.

[0032] Alternative methods for detecting CMS1 tumors, also referred to as MSI-H tumors, may be used. Genomic variants of POLE, MLH1, MSH2, MSH6 and PMS2, which are involved in DNA mismatch repair deficiency, are used as predictive biomarkers in clinical trials of immune checkpoint blockade therapy. Gene expression profiles focusing on the expression of immune inhibitory molecules, such as PD-1, PD-L1, CTLA-4, LAG-3 and IDO, can further be used to predict the increased immunogenicity of the microenvironment of MSI-H tumors to further predict the eligibility of patients to benefit from checkpoint immunotherapy.

[0033] Provided herein are stool-derived eukaryotic RNA biomarkers and stool-derived eukaryotic RNA biomarker panels for use in diagnosing colorectal neoplasia or specific subtypes of precancerous lesions or colorectal cancer.Biomarkers are generally characteristics that can be objectively measured, quantified, and used to evaluate biological processes, such as the occurrence, progression, remission, or recurrence of colorectal neoplasia.Biomarkers can take many forms, such as nucleic acids, polypeptides, metabolites, or physical or physiological parameters.

[0034] In general, biomarkers from eukaryotic cells may include: a) deoxyribonucleic acid (DNA) sequences, b) ribonucleic acid (RNA) sequences, c) predicted amino acid sequences that constitute the backbone of a protein, d) expression levels of ribonucleic acid biomarkers, e) predicted expression levels of amino acid sequences, or f) any combination of the above. In some embodiments, biomarkers may be fragments of large sequences, such as fragments of long RNA sequences, long DNA sequences, or long polypeptide sequences. In some embodiments, biomarkers, such as GAPDH, ACTB, etc., may be used for normalization of other biomarkers. In other embodiments, features such as total RNA counts, total RNA input, etc. may be used as biomarkers or for normalization of other biomarkers.

[0035] Stool-derived eukaryotic RNA biomarkers can be quantified using amplicons. Amplicons can contain zero, one, two, or more unique sequences. Amplicons of the same stool-derived eukaryotic RNA biomarker can differ in percent sequence identity. Amplicons can be designed to target different loci. Targeted loci can include: a) geographically similar loci on the same transcript from the same gene, b) geographically unique loci on the same transcript from the same gene, c) geographically unique loci on different transcripts from the same gene, or d) geographically unique loci on different transcripts from different genes. In some embodiments, amplicons designed to target different loci can reflect structural features of specific RNAs, such as sequences or secondary structures, that may be preserved or preferentially degraded in stool. In some embodiments, amplicons designed to target different loci can reflect specific disease parameters, such as those of diseases in which a particular alternative splicing transcript is increased or decreased.

[0036] A biological sample may be a sample containing cells or other cellular material from which nucleic acids or other analytes can be harvested. A biological sample may be a control or experimental sample. A biological sample may be a stool sample. A biological sample may be collected immediately after defecation into a toilet, on the ground, in a litter box, or into a collection device. In some embodiments, a biological sample may be collected after or during a procedure such as an enema, fecal swab, or endoscopy. A biological sample may be tested immediately. Alternatively, a biological sample may be stored in a buffer prior to testing, such as an aqueous buffer, a glycerol-based buffer, a polar solvent-based buffer, an osmolarity-balanced buffer, or other buffer sufficient to preserve the biological sample. Additionally or alternatively, a biological sample may be collected and stored in a refrigerated state, such as at 4°C, or in a frozen state, such as at 0°C, -20°C, -80°C, -140°C, or below, prior to testing. A biological sample may be stored for one month, two months, four months, six months, one year, two years, or longer prior to testing.

[0037] Biological samples can be derived from eukaryotic organisms, such as mammals.Mammals can be human or non-human animals, such as human, dog, cat, non-human primate, ruminant, ursine, horse, pig, sheep, goat, camel, buffalo, deer, elk, moose, weasel, rabbit, guinea pig, hamster, rat, mouse, pachyderm, rhinoceros or chinchilla.Thus, stool samples can be taken from human or non-human animals, such as human, dog, cat, non-human primate, ruminant, ursine, horse, pig, sheep, goat, camel, buffalo, deer, elk, moose, weasel, rabbit, guinea pig, hamster, rat, mouse, pachyderm, rhinoceros or chinchilla.

[0038] Provided herein is a useful method for the isolation of eukaryotic nucleic acid enriched nucleic acid from a biological sample, such as a stool sample. The method may include a step of disintegrating a stool sample with a buffer. The sample may be subjected to vortexing, shaking, stirring, rolling or other agitation methods sufficient to disperse solids and stool bacteria. The temperature at which the agitation and centrifugation steps are performed may vary, for example, from about 4°C to about 20°C, from about 4°C to about 1°C, from about 4°C to about 10°C, from about 4°C to about 6°C. After disintegration, the sample may be subjected to one or more centrifugations. In some embodiments, the disintegration and centrifugation steps may be repeated one, two, three or even more times. Commercially available reagents, such as Nuclisens® EasyMag® ​​reagent, may be used for disintegration, washing and cell lysis of stool. Lysis buffers may also be used to lyse eukaryotic cells. The lysate may be further centrifuged at any temperature, for any period of time, any number of times. After centrifugation, the supernatant can be used as input into an automated RNA isolation machine, such as an EasyMag® ​​instrument. In some embodiments, the extracted nucleic acids can be treated with DNase to degrade DNA in solution. Other RNA purification methods can be used; for example, mechanical or enzymatic cell disruption followed by solid-phase methods, such as column chromatography or extraction with organic solvents, such as phenol-chloroform or thiocyanate-phenol-chloroform extraction. In some embodiments, the nucleic acids can be extracted onto functionalized beads. In some embodiments, the functionalized beads can further comprise a magnetic core ("magnetic beads"). In some embodiments, the functionalized beads can comprise a surface functionalized with a charged moiety. The charged moiety can be selected from amines, carboxylic acids, carboxylates, quaternary amines, sulfates, sulfonates, or phosphates.

[0039] For the extraction of nucleic acid, the stool sample can be disintegrated in the presence of one or more of a buffer, a surfactant and a ribonuclease inhibitor to form a suspension.The buffer can be a biologically compatible buffer, such as Hanks' balanced salt solution, Alsever's solution, Earle's balanced salt solution, Gay's balanced salt solution, phosphate buffered saline, Puck's balanced salt solution, Ringer's balanced salt solution, Simm's balanced salt solution, TRIS buffered saline or Tyrode's balanced salt solution.The surfactant can be an ionic or non-ionic surfactant, such as Tween-20 or Triton-X-100. The ribonuclease inhibitor can be solvent-based, protein-based, or another type of method to suppress RNA destruction, such as Protector RNase Inhibitor (Roche), RNasin® (Promega), SUPERase-In™ (Thermo Fisher Scientific), RNaseOUT™ (Thermo Fisher Scientific), ANTI-RNase, Recombinant RNase Inhibitor, or cloned RNase Inhibitor. The stool sample can be disrupted in various ways, such as by vortexing, shaking, stirring, rolling, or other agitation methods sufficient to disperse solids and stool bacteria. In some embodiments, the stool sample can be disrupted using coated beads, magnetic beads, or stirring devices, such as glass rods, metal rods, wooden rods, or wooden blades.

[0040] The suspension can then be separated into a liquid portion and a solid portion. Separation can be performed, for example, by centrifugation, filtration, targeting probes that specifically bind to eukaryotic cells, antibodies, column-based filtration, bead-based filtration, or chromatography methods. The liquid portion is enriched in bacterial nucleic acids and can be discarded. The solid portion can be resuspended in a buffer, either in the presence or absence of a detergent and in the presence or absence of a ribonuclease. The separation process can be repeated 1, 2, 3, 4, 5, 6, 7, 8, or even more times.

[0041] The temperature at which the disintegration and separation steps are carried out can vary, for example, from about 4°C to about 20°C, from about 4°C to about 15°C, from about 4°C to about 10°C, from about 4°C to about 6°C.

[0042] The resulting pellet from the separation step can be suspended in a lysis buffer, for example, a buffer containing a chaotropic agent and optionally a detergent, to form a lysate. In some embodiments, the chaotropic agent can be guanidinium thiocyanate and the detergent can be Triton-X-100. In some embodiments, the lysis buffer can contain or not contain Tris-HCl, ethylenediaminetetraacetic acid (EDTA), sodium dodecyl sulfate (SDS), Nonidet P-40, sodium deoxycholate, or dithiothreitol.

[0043] The lysate can be fractionated into a portion enriched in eukaryotic nucleic acids. Fractionation can be performed, for example, by centrifugation, filtration, targeting probes that specifically bind eukaryotic nucleic acids, antibodies, column-based filtration, bead-based filtration, or chromatographic methods. In some embodiments, fractionation by centrifugation can result in the formation of a bottom layer (pellet) containing cell debris, a hydrophilic middle layer containing eukaryotic nucleic acids, and a hydrophobic top layer containing lipids and membrane fractions. The middle layer can be collected. In some embodiments, the middle layer and the top layer can be collected together. The middle layer can be collected through a narrow orifice. The narrow orifice can be a pipette tip, or a syringe with a needle attached. The pipette tip can be, for example, a 1 uL, 5 uL, 10 uL, 20 uL, or 100 uL pipette tip. The needle can be, for example, an 18-gauge or 15-gauge needle.

[0044] The layer containing the collected eukaryotic nucleic acid can be subjected to further extraction. The method of further extraction can be various. Exemplary methods include magnetic particle-based methods, column-based methods, filter-based methods, bead-based methods or organic solvent-based methods. Such exemplary methods can include commercially available reagents, such as Nuclisens® EasyMag® ​​reagent (bioMerieux).

[0045] The extracted nucleic acids can be analyzed for eukaryotic biomarkers associated with gastrointestinal disorders or gastrointestinal cells. Biomarkers can provide information about the health of an individual, i.e., a subject. Such biomarkers from eukaryotic cells can include: a) deoxyribonucleic acid (DNA) sequences, b) ribonucleic acid (RNA) sequences, c) predicted amino acid sequences that constitute the backbone of a protein, d) expression levels or percentages of expression levels of RNA biomarkers, e) predicted expression levels or percentages of predicted expression levels of amino acid sequences, or f) any combination of the above. Isolation of biomarkers from eukaryotic cells can allow comparison between experimental samples and controls. Isolation of such biomarkers from eukaryotic cells can provide a method for the detection of intestinal diseases in experimental samples. Comparisons include evaluation of: a) variations in DNA sequences, b) variations in RNA sequences, c) variations in predicted amino acid sequences, d) variations in expression levels or percentages of expression levels of RNA biomarkers, e) variations in predicted expression levels or percentages of predicted expression levels of amino acid sequences, or f) variations that constitute any combination of the above. Variation can be determined when a biomarker measured in an experimental sample differs from a biomarker measured in a control.

[0046] The method may include collecting an experimental sample and a control, for example, a stool sample. The stool sample contains exfoliated eukaryotic cells, which can be evaluated for biomarkers. In some embodiments, the eukaryotic cells can be intestinal cells, lymphocytes, enterochromiffin-like cells, enteroendocrine cells, neuroendocrine cells, pancreatic cells, hepatic cells, gastric cells or other cells. The method provides a means by which eukaryotic cells in a stool sample can be evaluated for eukaryotic biomarkers. Biomarkers can include DNA sequences, RNA sequences, predicted amino acid sequences, expression levels or percentages of expression levels of RNA biomarkers, predicted expression levels or percentages of predicted expression levels of amino acid sequences, or any combination of the above. In a specific embodiment, the biomarker is a stool-derived eukaryotic RNA biomarker. In some embodiments, the evaluation step includes any type of microarray sequencing, polymerase chain reaction (PCR), nucleic acid sequencing, amplicon sequencing, molecular barcoding, or probe-capture.

[0047] The method and composition are also useful for selecting a clinical plan for individuals suffering from gastrointestinal disorders, such as colorectal neoplasms or colorectal cancer.Through this method, the clinical plan can include additional diagnostic procedures, such as colonoscopy.In some embodiments, the clinical plan can include treatment methods.

[0048] The level of stool-derived eukaryotic RNA biomarkers can be evaluated using a variety of methods. Expression levels can be measured at either the nucleic acid level, such as the RNA level, or the polypeptide level. RNA expression can include the expression of seRNA, total RNA, mRNA, tRNA, rRNA, ncRNA, smRNA, miRNA, and snoRNA. Expression at the RNA level can be measured directly or indirectly by measuring the level of cDNA corresponding to the relevant RNA. Alternatively or additionally, the levels of the polypeptides encoded by the RNA, the RNA regulators of the genes encoding the relevant transcription factors, and the transcription factor polypeptides can be assayed. Methods for investigating gene expression at the mRNA level include, for example, microarray analysis, sequence analysis of gene expression (SAGE), RT-PCR, blotting, hybridization based on digital barcode quantitative assay, multiplex RT-PCR, droplet digital PCR (ddPCR), digital PCR (dPCR), NanoDrop spectrophotometer, RT-qPCR, qPCR, UV spectroscopy, amplicon sequencing, RNA sequencing, next-generation sequencing, lysate-based hybridization assay using branched DNA signal amplification, such as QuantiGene 2.0 Single Plex, and branched DNA analysis.Digital barcode quantitative assay can include BeadArray (Illumina), xMAP system (Luminex), nCounter (NanoString), HTG EdgeSe (High Throughput Genomics), BioMark (Fluidigm) or Wafergen microarray. Assays may include DASL (Illumina), RNA-Seq (Illumina), TruSeq (Illumina), SureSelect (Agilent), Bioanalyzer (Agilent), TaqMan (ThermoFisher), GeneReader (Qiagen) or QIAseq (Qiagen).

[0049] The inventors may use the terms "nucleic acid" and "polynucleotide" interchangeably to refer to both RNA and DNA, such as DNA (or RNA), including cDNA, genomic DNA, synthetic DNA, and nucleic acid analogs, any of which may encode the polypeptides of the invention, all of which are encompassed by the present invention. Polynucleotides may have essentially any three-dimensional structure. Nucleic acids may be double-stranded or single-stranded (i.e., sense or antisense). Non-limiting examples of polynucleotides include genes, gene fragments, exons, introns, messenger RNA (mRNA) and portions thereof, transfer RNA, microRNA, ribosomal RNA, siRNA, micro-RNA, ribozymes, cDNA, recombinant polynucleotides, branched polynucleotides, plasmids, vectors, isolated DNA of any sequence, isolated RNA of any sequence, nucleic acid probes and primers, and nucleic acid analogs. In the context of the present invention, the nucleic acid may encode a biomarker, for example, a fragment of a stool-derived eukaryotic RNA biomarker or a variant thereof from any of the biomarkers listed in Tables 1 and 2, or listed in Table 3, or listed in Table 4.

[0050] An "isolated" nucleic acid can be, for example, a DNA molecule or a fragment thereof, provided that at least one of the nucleic acid sequences normally found directly flanking the DNA molecule in a genome has been removed or is absent. Thus, an isolated nucleic acid includes, but is not limited to, a DNA molecule that exists as a separate molecule independent of other sequences (e.g., a chemically synthesized nucleic acid, or a cDNA or genomic DNA fragment produced by polymerase chain reaction (PCR) or restriction endonuclease treatment). Isolated nucleic acid also refers to a DNA molecule that is incorporated into a vector, an autonomously replicating plasmid, a virus, or into the genomic DNA of a prokaryote or eukaryote. Isolated nucleic acid can also include a DNA molecule that is part of an engineered nucleic acid, such as a hybrid or fusion nucleic acid. A nucleic acid that is present among many (e.g., dozens, or hundreds to millions) of other nucleic acids, for example, in a cDNA or genomic library or in a gel slice containing a genomic DNA restriction digest, is not an isolated nucleic acid.

[0051] Isolated nucleic acid molecules may be produced in various ways. For example, polymerase chain reaction (PCR) techniques can be used to obtain isolated nucleic acids that contain nucleotide sequences described herein, such as nucleotide sequences encoding polypeptides described herein. PCR can be used to amplify specific sequences from DNA as well as RNA, such as sequences from total genomic DNA or total cellular RNA. Generally, sequence information from both ends of the region of interest or beyond is used to design oligonucleotide primers that are identical or similar in sequence to the strand opposite the template to be amplified. Also available are various PCR strategies that can introduce site-specific nucleotide sequence modifications into template nucleic acid.

[0052] Alternatively, isolated nucleic acids can be chemically synthesized either as a single nucleic acid molecule (e.g., using automated DNA synthesis in the 3' to 5' direction using phosphoramidite technology) or as a series of oligonucleotides. For example, one or more pairs of long oligonucleotides (e.g., >50-100 nucleotides) containing the desired sequence can be synthesized, each pair containing a short segment of complementarity (e.g., about 15 nucleotides) that forms a double strand when the oligonucleotide pair is annealed. DNA polymerase is used to extend the oligonucleotides, resulting in one double-stranded nucleic acid molecule per oligonucleotide pair, which can then be ligated into a vector.

[0053] Two nucleic acids or the polypeptides they encode may be described as having a certain degree of identity to one another. For example, the stool-derived eukaryotic RNA biomarkers selected from Table 1 or Table 2 or a combination of Tables 1 and 2, or Table 3 or Table 4, and biologically active variants thereof, may be described as exhibiting a certain degree of identity. Alignments may be assembled by placing short sequences into the Protein Information Research (PIR) site (http: / / pir.georgetown.edu) and then analyzing using the "short nearly identical sequences" Basic Local Alignment Search Tool (BLAST) algorithm at the NCBI website (http: / / www.ncbi.nlm.nih.gov / blast).

[0054] As used herein, the term "percent sequence identity" refers to the degree of identity between any given query sequence and a subject sequence. For example, a stool-derived eukaryotic RNA biomarker sequence listed in Table 1 or Table 2 or a combination of Tables 1 and 2 or Table 3 or Table 4 can be the query sequence, and a fragment of a stool-derived eukaryotic RNA biomarker sequence listed in Table 1 or Table 2 or a combination of Tables 1 and 2 or Table 3 or Table 4 can be the subject sequence. Similarly, a fragment of a stool-derived eukaryotic RNA biomarker sequence listed in Table 1 or Table 2 or a combination of Tables 1 and 2 or Table 3 or Table 4 can be the query sequence, and a biologically active variant thereof can be the subject sequence.

[0055] To determine sequence identity, a query nucleic acid sequence or a query amino acid sequence can be aligned with one or more subject nucleic acid sequences or subject amino acid sequences, respectively, using computer programs such as ClustalW (version 1.83, default parameters), HISAT, HISAT2 or SAMTools, which are capable of aligning nucleic acid or protein sequences over their entire length (global alignment).

[0056] Nucleic acids and polypeptides described herein may be referred to as "foreign". The term "foreign" indicates that the nucleic acid or polypeptide is part of or encoded by a recombinant nucleic acid construct, or is not present in its natural environment. For example, a foreign nucleic acid may be a sequence from one species introduced into another species, i.e., a heterologous nucleic acid. Typically, such foreign nucleic acids are introduced into other species by recombinant nucleic acid constructs. Foreign nucleic acids may also be sequences that are native to an organism and reintroduced into the cells of the organism. Foreign nucleic acids that include native sequences can often be distinguished from the native sequence by the presence of non-native sequences linked to the foreign nucleic acid, such as non-native regulatory sequences flanking the native sequence in a recombinant nucleic acid construct. Stably transformed foreign nucleic acids are also typically integrated into a location other than the location where the native sequence is found.

[0057] Nucleic acids of the invention can include nucleic acids having the nucleotide sequence of any one of the stool-derived eukaryotic RNA biomarkers listed in Table 1 or Table 2, or a combination of Tables 1 and 2, or Table 3 or Table 4, or nucleic acids having a nucleic acid sequence that is at least about 70%, at least about 75%, at least about 80%, at least about 85%, at least about 90%, at least about 95%, at least about 99% identical to the nucleic acid sequence of any one of the stool-derived eukaryotic RNA biomarkers listed in Table 1 or Table 2, or a combination of Tables 1 and 2, or Table 3 or Table 4.

[0058] A nucleic acid specific to a target nucleic acid, such as an oligonucleotide (e.g., a probe or primer), hybridizes to the target nucleic acid under suitable conditions. We may refer to the process in which a single oligonucleotide strand anneals to a complementary strand by base pairing under defined hybridization conditions as hybridization or hybridization. This is a specific, i.e., non-random, interaction between two complementary polynucleotides. Hybridization and the strength of hybridization (i.e., the strength of the association between nucleic acids) are influenced by factors such as the degree of complementarity between the nucleic acids, the stringency of the conditions involved, and the melting temperature (Tm) of the hybrid formed. Hybridization products can be duplexes or triplexes formed with targets in solution or on solid support.

[0059] In some embodiments, the nucleic acid may include a short nucleic acid sequence useful for the analysis and quantification of stool-derived eukaryotic RNA biomarkers listed in Table 1 or Table 2 or a combination of Table 1 and Table 2 or Table 3 or Table 4. Such isolated nucleic acid may be an oligonucleotide primer. In general, an oligonucleotide primer is an oligonucleotide that is complementary to a target nucleotide sequence, such as the nucleotide sequence of any stool-derived eukaryotic RNA biomarker listed in Table 1 or Table 2 or a combination of Table 1 and Table 2 or Table 3 or Table 4, and can serve as an initiation point for DNA synthesis by addition of a nucleotide to the 3' end of the primer in the presence of a DNA or RNA polymerase. The 3' nucleotide of the primer should generally be identical to the target sequence at the corresponding nucleotide position for optimal extension and / or amplification. Primers may take many forms, such as peptide nucleic acid primers, locked nucleic acid primers, unlocked nucleic acid primers and / or phosphorothioate modified primers. In some embodiments, the forward primer can be a primer complementary to the antisense strand of dsDNA, and the reverse primer can be a primer complementary to the sense strand of dsDNA. We may also refer to primer pairs. In some embodiments, the 5' target primer pair can be a primer pair comprising at least one forward primer and at least one reverse primer that amplifies the 5' region of the target nucleotide sequence. In some embodiments, the 3' target primer pair can be a primer pair of at least one forward primer and at least one reverse primer that amplifies the 3' region of the target nucleotide sequence. In some embodiments, the primers can comprise a detectable label as discussed below. In some embodiments, the detectable label can be a quantifiable label.

[0060] The oligonucleotide primers provided herein are useful for amplifying any stool-derived eukaryotic RNA biomarker listed in Tables 1 and 2 or Tables 3 or 4. In some embodiments, the oligonucleotide primers can be complementary to 2, 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 or more of the stool-derived eukaryotic RNA biomarkers disclosed herein, such as those listed in Tables 1 and 2 or Tables 3 or 4. The length of the primers can vary depending on the sequence and composition of nucleotide bases of the particular nucleic acid sequence of the probe and the particular method in which the probe is used. Generally, useful primers can be about 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30 nucleotide bases in length. Useful primer lengths can range from 8 nucleotide bases to about 60 nucleotide bases; from about 12 nucleotide bases to about 50 nucleotide bases; from about 12 nucleotide bases to about 45 nucleotide bases; from about 12 nucleotide bases to about 40 nucleotide bases; from about 12 nucleotide bases to about 35 nucleotide bases; from about 15 nucleotide bases to about 40 nucleotide bases; from about 15 nucleotide bases to about 35 nucleotide bases; from about 18 nucleotide bases to about 50 nucleotide bases; from about 18 nucleotide bases to about 40 nucleotide bases; from about 18 nucleotide bases to about 35 nucleotide bases; from about 18 nucleotide bases to about 30 nucleotide bases; from about 20 nucleotide bases to about 30 nucleotide bases; from about 20 nucleotide bases to about 25 nucleotide bases.

[0061] Also provided are probes, i.e., isolated nucleic acid fragments, that selectively bind to and are complementary to any of the stool-derived eukaryotic RNA biomarkers listed in Table 1 and Table 2 or Table 3 or Table 4. The probes can be oligonucleotides or polynucleotides, DNA or RNA, single-stranded or double-stranded, and natural or modified, and can be present either in the nucleotide base or in the backbone. The probes can be made by a variety of methods, such as chemical synthesis or enzymatic synthesis.

[0062] The length of the probe can vary depending on the sequence and composition of the nucleotide bases of the specific nucleic acid sequence of the probe and the specific method in which the probe is used.In general, the length of a useful probe can be about 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 35, 40, 50, 55, 60, 65, 70, 75, 80, 85, 90, 100, 110, 120, 140, 150, 175 or 200 nucleotide bases. In general, useful probe lengths range from about 8 to about 200 nucleotide bases; about 12 to about 175 nucleotide bases; about 15 to about 150 nucleotide bases; about 15 to about 100 nucleotide bases about 15 to about 75 nucleotide bases; about 15 to about 60 nucleotide bases; about 20 to about 100 nucleotide bases; about 20 to about 75 nucleotide bases; about 20 to about 60 nucleotide bases; about 20 to about 50 nucleotide bases in length. In some embodiments, the probe set may include probes directed to 2, 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 or more of the stool-derived eukaryotic RNA biomarkers listed in Table 1 or Table 2, or a combination of Tables 1 and 2, or Table 3 or Table 4.

[0063] The primers and probes disclosed herein may be detectably labeled. The label may be a molecular moiety or compound that can be detected or produce a detectable response and can be directly or indirectly bound to a nucleic acid. Direct labeling may use a bond or interaction to link the label and the probe, including covalent bonds, non-covalent interactions (hydrogen bonds, hydrophobic interactions, and ionic interactions), or chelates or coordination complexes. Indirect labeling may use a bridging moiety or linker (e.g., an antibody, an oligomer, or another compound) that is directly or indirectly labeled and can amplify the signal. Labels may include any detectable moiety, such as a radionuclide, a ligand such as biotin or avidin, an enzyme, an enzyme substrate, a reactive group, a chromophore (a detectable dye, particle, or bead), a fluorophore, or a luminescent compound (a bioluminescent label, a phosphorescent label, or a chemiluminescent label). The label may be detectable in a homogeneous assay in which the bound labeled probe in a mixture shows a detectable change compared to that of the unbound labeled probe, such as differences in stability or degradation, without requiring physical separation of the bound form from the unbound form.

[0064] Suitable detectable labels may include molecules that are detectable themselves (e.g., fluorescent moieties, electrochemical labels, metal chelates, etc.) as well as molecules that can be indirectly detected by generating detectable reaction products (e.g., enzymes, such as horseradish peroxidase, alkaline phosphatase, etc.) or molecules that can be indirectly detected by specific binding molecules that can be detectable themselves (e.g., biotin, digoxigenin, maltose, oligohistidine, 2,4-dintrobenzene, phenylarsenate, ssDNA, dsDNA, etc.).As discussed above, the coupling of one or more ligand motifs and / or ligands with detectable labels may be direct or indirect.Detection may be in situ, in vivo, in vitro in tissue sections or in solution, etc.

[0065] In some embodiments, the method comprises the use of alkaline phosphatase conjugated polynucleotide probe.When alkaline phosphatase (AP)-conjugated polynucleotide probe is used, after the sequential addition of suitable substrate, such as fast blue or fast red substrate, AP degrades the substrate and forms a precipitate that allows the in situ detection of specific target RNA molecule.Alkaline phosphatase can be used with some substrates, such as fast blue, fast red or 5-bromo-4-chloro-3-indolyl-phosphate (BCIP).

[0066] In some embodiments, the fluorophore-conjugated probe may be a fluorescent dye-conjugated labeled probe, or may use other enzymatic approaches for chromogenic detection pathways in addition to alkaline phosphatase, such as the use of horseradish peroxidase-conjugated probes with substrates such as 3,3'-diaminobenzidine (DAB).

[0067] Fluorescent dyes used in conjugated labeled probes can typically be divided into families, such as, for example, fluorescein and its derivatives; rhodamine and its derivatives; cyanine and its derivatives; coumarin and its derivatives; Cascade Blue™ and its derivatives; Lucifer Yellow and its derivatives; BODIPY and its derivatives; etc. Exemplary fluorophores include indocarbocyanine (C3), indodicarbocyanine (C5), Cy3, Cy3.5, Cy5, Cy5.5, Cy7, Texas Red, Pacific Blue, Oregon Green 488, Alexa Fluor®-355, Alexa Fluor 488, Alexa Fluor 532, Alexa Fluor 546, Alexa Fluor-555, Alexa Fluor 568, Alexa Fluor 594, Alexa Fluor 647, Alexa Fluor 660, Alexa Fluor 570, Alexa Fluor 590, Alexa Fluor 592, Alexa Fluor 591 ... 680, JOE, Lissamine, rhodamine green, BODIPY, fluorescein isothiocyanate (FITC), carboxy-fluorescein (FAM), phycoerythrin, rhodamine, dichlororhodamine (dRhodamine™), carboxytetramethylrhodamine (TAMRA™), carboxy-X-rhodamine (ROX™), LIZ™, VIC™, NED™, PET™, SYBR, PicoGreen, RiboGreen, etc. Near infrared dyes are expressly included within the intended meaning of the terms fluorophore and fluorescent reporter group.

[0068] In some embodiments, the level of eukaryotic biomarkers can be analyzed in gene arrays. Microarray analysis can be performed in custom gene arrays. Alternatively or additionally, microarray analysis can be performed using commercially available systems according to the manufacturer's instructions and protocols. Exemplary commercially available systems include Affymetrix GENECHIP® technology (ThermoFisher, Walthum, MA), Agilent microarray technology, NCOUNTER® Analysis System (NanoString® Technologies, Seattle, WA) and BeadArray Microarray Technology (Illumina, San Diego, CA). Nucleic acid extracted from stool samples can be hybridized to the probes on the gene array. Probe-target hybridization can be detected by chemiluminescence, and the relative abundance of specific sequences can be examined. The relative abundance of specific sequences can be normalized between gene arrays or within gene arrays.

[0069] In some embodiments, the probe and probe set can be configured as a gene array. Gene arrays, also known as microarrays or gene chips, are ordered arrays of nucleic acids that allow for parallel analysis of complex biological samples. Typically, gene arrays include probes bound to a solid substrate, such as a microchip, glass slide, or bead. The binding generally involves chemical coupling that results in a covalent bond between the substrate and the probe. The number of probes in the array can vary, but each probe is fixed to a specific addressable location on the array or microchip. In some embodiments, the probe can be about 18 nucleotide bases, about 20 nucleotide bases, about 25 nucleotide bases, about 30 nucleotide bases, about 35 nucleotide bases, or about 40 nucleotide bases in length. In some embodiments, the probe set comprises probes directed to at least 2, 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 or more of the stool-derived eukaryotic RNA biomarkers listed in Table 1 or Table 2, or a combination of Tables 1 and 2, or Table 3 or Table 4. The probe sets can be assembled into high density arrays containing 5,000, 10,000, 20,000, 50,000, 100,000, 200,000, 300,000, 400,000, 500,000, 600,000, 700,000, 800,000, 900,000, 1,000,000, 2,000,000, 3,000,000, 4,000,000, 5,000,000, 6,000,000, 7,000,000, 8,000,000 or more different probes.

[0070] Methods of gene array synthesis can vary. Exemplary methods include synthesis of probes followed by deposition onto the array surface by "spotting," e.g., in situ synthesis on a microelectrode array using photolithography or electrochemistry.

[0071] In some embodiments, the probes and probe sets can be configured as reagents, i.e., nucleic acid pools that allow for parallel analysis of complex biological samples. The reagents can be, for example, a set of amplification probes, library preparations, amplicon panels, or capture panels. Typically, the reagents include targeting probes suspended in a solution. In some embodiments, the probes are designed to target specific regions. The probes can be configured in such a way that they allow for the capture of specific nucleic acids. They can also be configured to allow for the amplification of specific nucleic acids. The number of probes in a reagent can vary, but each probe is designed for a specific sequence. In some embodiments, the probes can be about 10 nucleotide bases, about 15 nucleotide bases, about 20 nucleotide bases, about 25 nucleotide bases, about 30 nucleotide bases, about 35 nucleotide bases, or about 40 nucleotide bases in length. In some embodiments, the probe set comprises probes directed to at least 2, 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 or more stool-derived eukaryotic RNA biomarkers listed in Table 1 or Table 2 or a combination of Tables 1 and 2, or Table 3 or Table 4. Probe sets can be assembled into high density reagents that contain 5,000, 10,000, 20,000, 50,000, 100,000, 200,000, 300,000, 400,000, 500,000, 600,000, 700,000, 800,000, 900,000, 1,000,000, 2,000,000, 3,000,000, 4,000,000, 5,000,000, 6,000,000, 7,000,000, 8,000,000 or more different probes.

[0072] Methods of reagent synthesis can be varied. Exemplary methods include synthesis of nucleic acid probes followed by suspension in a stabilizing solution. Probe reagents can include unique regions that act as molecular identifiers. The reagents can be used for methods such as PCR, rtPCR ddPCR, dPCR, next generation sequencing, amplicon sequencing, RNA-se, and other methods.

[0073] The level of eukaryotic biomarkers can also be analyzed by DNA sequencing. DNA sequencing can be performed by sequencing methods such as target sequencing, whole genome sequencing, amplicon sequencing or exome sequencing. Sequencing methods can include: Sanger sequencing or high-throughput sequencing. High-throughput sequencing can involve reading-by-synthesis, pyrosequencing, sequencing by ligation, real-time sequencing, nanopore sequencing or Sanger sequencing. In some embodiments, isolated RNA can be used to generate corresponding cDNA, and this cDNA can be sequenced.

[0074] The sequencing methods described herein can be performed in a multiplex format, such that multiple different target nucleic acids are operated simultaneously. In some embodiments, different target nucleic acids can be processed in a common reaction vessel or on the surface of a specific substrate, allowing for convenient delivery of sequencing reagents, removal of unreacted reagents, and detection of incorporation events in a multiplex manner. In some embodiments involving surface-bound target nucleic acids, the target nucleic acid can be in an array format. In an array format, the target nucleic acid can typically be coupled to a surface in a spatially distinguishable manner. For example, the target nucleic acid can be bound by covalent direct binding, binding to beads or other particles, or can be associated with a polymerase or other molecules that are surface-bound. The array can include a single copy of the target nucleic acid at each site (also referred to as a feature), or multiple copies with the same sequence can be present at each site or feature. The multiple copies are generated by amplification methods such as bridge amplification, amplicon amplification, PCR, or emulsion PCR.

[0075] In some embodiments, a normalization step can be used to control nucleic acid recovery and variation between samples. In some embodiments, a predetermined amount of exogenous control nucleic acid can be added ("spiked in") to the extracted eukaryotic nucleic acid. The exogenous control nucleic acid can be a nucleic acid having a sequence corresponding to one or more eukaryotic or non-eukaryotic sequences, such as PhiX. Alternatively or additionally, the exogenous control nucleic acid can have a sequence corresponding to a sequence found in another species, such as a bacterial sequence, such as a Bacillis subtilis sequence. In some embodiments, the method can include examining the level of one or more housekeeping genes. In some embodiments, the method can include normalizing the expression level of the biomarker to the level of the housekeeping gene.

[0076] The method includes determining whether the measured expression levels of 2, 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 or more stool-derived eukaryotic RNA biomarkers in an experimental sample differs from the measured expression levels of the same 2, 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 or more stool-derived eukaryotic RNA biomarkers in a control. In another embodiment, the method comprises determining whether the ratio of expression levels of 2, 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 or more stool-derived eukaryotic RNA biomarkers in an experimental sample differs from the ratio of measured expression levels of the same 2, 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 or more stool-derived eukaryotic RNA biomarkers in a control. The difference in expression levels or ratio of expression levels may be an increase or a decrease.

[0077] The compositions disclosed herein are generally and variously useful for the detection, diagnosis and treatment of colorectal neoplasms. Detection methods include measuring the expression levels of 2, 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 or more stool-derived eukaryotic RNA biomarkers selected from the biomarkers listed in Table 1 or Table 2 or a combination of Tables 1 and 2 or Table 3 or Table 4 in a stool sample, and detecting the expression levels of 2, 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 or more stool-derived eukaryotic RNA biomarkers selected from the biomarkers listed in Table 1 or Table 2 or a combination of Tables 1 and 2 or Table 3 or Table 4 in the sample. The method may include comparing the measured expression levels of 4, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30 or more stool-derived eukaryotic RNA biomarkers to the measured expression levels of said 2, 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 or more stool-derived eukaryotic RNA biomarkers selected from the biomarkers listed in Table 1 or Table 2 or a combination of Tables 1 and 2 or Table 3 or Table 4 in a control. 2. The measured expression levels of 2, 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 or more stool-derived eukaryotic RNA biomarkers selected from the biomarkers listed in Table 1 or Table 2, or a combination of Tables 1 and 2, in a patient sample. A difference in the measured expression levels of the 2, 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 or more stool-derived eukaryotic RNA biomarkers selected from the biomarkers listed in a combination of Tables 1 and 2 is an indication that the patient has a colorectal neoplasm, or more specifically a high-risk adenoma.In some embodiments, the measured expression levels of 2, 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 or more stool-derived eukaryotic RNA biomarkers selected from the biomarkers listed in Table 1 or Table 2, or a combination of Tables 1 and 2, in a patient sample are compared to the measured expression levels of 2, 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 or more stool-derived eukaryotic RNA biomarkers selected from the biomarkers listed in Table 1 or Table 2, or a combination of Tables 1 and 2, in a control. A difference in the measured expression levels of said 2, 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 or more stool-derived eukaryotic RNA biomarkers selected from the biomarkers listed in Tables 1 and 2 or a combination of Tables 1 and 2 is indicative of said patient being at risk for colorectal neoplasia, or more specifically high risk adenoma. Such methods may further include identifying a subject (e.g., a patient, more specifically a human patient) having a colorectal neoplasia, e.g., colorectal cancer or a precancerous lesion, or at risk for developing a colorectal neoplasia.

[0078] A difference in the variant allele frequency of one, two, three, four, five, six, seven, eight, nine, ten, eleven, twelve, thirteen, fourteen, fifteen, six, seven, seven, or eighteen stool-derived eukaryotic RNA variant biomarkers selected from the biomarkers listed in Table 3 in the subject sample relative to the variant allele frequency of said one, two, three, four, five, six, seven, eight, nine, ten, eleven, twelve, thirteen, fourteen, fifteen, six, seventeen, or eighteen stool-derived eukaryotic RNA variant biomarkers selected from the biomarkers listed in Table 3 in the control is an indication that the patient has a colorectal neoplasm. In some embodiments, a difference in the measured variant allele frequency of 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, or 18 stool-derived eukaryotic RNA biomarkers selected from the biomarkers listed in Table 3 in a patient sample relative to the measured variant allele frequency of 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, or 18 stool-derived eukaryotic RNA biomarkers selected from the biomarkers listed in Table 3 in a control is an indication that the patient is at risk for colorectal neoplasia. Such methods may further include identifying a subject (e.g., a patient, more particularly a human patient) having colorectal neoplasia, e.g., colorectal cancer or precancerous lesion, or at risk of developing a colorectal neoplasia.

[0079] Measured expression levels of 1, 2, 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 or more stool-derived eukaryotic RNA biomarkers selected from the colorectal neoplasia molecular subtype biomarker genes listed in FIG. 4 in patient samples versus the colorectal neoplasia molecular subtype biomarker genes listed in FIG. A difference in the measured expression levels of the 1, 2, 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 or more stool-derived eukaryotic RNA biomarkers selected from intestinal neoplasia molecular subtype biomarker genes is an indication that the patient has a molecular subtype of colorectal cancer, e.g., CMS1. In some embodiments, the results listed in FIG. 4 are used in patient samples relative to measured expression levels of 2, 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 or more stool-derived eukaryotic RNA biomarkers selected from the colorectal neoplasia molecular subtype biomarker genes listed in FIG. 4 in a control. The difference in the measured expression levels of the 1, 2, 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 or more stool-derived eukaryotic RNA biomarkers selected from colorectal neoplasia molecular subtype biomarker genes is indicative of the patient being at risk for a molecular subtype of colorectal cancer, such as CMS1. Such methods may further include identifying a subject (e.g., a patient, more particularly a human patient) having colorectal neoplasia, such as colorectal cancer or a precancerous lesion, or at risk of developing a colorectal neoplasia.

[0080] Colorectal neoplasia may include any form of colorectal cancer. Colorectal neoplasia may also include polyps, e.g., precancerous lesions. Colorectal cancer typically begins as growths called polyps in the luminal lining of the colon or rectum. Colorectal polyps are generally divided into two categories: adenomatous polyps and benign polyps. Adenomatous polyps may also be referred to as adenomas. Benign polyps may also be referred to as hyperplastic polyps, hamartomatous polyps, or inflammatory polyps. Patients with one adenomatous polyp or multiple adenomatous polyps may be classified as having high-risk adenoma, intermediate-risk adenoma, or low-risk adenoma. High-risk adenomas include adenomas of any size with intraepithelial carcinoma or high-grade dysplasia, adenomas of any size with a villous growth pattern of 25% or more, any adenomas of 1.0 cm or more, or any serrated lesions of 1.0 cm or more. Medium-risk adenomas include one or two non-high-risk adenomas ranging in size from 5.0 mm to 1.0 cm or three or more non-high-risk adenomas of less than 1.0 cm. Low-risk adenomas include one or two non-high-risk adenomas of 5.0 mm or less. Adenomatous polyps can give rise to colorectal cancer. Adenocarcinoma, the most common form of colorectal cancer, occurs in intestinal glandular cells lining the inner colon and / or rectum. Adenocarcinomas can include tubular adenocarcinomas, which are adenocarcinomas on the pedunculated portion. Adenocarcinoma may also include choriocarcinoma, which is an adenocarcinoma that lies flat on the surface of the colon. Other colorectal cancers are differentiated by their tissue of origin. These include gastrointestinal stromal tumor (GIST), which arises from interstitial cells of Cajal; primary colorectal lymphoma, which arises from blood lineage cells; leiomyosarcoma, which is a sarcoma arising from connective tissue or smooth muscle; melanoma, which arises from melanocytes; squamous cell carcinoma, which arises from stratified squamous epithelial tissue and is localized to the rectum; and mucinous carcinoma, which is an epithelial cancer that is generally associated with a poor prognosis.

[0081] Symptoms of colorectal neoplasia or colorectal cancer may include, but are not limited to, bowel habits change, such as diarrhea or constipation, or stool consistency change that lasts for more than 4 weeks, rectal bleeding or blood in stool, persistent abdominal discomfort, such as sudden abdominal pain, gas or pain, feeling that bowel does not empty completely, weakness or fatigue, and unexplained weight loss.Patients suspected of having colorectal neoplasia or colorectal cancer may undergo peripheral blood tests, such as complete blood count (CBC), fecal occult blood test (FOBT), liver function analysis, immunochemical stool test (FIT) and / or other analysis of specific tumor markers, such as carcinoembryonic antigen (CEA) and CA19-9.Colorectal neoplasia or colorectal cancer is often diagnosed based on colonoscopy. During colonoscopy, any identified polyps are removed, biopsied, and analyzed to determine whether they contain colorectal cancer cells or cells undergoing precancerous changes. Each of the specific cancers listed above may have a different appearance when viewed through an endoscope. Villous adenomas Melanomas and squamous cell carcinomas are typically flat or sessile, while tubular adenomas, lymphomas, leiomyosarcoma, and GIST tumors are typically pedunculated. However, flat and sessile adenomas may be missed by gastroenterologists during colonoscopy. Biopsy samples may be subjected to further analysis based on genetic alterations or microsatellite instability in specific genes.

[0082] Other diagnostic methods may include sigmoidoscopy; imaging tests, such as computed tomography (CT or CAT) scans; ultrasound, such as abdominal, rectal or intraoperative ultrasound; or magnetic resonance imaging (MRI) scans, such as transrectal MRI. Other tests, such as angiograms and chest x-rays, may be performed to determine whether colorectal cancer has metastasized.

[0083] Various methods have been developed for staging colorectal cancer. The most commonly used system, the TNM system, is based on three components: 1) how far the primary tumor has grown within the wall of the bowel and into adjacent areas (T); 2) whether the tumor has spread to adjacent regional lymph nodes (N); and 3) whether the cancer has metastasized to other organs (M). Other staging methods include the Dukes staging system and the Astra-Coller system.

[0084] The TNM system provides four stage classifications for colorectal cancer. In stage 1 (T1) colorectal cancer, the tumor has grown within the layers of the colon wall but has not spread outside the colon wall or into the lymph nodes. If the cancer is part of a tubular adenoma polyp, a simple resection is performed and the patient may continue to receive regular checkups for future cancer development. If the cancer is high-grade or part of a flat / sessile polyp, further surgery may be required and wide margins are taken; this may include a partial colectomy in which a piece of the colon is removed. In stage 2 (T2) colorectal cancer, the tumor has grown within the wall of the colon and possibly into adjacent tissues but has not spread to the adjacent lymph nodes. Surgical removal of the tumor and a partial colectomy are typically performed. Adjuvant therapy, such as chemotherapy with drugs such as 5-fluorouracil, leucovorin, or capecitabine, may be administered. Such tumors are unlikely to recur, but increased screening for patients is generally warranted. In stage 3 (T3) colorectal cancer, the tumor has spread to nearby lymph nodes but not to other parts of the body. Surgery is required to remove a section of the colon and all affected lymph nodes. Chemotherapy using drugs such as 5-fluorouracil, leucovorin, oxaliplatin or capecitabine in combination with oxaliplatin is typically recommended. Radiation therapy may also be used depending on the age of the patient and the aggressive nature of the tumor. In stage 4 (T4) colorectal cancer, the tumor has spread from the colon via the blood to distant organs. Colorectal cancer frequently metastasizes to the liver, lungs and / or peritoneum. Surgery is unlikely to cure such cancer, and chemotherapy and / or radiation are generally required to improve survival.

[0085] The method disclosed herein is generally useful for the diagnosis and treatment of colorectal neoplasia.The expression levels of 2, 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 or more stool-derived eukaryotic RNA biomarkers, such as stool-derived eukaryotic RNA biomarkers selected from Table 1 or Table 2 or a combination of Table 1 and Table 2, or stool-derived eukaryotic RNA biomarkers in Table 3 or Table 4, are measured in biological samples, such as stool samples from subjects.The subject may be a patient with one or more of the above symptoms that may indicate that the patient is at risk of colorectal cancer. A subject may also be a patient who does not have symptoms but is believed to be at risk for colorectal neoplasia, inflammatory bowel disease, or genetic syndromes associated with high risk of colorectal cancer based on age (e.g., over 50 years of age), family history, obesity, diet, alcohol consumption, tobacco use, history of diagnosis of colorectal polyps, racial and ethnic background, such as familial adenomatous polyposis, Gardner's syndrome, Lynch syndrome, Turcot's syndrome, Peutz-Jeghers syndrome, and MUTYH-associated polyposis. The methods disclosed herein are also useful for monitoring patients who have previously been diagnosed and treated with colorectal neoplasia or colorectal cancer to monitor remission and to detect recurrence of the disease.

[0086] In some embodiments, the disease state of a subject, i.e., a human or non-human animal patient, is determined by pathological evaluation. For example, in one type of disease, such as colorectal cancer, the extent of disease is classified as stage 1 (T1), stage 2 (T2), stage 3 (T3) and stage 4 (T4). Colorectal cancer can be tubular adenocarcinoma, villous adenocarcinoma, gastrointestinal stromal tumor, primary colorectal lymphoma, leiomyosarcoma, melanoma, squamous cell carcinoma or mucinous carcinoma. In another type of disease, such as inflammatory bowel disease, the disease state is determined by the location of the disease along the intestinal tract and histological features, such as granuloma, leukocyte infiltration and / or crypt abscess. Other methods for determining disease state, such as physician's judgment, physical symptoms, fecal occult blood test, immunochemical stool test, sigmoidoscopy, FIT-DNA, CT colonography or colonoscopy, can also be used in conjunction with the methods disclosed herein.

[0087] Also provided is a method for determining whether a subject is at risk for bowel disease.Bow disease can include intestinal cancer, colorectal cancer, adenomatous polyps that show precancerous changes, irritable bowel syndrome, necrotizing enterocolitis, ulcerative colitis, Crohn's disease, celiac disease or other bowel disease.The method for determining whether a subject is at risk for bowel disease can be determined by using the present invention to detect a) deoxyribonucleic acid (DNA) sequence, b) ribonucleic acid (RNA) sequence, c) predicted amino acid sequence that constitutes protein backbone, d) expression level of ribonucleic acid biomarker, e) predicted amino acid sequence variation, or f) any combination of the above, where the difference between control and experimental sample can indicate that the subject is at risk for bowel disease.

[0088] The methods and compositions are also useful for selecting a clinical plan for a subject with a bowel disease. Through this method, the clinical plan may include the administration of additional diagnostic procedures. In some embodiments, the clinical plan may include a treatment method.

[0089] For example, an algorithm for determining the diagnosis, state, or response to treatment can be determined for a specific clinical condition.The algorithm used in the methods provided herein can be a mathematical function incorporating multiple parameters that can be quantified using, but not limited to, medical devices, clinical assessment scores, or biological / chemical / physical tests of biological samples.Each mathematical function can be a weight-adjusted representation of the parameter level (e.g., measured level) that is determined to be relevant to the selected clinical condition.Because the method involves weighting and evaluation of multiple marker panels, a computer with reasonable computing power can be used to analyze data.

[0090] Thus, a diagnostic method may include obtaining a stool sample from a patient at risk of or suspected of having a colorectal neoplasm; determining the expression of 2, 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 or more stool-derived eukaryotic RNA biomarkers selected from those listed in Table 1 or Table 2 or a combination of Tables 1 and 2 or Table 3 or Table 4; and obtaining a test value by a machine learning algorithm incorporating the plurality of stool-derived eukaryotic RNA biomarkers with predetermined coefficients. Exemplary machine learning algorithms include support vector machines, gradient boosting, adaptive boosting, random forest, naive Bayes, decision trees, or k-nearest neighbors. A significant change in the expression of the plurality of colorectal neoplasia biomarkers relative to a control, e.g., a population of healthy individuals, indicates that the patient is likely to have colorectal neoplasia. In some embodiments, the expression levels measured in the sample are used to derive or calculate a probability or confidence score. This value can be derived from the expression level. Alternatively or additionally, the value can be derived from a combination of the expression level and other factors, such as the patient's medical history, ethnicity, sex, age, smoking status, previous genomic test results, previous histopathology test results, and genetic background. Alternatively or additionally, the value can be derived from a combination of the expression level and a fecal immunochemical test (FIT). In some embodiments, the method can further comprise a step of communicating the test value to the patient. This method could include, for example, a visual display of the marker, a numerical output of the marker, or other communication method.

[0091] In some embodiments, predictions for one or more patients may be made using a model-based approach. For example, in some embodiments, a random forest model may be configured to predict the absence, presence, and / or severity of disease in one or more groups, such as colorectal cancer, HRA, MRA, LRA, benign polyps, or no findings. In some embodiments, a validation data set and / or a test data set may be applied to test or refine the model. Once the model is created, it is used to predict the absence, presence, and / or severity of disease for one or more specific patients based on the provided input, such as a plurality of amplicons. Although specific embodiments are discussed herein, it will be recognized that any suitable model may include any number of decision trees, nodes, input layers, output layers, hidden layers, or other various parameters. In some embodiments, a random forest model may be created that uses a greater and / or lesser number of decision trees, a greater and / or lesser number of candidate features, and the like.

[0092] In some embodiments, the one or more models can be created, tested and / or executed using a system configured for disease detection. In some embodiments, the system includes a computer system having one or more processors. Each processor is connected to a communication infrastructure (e.g., a communication bus, a cross-over bar or a network). The processor can be implemented as a central processing unit, an embedded processor or microcontroller, an application specific integrated circuit (ASIC) and / or any other circuit configured to execute computer executable instructions to perform one or more steps. The processor is similar to the processor discussed above, and the same description will not be repeated here. The computer system can include a display interface that transfers diagrams, text and other data to a user from the display communication infrastructure (or from a frame buffer) of a display unit.

[0093] The computer system may also include a primary storage device, such as a random access memory (RAM), and a secondary storage device. The primary storage device and / or the secondary storage device may include a dynamic random access memory (DRAM). The secondary storage device may include, for example, a hard disk drive (HDD), which may be a solid-state memory, and / or a removable storage drive, an optical disk drive, a flash drive, a magnetic tape drive, and the like. The removable storage drive reads from and / or writes to a removable storage unit. The removable storage unit may be an optical disk, a magnetic disk, a floppy disk, a magnetic tape, and the like. The removable storage unit may include a computer-readable storage medium on which data is tangibly stored (or embodied) and / or computer-executable software instructions, for example, for causing a processor(s) to perform various operations and / or one or more steps.

[0094] In alternative embodiments, secondary storage devices may include other devices for allowing computer programs or other instructions to be loaded into a computer system. Secondary storage devices may be similar to removable storage drives and may include removable storage units having their own removable storage units and corresponding removable storage interfaces. Examples of such removable storage units include, but are not limited to, Universal Serial Bus (USB) or flash drives that allow software and data to be transferred from the removable storage units to the computer system.

[0095] A computer system may also include a communications interface (e.g., a networking interface). A communications interface allows instructions and data to be transferred between the computer system and one or more additional systems. A communications interface also provides for communication with other external devices. Examples of communications interfaces may include modems, Ethernet interfaces, wireless network interfaces (e.g., radio frequency, IEEE 802.11 interfaces, Bluetooth interfaces, etc.), Personal Computer Memory Card International Association (PCMCIA) slots and cards, etc. The instructions and data transferred through the communications interface may be in the form of signals that may be received by the communications interface, and may be electronic, electromagnetic, optical, etc. Such signals may be provided to the communications interface over a communications path (e.g., channel) that may be implemented using wire, cable, optical fiber, telephone line, mobile link, radio frequency (RF) link, and other communications channels.

[0096] The methods and systems described herein may be embodied at least in part in the form of a computer-executed process and an apparatus for carrying out the method. The methods disclosed herein may also be embodied at least in part in the form of a tangible non-transitory machine-readable storage medium coded with a computer-executable program code. Such medium may include, for example, a RAM, a ROM, a CD-ROM, a DVD-ROM, a BD-ROM, a hard disk drive, a flash memory, or any other non-transitory machine-readable storage medium, in which the computer program code is loaded and executed in a computer, making the computer an apparatus for carrying out the method. The methods may also be embodied at least in part in the form of a computer in which the computer program code is loaded and / or executed, making the computer a dedicated computer for carrying out the method. The computer program code segments, when executed in a general-purpose processor, configure the processor to create specific connections, circuits, and algorithms for carrying out the method disclosed herein.

[0097] Standard computing devices and systems can be used, implemented, e.g., appropriately programmed, to perform the calculations required to perform the methods described herein, e.g., to determine the values ​​described herein.Computing devices include various forms of digital computers, such as laptops, desktops, mobile devices, workstations, personal digital assistants, servers, blade servers, mainframes, and other suitable computers.In some aspects, the computing device is a mobile device, such as a personal digital assistant, a mobile phone, a smartphone, a tablet, or other similar computing device.

[0098] In some embodiments, a computer may be used to communicate information, for example, to a health care professional. Information may be communicated to the professional by making the information available electronically (e.g., in a secure manner). For example, information may be placed on a computer database so that the health care professional can access the information. Information may also be communicated to a hospital, clinic, or research facility that acts as an agent for the professional. Information transferred over an open network (e.g., the Internet or e-mail) may be encrypted. Patient gene expression data and analysis may be encrypted and stored in the cloud. A 256-bit AES method with tamper resistance may be used for disk encryption; an SSL protocol may preferably ensure protection during data transmission, and a SHA2-HMAC key management method may allow authenticated access to data. Other secure data storage means may also be used.

[0099] The results of such an analysis as above, e.g., a probability or confidence score derived from the expression level in combination with other factors, e.g., the patient's medical history, ethnicity, sex, age, smoking status, previous genomic test results, previous histopathology test results, genetic background, or fecal immunochemical test (FIT), may be the basis for follow-up and treatment by the attending physician. If the expression levels of 2, 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 or more stool-derived eukaryotic RNA biomarkers, e.g., stool-derived eukaryotic RNA biomarkers selected from Table 1 or Table 2 or a combination of Tables 1 and 2, or stool-derived eukaryotic RNA biomarkers in Table 3 or Table 4, are not significantly different from the expression levels of the same stool-derived eukaryotic RNA biomarkers in a control, the physician may determine that the patient is not currently at risk for colorectal neoplasia. Such patients may be invited to return for rescreening in the future. The extent to which the expression levels of 2, 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 or more stool-derived eukaryotic RNA biomarkers, such as those selected from Table 1 or Table 2 or a combination of Tables 1 and 2, or those in Table 3 or Table 4, are not significantly different from the expression levels of the same stool-derived eukaryotic RNA biomarkers in a control, may be used to determine the required follow-up time. In some embodiments, the physician may recommend that the patient return for follow-up after 1 month, 2 months, 3 months, 6 months, 1 year, 2 years, 3 years, 5 years, or 10 years. The methods disclosed herein can be used to monitor any changes in the levels of colorectal neoplasia markers over time. Subjects can be monitored for any length of time after initial screening and / or diagnosis.For example, subjects may be monitored for at least 2, 4, 6, 8, 10, 12, 14, 16, 18, 20, 25, 30, 35, 40, 45, 50, 55, or 60 months or more, or for at least 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20 years or more.

[0100] The methods and compositions disclosed herein are useful for selecting a clinical plan for subjects who are at risk of colorectal neoplasia or colorectal cancer or suffer from colorectal neoplasia or colorectal cancer.Clinical plan may include additional diagnostic procedures, such as fecal occult blood test, immunochemical stool test, or colonoscopy to remove cancer, polyp or precancerous lesion.In some embodiments, clinical plan may include a treatment method.In some embodiments, the method includes the step of selecting a treatment for subjects with colorectal neoplasia or colorectal cancer. If the expression level of 2, 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 or more of the stool-derived eukaryotic RNA biomarkers selected from Table 1 or Table 2 or a combination of Tables 1 and 2, or Table 3 or Table 4, is significantly different from the expression level of the same 2, 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 or more of the stool-derived eukaryotic RNA biomarkers in a control, the patient may have colorectal neoplasia or cancer. In such cases, further screening may be recommended, such as increased frequency of screening using the methods disclosed herein, as well as fetal occult blood testing, immunochemical stool testing and / or colonoscopy.The expression levels of 2, 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 or more of the stool-derived eukaryotic RNA biomarkers selected from Table 1 or Table 2 or a combination of Tables 1 and 2 or Table 3 or Table 4 in a control. If the expression levels of the same 2, 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 or more of the combinations or stool-derived eukaryotic RNA biomarkers selected from Table 3 or Table 4 are significantly different, the patient may have a particular type of colorectal neoplasm, e.g., high-risk adenoma. In some embodiments, treatments such as, for example, colonoscopy with removal of polyps, chemotherapy, immunotherapy, or surgery, e.g., bowel resection, may be recommended. Thus, the method can be used to determine the level of expression of 2, 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 or more stool-derived eukaryotic RNA biomarkers, such as 2, 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 or more biomarkers or variants thereof ... and then to determine a course of treatment. A subject, i.e., a patient, is effectively treated whenever it produces clinically beneficial results.This can mean, for example, the complete elimination of disease symptoms, the reduction of disease symptoms severity, or the slowing down of disease progression.Such a method can further comprise: a) identifying the subject (e.g., patient, more specifically, human patient) with colorectal neoplasia or colorectal cancer, and b) providing the subject with anti-cancer treatment, for example, therapeutic agents and, for example, immunotherapeutic agents, surgery or radiation therapy.The amount of therapeutic agent provided to the subject that causes complete elimination of disease symptoms, reduction of disease symptoms severity, or slowing down disease progression is considered to be therapeutically effective amount.The method of the present invention can also include a monitoring step to help optimize dosing and scheduling and predict outcome.Monitoring can also be used to detect the development of drug resistance, to quickly distinguish responding patients from non-responding patients, or to evaluate cancer recurrence.If signs of resistance or non-response are found, physicians can select alternative or supplementary agents before tumors develop additional escape mechanisms.

[0101] The method disclosed herein can also be used in combination with conventional methods for the diagnosis and treatment of colorectal cancer.Therefore, the diagnostic method can be used in combination with standard diagnostic methods for colorectal cancer.For example, the method can be used in combination with fecal occult blood test, immunochemical stool test or colonoscopy.The method can also be used with other colorectal cancer markers, such as KRAS, NRAS, BRAF, CEA, CA 19-9, p53, MSL, DCC, MSI and MMR.

[0102] The diagnostic methods disclosed herein may also be used in conjunction with the treatment of colorectal cancer. Treatment methods for colorectal cancer fall into several general categories: surgery, chemotherapy, radiation therapy, targeted therapy, and immunotherapy. Surgery may include colectomy, colostomy with partial hepatectomy, or protectomy. Chemotherapy may be systemic chemotherapy or local chemotherapy, where chemotherapeutic agents are placed in close proximity to the affected organ. Exemplary chemotherapeutic agents may include 5-fluorouracil, oxaliplatin or its derivatives, irinotecan or its derivatives, leucovorin or capecitabine, mitomycin C, cisplatin, and doxorubicin. Radiation therapy may be external radiation therapy, which uses a machine to direct radiation to the cancer, or internal radiation therapy, where a radioactive material is placed directly in or near the colorectal cancer. Targeted agents may include antiangiogenic agents, such as bevacizumab) or EGFR-inhibiting monoclonal antibodies (cetuximab, panitumumab), ramucilamab (anti-VEGFR2), aflibercept, regorafenib, tripfluridine-tipiracil, or combinations thereof. Targeted agents may also be combined with standard chemotherapy agents. Immunotherapy may include administration of specific antibodies, such as anti-PD-1 antibodies, anti-PD-L-1 antibodies, and time-CTLA-4 antibodies, anti-CD27 antibodies; cancer vaccines, adoptive cell therapy agents, oncolytic virotherapy agents, adjuvant immunotherapy agents, and cytokine-based therapy agents. Exemplary immunotherapy agents may include Keytruda, Opdiva, and iprimumab. Other treatment methods include stem cell transplantation, hyperthermia, photodynamic therapy, blood product donation and transfusion, or laser treatment.

[0103] We may use the terms "increased / increased," "increase / increase" or "upregulated" to generally contemplate a statistically significant amount of increase / increase in the level of a biomarker. In some embodiments, the increase / increase may be at least a 10% increase / increase compared to a control, such as at least about a 20%, or at least about a 30%, or at least about a 40%, or at least about a 50%, or at least about a 60%, or at least about a 70%, or at least about a 80%, or at least about a 90% increase / increase compared to a control, or up to 100%, including 100%, or any increase / increase between 10-100%, or at least about a 0.5-fold increase / increase, or at least about a 1.0-fold increase, or at least about a 1.2-fold increase, or at least about a 1.5-fold increase, or at least about a 2-fold increase, or at least about a 3-fold increase, or at least about a 4-fold increase, or at least about a 5-fold increase, or at least about a 10-fold increase, or any increase / increase between 1.0-fold and 10-fold or more.

[0104] We may use the terms "reduce," "reduced," "reduced," "reduction," or "downregulated" to indicate a statistically significant amount of reduction in the level of a eukaryotic biomarker. In some embodiments, the reduction may be at least a 10% reduction compared to a control, such as at least about a 20% reduction compared to a control, or at least about a 30% reduction, or at least about a 40% reduction, or at least about a 50% reduction, or at least about a 60% reduction, or at least about a 70% reduction, or at least about a 80% reduction, or at least about a 90% reduction, or up to 100% reduction (i.e., the absence of levels compared to a control), or any reduction between 10 and 100%, or at least about a 0.5-fold reduction compared to a control, or at least about a 1.0-fold reduction, or at least about a 1.2-fold reduction, or at least about a 1.5-fold reduction, or at least about a 2-fold reduction, or at least about a 3-fold reduction, or at least about a 4-fold reduction, or at least about a 5-fold reduction, or at least about a 10-fold reduction, or any reduction between 1.0-fold and 10-fold or more.

[0105] The statistical significance of the increase / increase in eukaryotic biomarker or the decrease in eukaryotic biomarker can be expressed as p-value or q-value.Depending on the specific eukaryotic biomarker, p-value or q-value can be less than 0.05, less than 0.01, less than 0.005, less than 0.002, less than 0.001 or less than 0.0005.Q-value can be a value derived from p-value.In some embodiments, q-value can be the p-value adjusted for false discovery rate.

[0106] A control may be a biological sample taken from a patient or a group of patients. In some embodiments, a control may be a reference value. A control may be from an individual or a population of individuals who have been diagnosed as healthy. A healthy individual may include, for example, an individual who has tested negative in a fecal parasite test, fecal bacterial test, colonoscopy or endoscopy within the last year. A control may be from an individual or a population of individuals who have been diagnosed with a disease state. An individual who has tested positive in a fecal parasite test, fecal bacterial test, colonoscopy or endoscopy within the last year. A control may be from an individual or a population of individuals who have previously been diagnosed with a disease but are now in remission, do not have active disease, or are not currently suffering from the disease. A control may be from an individual at one, two, or more time points. For example, a control may be a biological sample taken from a subject at an earlier time point. A control may be a baseline reference value for a specific biomarker. A normative reference value may be derived based on evaluation of individuals of similar age, sex, gender, build, ancestry, ethnic background, or general health status. In some embodiments, the control may be an algorithmically derived value or values.

[0107] The experimental sample can be a biological sample taken from a subject.The experimental sample can be taken from a subject with known or unknown health status.In some embodiments, the health status of the subject can be examined, for example, by analyzing the experimental sample, biopsy, physical examination, laboratory findings, visual examination, or genetic analysis.The health status of the subject that can be examined by the experimental sample can be diseased, at risk of disease, or healthy.

[0108] manufactured goods Also provided is a kit for detecting and quantifying selected stool-derived eukaryotic RNA biomarkers in biological samples, such as stool samples.Thus, packaged products (e.g., sterile containers containing one or more compositions described herein, packaged in concentrated or ready-to-use concentrations for storage, delivery or sale) and kits are also within the scope of the present invention.The product may include a container (e.g., vial, bottle, bag, microplate, microchip or bead) containing one or more compositions of the present invention.The article of manufacture may also include, for example, packaging materials, instructions for use, syringes, delivery devices, buffers or other control reagents.

[0109] The kit may include a compound or agent capable of detecting RNA corresponding to 2, 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 or more stool-derived eukaryotic RNA biomarkers, such as RNA corresponding to stool-derived eukaryotic RNA biomarkers selected from Table 1 or Table 2 or a combination of Tables 1 and 2 or Table 3 or Table 4, in a biological sample; and a standard; and, optionally, one or more reagents necessary for detection, quantification or amplification. In some embodiments, the kit may include a compound or agent capable of detecting RNA corresponding to 2, 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 or more stool-derived eukaryotic RNA biomarkers in a biological sample, such as RNA corresponding to stool-derived eukaryotic RNA biomarkers selected from Table 1 or Table 2 or a combination of Table 1 and Table 2 or Table 3 or Table 4; and a standard; and optionally one or more reagents required for detection, quantification or amplification. The compound, agent and / or reagent may be packaged in a suitable container. The kit may further include instructions for using the kit to detect and quantify nucleic acids. The kit may also include a control or a series of controls that can be assayed and compared to the test sample contained therein. Each component of the kit may be enclosed in an individual container, and all of the various containers may be included in a single package, together with instructions for interpreting the results of the assay performed using the kit. In some embodiments, the kit may include primers or oligonucleotide probes specific for one or more control markers.In some embodiments, the kit includes reagents specific for the quantification of 2, 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 or more stool-derived eukaryotic RNA biomarkers, such as those selected from Table 1 or Table 2 or a combination of Tables 1 and 2, or Table 3 or Table 4.

[0110] In some embodiments, the kit may include reagents for separating eukaryotic cells from bacterial cells and other components in stool, and for extracting stool-derived eukaryotic RNA from a patient's stool sample, for example, from a human patient's stool sample. Thus, the kit may include buffers, emulsion beads, silica beads, stabilizing reagents, and various filters and centrifugation vessels. The kit may also include instructions for handling stool to minimize sample contamination and ensure the stability of stool-derived eukaryotic RNA in the stool sample. The kit may also include materials to ensure sample preservation, such as stabilizing buffers, ice packs, or heat packs. In some embodiments, the kit may include a stool collection device.

[0111] The product may also include instructions (e.g., a printed label or insert or other media (e.g., audio- or videotape or computer-readable media) that explain the use of the product. The instructions may be associated with the container (e.g., affixed to the container) and may describe the manner in which the reagent may be used. The reagent may be ready for use (e.g., present in a suitable unit) and may include one or more additional adjuvants, carriers or other diluents. Alternatively, the reagent may be provided in concentrated form, along with a diluent and instructions for dilution. EXAMPLES

[0112] Example 1: Obtaining human stool samples Collection of human stoolPatients were asked to defecate into a bucket attached to a toilet seat, and the resulting samples were stored in a freezer until transport to Kharkiv National Medical University (Kharkov, Ukraine). Stool was aliquoted into 50 mL conical tubes and stored at -80°C. The samples were transported from Kharkiv National Medical University under dry ice to Capital Biosciences (Gaithersburg, MD) and immediately transferred to a -80°C freezer. From there, the samples were transported under dry ice to BioGenerator Labs (St. Louis, MO), where they were stored in a -80°C freezer until extraction.

[0113] Human sample typeStool samples were obtained from 195 patients with colorectal cancer (stages I-IV), 126 patients with precancerous adenomas, 8 patients with benign polyps, and 125 patients with negative colonoscopy findings, resulting in a pooled sample of 454 cases. Healthy individuals were those with no history of colorectal cancer, inflammatory bowel disease, celiac disease, irritable bowel syndrome, diarrhea within the last 20 days, or any other gastrointestinal disease. Patients with benign polyps provided stool samples before undergoing colonoscopy, during which physician-detected polyps were determined to be benign by subsequent biopsy and histological evaluation. Diseased individuals were those diagnosed with colorectal cancer or precancerous adenomas. Colorectal cancer patients had been diagnosed with stage I-IV colorectal cancer by colonoscopy and subsequent biopsy within the last month and had not yet undergone any post-biopsy treatment, which may include chemotherapy, radiation, and / or surgery. Patients with precancerous adenomas provided stool samples prior to undergoing colonoscopy, during which physicians detected polyps that were determined to be precancerous by subsequent biopsy and histological evaluation. Healthy and benign polyp individuals were matched with adenoma and cancer patients based on gender and age group (50-60 years, 60-70 years, 70-80 years, and 80-90 years). Patients used in this collection provided consent through Capital Biosciences. The Schulman Internal Review Board provided ethical oversight of this collection.

[0114] Example 2: Extraction of human nucleic acids Total Nucleic Acid Extraction:Each stool sample was placed in a 50 mL conical tube. Approximately 1,000-25,000 mg of stool was added to each tube. An additional 20-40 mL of solution was added to each tube. The solution contained a mixture of Hank's Balanced Salt Solution (HBSS) (Sigma-Aldrich) with 0.05% Tween-20 (Sigma-Aldrich) and 0.0002% RNAse Inhibitor (Sigma-Aldrich). The stool was suspended in the solution and spun at approximately 0-10°C for 0-10 minutes. The solution was centrifuged at 1000 rpm at 4°C for 10 minutes and the supernatant was discarded. Approximately 4-10 mL of EasyMag® ​​Lysis Buffer (bioMerieux) was added to the pellet and the pellet was resuspended in the solution. The solution was centrifuged at 2500-3500 rpm at 20-25°C for 10-15 minutes. During differential centrifugation, the solution separated into three layers. The bottom layer contained solid intracellular debris, the middle layer was a hydrophilic layer enriched in human nucleic acids, and the top layer was a hydrophobic lipid layer. The top two layers were transferred to a new 15 mL conical tube and the solution was centrifuged again at 2500 rpm for 10 minutes at 20-25°C. This centrifugation step resulted in separation into three layers: the bottom layer was solid intracellular debris, the middle layer was a hydrophilic layer enriched in human nucleic acids, and the top layer was a hydrophobic lipid layer. To screen for large debris from the solution, a 20 uL pipette tip was placed onto a 1 mL pipette tip and 2 mL of the hydrophilic layer was pipetted from the 15 mL tube and transferred to an EasyMag® ​​Disposable cartridge (bioMerieux). Additionally, 60 uL of EasyMag® ​​Magnetic Silica (bioMerieux) was added to the cartridge. The beads were mixed in the solution for 0.5-1 min using a pipette. The nucleic acid bound to the beads was eluted in a buffer solution using Specific A Protocol according to the manufacturer's instructions. The volume of eluted nucleic acid was 70 uL. The nucleic acid solution was pipetted into a 1.5 mL tube and placed on ice.The same EasyMag® ​​Disposable Cartridge (bioMerieux) used in the previous step was then reloaded with an additional 2 mL of hydrophilic layer of the same solution in the 15 mL tube used previously using the same technique to screen out large debris. An additional 20 uL of EasyMag® ​​Magnetic Silica (bioMerieux) was added to the cartridge. The beads were mixed in solution for 0.5-1 min using a pipette. The nucleic acid bound to the beads was eluted in a buffer solution as described above using Specific A Protocol according to the manufacturer's instructions. The volume of eluted nucleic acid was 70 uL. This nucleic acid solution was pipetted into the original 1.5 mL tube already containing the initial 70 uL of elution solution and the combined solution was placed on ice.

[0115] DNAse treatment: The 140 uL solution was treated with Baseline-Zero-DNase (Epicentre) at 35-40°C for 20-40 min. A 1-2 mL aliquot of EasyMag® ​​Lysis Buffer was added to the DNAse-treated solution and the sample was transferred to a new EasyMag® ​​Disposable cartridge. The entire solution was added to the new cartridge along with 60 uL of EasyMag® ​​Magnetic Silica. The nucleic acid bound to the beads was eluted in the buffer solution using the EasyMag® ​​Generic Protocol according to the manufacturer's instructions. The volume of the eluted nucleic acid was 25 uL. The nucleic acid solution was pipetted into a 1.5 mL tube and stored at 0-6°C.

[0116] Example 3: Measuring human nucleic acid levels in human stool samples Extraction result1-2 uL of each sample extracted above was assessed for total nucleic acid and RNA integrity using an Agilent 2100 Bioanalyzer. The samples were analyzed qualitatively and quantitatively. Electrophoretic analysis was used to confirm the quality of the extracted RNA. Electrophoretic files were read by comparing the bands of each sample with those indicated by size markers of the RNA ladder and identifying 18S and 28S ribosomal RNA (rRNA) bands. The rRNA bands are two thick prominent bands near the 2,000 nucleotide marker in the standardization ladder. Qualitatively, sufficient banding and strong band intensity indicated that sufficient intact nucleic acid was available for further analysis, e.g., microarray sequencing, polymerase chain reaction (PCR), nucleic acid sequencing, molecular barcoding or probe-capture. The electrophoretic separation diagram is a graphical representation of each electrophoretic file with quantification of RNA integrity number (RIN), total RNA mass and total rRNA mass. Quantitatively, the higher the RIN, the higher the total RNA mass, and the higher the total rRNA mass, the more likely the sample will be useful for further analysis, such as microarray sequencing, polymerase chain reaction (PCR), nucleic acid sequencing, molecular barcoding, or probe-capture.

[0117] Figure 1A shows the electrophoretic files of six samples and the electrophoretic separation diagram of one sample extracted according to the method described in the literature. Figure 1B shows the electrophoretic files of six samples and the electrophoretic separation diagram of one sample extracted as described above. The sample extracted as described above yielded a higher RIN and more eukaryotic mass. The higher quality of seRNA extracted as described above was also shown by clearer ribosomal RNA bands (18S and 28S) and less bacterial noise as evidenced by minimal banding below the 18S band.

[0118] Example 4: Incubation of seRNA in stabilizing buffer Eleven samples were selected to perform the test with the stabilization buffer. These samples were divided into 5 gram aliquots into three cohorts: Cohort 1 (n=11), Cohort 2 (n=11) and Cohort 3 (n=8). Cohort 1 samples were extracted immediately using the method described above (Figure 2A). Cohort 2 samples were incubated in the stabilization buffer and stored at room temperature for 24 hours before being extracted using the method described above (Figure 2B). Cohort 3 samples were incubated in the stabilization buffer and stored at room temperature for 48 hours before being extracted using the method described above (Figure 2C).

[0119] After extraction, all samples (n=30) were qualitatively analyzed using an Agilent Bioanalyzer. Clear, distinct, intact ribosomal 18S and 28S bands are detected in the isolated RNA of all samples. The intensities of 18S and 28S, which can be used to estimate the amount of eukaryotic RNA, were increased by incubation in the stabilizing buffer. Furthermore, the bacterial noise, indicated by banding below 18S, was reduced by incubation in the stabilizing buffer.

[0120] The Agilent Bioanalyzer data also allowed for quantification of RNA Integrity Number (RIN) and eukaryotic mass. RIN was sufficient for all samples. Total RIN for each cohort was increased by incubation in stabilization buffer, with mean RINs of 4.6, 5.9, and 7.1 for cohort 1, cohort 2, and cohort 3, respectively. Eukaryotic mass was sufficient for all samples. Total eukaryotic mass was increased by incubation in stabilization buffer, with mean masses of 11.1 ng, 39.7 ng, and 78.4 ng for cohort 1, cohort 2, and cohort 3, respectively.

[0121] Example 5: Analysis of RNA transcripts 330 samples were selected for analysis using the Affymetrix GeneChip™ Human Transcriptome Array 2.0 (Santa Clara, CA). Approximately 100 ng of DNase-free fecal RNA was amplified using the Ambio WT-pico kit with subsequent hybridization to the Affymetrix GeneChip™ Human Transcriptome Array 2.0 according to the manufacturer's protocol. All samples were normalized in the Affymetrix Expression Console™ using Signal Space Transformation-Robust Multiarray Analysis (SST-RMA).

[0122] Of the 70,523 transcript clusters in the Affymetrix Microarray, a subset of 5,149 transcript clusters corresponding to 3,977 genes were preselected and assessed for differential expression. This initial selection reduced the false discovery rate and filtered out genes with no known function in cancer initiation and progression.

[0123] The 330 individuals were split into a training set of 265 individuals and a test set of 65 individuals. The training set was used to identify differentially expressed genes and to build a computational model, while the test set was used to test the detection accuracy of the computational model. The standard LIMMA package was used to identify a subset of RNA transcript clusters that were differentially expressed between individuals with either precancerous adenomas or CRC and those with no colonoscopy findings. All biomarkers were ranked according to their log-odds scores, and the top 200 ranked biomarkers (p<0.05) were used as features in building the machine learning model. The Support Vector Machine Model (ν-SVM) with RBF kernel was chosen for model development. This kernel function allows the calculation of distances between individuals by expanding the features into a higher dimensional space that is not explicitly computed. The SVM finds the maximum margin hyperplane that separates the label groups. The parameter ν defines the lower bound on the proportion of individuals used to examine the maximum margin. The SVM model was trained using expression levels of 200 transcripts from all 265 individuals in the training set. Internal validation of the SVM yielded an overall ROC AUC of 0.776. The model yielded ROC AUCs of 0.829 and 0.788 when assessing CRC and adenoma, respectively (Figure 3A).

[0124] We also used this multi-target RNA biomarker algorithm on 65 individuals in an independent test set. The model correctly identified 79% (34 of 43) of all individuals with positive screening colonoscopy findings, 95% of individuals with precancerous adenomas, and 65% of individuals with cancer. Model sensitivity in CRC correlated directly with size, such that 72% of tumors >4 cm in diameter were correctly detected. Model sensitivity to size in adenomas was agnostic, with 100% predictive accuracy for both small (<5 mm) and large (>1 cm) lesions (Figure 3B).

[0125] Example 6: CRC molecular subtyping using seRNA expression signatures Of the 70,523 transcript clusters in the Affymetrix Microarray, a subset of transcript clusters corresponding to 274 genes was selected to annotate patient samples from individuals diagnosed with colorectal cancer using the consensus molecular subtypes (CMS) of CRC defined by the Colorectal Cancer Subtyping Consortium (CRCSC) (Figure 4A). The CRCSC classifier is organized based on the importance of each gene with respect to its ability to improve the accuracy of molecular subtype classification. Expression of transcript clusters was summarized at the gene level using the median luminescence of the transcript clusters associated with each gene. Gene expression data were normalized at the gene level and across the entire cohort using the median expression levels. The normalized data was used as input to a random forest classifier defined in the R package CMS Classifier to label consensus molecular subtypes.

[0126] The output from the CMS Classifier includes four values, each of which is a posterior probability of how likely the sample is to be associated with CMS1-4. CMS1 includes tumors with a signature associated with high microsatellite instability (MSI-H) and immune infiltration. Figure 4B shows 25 exemplary colorectal neoplasia molecular subtype biomarker genes useful for identifying colorectal cancer subtype CMS1. CMS2-4 are associated with canonical, metabolic or mesenchymal gene expression signatures, respectively. Based on the CMS classifier, 14 of 117 (12%) individuals were classified as CMS1, 100 of 117 (85%) were classified as CMS2-4 (canonical, metabolic and mesenchymal), and 3 of 117 (3%) were classified as CMS1 / CMS2 mixed (Figure 5).

[0127] Example 7: Obtaining, extracting and measuring human stool samples Human stool collection:Patients were asked to defecate into a bucket attached to a toilet seat and the resulting samples were delivered to a courier for transport to the Digestive Diseases Research Core Center at Washington University School of Medicine (St. Louis, MO). Stool was aliquoted into 50 mL conical tubes and stored at -80°C. From there, samples were shipped under dry ice to BioGenerator Labs (St. Louis, MO), where samples were stored in a -80°C freezer until extraction. Patients used in this collection provided consent through Washington University School of Medicine, and the Internal Review Board of Washington University School of Medicine provided ethical oversight of this collection.

[0128] Human sample type Stool samples were obtained from 6 patients with colorectal cancer (stages I-IV), 4 patients with precancerous adenomas, and 14 patients with negative colonoscopy findings, resulting in a pooled sample of 24 cases. These samples were from the Human Stool Collections at both Kharkiv National Medical University and Washington University School of Medicine. Sample labels were verified and matched in a manner consistent with the criteria outlined above for types of human samples.

[0129] Total Nucleic Acid Extraction : seRNA was extracted from samples in a manner consistent with the methods outlined above for total nucleic acid extraction, including DNAse treatment, and the quality of the seRNA was analyzed in a manner consistent with the methods outlined above for extraction results.

[0130] Example 8: Analysis of RNA transcripts Library preparation:Libraries of seRNAs were generated using the Illumina Targeted RNA Custom Panel, consisting of 398 custom amplicons. Library preparation relied on an initial cDNA synthesis step using ProtoScript II Reverse Transcriptase (Illumina), hybridization of oligo pools with the targeted seRNA, extension of oligos using Illumina reagents (AM1, ELM4, RSB, UB1), and amplification by polymerase chain reaction (PCR). Total mass input ranged from 200 to 400 ng, and the number of PCR cycles used ranged from 26 to 28. After library amplification, the cDNA capture was cleaned using Illumina reagents (RSB, AMPure, XP beads EtOH). Library preparations were analyzed for quantity and quality using an Agilent BioAnalyzer and Qubit Fluorometric Quantitation (Thermo Fisher). All samples described in this analysis passed initial quality checks and were candidates for next-generation sequencing.

[0131] Sequencing: A unique index was used for each sample, allowing pooling of library preparations and multiplexing of all samples on the same flow cell of an Illumina NextSeq System. All 24 samples were pooled in one lane on a medium-power flow cell (Illumina). The first 150 base pairs of each end of the reads were sequenced (2 × 150), the sequenced reads were appended, and FASTQ files were output. A quality check of the FASTQ files showed that 19 samples had sufficient total reads and sufficient quality for bioinformatics analysis.

[0132] alignment:After sequencing, sequences were trimmed of custom primer sequences and the trimmed reads were aligned to the latest reference genome (GRCh38). Transcript expression was obtained by calculating the average coverage at gene loci. Transcript expression was normalized by the average coverage at two housekeeping genes (GAPDH and ACTB).

[0133] Example 9: Biological replicates performed on various platforms Four samples were evaluated by both microarray and sequencing. Linear regression of 398 transcripts across multiple platforms showed moderate reproducibility (Pearson's r range = 0.48-0.63). Sequencing showed increased resolution compared to microarray, as evidenced by the signal range of low-emitting transcripts (Figure 6).

[0134] Example 10: Hierarchical clustering analysis using seRNA Unsupervised principal component analysis (PCA) was performed on the RNA sequencing data of all 13 unique samples. Clustering was observed between patients with CRC, patients with adenomas, and patients without neoplastic findings. Samples from patients with cancer showed the greatest deviation and separation from other patient populations, while samples from patients without neoplastic findings showed narrower clustering (Figure 7).

[0135] Example 11: Evaluation of variants in seRNA-based sequencing Variant calling and annotation:Integrative Genomics Viewer was used to identify variants involved in CRC tumorigenesis. The amplicon panel covered approximately 3% of the genomic space of 398 captured genes. Exemplary driver mutations are shown in Figure 9. As shown in Figure 8, we identified several potential driver mutations. Such mutations included missense mutations in APC in patients with high-risk adenomas (13% variant allele frequency (VAF)), missense mutations in SMAD4 in patients with high-risk adenomas (17% VAF), 3' deletions in the regulatory region of MAPK3 in patients with stage I CRC (7% VAF), missense mutations in PIK3CA in patients with no colonoscopy findings (12% VAF), missense mutations in KRAS in patients with high-risk adenomas (3% VAF), and missense mutations in CDH1 in patients with high-risk adenomas (2% VAF) (Figure 8).

[0136] Example 12: Obtaining human stool samples Collection of human stool Stool samples were obtained at the Digestive Disease Research Core Center (DDRCC) at Washington University School of Medicine (St. Louis, MO). All patients were mailed a stool sample collection kit, which was returned to the DDRCC by courier. Clinical data (e.g., demographic information, colonoscopy results, etc.) were collected by the DDRCC. Each sample was tested for fecal blood using a commercially available immunochemical stool test (FIT) (Polymedco, OC-Light S FIT) and then frozen at -80°C. Each patient recruited for this study underwent colonoscopy, and those with positive findings underwent biopsy and subsequent histopathological review to determine neoplasm classification. Adenoma classification was stratified based on histopathology (benign vs. precancerous), number of polyps, polyp size, and degree of differentiation. Cancer classification was stratified based on the American Joint Committee on Cancer (AJCC) 7 TNM system. Patients were labeled as healthy if they had unremarkable colonoscopy findings.

[0137] Human sample type : In total, stool samples from 275 individuals were collected for this study. Sequencing data, FIT, demographic information (i.e., sex, age, ethnicity, smoking status, and family history) and colonoscopy results with histopathology information were obtained for all patients, where applicable. In this study, 11 patients had CRC (stages I-IV), 26 patients had high-risk adenomas (HRAs), 37 patients had medium-risk adenomas (MRAs), 61 patients had low-risk adenomas (LRAs), 50 patients had benign polyps, and 90 patients had no colonoscopy findings. Patient types, demographics, and processing information are summarized in Figure 10. Healthy individuals were those with no colonoscopy findings and no history of colorectal cancer, inflammatory bowel disease, celiac disease, irritable bowel syndrome, diarrhea within the last 20 days, or any other gastrointestinal disease. Benign polyp patients provided a stool sample prior to undergoing colonoscopy, during which the polyps detected by the physician were determined to be benign by subsequent biopsy and histological evaluation. Diseased individuals were patients diagnosed with colorectal cancer or precancerous adenoma. Colorectal cancer patients were diagnosed with stage I-stage IV colorectal cancer by colonoscopy and subsequent biopsy within the last month, and had not yet undergone any post-biopsy treatment, which may include chemotherapy, radiation, and / or surgery. Precancerous adenoma patients (HRA, MRA, and LRA) provided a stool sample prior to undergoing colonoscopy, during which the polyps detected by the physician were determined to be precancerous by subsequent biopsy and histological evaluation. Adenoma risk stratification was based on polyp size, number of polyps, degree of dysplasia, and cellular morphology. The patient population was overrepresented in colorectal cancer patients, but the remaining samples were representative of the asymptomatic screening population. Patients used in this collection were consented by Washington University School of Medicine. The Washington University School of Medicine Internal Review Board provided ethical oversight for this collection (IRB #20111107).

[0138] Separation into training and test sets 154 prospectively collected stool samples were used as the training set, and 110 prospectively collected stool samples were used as the hold-out test set. Eleven retrospectively collected stool samples from CRC patients were also included in the hold-out test set. The training and hold-out test sets were evaluated for categorical, demographic and handling differences using t-tests (population means) or z-tests (population frequencies), with p-values ​​less than 0.05 indicating significance. Two statistically significant differences were found between the characteristics of the training set and the hold-out test set. First, the training set did not include retrospectively collected samples (i.e., samples from patients with CRC). Second, the hold-out test set had different processing quality compared to the training set. Specifically, there was a reduction in the mean stool input used for stool-derived eukaryotic RNA extraction (12.9 grams vs. 12.0 grams; p-value=0.03), a reduction in the mean stool-derived eukaryotic RNA concentration (168.6 ng / uL vs. 56.1 ng / uL; p-value<0.01), and a reduction in the mean library prep fragment size (200.6 base pairs vs. 192.2 base pairs; p-value<0.01) (Figure 10).

[0139] Example 13: Development of a custom capture panel Panel Transcript: A custom capture panel of 639 amplicons was developed in Illumina DesignStudio for library preparation. Custom capture probes were associated with 408 transcripts selected using previous studies and the literature.

[0140] Microarray transcripts: Transcripts were selected based on microarray experiments, in which total seRNA was extracted from stool samples and expression was assessed using Affymetrix Human Transcriptome Array 2.0 (Thermo Fisher Scientific, Waltham, MA). Microarray expression profiles obtained in 177 patients with CRC or precancerous adenomas (diseased cohort) were compared with those of 88 patients without colonoscopy findings (healthy cohort). 214 transcripts were identified as differentially expressed (p<0.03) and selected for the capture panel.

[0141] NanoString Transcripts : Transcripts were selected based on NanoString experiments, in which total seRNA was extracted from stool samples and expression was assessed using the nCounter® PanCancer Pathways Panel (NanoString, Seattle, WA) and the nCounter® PanCancer Progression Panel (NanoString, Seattle, WA). NanoString expression profiles from 59 patients with CRC or precancerous adenomas (disease cohort) were compared to those from 26 patients with no colonoscopy findings (healthy cohort). 123 transcripts were identified as differentially expressed and selected for the capture panel.

[0142] Other transcripts The literature was evaluated for additional transcripts implicated in CRC, including searches of GeneCards, ClinVar, the Catalogue of Somatic Mutations in Cancer (COSMIC), Clinical Interpretations of Variants in Cancer (CIViC), the Colorectal Cancer Subtyping Consortium classifier, and other relevant studies. Using this literature, 71 transcripts were selected for a custom capture panel.

[0143] Example 14: Extraction of human nucleic acids Total Nucleic Acid Extraction:Each stool sample was placed in a 50 mL conical tube. Approximately 6,000-25,000 mg of stool was added to each tube. An additional 20-40 mL of solution was added to each tube. The solution contained a mixture of 10 mM Trizma Base (Sigma-Aldrich, St. Louis, MO), 1 mM EDTA (Sigma Aldrich) with 0.05% Tween-20 (Sigma-Aldrich), and 0.0002% RNase Inhibitor (Sigma-Aldrich), pH 7.5. The solution was centrifuged at 1000 rpm for 10 minutes at 4°C, and the supernatant was discarded. Approximately 4-10 mL of EasyMag® ​​Lysis Buffer (bioMerieux, Durham, NC) was added to the pellet, and the pellet was resuspended in the solution. The solution was centrifuged at 2500-3500 rpm at 20-25°C for 10-15 minutes. During differential centrifugation, the solution separated into three layers. The bottom layer contained solid intracellular debris, the middle layer was a hydrophilic layer enriched in human nucleic acids, and the top layer was a hydrophobic lipid layer. The top two layers were transferred to a new 15 mL conical tube and the solution was centrifuged again at 2500 rpm at 20-25°C for 15 minutes. This centrifugation step resulted in separation into three layers: the bottom layer was solid intracellular debris, the middle layer was a hydrophilic layer enriched in human nucleic acids, and the top layer was a hydrophobic lipid layer. To screen for large debris from the solution, a 10 uL pipette tip was placed onto a 1 mL pipette tip and 2 mL of the hydrophilic layer was pipetted from the 15 mL tube and transferred to an EasyMag® ​​Disposable cartridge (bioMerieux). Additionally, 50 uL of EasyMag® ​​Magnetic Silica (bioMerieux) was added to the cartridge. The beads were mixed in the solution for 0.5-1 min using a pipette. The nucleic acid bound to the beads was eluted in a buffer solution using Specific A Protocol according to the manufacturer's instructions. The volume of the eluted nucleic acid was 70 uL. The nucleic acid solution was pipetted into a 1.5 mL tube and placed on ice.The same EasyMag® ​​Disposable Cartridge (bioMerieux) used in the previous step was then reloaded with an additional 2 mL of hydrophilic layer of the same solution in the 15 mL tube used previously using the same technique to screen out large debris. An additional 20 uL of EasyMag® ​​Magnetic Silica (bioMerieux) was added to the cartridge. The beads were mixed in the solution for 0.5-1 min using a pipette. The nucleic acid bound to the beads was eluted in a buffer solution as described above using Specific A Protocol according to the manufacturer's instructions. The volume of eluted nucleic acid was 70 uL. This nucleic acid solution was pipetted into the original 1.5 mL tube already containing the initial 70 uL of elution solution and the combined solution was placed on ice. An additional 2 mL of hydrophilic layer from the same 15 mL solution used above was added to a new EasyMag® ​​Disposable Cartridge (bioMerieux) using the same technique and large debris was screened out. An additional 20 uL of EasyMag® ​​Magnetic Silica (bioMerieux) was added to the cartridge. The beads were mixed in the solution for 0.5-1 min using a pipette. The nucleic acid bound to the beads was eluted in a buffer solution using Specific A Protocol according to the manufacturer's instructions. The volume of eluted nucleic acid was 70 uL. This nucleic acid solution was pipetted into the 1.5 mL tube containing the first two elutions and the combined solution was placed on ice. The same EasyMag® ​​Disposable Cartridge (bioMerieux) used in the previous step was then reloaded with an additional 2 mL of hydrophilic layer from the same solution in the 15 mL tube used above using the same technique and large debris was screened out. An additional 20 uL of EasyMag® ​​Magnetic Silica (bioMerieux) was added to the cartridge, and the beads were mixed in the solution for 0.5-1 min using a pipette.The nucleic acids bound to the beads were eluted in a buffer solution using Specific A Protocol according to the manufacturer's instructions as described above. The volume of eluted nucleic acid was 70 uL. This nucleic acid solution was pipetted into the original 1.5 mL tube that already contained the first three 70 uL elutions, and the combined solution was placed on ice.

[0144] DNase treatment: The 280 uL solution was treated with Baseline-Zero-DNase (Epicentre) at 35-40°C for 20-40 min. A 1-2 mL aliquot of EasyMag® ​​Lysis Buffer was added to the DNase-treated solution and the sample was transferred to a new EasyMag® ​​Disposable cartridge. The entire solution was added to the new cartridge along with 85 uL of EasyMag® ​​Magnetic Silica. The nucleic acid bound to the beads was eluted in the buffer solution using the EasyMag® ​​Generic Protocol according to the manufacturer's instructions. The volume of the eluted nucleic acid was 25 uL. The nucleic acid solution was pipetted into a 1.5 mL tube and stored at -80°C.

[0145] Example 15: Measurement of human nucleic acid levels in human stool samples Extraction result:1-2 uL of each sample extracted above was assessed for total nucleic acid and RNA integrity using an Agilent 2100 Bioanalyzer. The samples were analyzed qualitatively and quantitatively. Electrophoretic analysis was used to confirm the quality of the extracted RNA. Electrophoretic files were read by comparing the bands of each sample with those indicated by size markers of the RNA ladder and identifying 18S and 28S ribosomal RNA (rRNA) bands. The rRNA bands are two thick prominent bands near the 2,000 nucleotide marker in the standardized ladder. Qualitatively, sufficient banding and strong band intensity indicated that sufficient intact nucleic acid was available for further analysis, such as microarray sequencing, polymerase chain reaction (PCR), nucleic acid sequencing, molecular barcoding, amplicon sequencing or probe-capture. The electrophoretic separation diagram is a graphical representation of each electrophoretic file with quantification of the RNA integrity number (RIN), total RNA mass and total rRNA mass. Quantitatively, the higher the RIN, the higher the total RNA mass, and the higher the total rRNA mass, the more likely the sample will be useful for further analysis, such as microarray sequencing, polymerase chain reaction (PCR), nucleic acid sequencing, molecular barcoding, amplicon sequencing, or probe-capture. Samples were also assessed for RNA concentration using a Qubit 4.0 Fluorometer. RNA concentration is determined by quantifying the fluorescence generated by components in the Qubit assay that selectively bind RNA present in the eluate. Quantitatively, the higher the RNA concentration, the more likely the sample will be useful for further analysis, such as microarray sequencing, polymerase chain reaction (PCR), nucleic acid sequencing, molecular barcoding, amplicon sequencing, or probe-capture.

[0146] Example 16: Analysis of RNA transcripts Library preparationLibraries of seRNAs were generated using the Illumina Targeted RNA Custom Panel, which consisted of 639 custom amplicons. Library preparation relied on an initial cDNA synthesis step using ProtoScript II Reverse Transcriptase (Illumina, San Diego, CA), hybridization of oligo pools with the targeted seRNA, extension of oligos using Illumina reagents (AM1, ELM4, RSB, UB1), and amplification by polymerase chain reaction (PCR). Total mass input ranged from 200 to 400 ng, and the number of PCR cycles used ranged from 28 to 30. After library amplification, the cDNA capture was cleaned using Illumina reagents (RSB, AMPure, XP beads EtOH). Library preparations were analyzed for quantity and quality using an Agilent 2100 BioAnalyzer and Qubit 4.0 Fluorometer (Thermo Fisher). All samples described in this analysis passed initial quality checks and were candidates for downstream analysis.

[0147] Sequencing Analysis A unique index was used for each sample, allowing pooling of library preparations and multiplexing of samples on the flow cell of an Illumina NextSeq 550 System. PhiX spike-in was used for quality control. 275 samples were pooled in eight individual high-output flow cell runs (Illumina). Up to 150 base pairs on each end of the reads were sequenced (2x150), the sequenced reads were appended, and FASTQ files were output. A quality check of the FASTQ files showed that all 275 samples had sufficient total reads (>100,000) and sufficient quality for bioinformatics analysis.

[0148] alignmentAfter sequencing, reads were trimmed of custom primer sequences and aligned to the latest reference genome (GRCh38) by HISAT2.3.0. Transcript expression was obtained by calculating the average coverage at the locus. For each transcript, raw amplicon expression was normalized to an internal housekeeping gene, GAPDH, such that reported expression was equal to the number of amplicon reads per million mapped GAPDH reads.

[0149] Transcript Selection Normalized expression of 639 amplicons was evaluated for all samples in the training set (n=154 samples). Of these 639 amplicons, 48 ​​amplicons were not expressed in any sample, and an additional 71 amplicons were not expressed in >95% of all samples; these amplicons were excluded from the analysis. For the remaining amplicons, a bootstrap analysis was performed by dividing the training set into 100 different 9:1 splits, whereby each split was evaluated for informative amplicons. Amplicons were expressed as absolute log 2Fold changes were considered informative if they were greater than 1 in both contrast groups (HRA vs. LRA, benign polyps, absent on colonoscopy; MRA vs. LRA, benign polyps, absent on colonoscopy) and if the ANOVA between contrast groups had a p-value <0.05. The transcript selection process is further illustrated in Figure 11A. In total, 40 amplicons from 29 genes were identified as informative in at least one of the 100 splits (Figure 11B). If an amplicon was determined to be informative in at least 33% of all bootstrap splits, it was considered differentially expressed and was made a candidate feature for model development. There were 10 amplicons identified as differentially expressed (informative in at least 33 of the 100 splits) (Figure 11B). Raw GAPDH values ​​are considered a measure of total eukaryotic RNA in each sample. It was observed that raw GAPDH values ​​were higher in patients with MRA, HRA and CRC compared to healthy patients (Figure 12). Demographic features (age, smoking status, family history, ethnicity and gender) were also considered for model development. Finally, 10 differentially expressed transcripts, raw GAPDH values ​​and two demographic identifiers (age and smoking status) were selected as candidate features for model development.

[0150] Example 17: Development of a Random Forest Model A random forest model was constructed using a training set of 154 patients and all 13 candidate features. 5,000 decision trees were constructed from bootstrap training samples; each node split was optimized by Gini importance; each tree was constructed until maximum depth was reached. Although specific embodiments are discussed herein, it will be appreciated that any suitable model, such as a random forest model, can be constructed using a greater and / or lesser number of decision trees, a greater and / or lesser number of candidate features, etc. Furthermore, other types of models, such as deep learning models or support vector models, could also be used with a variety of parameters. The random forest model used candidate features such as differentially expressed transcripts, raw GAPDH values, age, and smoking status. Although specific embodiments are discussed herein, it will be appreciated that any suitable model, such as a random forest model, can be constructed using all informative features and / or a selected subset of informative features.

[0151] The model output was configured to give a prediction between 0 and 1, with higher numbers reflecting higher confidence in a neoplastic or positive finding. In some embodiments, an immunochemical fecal test (FIT) was used to modify the confidence in a neoplastic or positive finding. For example, a FIT-positive sample would have a higher prediction score of 1. A three-fold internal cross-validation was used to evaluate the performance of the training model. Three-fold internal cross-validation used three different 2:1 splits, whereby a model was built using the larger split and used in the smaller split. The model predictions were used to generate a receiver operating characteristic (ROC) curve, and the area under the curve (AUC) was used to measure the model performance. The median of the ROC curves of the three splits was used to approximate the cross-validation performance. The ROC curves were plotted with and without the incorporation of the FIT feature. For the ROC curve plotted with the FIT feature, the FIT forced positive model prediction value is equal to 1. In this example shown, internal cross-validation without the FIT feature gave a ROC AUC of 0.65 for HRA vs. all other categories (MRA, LRA, benign polyps, and colonoscopy null). In this example shown, internal cross-validation with the FIT feature gave a ROC AUC of 0.70 for HRA vs. all other categories (MRA, LRA, benign polyps, and colonoscopy null) (Figure 13).

[0152] Holdout test set: The final random forest model was constructed using all 154 samples in the training set. For the model created, the most influential features as measured by Gini importance were ACY1 and TNFRSF10B (Gini importance ≧0.13), and the least important feature was PER3 (Gini importance <0.05). Raw GAPDH values ​​were the fourth most important feature in constructing the random forest model (Figure 14). This model was used on the hold-out test set of 110 prospectively collected stool samples. ROC curves were plotted with and without the FIT feature, and the area under the curve (AUC) was used to measure the performance of the model. The model yielded a ROC AUC of 0.67 without the FIT feature and 0.78 with the FIT feature (Figure 15).

[0153] Model Predictions : The model predictions in the holdout test set were correlated with disease severity (Figure 16). The correlation of model output with disease severity is a direct reflection of biology and was not specifically trained as part of the model. In the above embodiment, the feature selection and model input included the use of three categories (HRA, MRA and all others), but disease subtypes (e.g., subsets of HRA) and disease extent (e.g., HRA is more severe than MRA) were not used as features for model training. Given that the model output is correlated with disease severity, this allows for the prospective identification of specific subtypes and disease severity using the model output. Furthermore, stratification of positive and negative findings was improved by modifying the model parameters to provide disease severity information to the model.

[0154] Downsampling Analysis:To understand the extent of model training, downsamplings of 154 samples from the training set were selected and performance was evaluated using the hold-out test set. Downsamplings ranged from 30% to 100% in 10% increments. For each downsampling, feature selection was performed using bootstrap, a random forest model was trained with the candidate features, and the model was used on the hold-out test set. Model performance was evaluated using the ROC AUC of the hold-out test set. This process was repeated 10 times for each downsampling to reduce selection bias in subsampling, and model performance was evaluated with and without the incorporation of FIT features. Downsampling analysis showed a direct relationship between the total number of samples used for training and performance on the hold-out test set. When FIT features were excluded, the median ROC AUC for HRA versus all other categories increased from 0.55 (30% of training data) to 0.67 (100% of training data) (Figure 17A). When the FIT feature was included, the median ROC AUC for HRA versus all other categories increased from 0.72 (30% of training data) to 0.78 (100% of training data) (Figure 17B).

[0155] Final AccuracyContinuing with the above, the random forest model was also used on 11 retrospectively collected stool samples from CRC patients. The model output showed a prediction between 0 and 1 and a FIT forced positive model predictive value equal to 1. Samples with a positive immunochemical stool test (FIT+) or a positive model predictive value (Model+) were considered positive and all other samples were considered negative. A ROC curve was plotted considering only CRC samples as positive and other categories (HRA, MRA, LRA, benign polyps and no findings on colonoscopy) as negative. Using all 121 samples from this imputed holdout test set, the model yielded an ROC AUC of 0.94. Separate ROC curves were plotted considering CRC and HRA samples as positive and all other categories (MRA, LRA, benign polyps and no findings on colonoscopy) as negative. Using all 121 samples from the imputed holdout test set, the model had a ROC AUC of 0.87 (Figure 18). The point on the ROC curve where accuracy was at its maximum was selected and sensitivity and specificity were calculated. At this point, the model showed a specificity of 89% (n=99 samples) with a sensitivity of 91% for CRC (n=11 samples) and a sensitivity of 73% for HRA (n=11 samples) (Figure 18).

[0156] Extrapolation to screening populations : To better approximate the ultimate model performance, the accuracy profile observed in the imputed holdout test set was extrapolated to the predicted relative frequencies in the prospectively screened population. The ROC curves above were plotted to show the model performance. When weighting cancer and HRA against the predicted morbidity in the prospectively screened population, the model yielded an ROC AUC of 0.80 for CRC and HRA samples versus all other categories (Figure 19). Extrapolation of the results to the prospectively screened population also allows for the calculation of the combined sensitivity of neoplastic findings, negative predictive value (NPV) and positive predictive value (PPV). The extrapolated accuracy profile showed a combined sensitivity of 74% for CRC and HRA, a positive predictive value of 37% and a negative predictive value of 98% (Figure 19).

Claims

1. a) measuring the expression level of each of a plurality of stool-derived eukaryotic RNA biomarkers in eukaryotic nucleic acid extracted from a stool sample from a subject; and b) comparing a first score calculated from an aggregation of the measured expression levels from step a) with a second score calculated from an aggregation of the measured expression levels of the plurality of stool-derived eukaryotic RNA biomarkers in controls from a population of healthy individuals, wherein an increase in the first score compared to the second score indicates that the subject has colorectal neoplasia.

1. A method for detecting colorectal neoplasia in a subject, comprising: the plurality of stool-derived eukaryotic RNA biomarkers consists of i) ACY1, TNFRSF10B, EGLN2, SMAD4, KRAS, AREG, CDH1, and GAPDH; or ii) ACY1, TNFRSF10B, EGLN2, SMAD4, KRAS, AREG, and CDH1; The method.

2. The method of claim 1, wherein the multiple stool-derived eukaryotic RNA biomarkers consist of ACY1, TNFRSF10B, EGLN2, SMAD4, KRAS, AREG, and CDH1.

3. The method of claim 1, wherein the multiple stool-derived eukaryotic RNA biomarkers consist of ACY1, TNFRSF10B, EGLN2, SMAD4, KRAS, AREG, CDH1, and GAPDH.

4. 2. The method of claim 1, wherein the colorectal neoplasia is selected from the group consisting of colorectal cancer, high-risk adenoma, medium-risk adenoma and low-risk adenoma.

5. The method of claim 1 , wherein the nucleic acid comprises mRNA.

6. 2. The method of claim 1, wherein the expression level is measured by nucleic acid sequencing, microarray sequencing, molecular barcoding, amplicon sequencing, probe capture, polymerase chain reaction (PCR), ddPCR, dPCR, RT-PCR or RT-qPCR.

7. 10. The method of claim 1, further comprising determining demographic information of the subject, wherein the demographic information comprises smoking status.

8. The method of claim 1, further comprising administering a stool immunochemical test (FIT) to the subject.

9. 1. A method for assisting in selecting a clinical regimen for a subject having or at risk for colorectal neoplasia, comprising: a) determining the expression level of each of a plurality of stool-derived eukaryotic RNA biomarkers in eukaryotic nucleic acid extracted from a stool sample from the subject; and b) comparing a first score calculated from an aggregation of the measured expression levels from step a) with a second score calculated from an aggregation of the measured expression levels of the plurality of stool-derived eukaryotic RNA biomarkers in controls from a population of healthy individuals, wherein an increase in the first score compared to the second score indicates that the subject has colorectal neoplasia. Including, The plurality of stool-derived eukaryotic RNA biomarkers consists of i) ACY1, TNFRSF10B, EGLN2, SMAD4, KRAS, AREG, CDH1, and GAPDH; or ii) ACY1, TNFRSF10B, EGLN2, SMAD4, KRAS, AREG, and CDH1; The method.

10. The method of claim 9, wherein the multiple stool-derived eukaryotic RNA biomarkers consist of ACY1, TNFRSF10B, EGLN2, SMAD4, KRAS, AREG, and CDH1.

11. The method of claim 9, wherein the multiple stool-derived eukaryotic RNA biomarkers consist of ACY1, TNFRSF10B, EGLN2, SMAD4, KRAS, AREG, CDH1, and GAPDH.

12. 10. The method of claim 9, wherein the colorectal neoplasia is selected from the group consisting of colorectal cancer, high-risk adenoma, intermediate-risk adenoma and low-risk adenoma.

13. The method of claim 9, wherein the nucleic acid comprises mRNA.

14. 10. The method of claim 9, wherein the expression level is measured by nucleic acid sequencing, microarray sequencing, molecular barcoding, amplicon sequencing, probe capture, polymerase chain reaction (PCR), ddPCR, dPCR, RT-PCR or RT-qPCR.

15. 10. The method of claim 9, further comprising determining demographic information of the subject, wherein the demographic information comprises smoking status.

16. 10. The method of claim 9, further comprising administering a fecal immunochemical test (FIT) to the subject.

17. 10. The method of claim 9, wherein the clinical plan comprises treatment for a subject.

18. 20. The method of claim 17, wherein the treatment comprises surgery, chemotherapy, radiation therapy, targeted therapy, immunotherapy, or any combination thereof.

Images