Materials And Methods For Determining Cancer Risk

  *US20120094289A1*
  US20120094289A1                                 
(19)United States 
(12)Patent Application Publication(10)Pub. No.: US 2012/0094289 A1
 Garrity-Park et al.(43)Pub. Date:Apr.  19, 2012

(54)MATERIALS AND METHODS FOR DETERMINING CANCER RISK 
    
(76)Inventors: Megan Garrity-Park,  Pine Island, MN (US); 
  Thomas C. Smyrk,  Rochester, MN (US); 
  Edward V. Loftus, JR.,  Rochester, MN (US); 
  William J. Sandborn,  LaJolla, CA (US) 
(21)Appl. No.: 13/272,044 
(22)Filed: Oct.  12, 2011 
 Related U.S. Application Data 
(60)Provisional application No. 61/392,342, filed on Oct.  12, 2010.
 
 Publication Classification 
(51)Int. Cl. C12Q 001/68 (20060101); G01N 033/574 (20060101)
(52)U.S. Cl. 435/6.11; 435/7.1

        

(57)

Abstract

This document relates to materials and methods involved in assessing inflammatory bowel disease patients at risk for developing cancer. For example, materials and methods for monitoring colorectal cancer risk in ulcerative colitis patients are provided.
 Claim(s),  Drawing Sheet(s), and Figure(s)
 
 


CROSS-REFERENCE TO RELATED APPLICATIONS

[0001] This application claims the benefit of U.S. Provisional Application Ser. No. 61/392,342, filed Oct. 12, 2010. The disclosure of the prior application is considered part of (and is incorporated by reference in) the disclosure of this application.

BACKGROUND

[0002] 1. Technical Field
[0003] This document relates to materials and methods involved in assessing inflammatory bowel disease patients at risk for developing cancer. For example, this document relates to materials and methods for monitoring colorectal cancer risk in ulcerative colitis patients.
[0004] 2. Background Information
[0005] Inflammatory bowel disease (IBD) refers to chronic diseases that cause inflammation in the intestine. The major types of IBD are Crohn's disease and ulcerative colitis (UC). Crohn's disease and UC differ in the location and nature of the inflammation. Crohn's disease can affect any part of the gastrointestinal tract, though it most commonly affects the terminal ileum and parts of the large intestine. Ulcerative colitis is an idiopathic inflammatory bowel disease characterized by chronic, relapsing mucosal inflammation primarily limited to the colon and rectum. Patients with longstanding and extensive IBD are at increased risk to develop colorectal cancer (CRC). Because of this, patients with IBD are advised to undergo surveillance colonoscopy and biopsy, every one to two years, wherein biopsy samples are histologically evaluated for the presence of pre-cancerous changes (colorectal dysplasia) or CRC.

SUMMARY

[0006] This document provides methods and materials for assessing inflammatory bowel disease patients at risk for developing cancer. For example, this document provides materials and methods that can be used to monitor colorectal cancer risk in ulcerative colitis patients. As described herein, markers (e.g., nucleic acid markers, polypeptide markers, epigenetic markers, or combinations thereof) can be used to screen UC patients to determine risk for developing CRC. Detection of such markers may allow a physician to more closely monitor those patients deemed to be at a higher risk of developing CRC.
[0007] Patients with UC have an increased risk of developing CRC as compared with the general population (Ekbom et al., N. Engl. J. Med., 323:1228-1233 (1990)). The exact mechanism by which the extent and duration of UC contribute to the pathogenesis of CRC is unclear, but studies measuring colonic inflammation and CRC risk have found a correlation between increased severity of histologic inflammation and risk for CRC (Rutter et al., Gastroenterology, 126(2):451-459 (2004) and Gupta et al., Gastroenterology, 133(4):1099-1105 (2007)). Rutter et al. assessed disease activity using a four-point grading scale ranging from 0 (inactive) to 3 (severely active) to quantify levels of neutrophil infiltration on hematoxylin and eosin-stained tissue sections (H&E).
[0008] This document is based, in part, on the discovery that hemotoxylin and eosin-stained tissue section (H&E) examinations alone may underestimate the level of disease activity present in the colonic tissue of patients with UC, and that patients with UC-CRC may have higher levels of disease activity at the tissue level even when they have what would currently be defined as inactive disease. For example, nucleic acid markers, epigenetic markers, polypeptide markers, or combinations of markers can be used to identify patients with higher levels of immune cell infiltrate associated with UC-CRC and can be detected even during what is currently defined as inactive disease (e.g., no neutrophil infiltration seen on H&E stained tissue slides). Measuring polypeptide levels of MPO (myeloperoxidase), and/or the methylation status of MINT1, COX-2, and/or RUNX3 nucleic acids in patient samples can provide useful information about the risk of developing colorectal cancer in inflammatory bowel disease patients. In some cases, genetic associations in TNF-alpha nucleic acids or in other biomolecules regulated by NFκB can provide additional useful information about cancer risk.
[0009] In general, one aspect of this document features a method for assessing a mammal diagnosed with inflammatory bowel disease for the presence of or an increased risk of developing colorectal cancer. The method comprises, or consists essentially of, determining whether or not the mammal comprises at least two markers from the group consisting of elevated MPO polypeptide levels, elevated RUNX3 methylation status, elevated MINT1 methylation status, and reduced COX-2 methylation status as compared to a normal control, wherein the presence of the at least two markers is indicative of an increased risk of developing colorectal cancer. The method can further comprise determining whether or not the mammal comprises at least one polymorphism in a TNF alpha nucleic acid, wherein the presence of the polymorphism is indicative of an increased risk of developing colorectal cancer. The inflammatory bowel disease can be ulcerative colitis. The determining step can comprise performing an immunoassay. The determining step can comprise performing methylation-specific PCR. The mammal can be a human.
[0010] In another aspect, this document features a method for assessing a mammal with histologically inactive inflammatory bowel disease for the presence of colorectal cancer or an increased risk of developing colorectal cancer. The method comprises, or consists essentially of, determining whether or not the mammal comprises the presence of at least two markers selected from a group consisting of the presence of at least one polymorphism in a TNF alpha nucleic acid, an elevated MPO polypeptide level, an elevated methylation level of a RUNX3 nucleic acid, an elevated methylation level of a MINT1 nucleic acid, and a reduced methylation level of a COX-2 nucleic acid as compared to a normal control, wherein the presence of the at least two markers is indicative of the presence of colorectal cancer or an increased risk of developing the colorectal cancer. The mammal can be assessed as having the increased risk of developing colorectal cancer and can be categorized as a mammal needing more frequent monitoring than a mammal assessed as not having the increased risk of developing colorectal cancer.
[0011] In another aspect, this document features a method for assessing a mammal diagnosed with inflammatory bowel disease for the presence of or an increased risk of developing colorectal cancer. The method comprises, or consists essentially of, (a) determining the methylation status in a RUNX3 nucleic acid and a COX-2 nucleic acid in the human, (b) classifying the human as having or as having an increased risk of developing the colorectal cancer if the RUNX3 nucleic acid methylation status is elevated and the COX-2 nucleic acid methylation status is reduced as compared to a normal control, and (c) classifying the human as not having or as not having at an increased risk of developing the colorectal cancer if the RUNX3 nucleic acid methylation status is not elevated and the COX-2 nucleic acid methylation status is not reduced. The inflammatory bowel disease can be histologically inactive. The method can further comprise determining the level of an MPO polypeptide in the mammal, wherein an elevated level of the MPO polypeptide is indicative of the presence of or an increased risk of developing colorectal cancer. The method can further comprise determining the methylation status of a MINT1 nucleic acid in the mammal, wherein an elevated level of the MINT1 nucleic acid methylation status is indicative of the presence of or an increased risk of developing the colorectal cancer. The method can further comprise determining whether or not the mammal contains a polymorphism in a nucleic acid encoding a TNF-alpha protein, wherein the presence of the polymorphism is associated with the presence of or an increased risk of the colorectal cancer. The polymorphism can be rs1800629.
[0012] In another aspect, this document features a method for assessing a biopsy sample from an ulcerative colitis patient having the presence of a polymorphism in a TNF-alpha nucleic acid. The method comprises, or consists essentially of, analyzing the biopsy sample for at least two markers selected from the group consisting of an MPO polypeptide level, methylation status of a RUNX3 nucleic acid, methylation status of a MINT1 nucleic acid, and methylation status of a COX-2 nucleic acid.
[0013] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains. Although methods and materials similar or equivalent to those described herein can be used in the practice or testing of the present invention, suitable methods and materials are described below. All publications, patent applications, patents, and other references mentioned herein are incorporated by reference in their entirety. In case of conflict, the present specification, including definitions, will control. In addition, the materials, methods, and examples are illustrative only and not intended to be limiting.
[0014] Other features and advantages of the invention will be apparent from the following detailed description, and from the claims.

DESCRIPTION OF DRAWINGS

[0015] FIG. 1 contains a sequence listing of human TNF-alpha nucleic acid promoter region (GenBank Accession No. AB048818; GI No. 13365764; SEQ ID NO: 1).
[0016] FIG. 2 contains a sequence listing of a human clone MINT1 colon cancer differentially methylated CpG island genomic sequence (GenBank Accession No. AF135501; GI No: 4914684; SEQ ID NO: 51).
[0017] FIG. 3 contains a sequence listing of human cyclooxygenase nucleic acid, promoter region and exon 1 (GenBank Accession No. AF044206; GI: 3282785; SEQ ID NO: 52).
[0018] FIG. 4 contains the 3′ end of human runt-related transcription factor 3 coding and non-coding regions and a CpG island, complete sequence. (GenBank Accession No. AL023096; GI: 3900882; SEQ ID NO: 53).
[0019] FIG. 5 contains a sequence listing of a human TNF-alpha nucleic acid promoter region (SEQ ID NO: 4). The underlined regions represent the area of the −308 and −238 SNP, respectively, and the parenthetic bases indicate the polymorphism sites.
[0020] FIG. 6 is a graph of histologic disease activity and cell surface marker levels.
[0021] FIG. 7 contains data showing the association of MPO expression levels with a TNF-alpha polymorphism (FIG. 7A) and RUNX3 methylation status (FIG. 7B).
[0022] FIG. 8 contains Receiver operating characteristic (ROC) curves for combined markers. The area under the curve (AUC) for any combination of markers was higher than for TNF-α (72.0%), MPO (72.3%) or RUNX3 (66.9%) alone. The AUC for RUNX3 & MPO was 84.1, for TNF-α & MPO was 82.2% and for TNF-α, MPO & RUNX3 was 88.1%.

DETAILED DESCRIPTION

[0023] This document provides materials and methods related to assessing inflammatory bowel disease patients at risk for developing cancer. For example, this document relates to materials and methods for monitoring colorectal cancer risk in ulcerative colitis patients.
[0024] In general, this document provides methods for determining the risk of inflammatory bowel disease patients of developing a cancer by determining the methylation status, genetic polymorphism status, or level of one or more biomolecules in a test sample from a mammal. The methylation status, genetic polymorphism status, or level of one or more biomolecules can be correlated with the presence of or the risk of developing cancer. Identifying cancers at an early stage can help a physician properly diagnose and treat a cancer patient. Typically, a properly diagnosed and treated cancer patient can experience an improvement in general health and survival.
[0025] As described herein, methods and materials to stratify risk of inflammatory bowel disease patients developing CRC have been identified that may identify patients at risk of developing CRC even in patients deemed to have histologically inactive disease (0 neutrophils on H&E). Patients found to have an increased risk of developing colorectal cancer may benefit from more intensive surveillance and/or different treatment strategies.
[0026] The term “biomolecule” as used herein refers to DNA, RNA, or polypeptides. This document provides methods for measuring biomolecules related to, without limitation, markers of immune cell infiltration into gastrointestinal tissue such as markers of neutrophil granulocytes (e.g., myeloperoxidase; MPO), T-cells (e.g., CD3), natural killer cells (e.g., CD16, CD56, CD8), B-cells (e.g., CD19, CD20), and macrophages (e.g., CD68). This document also provides methods for measuring biomolecules related to inflammatory markers (e.g., tumor necrosis factor alpha; TNF-alpha, cyclooxygenase 2; COX-2), runt-related transcription factors (e.g., RUNX3), and other factors such as Methylated-in-tumor 1 (MINT1). In some cases, this document provides methods for measuring biomolecules that are regulated by nuclear factor kappa beta (NFκB).
[0027] The term “marker level” as used herein refers to a test level of a biomolecule that is either altered or normal compared to a control level. The level of a particular biomolecule can be measured in a test sample from a mammal. The resulting test level then can be compared to a control level of the corresponding biomolecule. If a test level is altered compared to a control level, then the potential for the presence of or the risk of developing cancer in the mammal corresponding to that test sample can be classified as increased. For example, if the level of an MPO polypeptide measured in a colorectal biopsy sample from a patient is elevated compared to a control level of MPO polypeptide, then that patient can be classified as having an increased risk of developing colorectal cancer. In another example, if the methylation status of a MINT1 or RUNX3 nucleic acid measured in a colorectal tissue biopsy is elevated compared to a control level of MINT1 or RUNX3 methylation, then that patient can be classified as having an increased risk of developing colorectal cancer. In yet another example, if the methylation status of a COX-2 nucleic acid measured in a colorectal tissue biopsy is reduced compared to a control level of COX-2 methylation, then that patient can be classified as having an increased risk of developing colorectal cancer.
[0028] In some cases, if a test level is normal compared to a control level, then the risk of developing cancer in a patient corresponding to that test sample can be classified as decreased. For example, if the level of an MPO polypeptide measured in a colorectal biopsy sample is normal compared to a control level of MPO polypeptide, then the patient corresponding to that tissue biopsy sample can be classified as having a decreased risk of developing colorectal cancer.
[0029] In another embodiment, detecting the presence, absence, levels, or status of multiple biomarkers can be used to determine the risk for developing a cancer. In some cases, the presence of one or more polymorphisms in the promoter region of a TNF-alpha nucleic acid can be determined in combination with determining the methylation levels of one or more MINT1, COX-2, and RUNX3 nucleic acids and/or polypeptide levels of biomarkers associated with immune cell infiltration (e.g., MPO). In some cases, a combination of biomarkers that do not include TNF-alpha polymorphism detection can be used as described herein. For example, the presence of a polymorphism (e.g., −308G>A, −301G>A, −293C>T) in a TNF-alpha nucleic acid, an elevated level of an MPO polypeptide, and an elevated level of methylation in a RUNX3 nucleic acid in a sample or samples from a mammal can indicate that that mammal has an increased risk of developing colorectal cancer. In some cases, determining the presence or absence of a polymorphism in a nucleic acid of a biomolecule regulated by NFκB (e.g., IL1B) in combination with determining the methylation status of one or more MINT1, COX-2, and RUNX3 nucleic acids can be used to determine the risk for developing cancer. Other non-limiting examples of suitable combinations of markers include determining an MPO polypeptide level in a patient sample in combination with determining the methylation status of one or more RUNX3, MINT1, or COX-2 nucleic acids.
[0030] In some cases, the presence, absence, level, or status of one or more biomarkers can be determined prior to testing for the presence, absence, level, or status of additional biomarkers. For example, the presence of one or more polymorphisms in a promoter region of a TNF-alpha nucleic acid (or other biomarkers regulated by NFκB) can be determined in an initial screening assay from a patient with UC. Genomic screening tools such as single nucleotide polymorphism (SNP) analysis are particularly useful in inflammatory disease settings because these markers are not affected by disease activity and thus do not change over time. Patients determined to have a SNP present in a TNF alpha nucleic acid could then undergo additional testing to determine the status or levels of other biomarkers. For example, MPO polypeptide levels could be determined in a biopsy sample. In another example, methylation levels of one or more RUNX3, MINT1, and COX-2 nucleic acids can be determined in patients with a SNP present in a TNF alpha nucleic acid. Other non-limiting examples of suitable screening/reflex tests include determining MPO polypeptide levels in a blood or biopsy tissue sample followed by determining methylation status of one or more RUNX3, MINT1, and COX-2 nucleic acids in biopsy samples from patients with increased MPO polypeptide levels.
[0031] In some cases, it may be useful to determine the presence, absence, level, or status of one or more biomarkers in a patient that has previously been deemed to have histologically inactive disease (e.g., 0 neutrophils on H&E). For example, determining the presence or absence of a polymorphism in a nucleic acid of a biomolecule (e.g., TNF-alpha, IL1B), optionally in combination with determining the methylation status of one or more MINT1, COX-2, and RUNX3 nucleic acids in a biopsy sample from an inactive area of the colon (e.g., 0 neutrophils on H&E), can be used to determine the risk for developing cancer in a patient previously found to have histologically inactive disease. Other non-limiting examples of suitable combinations of markers include determining an MPO polypeptide level in a patient sample in combination with determining the methylation status of one or more RUNX3, MINT1, or COX-2 nucleic acids in a patient previously found to have histologically inactive disease.
[0032] Various types of samples can be used when measuring a biomolecule. Such samples include, without limitation, tissue samples, neoplastic tissue biopsies, non-neoplastic tissue biopsies, blood, plasma, serum, surgical waste, and whole organs. Biopsy specimens can be frozen, embedded, sectioned, and stained to identify regions of cellular infiltration. Samples can also include those that have been manipulated in any way after their procurement, such as by treatment with reagents, solubilization, or enrichment for certain components, such as polynucleotides or polypeptides.
[0033] Various appropriate methods can be used to measure a biomolecule level in a sample. Such methods can vary depending on the type of biomolecule measured. For example, methods for measuring polypeptide levels include, without limitation, ELISA, immunohistochemistry, and immunofluorescence-based techniques. Such methods typically involve using antibodies having specific binding affinity for a particular polypeptide.
[0034] The term “antibody” as used herein refers to intact antibodies as well as antibody fragments that retain some ability to bind an epitope. Such fragments include, without limitation, Fab, F(ab′)2, and Fv antibody fragments. The term “epitope” refers to an antigenic determinant on an antigen to which the paratope of an antibody binds. Epitopic determinants usually consist of chemically active surface groupings of molecules (e.g., amino acid residues, amino acid-nucleic linkages) and usually have specific three dimensional structural characteristics as well as specific charge characteristics.
[0035] The antibodies provided herein can be any monoclonal or polyclonal antibody having specific binding affinity for an MPO polypeptide. “Specific binding affinity” refers to an antibody's ability to interact specifically with a particular polypeptide without significantly cross-reacting with other different polypeptides in the same environment. An antibody having specific binding affinity for MPO can interact with MPO polypeptides specifically in the presence of multiple different polypeptides, for example, multiple different markers expressed by neutrophils. MPO antibodies can have specific binding affinity for full-length or fragments of an MPO polypeptide from any suitable species, including, without limitation, mouse, rat, chimpanzee, and human. For example, MPO antibodies can have specific binding affinity for a full-length human MPO polypeptide or fragments of a human MPO polypeptide including.
[0036] Antibodies used for measuring polypeptide levels can include a detectable label. A detectably labeled antibody can refer to an antibody (or antibody fragment which retains binding specificity for a target polypeptide or epitope), having an attached detectable label. The detectable label is normally attached by-chemical conjugation, but where the label is a polypeptide, it could alternatively be attached by genetic engineering techniques. Methods for production of detectably labeled proteins are well known in the art. Detectable labels may be selected from a variety of such labels known in the art, including, but not limited to, radioisotopes, fluorophores, paramagnetic labels, enzymes (e.g., horseradish peroxidase), or other moieties or compounds which either emit a detectable signal (e.g., radioactivity, fluorescence, color) or emit a detectable signal after exposure of the label to its substrate. Various detectable label/substrate pairs (e.g., horseradish peroxidase/diaminobenzidine, avidin/streptavidin, luciferase/luciferin), methods for labeling antibodies, and methods for using labeled antibodies are well known in the art (see, for example, Harlow and Lane, eds. Antibodies: A Laboratory Manual (1988) Cold Spring Harbor Laboratory Press, Cold Spring Harbor, N.Y.).
[0037] MPO polypeptide levels in a colon tissue biopsy sample can, for example, be measured using a quantitative or semi-quantitative immunohistochemistry technique. For example, a section of a colorectal tissue biopsy sample can be treated with anti-MPO primary antibodies. Negative control sections can be incubated with pre-immune rabbit or mouse serum in lieu of primary antibodies. After antibody binding and subsequent washing, the primary antibodies can be detected with appropriate label-conjugated secondary antibodies (e.g., gold-conjugated or enzyme-conjugated antibodies). The label is then developed and quantitated using an image analysis system such as a computer-aided imaging system.
[0038] The resulting quantitated polypeptide levels can be correlated with the risk of having or developing colorectal cancer. Although samples can be processed individually, samples from different tissues or from a population of different patients can be processed simultaneously. Such processing methods include, without limitation, tissue microarrays as described elsewhere (Kononen et al., Nat. Med., 4:844-847 (1998)).
[0039] Immunofluorescence techniques represent another approach to measuring the level of a polypeptide. For example, MPO and CD68 polypeptides can be localized in the same colon biopsy sample section using polyclonal and monoclonal antibodies against MPO and CD68. The bound antibodies can be detected using different fluorescently conjugated antibodies. The levels of MPO and CD68 fluorescence can be quantitated using an image analysis system, and the resulting quantitated levels correlated with the risk of having or developing cancer.
[0040] Suitable antibodies for ELISA-, immunohistochemistry- and immunofluorescence-based methods can be obtained using standard techniques. In addition, commercially available antibodies to polypeptides associated with immune cell infiltration can be used.
[0041] As used herein, a “methylated nucleic acid marker” is a mammalian nucleic acid sequence that is methylated (e.g., hypermethylated or hypomethylated) in certain conditions (e.g., pre-cancer, cancer) as compared to the methylation status of the same mammalian nucleic acid under normal conditions (e.g., in an individual that does not have pre-cancer or cancer). In some cases, hypermethylated DNA markers can be particularly useful for detecting colorectal dysplasia or colorectal cancer. Such hypermethylated DNA markers can include, for example, CpG sequences from a methylated-in-tumor 1 (MINT1) nucleic acid and a runt-related transcript factor 3 (RUNX3) nucleic acid. In some cases, hypomethylated DNA markers can be particularly useful for detecting colorectal dysplasia or colon cancer. Such hypomethylated DNA markers can include, for example, CpG sequences from a cyclooxygenase 2 (COX-2) nucleic acid.
[0042] DNA methylation does not alter the coding function of a DNA, but has the potential to alter gene expression and thus can have profound developmental and genetic consequences. DNA methylation occurs at target cytosine residues that are found within CpG dinucleotides. The methylation reaction involves flipping a target cytosine out of an intact double helix to allow the transfer of a methyl group from S-adenosylmethionine to form 5-methylcytosine (Klimasauskas et al., Cell 76:357-369 (1994)). Areas of the genome containing long repeats of CpG dinucleotides are referred to as “CpG islands” (Bird, Nature 321:209-213 (1986) and Gardiner-Garden et al., J. Mol. Biol., 196:261-282 (1987)). CpG islands typically are between 0.2 to about 1 kb in length and are located upstream of many genes, but may also extend into gene coding regions.
[0043] Methylation of cytosine residues contained within CpG islands of certain genes typically correlates inversely with gene activity. For example, CpG islands of promotors are unmethylated if genes are expressed. Methylation can lead to decreased gene expression by a variety of mechanisms including, without limitation, disruption of local chromatin structure, inhibition of DNA binding by transcription factors, or by recruitment of proteins that interact specifically with methylated sequences and thus indirectly prevent transcription factor binding. Hypermethylation of CpG islands within tumor suppressor genes therefore can lead to progressive reduction of normal tumor suppressor expression, resulting in the selection of a population of cells having a selective growth advantage (i.e., neoplasm). Alterations in normal methylation processes also can be associated with genomic instability (see, e.g., Lengauer et al., Proc. Natl. Acad. Sci. USA, 94:2545-2550 (1997)). Such abnormal epigenetic changes may be found in many types of cancer and can therefore serve as potential markers for oncogenic transformation.
[0044] Any appropriate method can be used to detect a DNA methylation marker in a sample. Such methods can include isolating DNA from the sample, separating out one or more particular regions from the total DNA (e.g., CpG islands), subjecting the DNAs to bisulfite treatment, and determining whether the separated DNAs are abnormally methylated (e.g., hypermethylated). To analyze which residues within a DNA sample are methylated, the sequences of PCR products corresponding to samples treated with and without sodium bisulfite can be compared. The sequence from the untreated DNA will reveal the positions of all cytosine residues within the PCR product. Cytosines that were methylated will be converted to thymidine residues in the sequence of the bisulfite-treated DNA, while residues that were not methylated will be unaffected by bisulfite treatment.
[0045] In some cases, a test nucleic acid sample can be amplified with primers which amplify a sequence region known to include a CpG island region of interest. For example, primers specific for unmethylated and methylated nucleic acids such as those described in Example 1 can be used to amplify the sample DNA and determine the methylation status of the tested residues. In some cases, oligonucleotide primers can amplify a region of interest in a RUNX3, MINT1, or COX-2 nucleic acid. For example, the methylated primers of SEQ ID NO: 37 and SEQ ID NO: 38 amplify a 129 base pair fragment (SEQ ID NO: 45) and unmethylated primers of SEQ ID NO: 39 and SEQ ID NO: 40 can be used to amplify a 159 base pair fragment (SEQ ID NO 46) in the promoter region of a RUNX3 nucleic acid (Table 1). In some cases, alternate oligonucleotide primer sequences could be used to amplify all or part of the RUNX3 nucleic acid fragment of SEQ ID NO: 45 or SEQ ID NO: 46 or any fragment of SEQ ID NO: 53 (FIG. 4) that when amplified, can be analyzed to determine the methylation status of a RUNX3 nucleic acid. A patient diagnosed with IBD and containing a hypermethlated RUNX3 nucleic acid (e.g. elevated methylation status) can be classified as being at a higher risk of having or developing colorectal cancer as compared to a corresponding patient not containing a hypermethylated RUNX3 nucleic acid.
[0046] In another example, the methylated primers of SEQ ID NO: 29 and SEQ ID NO: 30 amplify a 81 base pair fragment (SEQ ID NO: 47) and unmethylated primers of SEQ ID NO: 32 and SEQ ID NO: 33 can be used to amplify a 112 base pair fragment (SEQ ID NO 48) in the promoter region of a MINT1 nucleic acid. In some cases, alternate oligonucleotide primer sequences could be used to amplify all or part of the MINT1 nucleic acid fragment of SEQ ID NO: 47 or SEQ ID NO: 48 or any fragment of SEQ ID NO: 51 (FIG. 2) that when amplified, can be analyzed to determine the methylation status of a MINT1 nucleic acid. A patient diagnosed with IBD and containing a hypermethlated MINT1 nucleic acid (e.g. elevated methylation status) can be classified as being at a higher risk of having or developing colorectal cancer as compared to a corresponding patient not containing a hypermethylated MINT1 nucleic acid.
[0047] In yet another example, the methylated primers of SEQ ID NO: 13 and SEQ ID NO: 14 amplify a 142 base pair fragment (SEQ ID NO: 49) and the unmethylated primers of SEQ ID NO: 15 and SEQ ID NO: 16 can be used to amplify a 138 base pair fragment (SEQ ID NO 50) in the promoter region of a COX-2 nucleic acid. In some cases, alternate oligonucleotide primer sequences could be used to amplify all or part of the COX-2 nucleic acid fragment of SEQ ID NO: 49 or SEQ ID NO: 50 or any fragment of SEQ ID NO: 52 (FIG. 3) that when amplified, can be analyzed to determine the methylation status of a COX-2 nucleic acid. In some cases, a patient diagnosed with IBD and containing a hypomethylated COX-2 nucleic acid (e.g. reduced methylation status) can be classified as being at a higher risk of having or developing colorectal cancer as compared to a corresponding patient not containing a hypomethylated COX-2 nucleic acid.
[0048] Other non-limiting examples of nucleic acids where analyzing the methylation status may be useful in determining the risk of developing colorectal cancer in IBD patients include p16, p14, e-cadherin, estrogen receptor and HPP1.
[00001] [TABLE-US-00001]
  TABLE 1
 
  Methylation Assay Amplification Products
  SEQ ID   Gene   Status   Nucleotide Sequence
 
  45   RUNX3   MethylatedCGTTTGCGTGGTTCGTTAGTACGTTTATTA
      TCGAGCGTATTTCGGGTCGGGCGCGTTTTT
      CGGGTTTTACGGTCGTTTGCGCGTTTAGCG
      CGTCGTTGTTTTCGTTTATTTTGTCGTCGT
      CGTCGTCGT
 
  46   RUNX3   UnmethylatedTTGGGTTTTATGGTTGTTTGTGTGTTTAGT
      GTGTTGTTGTTTTTGTTTATTTTGTTGTTG
      TTGTTGTTGTAGGGGAAGGTTGGGGAGGGA
      GGTGTGAAGTGGTGGTTGGTGTTTGGGTTT
      ATGGGAATATGTATAATAGTGGTTGTTAGG
      GTGTTGGGT
 
  47   MINT1   MethylatedTTTCGAAGCGTTTGTTTGGCGTTTAAGAGA
      GAGTAAGAGAGGGTTGGAGTGTAGGGGAGT
      TCGCGGGGTTGAGGTTT
 
  48   MINT1   UnmethylatedTGGAGAGTAGGGGAGTTTGTGGGGTTGAGG
      TTTTTTGTTAGTGTTTGTATTTTTTATGTT
      ATAATGTTTTTATTTAGTAAAAATTTTTTG
      GGTGTTTGTTGTGTGTTAGGTT
 
  49   COX-2   MethylatedAGGGGATTTTTTGCGTTTTCGGATTTTAGG
      GTCGTTTAGATTTTTGGAGAGGAAGTTAAG
      TGTTTTTTTGTTTTTTTTCGGTATTTTATT
      TAAGGCGATTAGTTTAGAATTGGTTTTCGG
      AAGCGTTCGGGTAAAGATTGCG
 
  50   COX-2   UnmethylatedGAGGGGATTTTTTGTGTTTTTGGATTTTAG
      GGTTGTTTAGATTTTTGGAGAGGAAGTTAA
      GTGTTTTTTTGTTTTTTTTTGGTATTTTAT
      TTAAGGTGATTAGTTTAGAATTGGTTTTTG
      GAAGTGTTTGGGTAAAGA
 
[0049] It is noted that a single sample can be analyzed for one DNA methylation marker or for multiple DNA methylation markers. For example, a sample can be analyzed using assays that detect a panel of different DNA methylation markers. In addition, multiple samples can be collected from a single mammal and analyzed as described herein. In some cases, PCR techniques can be used to detect the presence or absence of a methylated mammalian nucleic acid marker. Cottrell et al describe appropriate methods of methylation-specific PCR (MSP) and other DNA methylation techniques (Ann N Y Acad Sci 2003 March; 983:120-30).
[0050] Purified nucleic acid fragments from a sample or samples can be analyzed to determine the presence or absence of one or more polymorphisms, such as single nucleotide polymorphisms (SNPs). For example, a sample can be analyzed to determine the presence or absence of a polymorphism identified as rs1800629 (−308G>A) which can be viewed in the single nucleotide polymorphism section of the NCBI website and the TNF-alpha sequences carrying the major alleles disclosed as SEQ ID NO:1 in the present document. It is noted that the minor allele (e.g. A) of this SNP is associated with higher risk of having or developing colorectal cancer, whereas the major allele (e.g. G) is associated with lower risk of developing colorectal cancer. In some cases, a test sample can be analyzed to determine the presence or absence of one or more polymorphisms such as −301G>A, and −293C>T in a TNF-alpha nucleic acid. The exact position of the aforementioned variants may vary from individual to individual or from species to species, e.g., by from 1 to about 10 base pairs. Further description of these and other TNF-alpha polymorphisms are provided elsewhere (Garrity-Park et al., Am. J. Gastroenterol., 103:407-415 (2008)). In some cases, polymorphisms may occur in the promoter region of a TNF-alpha nucleic acid. In some cases, polymorphisms may occur in the coding or non-coding regions of a TNF-alpha nucleic acid.
[0051] A mammal diagnosed with IBD and containing one or more polymorphisms in a TNF-alpha nucleic acid can be classified as being at a higher risk of having or developing colorectal cancer as compared to a corresponding mammal containing wild-type TNF-alpha nucleic acid at one or both alleles. For example, detection of the rs1800629 polymorphism in a sample from an IBD patient indicates that the patient is at a higher risk of having or developing colorectal cancer. Detection of this polymorphism allows selection of a monitoring schedule or treatment plan that is most likely to be effective in early diagnosis and prevention of CRC.
[0052] In some embodiments, genomic DNA or mRNA can be used to detect polymorphisms. If mRNA is used, a cDNA copy may first be made. Genomic DNA or mRNA is typically extracted from a biological sample, such as a peripheral blood sample or a tissue sample. Standard methods can be used to extract genomic DNA or mRNA from a biological sample, such as phenol extraction. In some cases, genomic DNA or mRNA can be extracted using a commercially available kit (e.g., from Qiagen, Chatsworth, Calif.; Promega, Madison, Wis.; or Gentra Systems, Minneapolis, Minn.).
[0053] Any appropriate method of analysis can be used to detect a polymorphism in a nucleic acid. Methods of analysis can include conventional Sanger based sequencing, pyrosequencing, next generation sequencing, allele specific PCR, allele-specific restriction digests, microarrays, single molecule sequencing, sequencing by synthesis, single strand conformation polymorphism (SSCP) detection, restriction length polymorphism (RFLP) analysis, denaturing high performance liquid chromatography (DHPLC), and the like. The aforementioned techniques are well known in the art. Detailed description of these techniques can be found in a variety of publications, including, e.g., “Laboratory Methods for the Detection of Mutations and Polymorphisms in DNA” (1997) G. R. Taylor, ed., CRC Press, and references cited therein.
[0054] In some cases, a test nucleic acid sample can be amplified with primers which amplify a sequence region known to comprise the polymorphism(s) of interest. For example, oligonucleotide primers such as SEQ ID NO: 2 (ACCTGGTCCCCA-AAAGA) and SEQ ID NO: 3 (CGGGGATTTGGAAAGTTG) can be used to amplify a region of interest in a TNF-alpha nucleic acid. The primers of SEQ ID NO: 2 and SEQ ID NO: 3 amplify a 186 base pair fragment (SEQ ID NO: 4) in the promoter region of a TNF-alpha nucleic acid. In some cases, alternate oligonucleotide primer sequences could be used to amplify all or part of the TNF-alpha nucleic acid fragment of SEQ ID NO: 4 (FIG. 5) or any fragment of SEQ ID NO: 1 that when amplified, can be analyzed for association with increased TNF-alpha expression levels. The reference TNF-alpha promoter region nucleic acid sequence is provided in GenBank (Accession No. AB048818; GI No. 13365764); a portion of this sequence is provided in FIG. 1 and SEQ ID NO: 1.
[0055] In another example, commercially available kits can be used to amplify a region of interest. For example, a commercially available kit, such as a SNP genotyping kit from Applied Biosystems, can be used to amplify of region of interest in an Interleukin 1B (IL1B) nucleic acid. In some cases alternate kits or methods could be used to amplify all or part of an IL1B nucleic acid fragment that when amplified, can be analyzed for the presence or absence of a polymorphism identified as rs1143627 (−31T>C) which can be viewed in the single nucleotide polymorphism section of the NCBI website. It is noted that the T allele of this SNP is associated with higher risk of having or developing ulcerative colitis associated-colorectal cancer, whereas the C allele is associated with lower risk of developing ulcerative colitis associated-colorectal cancer. The exact position of the aforementioned variants may vary from individual to individual or from species to species, e.g., by from 1 to about 10 base pairs. In some cases, polymorphisms may occur in the promoter region of an IL1B nucleic acid. In some cases, polymorphisms may occur in the coding or non-coding regions of an IL1B nucleic acid.
[0056] A mammal diagnosed with IBD and containing one or more polymorphisms in an IL1B nucleic acid can be classified as being at a higher risk of having or developing colorectal cancer as compared to a corresponding mammal containing wild-type IL1B nucleic acid at one or both alleles. For example, detection of the rs1143627 polymorphism in a sample from an IBD patient indicates that the patient is at a higher risk of developing colorectal cancer. Detection of this polymorphism allows selection of a monitoring schedule or treatment plan that is most likely to be effective in early diagnosis and treatment of CRC. Other non-limiting examples of polymorphisms associated with a higher risk of developing colorectal cancer include SNP's found in an Interleukin-23 Receptor (IL-23R) nucleic acid such as rs10889677 (2284C>A) and rs1884444 (94G>T).
[0057] Polymorphisms in promoter sequences may affect gene expression. In some cases, serum levels of one or more of a TNF-alpha, IL1B, or IL-23R polypeptide can be measured to determine whether or not an IBD patient has an increased risk of developing CRC. For example, an IBD patient with an increased serum level of a TNF-alpha polypeptide as compared to a normal control may have an increased likelihood of developing CRC. Detection of an increased level of a TNF-alpha polypeptide allows selection of a monitoring schedule or treatment plan that is most likely to be effective in early diagnosis and treatment of CRC. Any known method for measuring polypeptide levels can be used, such as denaturing high performance liquid chromatography (DHPLC, Underhill et al. (1997) Genome Res. 7:996-1005), infrared matrix-assisted laser desorption/ionization (IR-MALDI) mass spectrometry (WO 99/57318), and combinations of such methods. Other useful detection techniques include, but are not limited to surface-enhanced laser desorption/ionization (SELDI) mass spectrometry, immunoassays, and array-based technologies. Other non-limiting examples of increased polypeptide levels associated with a higher risk of developing CRC include increased IL-23R and IL1B polypeptide levels.
[0058] It is understood that the term “specifically amplifies” refers to the ability of an oligonucleotide primer to interact specifically with a particular nucleic acid without significantly cross-reacting with other different nucleic acids in the same environment and facilitate or promote the amplification of that particular nucleic acid. Likewise, the term “specifically hybridizes” refers to the ability of an oligonucleotide probe to interact specifically with a particular nucleic acid without significantly cross-reacting with other different nucleic acids in the same environment and facilitate or promote the detection of that particular nucleic acid.
[0059] The term “elevated level” as used herein with respect to the level of an MPO polypeptide is any level that is above a median polypeptide level in a sample from a random population of mammals (e.g., a random population of 10, 20, 30, 40, 50, 100, or 500 mammals) that do not have UC-CRC. Elevated MPO polypeptide levels can be any level provided that the level is greater than a corresponding reference level. For example, an elevated level of MPO polypeptide can be 0.5, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, or more fold greater than the reference level MPO polypeptide observed in a normal colon biopsy or blood sample. It is noted that a reference level can be any amount. For example, a reference level can be zero. In some cases, an elevated level of an MPO polypeptide can be any detectable level of an MPO polypeptide in a tissue biopsy sample.
[0060] The term “elevated level” as used herein with respect to the methylation status of MINT1 or RUNX3 nucleic acid is any methylation level that is above a median methylation level in a sample from a random population of mammals (e.g., a random population of 10, 20, 30, 40, 50, 100, or 500 mammals) that do not have UC-CRC. Elevated MINT1 or RUNX3 methylation levels can be any level provided that the level is greater than a corresponding reference level. For example, an elevated level of MINT1 or RUNX3 methylation can be 0.5, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, or more fold greater than the reference level methylation observed in a normal colon biopsy sample. It is noted that a reference level can be any amount.
[0061] The term “reduced level” as used herein with respect to the level of COX-2 methylation status is any level that is below a median methylation level in a sample from a random population of mammals (e.g., a random population of 10, 20, 30, 40, 50, 100, or 500 mammals) that do not have UC-CRC. Reduced COX-2 methylation levels can be any level provided that the level is lesser than a corresponding reference level. For example, a reduced level of COX-2 methylation can be 0.5, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, or more fold lesser than the reference level methylation observed in a normal colon biopsy sample. It is noted that a reference level can be any amount.
[0062] As used herein, the terms “treatment,” “treating,” and the like, refer to obtaining a desired pharmacologic and/or physiologic effect. The effect may be prophylactic in terms of completely or partially preventing a disease or symptom thereof and/or may be therapeutic in terms of a partial or complete cure for a disease and/or adverse affect attributable to the disease. “Treatment,” as used herein, includes any treatment of a disease in a mammal, particularly in a human, and includes: (a) preventing the disease from occurring in a subject which may be predisposed to the disease but has not yet been diagnosed as having it; (b) inhibiting the disease, i.e., arresting its development; and (c) relieving the disease, i.e., causing regression of the disease.
[0063] This document provides kits that can be used to determine the level of one or more biomolecules in a sample. Kits can contain an oligonucleotide primer pair that specifically amplifies all or a portion of a target region of a nucleic acid. For example, kits can contain oligonucleotide primers that specifically amplify a TNF-alpha, COX-2, MINT1, or a RUNX3 nucleic acid. Target regions can be defined at any place along a TNF-alpha, COX-2, MINT1, or RUNX3 nucleic acid. For example, a target region can be defined by nucleotides 1-500 of the 5′ portion of a TNF-alpha nucleic acid. In this case, a kit of the invention can contain an oligonucleotide primer pair that specifically amplifies all 500 nucleotides defining that target region, or a portion (e.g., nucleotides 80-188) of that target region.
[0064] Components and methods for producing kits are well known. Kits can contain multiple oligonucleotide primer pairs that specifically amplify TNF-alpha, COX-2, MINT1, or RUNX3-related nucleic acids, or probes that specifically hybridize TNF-alpha, COX-2, MINT1, or RUNX3-related nucleic acids. In addition, kits can contain antibodies for detecting MPO-related polypeptides. The kits provided herein also can contain a reference chart that indicates a reference level or baseline for MPO polypeptides or TNF-alpha, COX-2, MINT1, or RUNX3 nucleic acids. Kits can be configured in any type of design (e.g., microtiter plate design) and can be made of any type of material (e.g., plastic).
[0065] In some cases, a human may have a family history of primary sclerosing cholangitis (PSC) or CRC. Family history or relatives with PSC or CRC can be identified by examining medical records or family tree history. The methods provided in this document can also be used to identify CRC risk in relatives of affected mammals likely to have IBD or UC. Thus, these methods can facilitate decisions regarding the course of evaluation and treatment in humans with and without altered methylation in MINT1, COX-2, or RUNX3 nucleic acids, with and without polymorphisms in a TNF-alpha nucleic acid, or with and without increased MPO polypeptide levels.
[0066] This document also provides materials and methods to assist a medical professional in determining the risk of having or developing colorectal cancer in a mammal. Such a medical professional can be, for example, a physician, a nurse, a medical laboratory technologist, or a pharmacist. A person can be assisted by (1) determining the presence or absence of a nucleic acid polymorphism in a nucleic acid such as a TNF-alpha nucleic acid, determining the methylation status of a nucleic acids such as a RUNX3 nucleic acid in a test sample, and/or determining the level of a polypeptide such as an MPO polypeptide, and (2) communicating information about the presence, absence, or level of that marker to that medical professional.
[0067] After the presence, absence, level, or status of a particular biomolecule or biomolecules is reported, a medical professional can take one or more actions that can affect patient care. For example, a medical professional can record the results in a patient's medical record. In some cases, a medical professional can record that a patient is at an increased risk of developing colorectal cancer, or otherwise transform the patient's medical record to reflect the patient's medical condition. In some cases, a medical professional can review and evaluate a patient's entire medical record and assess multiple strategies for clinical intervention of a patient's condition. In some cases, a medical professional can recommend a change in therapy or a change in frequency or type of surveillance.
[0068] Any appropriate method can be used to communicate information to another person. For example, information can be given directly or indirectly to a person. In addition, any type of communication can be used to communicate the information. For example, mail, e mail, telephone, and face-to-face interactions can be used. The information also can be communicated to a person by making that information electronically available to the person. For example, the information can be communicated to a person by placing the information on a computer database such that the person can access the information. In addition, the information can be communicated to a hospital, clinic, or research facility at which the person is located.
[0069] The invention will be further described in the following examples, which do not limit the scope of the invention described in the claims.

EXAMPLES

Example 1

Methylation Status of Genes in Non-Neoplastic Mucosa from Patients with Ulcerative Colitis-Associated Colorectal Cancer

Patient Selection

[0070] The Mayo Clinic Institutional Review Board approved this work. The UC-CRC cases and UC controls analyzed herein were described elsewhere (Garrity-Park et al., Am. J. Gastroenterol., (2008) and Garrity-Park et al., Gut (2009)). UC-CRC cases were selected from a review of 274 patients identified from the Mayo Clinic centralized diagnostic index of medical records (1976-2006). These patients had inflammatory bowel disease (either Crohn's or UC) and CRC. Patients with Crohn's disease were excluded. Medical records for the remaining UC-CRC patients were reviewed to establish a date of disease onset. For each case, pathology slides from the surgical resection were also recalled to confirm the diagnosis of UC and identify the best tumor and non-adjacent, non-neoplastic block for DNA extraction. Patients who did not have UC confirmed by review of the pathology or whose duration of disease was less than 10 years were excluded. After these exclusions, 114 UC-CRC cases were included.
[0071] Potential UC controls were identified through the Mayo pathology index (1994-2006), which indicated the patient age, gender, and extent of UC as well as the presence of other confounding pathologies such as dysplasia. The final pool of potential UC controls for this work included UC patients who did not develop CRC, who underwent either colectomy or colonoscopy with biopsy at the Mayo Clinic, and who did not have prior dysplasia. The Mayo Clinic centralized diagnostic index of medical records was used with these remaining controls to establish a date of diagnosis. Patients with less than ten years between the date of UC diagnosis and either colectomy or date of last biopsy were excluded as were patients with a prior dysplasia diagnosis. From the remaining list, 181 controls were selected that were most closely matched to the UC-CRC cases with regard to gender, age, ethnicity, duration, and extent of UC. The surgical resection or biopsy specimens from these 181 controls were re-reviewed to confirm histologically the diagnosis of UC. After final review, 114 UC controls were included.

DNA Extraction

[0072] All formalin-fixed, paraffin-embedded (FFPE) blocks and hematoxylin and eosin-stained (H&E) slides were reviewed on all cases and controls to determine the inflammatory activity level (as assessed by neutrophil infiltrates) for all non-neoplastic sections. Each section was scored as normal, inactive (0), mildly active (1), moderate (2), or severe (3). A total of three different DNA extractions were then completed: 1) UC control, 2) UC-CRC non-neoplastic, non-adjacent, and 3) UC-CRC tumor. For non-adjacent, non-neoplastic UC-CRC cases and UC controls, DNA was extracted from all non-neoplastic paraffin tissue sections that showed evidence of chronic disease (scores 0-3; n=1-6 blocks/patient). For the tumor DNA extraction, only the section with confirmed CRC was used. Any sections scored as “normal colon” or any dysplastic lesions located away from the CRC in UC-CRC cases were excluded from all extractions. DNA was extracted using Gentra Puregene Tissue kit (Qiagen, Valencia, Calif.). DNA pellets were suspended in TE (10 mM Tris, pH=7.5, 0.1 mM EDTA, Integrated DNA Technologies, Coralville, Iowa) and quantified using Quant-iT™ PicoGreen® (Invitrogen, Carlsbad, Calif.).

Bisulfite Treatment/Methylation Specific Polymerase Chain Reaction (MSP)

[0073] Methylation status of each gene was determined using MSP after bisulfite treatment of 500 ng of DNA using the EZ DNA Methylation-Gold Kit™ (Zymo Research, Orange, Calif.) following standard protocols. Primers were designed using Methyl Primer Express v1.0 software (Applied Biosystems, Foster City, Calif.) (Table 2). Most of the proposed genes have multiple methylation sites. Therefore, whenever possible, sites chosen for evaluation were selected based on published studies that indicated that methylation in that area altered protein expression in situ. Primers were designed for the following genes: p16, p14, cyclooxygenase-2 (COX-2), e-cadherin, estrogen receptor (ER), HPP1, methylated-in-tumor 1 (MINT1), MINT31, RUNX3, and sodium solute symporter family 5 member 8 protein (SLC5A8). Unmethylated and methylated PCR reactions were carried out in separate, 25 μL reactions.
[0074] Amplicons were run through ethidium-stained agarose gels and visualized using the BioRad Gel Doc™ (Bio-Rad, Hercules, Calif.). Positive and negative controls were included in each experimental set-up. A sample was considered positive if amplicon was produced using the methylated primer set. A sample was negative for methylation if amplicon was produced using only the unmethylated primer set. Samples that did not produce amplicons for either reaction were excluded from analyses. To ensure the specificity of each reaction and to validate the adequacy of the bisulfite modification, 25 methylated and 25 unmethylated amplicons were sequenced for each gene using the ABI PRISM™ (Applied Biosystems) after shrimp alkaline phosphatase (USB, Cleveland, Ohio) and exonuclease (USB) treatment of the amplicon. All of the MSP products demonstrated methylation of CpG sites. To test the sensitivity of each methylated/unmethylated assay, serial dilutions of positive control DNA (100% to 0%) were tested for each gene. All assays could detect a positive result with 5% positive control DNA.
[00002] [TABLE-US-00002]
  TABLE 2
 
  Methylation-specific PCR primers.
        Size  
  Gene   Forward (5′ → 3′)   Reverse (5′ → 3′)   (bp)   Temp
 
  p16
  MethylatedTGGGGCGGATCGCGTCGACCCCGAACCGC   140   60
  GCGTT (SEQ ID NO: 5)GACGGT (SEQ ID NO: 6)    
  UnmethylatedTGGGGTGGATTGTGTCCACCTCCAACAATA   172   60
  GTGTTTGGT (SEQ ID NO: 7)CCCATACCT (SEQ ID NO: 8)    
 
  p14
  MethylatedGGCGGCGAGAATATGACGACGAACGGCCC   137   60
  GTGC (SEQ ID NO: 9)TAACG (SEQ ID NO: 10)    
  UnmethylatedTTGGTGTTAAAGGGTGAAAAACCCTCACTC   126   60
  GT (SEQ ID NO: 11)ACAA (SEQ ID NO: 12)    
 
  COX-2
  MethylatedAGGGGATTTTTTGCGTCGCAATCTTTACCCG   142   55
  TTTC (SEQ ID NO: 13)AACGC (SEQ ID NO: 14)    
  UnmethylatedGAGGGGATTTTTTGTGTCTTTACCCAAACAC   138   60
  TTTTT (SEQ ID NO: 15)TTCCAA (SEQ ID NO: 16)    
 
  E-cadherin
  MethylatedTTAGAGGGTTATCGCG   ACCAAATAAACCCC    
  TTTATGC (SEQ ID NO: 17)GAAACACGG (SEQ ID NO: 18)   150   50
  Unmethylated   TAATTTTAGGTTAGAG   CACAACCAATCAAC    97   63
  GGTTATTGT (SEQ ID NO: 19)AACACA (SEQ ID NO: 20)    
 
  Estrogen receptor
  MethylatedCGTTCGGTTTTATCGG   AAAAACTCAAAAAC   138   55
  ATTC (SEQ ID NO: 21)CGACGA (SEQ ID NO: 22)    
  UnmethylatedTGAGTTGGAGTTTTTGACACATTAACAACA   149   60
  AATTGTTT (SEQ ID NO: 23)ACCACA (SEQ ID NO: 24)    
 
  HPP1
  MethylatedTTTCGGCGTAGTTTTTACTAAACATCCCGC   167   60
  TAGC (SEQ ID NO: 25)GAACG (SEQ ID NO: 26    
  UnmethylatedTGGTGTAGTTTTTTAGACAATAACAATAAC   127   60
  TGGATG (SEQ ID NO: 27)ACCCAACA (SEQ ID NO: 28)    
 
  MINT1
  MethylatedTTTCGAAGCGTTTGTT   CAAAAAACCTCAAC    81   55
  TGGC (SEQ ID NO: 29)CCCGGG (SEQ ID NO: 30)    
  Unmethylated   TGGAGAGTAGGGGAGAACCTAACACACAA   112   60
  TTTGT (SEQ ID NO: 31)CAAACA (SEQ ID NO: 32)    
 
  MINT3 1
  MethylatedTATTCGATTTATTTCGCTACGAAAAATAAA   105   55
  TC (SEQ ID NO: 33)CACG (SEQ ID NO: 34)    
  UnmethylatedGATTTTAATTTTTTGT   CTAAAACCATCACCC    95   60
  GGTGGT (SEQ ID NO: 35)CTAAACA (SEQ ID NO: 36)    
 
  RUNX3
  MethylatedCGTTTGCGTGGTTCGTACGACGACGACGAC   129   60
  TAGTAC (SEQ ID NO: 37)GACA (SEQ ID NO: 38)    
  UnmethylatedTTGGGTTTTATGGTTGACCCAACACCCTAA   159   60
  TTTGTGT (SEQ ID NO: 39)CAACCAC (SEQ ID NO: 40)    
 
  SLC5A8
  MethylatedACGGGGTATCGGTATTTACGATCATTCTACG   151   55
  TTC (SEQ ID NO: 41)ACCG (SEQ ID NO: 42)    
  UnmethylatedGGTTATTTTGGTTGTTCAAACACTACAATC   104   55
  ATT (SEQ ID NO: 43)ATTCTACA (SEQ ID NO: 44)
 
  bp, base pairs (bases in bold indicate methylation sites); COX, cyclooxygenase; HPP, hyperplastic polyposis gene; MINT, methylated-in-tumor; RUNX, runt-related transcript factor; SLC5A8, sodium solute symporter family-5 member-8.

Inflammation Scoring

[0075] All H&E slides from each case or control were reviewed by a pathologist, and the histologic disease activity was scored as inactive, mild, moderate, or severe based on the percentage of neutrophils. Each histologic activity level was given a corresponding number, such that inactive sections were scored as 0 and mildly active, moderate, or severe sections were scored as 1, 2, or 3, respectively. To obtain the final inflammation score for each case or control extracted, the values for all sections included in the extraction were summed and then divided by the total number of sections used. For instance, if a non-adjacent, non-neoplastic extraction for a case had four sections that were inactive and two sections that were mildly active, the inflammation score would be 0.33. Scores were obtained for all non-adjacent, non-neoplastic extractions for cases and for all extractions for controls.

Statistics

[0076] For initial identification of potential genes, a univariate analysis using the Fisher's exact test was done to compare the prevalence of methylation for each gene in UC-CRC cases versus UC controls. For genes identified as significant, another Fisher's Exact test was performed to determine its significance when comparing non-neoplastic DNA from UC-CRC cases versus UC controls. A multivariable analysis was then done to test for interactions between the genes found to be significant in the non-neoplastic comparison. Finally, logistic regression modeling was performed to determine the additive effect of these significant genes.

Results

Patient Selection

[0077] The summary of patient characteristics is given in Table 3. There were no significant differences between cases and controls with regard to age, gender, family history of CRC, or duration/extent of UC. Because there were no significant differences in distribution of disease extent between cases and controls (p=0.07), all subsequent analyses involving extent therefore used the broad categorization of extensive versus non-extensive (left-sided and proctitis) disease for cases and controls. Primary sclerosing cholangitis (PSC) was more prevalent in UC-CRC cases (p<0.0001). All cases and controls were Caucasian.
[00003] [TABLE-US-00003]
  TABLE 3
 
  Demographic and clinical features of
  114 UC-CRC cases and 114 UC controls.
  Demographic/clinical   UC-CRC   UC, no CRC  
  information   (n = 114)   (n = 114)   P value
 
  Mean age at index date,   47.8 (26-82)    48.8 (24-77)   0.35
years (range)a
  Gender, n (%)
  Female   36 (32)     36 (32)   
  Male   78 (68)     78 (68)    1.00
  Mean duration of UC at   20.3 (10-49)    19.5 (10-45)   0.37
  index date, years (range)
Maximal extent, n (%)b
  Proctitis   8 (7.0)   7 (6)  
  Left-sided   12 (10.5)   25 (22)
  Extensive   94 (82.5)   82 (72)0.07c
  PSC, n (% yes)   31 (27.2)     4 (3.5)<0.0001
  Family history of CRC, n (%)   19 (17)     15 (13)   0.58
 
aIndex date for UC-CRC was the age at CRC and for UC controls was the age at colectomy or most recent biopsy.
bExtent based on histological assessment of involvement.
cχ2-Test, P value represents the comparison between cases and controls with extensive colitis vs. left-sided vs. proctitis.
  Values in bold are significant.

DNA Extraction and Location

[0078] Sixty percent of UC-CRC non-adjacent, non-neoplastic DNA extractions included a tissue section that was from the same segment of the colon in which the tumor arose, i.e. the tumor was in the ascending colon, and a different block without neoplasia was also available from the ascending colon. The majority of UC-CRC non-neoplastic and UC control DNA extractions included tissues obtained from both the left and right side of the colon (69% versus 74%, p=0.57). Three cases and two controls had tissue available from the rectum only. The majority of tissues used for UC-CRC case and UC control (218/228) extractions were obtained from resected colons. The remaining 10 patients had only biopsy samples available.

UC-CRC DNA Versus UC Control DNA

Univariate Analyses

[0079] To identify targets to investigate in non-adjacent, non-neoplastic regions of the UC-CRC colon, initial univariate analyses focused on the level of gene methylation in DNA extracted from tumor sections only as compared to UC controls. The majority of DNA from UC controls (between 96 to 109) and UC-CRC tumors (between 83 to 100) were successfully amplified for each target. Table 4 summarizes the results of these analyses for all 10 genes included in this study. The prevalence of gene methylation for p16, RUNX3, MINT1, MINT31, and HPP1 was significantly increased in UC-CRC cases versus controls. Conversely, COX-2 and e-cadherin were more frequently methylated in controls as compared to cases. The difference in methylation for ER, p14, and SLC5A8 was not significantly different between cases and controls.
[00004] [TABLE-US-00004]
  TABLE 4
 
  Univariate analyses of gene methylation status in
  UC-CRC cases (tumor sections) vs. UC controls
  Gene (±for methylation)   UC-CRC (%)   UC controls (%)P value a
 
  (a) Methylated in UC-CRC cases
  p16
  Negative   72 (84.7)   107 (100)    
  Positive   13 (15.3)   0 (0)  <0.0001
  RUNX3
  Negative   46 (55.4)   97 (93.3)  
  Positive   37 (44.6)   7 (6.7)<0.0001
  MINT1
  Negative   46 (49.5)   87 (85.3)  
  Positive   47 (50.5)   15 (14.7)<0.0001
  MINT31
  Negative   39 (40.6)   76 (79.2)  
  Positive   57 (59.4)   20 (20.8)<0.0001
  HPP1
  Negative   19 (21.3)   53 (49.5)  
  Positive   70 (78.7)   54 (50.5)0.0001
  ESR1
  Negative   10 (10.8)   17 (15.9)  
  Positive   83 (89.2)   90 (84.1)   0.31
  p14
  Negative   75 (81.5)   95 (88.0)  
  Positive   17 (18.5)   13 (12.0)   0.24
  SLC5A8
  Negative   14 (14.7)   6 (5.8)  
  Positive   81 (85.3)   97 (94.2)   0.06
  (b) Methylated in UC controls
  COX-2
  Negative   64 (66.7)   43 (39.4)  
  Positive   32 (33.3)   66 (60.6)0.0001
  E-cadherin
  Negative   64 (64.0)   42 (38.9)  
  Positive   36 (36.0)   66 (61.1)0.0003
 
  COX, cyclooxygenase; CRC, colorectal cancer; HPP, hyperplastic polyposis gene; MINT, methylated-in-tumor; RUNX, runt-related transcript factor; SLC5A8, sodium solute symporter family-5 member-8; UC, ulcerative colitis.
a Calculated using Fisher's exact test.
  Values in bold are significant.

UC-CRC Non-Neoplastic DNA Versus UC-Control DNA

Univariate Analyses

[0080] Only genes that were significantly different between tumor and UC controls were tested for significance in non-adjacent, non-neoplastic normal tissue. The majority of DNA from UC controls (between 96 to 109) and non-adjacent, non-neoplastic areas from UC-CRC patients (between 66 to 88) were successfully amplified for each target. RUNX3, p16, MINT1, MINT31, e-cadherin, and COX-2 remained significantly associated with UC-CRC. The association involving HPP 1 was no longer significant (Table 5).
[00005] [TABLE-US-00005]
  TABLE 5
 
  Univariate analyses of gene methylation status in UC-CRC cases
  (non-adjacent, non-neoplastic sections) vs. UC controls.
  Gene (±for methylation)   UC-CRC (%)   UC controls (%)P value a
 
  (a) Methylated in UC-CRC cases
  p16
  Negative   53 (80.3)   107 (100)    
  Positive   13 (19.7)   0 (0)  <0.0001
  RUNX3
  Negative   37 (49.3)   97 (93.3)  
  Positive   38 (50.7)   7 (6.7)<0.0001
  MINT1
  Negative   48 (54.5)   87 (85.3)  
  Positive   40 (45.5)   15 (14.7)<0.0001
  MINT31
  Negative   47 (54.7)   76 (79.2)  
  Positive   39 (45.3)   20 (20.8)0.0005
  HPP1
  Negative   27 (36.0)   53 (49.5)  
  Positive   48 (64.0)   54 (50.5)   0.09
  (b) Methylated in UC controls
  COX-2
  Negative   54 (61.4)   43 (39.4)  
  Positive   34 (38.6)   66 (60.6)0.003
  E-cadherin
  Negative   46 (55.4)   42 (38.9)  
  Positive   37 (44.6)   66 (61.1)0.03
 
  COX, cyclooxygenase; CRC, colorectal cancer; HPP, hyperplastic polyposis gene; MINT, methylated-in-tumor; RUNX, runt-related transcript factor; SLC5A8, sodium solute symporter family-5 member-8; UC, ulcerative colitis.
a Calculated using Fisher's exact test.
  Values in bold are significant.

Multivariable Analyses

[0081] Multivariable logistic regression was performed with the univariately significant genes (p16, RUNX3, MINT1, MINT31, e-cadherin, and COX-2) to determine if each gene was independently significant for UC-CRC. Table 6 indicates p-values and odds ratios for the three genes that remained significant in this analysis. Methylation of RUNX3 and MINT1 in non-neoplastic sections remained strongly associated with the presence of CRC. Conversely, unmethylated COX-2 was an indication of CRC.
[00006] [TABLE-US-00006]
  TABLE 6
 
  Multivariate analyses of UC-CRC (non-adjacent,
  non-neoplastic sections) vs. UC controls.
    Logistic regression  
  Gene (±for methylation)   Odds ratio   CI   P value
 
  (a) Significantly methylated in UC-CRC cases
  RUNX3   12.6   4.4, 35.7<0.0001
  MINT1   9.0   3.4, 23.7<0.0001
  (b) Significantly methylated in UC controls
  COX-2   0.2   0.07, 0.4 0.0002
 
  CI, confidence interval; COX, cyclooxygenase; CRC, colorectal cancer; HPP, hyperplastic polyposis gene; MINT, methylated-in-tumor; RUNX, runt-related transcript factor; UC, ulcerative colitis.
  Values in bold are significant.
[0082] Given the association of methylation with inflammation (Kundu et al., Mutat Res (2008)), a multivariable logistic regression also was performed that included RUNX3, MINT1, COX-2, and the inflammation score to determine if the increased incidence of methylation merely reflected a higher inflammation score in cases versus controls. Table 7A summarizes these findings. The p-values and odds ratios all remained highly significant even when the degree of inflammation, as determined by H&E, was incorporated into the logistic regression. Interestingly, greater inflammation as determined by neutrophils on H&E was not associated with UC-CRC.
[0083] Because cases and controls varied with regard to the presence of PSC, logistic regression also was performed to ensure that the significance of RUNX3, MINT1, and COX-2 was independent of PSC (Table 7B). Although PSC remained highly associated with UC-CRC, the methylation status of these three genes was still significant in this analysis.
[0084] Given that for the majority of the UC-CRC cases (60%) the non-adjacent, non-neoplastic DNA sample included tissue procured from the same region in which the tumor arose, analysis was performed to determine whether proximity to the tumor affected methylation status. The presence or absence of a non-neoplastic section from within the corresponding tumor region did not affect the prevalence of methylation for RUNX3, COX-2, or MINT1 (P=0.17, 0.69, and 0.23, respectively).
[0085] Although there was no significant difference between the UC-CRC non-neoplastic and UC controls with regard to inclusion of tissue from both the right and the left colon (P=0.57), tests were performed to determine whether this could have affected the methylation status of a given gene. It was found that the prevalence of altered methylation was not significantly different between DNA samples containing tissue sections from both the left and the right side of the colon and those that did not (P=0.24, 0.87, and 0.48 for COX-2, MINT1, and RUNX3, respectively).
[0086] Finally, to interrogate whether these alterations in methylation were specific to UC, the prevalence of altered gene methylation in a cohort of non-UC patients described elsewhere (Garrity-Park et al., Gut, 58:1226-1233 (2009)) was assessed. In brief, this cohort included biopsies taken from 60 non-UC normal patients that are a part of the average risk CRC screening population at the Mayo Clinic. These were frequency matched for age to the UC patients (both CRC and non-CRC controls) in this study (average age for UC group, 48 years, vs. 49 years for non-UC patients). For RUNX3, COX-2, and MINT1, there was no significant difference between the UC controls and non-UC patients (P=0.53, 0.21, and 0.70, respectively), but there was a significance between the UC-CRC cases and non-UC patients (P<0.0001, 0.001, and <0.001, respectively).
[00007] [TABLE-US-00007]
  TABLE 7
 
  Effect of inclusion of inflammation and PSC
  in the multivariable model of UC-CRC risk
    Odds ratio   CI   P value
   
    (a) Inflammation      
    Inflammation score   0.3   0.1, 0.6 0.001
    RUNX3   11.9   3.9, 36.0<0.0001
    MINT1   9.7   3.4, 27.7<0.0001
    COX-2   0.2   0.1, 0.5 0.002
    (b) PSC
    PSC   9   2.2, 37.80.003
    RUNX3   11.7   3.9, 35.3<0.0001
    MINT1   10.4   3.7, 28.8<0.0001
    COX-2   0.2   0.06, 0.4 0.0002
   
    CI, confidence interval; COX, cyclooxygenase; CRC, colorectal cancer; MINT, methylated-in-tumor; PSC, primary sclerosing cholangitis; RUNX, runt-related transcript factor; UC, ulcerative colitis.
    Values in bold are significant.

Diagnostic Modeling

[0087] Logistic regression modeling was undertaken to determine if RUNX3, MINT1, and COX-2 interacted to have an additive effect, i.e. did the odds of having a synchronous CRC increase as the number of genes altered increased (Table 8). These analyses indicated that having both RUNX3 methylated and COX-2 unmethylated greatly increased the likelihood of a CRC elsewhere in the colon. This also was true of the concurrent presence of MINT1 methylation and COX-2 unmethylation, although the increase was not as dramatic. Although informative, it is important to note that the number of samples available for this analysis was small, as reflected by the wide confidence intervals.
[00008] [TABLE-US-00008]
  TABLE 8
 
  Logistic regression model of gene methylation on UC-CRC.
    Logistic   Exact odds ratio CI
  Gene combination   regression   estimation method
  (M = methylated;   Odds   (StatExact) a
  U = unmethylated)   Ratio   95% CI   Exact   95% CI
 
  RUNX3 (M) + MINT1   1.0   Referent   1.0   Referent
  (U) + COX-2 (M)
  RUNX3 (M) + COX-2 (M)   4.4   0.8, 23.2   4.2   0.5, 29.6
  MINT1 (M) + COX-2 (M)   6.1   1.7, 21.4   5.9   1.5, 26.8
  RUNX3 (U) + MINT1   4.7   1.6, 14.0   4.6   1.4, 17.7
  (U) + COX-2 (U)
  RUNX3 (M) + MINT1bbb29.5, b
  (M) + COX-2 (M)
  RUNX3 (M) + COX-2 (U)   61.2   6.2, 608.5   53.5   5.4, 2, 833.0
  MINT1 (M) + COX-2 (U)   17.6   2.5, 121.6   16.0    1.8, 219.4
  RUNX3 (M) + MINT1bbb19.6, b
  (M) + COX-2 (U)
 
a Used to establish the lower confidence interval of the effect.
b Unable to calculate because there is a 0 in the control group.

Example 2

Myeloperoxidase as a Measure of Disease Activity in UC: Association with UC-CRC, TNF Polymorphism, and RUNX3

Patient Selection

[0088] The Mayo Clinic Institutional Review Board approved this work. Patients with UC for >10 years who developed CRC were identified from the Mayo Clinic centralized diagnostic index of medical records (1986-2006). For each case, pathology slides from the surgical resection were recalled to confirm the diagnosis of UC and to identify the best non-adjacent, non-neoplastic block for immunostaining Complete patient chart reviews were completed on all UC-CRC cases. Patients who did not have UC confirmed by review of the pathology or whose duration of disease was less than 10 years as documented in the clinical chart were excluded. A total of 50 UC-CRC cases, representing a subset of the UC-CRC cases described elsewhere (Garrity-Park et al., Gut., 58:1226-1233 (2009); Garrity-Park et al., Am. J. Gastroenterol., 103(2):407-15 (2008); and Garrity-Park et al., Am. J. Gastroenterol., 107(7):1610-9 (2010)), were examined in this study. UC controls were identified through the Mayo pathology index (1994-2006), which indicated the patient age, gender, and extent of UC as well as any other confounding pathologies such as dysplasia. Complete patient chart reviews were completed on all potential UC-controls. All potential UC-controls included UC patients who did not develop CRC, who had greater than 10 years of disease, who underwent either colectomy or colonoscopy with biopsy at the Mayo Clinic, and who did not have prior dysplasia. The final selection of UC-controls was based on frequency matching to UC-CRC cases for age, gender, extent, and duration of UC. A representative non-neoplastic section for each control was selected for analyses. A total of 50 UC-controls, a subset of the UC-control group described elsewhere (Garrity-Park et al., Gut., 58:1226-1233 (2009); Garrity-Park et al., Am. J. Gastroenterol., 103(2):407-15 (2008); and Garrity-Park et al., Am. J. Gastroenterol., 107(7):1610-9 (2010)), were analyzed in this study.

H&E Scoring

[0089] A board certified pathologist reviewed and scored all sections. Histologic disease activity level was determined for the entire resection specimen using standard clinical methodologies utilizing H&E-stained sections. A disease activity score was assigned to each case or control using the following cut offs: 0 (inactive)—no neutrophils; 1 (mild)—rare neutrophils in crypt or surface epithelium; 2 (moderate)—neutrophils in up to 25% of crypts; and 3 (severe)—neutrophils in more than 25% of crypts.
[0090] For cases, slides from all available sections were examined to identify the best non-adjacent, non-neoplastic section for immunostaining. Whenever possible, this section was from an area distinct from where the tumor arose. Selection criteria also included 1) a well-oriented, full thickness section with generous amounts of mucosa for improved likelihood of informative IHC scoring and 2) a section reflective of overall disease state, i.e. if the patient had colitis to the hepatic flexure, sections were not chosen from the cecum.
[0091] For controls, slides from all available sections were examined to represent the best normal section for immunostaining. This included a well-oriented, full thickness section with generous amounts of mucosa that was reflective of the overall disease state.

Immunohistochemistry

[0092] Serial 4-micron sections were cut from each selected block. CD3 (DAKO, Carpinteria, Calif.), CD68 (DAKO), and MPO (Abcam, Cambridge, Mass.) antibodies were applied and developed using the DAKO Envision+system (DAKO) after heated antigen retrieval. Whole sections were then digitally scanned using the NanoZoomer (Hamamatsu, Bridgewater, N.J.). Scans were downloaded and analyzed using ImageJ software (available at http://rsbweb.nih.gov/ij/). Using 15 different tissue sections, optimal thresholding was established that accurately distinguished positive cellular area (stained brown) from negative area (stained purple). Once established, this threshold was used for all subsequent slides to avoid biasing results. A total of four to six areas of mucosa were measured to determine the % area positive (scans were analyzed at 5× magnification). The average of the areas was recorded and used for statistical analyses.

TNF-α Polymorphism and RUNX3 Data

[0093] Prior studies included runt-related transcription factor 3 (RUNX3) methylation status and single nucleotide polymorphism testing for TNF-α (Garrity-Park et al., Am. J. Gastroenterol., 103(2):407-15 (2008); and Garrity-Park et al., Am. J. Gastroenterol., 107(7):1610-9 (2010)). These previously derived data were used for associations in the current study. Cases and controls were selected without knowledge of TNF-α or RUNX3 status.

Statistical Analyses

[0094] Four statistical analyses were performed: 1) a Fisher exact test or Chi-square test was used to determine if the demographic/clinical selection criteria between UC-CRC and UC-control groups were appropriately matched; 2) a Fisher exact test was used to test for significant differences in % area between UC-CRC cases and UC-controls for CD3, CD68, and MPO, and for the association between the % area and the presence/absence of TNF-α SNP and RUNX3; 3) a Chi-square test was used to determine the association between case/control status and TNF-α SNP and RUNX3 methylation; and 4) logistic regression and receiver operating characteristic (ROC) analyses were performed to determine if the combination of any significant variables improved the prediction of case/control status.

Results

Patient and Sample Characteristics

[0095] UC-CRC cases and controls did not have any significant differences with regard to clinical characteristics (Table 9). Similarly, the UC-CRC or UC-control tissue sections selected for analyses were matched for location. However, UC-controls demonstrated significantly higher levels of histologic disease activity as determined by H&E than UC-CRC cases (Table 10).
[00009] [TABLE-US-00009]
  TABLE 9
 
  Characteristics of UC-CRC cases versus UC-controls.
    Cases   Controls  
  Characteristic   (n = 50)   (n = 50)   p-value
 
  Average age, years (range)   49.4 (26-80)   50.8 (28-77)   0.55
  Average duration of UC, years   20.6 (10-49)   20.9 (10-45)   0.86
  (range)
  Extent of UC
  Extensive/pancolitis   82%   68%
  Left-sided   18%   32%   0.11
  Gender
  Male   68%   66%
  Female   32%   34%   0.83
  Ethnicity
  Caucasian   100%    100%    1.00
 
[00010] [TABLE-US-00010]
  TABLE 10
 
  Pathological characteristics of UC-CRC cases versus UC-controls.
    Cases   Controls  
  Non-neoplastic tissue location   (n = 50)   (n = 50)   p-value*
 
  Rectum   21%   26%   0.12
  Sigmoid   32%   19%
  Descending   15%   33%
  Transverse    9%   14%
  Ascending    9%    5%
  Cecum   15%    2%
  Histologic disease activity level
  of non-neoplastic tissue section
  0   66%   41%   0.01
  1   26%   35%
  2    8%   10%
  3    0%   14%
 
  *p-value calculated using chi-square test.

Activity Level Determined Using Cell Surface Markers

[0096] Analysis of all cases and controls indicated detectable staining of all three cell surface markers, demonstrating that current H&E scoring, in general, underestimates cellular infiltrate (FIG. 6). Determination of the possible significance of a given cell surface marker in discriminating between UC-CRC cases and UC-controls is summarized in Table 11. For cases, MPO staining was significantly higher than that of UC-controls regardless of H&E activity level (p<0.0001). There were limited UC-CRC cases with active disease as measured by H&E so to facilitate subgroup analyses, cases and controls were pooled and categorized as either inactive (H&E=0) or active (H&E=1, 2 or 3). The significance of MPO was maintained in subgroup analyses of inactive cases and controls (H&E score of 0, p=0.002) as well as those classified as active (H&E score of 1-3, p=0.02). CD68 staining was slightly elevated in the overall analysis of UC-CRC cases versus controls (p=0.04), but this finding did not persist in subgroup analyses. CD3 staining did not vary between UC-CRC cases and UC-controls. The % area of positive staining for FOXP3, a marker of T-regulatory cells (Treg) involved in suppression of inflammation (Kamikozuru et al., Clin. Exp. Immunol., 156(2):320-327 (2009); Yu et al., Inflamm. Bowel Dis., 13(2):191-199 (2007)), on a subset of cases and controls (n=25 for both) was subsequently investigated to see if this could account for the lack of difference in CD3. Analysis indicated that there was no significant difference (p=0.624, data not shown).
[00011] [TABLE-US-00011]
  TABLE 11
 
  Disease activity level versus cell surface markers in cases and controls.
  Activity   CD68   p-   MPO   p-   CD3   p-
  Level   Cases   Controls   value   Cases   Controls   value   Cases   Controls   value
 
  All   2.668   2.0310.04   6.06   3.41<0.0001   3.58   3.44   0.70
  0   2.76   1.95   0.14   6.44   2.770.002   3.65   2.69   0.07
  1-3   2.49   2.09   0.25   5.74   3.850.02   3.45   3.94   0.43
 
  Values in bold are significant.

MPO Expression Associated with Genetic and Epigenetic Changes
[0097] Increased MPO expression was significantly associated with the presence of the TNF-α-308 G>A SNP (5.95 vs 4.02, p=0.008) (FIG. 7A) as well as RUNX3 methylation (5.90 vs 4.30, p=0.03) (FIG. 7B). It is important to note that the RUNX3 methylation was detected in DNA extracted from non-adjacent, non-neoplastic of UC-CRC cases. Receiver operating characteristic (ROC) curves indicated that an analysis with combined markers was more informative than individual marker assessment (FIG. 7). The area under the curve (AUC) was higher for MPO combined with TNF-α and/or RUNX3. To further interrogate this association, logistic regression was performed. Odds ratios and p-values were either improved or remained highly significant in the presence of other variables (Table 12).
[00012] [TABLE-US-00012]
  TABLE 12
 
  Univariate and Multivariate analyses of variables.
    Odds Ratio   CI*   p-value
   
  Univariate      
  MPO   1.38   1.17, 1.67<0.0001
  RUNX3 methylated   8.07   2.47, 36.580.0003
  TNF-α   7.87   3.18, 21.36<0.0001
  Multivariate
  MPO&TNF-α
  MPO   1.36   1.13, 1.690.0005
  TNF-α   7.15   2.66, 21.06<0.0001
  MPO&RUNX3
  MPO   1.51   1.25, 1.90<0.0001
  RUNX3   15.90   4.20, 81.78<0.0001
  MPO, RUNX3 &TNF-α
  MPO   1.46   1.19, 1.87<0.0001
  RUNX3   14.29   3.46, 80.000.0001
  TNF-α   6.60   2.22, 21.630.0006
 
  *Confidence interval
  P-values in bold are significant.

Example 3

Nucleic Acid Markers and UC-CRC

Patient Selection

[0098] The Mayo Clinic Institutional Review Board approved this work. The UC-CRC cases and UC controls analyzed in this study have been described elsewhere (Garrity-Park et al., Am. J. Gastroenterol., 103(2):407-15 (2008); and Garrity-Park et al., Gut., 58:1226-1233 (2009)). UC-CRC cases were selected from a review of 274 patients identified from the Mayo Clinic centralized diagnostic index of medical records (1976-2006). These patients had inflammatory bowel disease (either Crohn's or UC) and CRC. Patients with Crohn's disease were excluded. Medical records for the remaining UC-CRC patients were reviewed to establish a date of disease onset.
[0099] For each case, pathology slides from the surgical resection also were recalled to confirm the diagnosis of UC and identify the best block for DNA extraction. Patients who did not have UC confirmed by review of the pathology or whose duration of disease was less than 10 years were excluded. After these exclusions, 114 UC-CRC cases were included in the study. Potential UC controls were identified through the Mayo pathology index (1994-2006), which indicated the patient age, gender, and extent of UC as well as the presence of other confounding pathologies such as dysplasia. The final pool of potential UC controls for this study included UC patients who did not develop CRC, who underwent either colectomy or colonoscopy with biopsy at the Mayo Clinic, and who did not have prior dysplasia. The Mayo Clinic centralized diagnostic index of medical records was used with these remaining controls to establish a date of diagnosis. Patients with less than ten years between the date of UC diagnosis and either colectomy or date of last biopsy were excluded as were patients with a prior dysplasia diagnosis. From the remaining list, 181 controls were selected that were most closely matched to the UC-CRC cases with regard to gender, age, ethnicity, duration, and extent of UC. The surgical resection or biopsy specimens from these 181 controls were re-reviewed to histologically confirm the diagnosis of UC. After final review, 114 UC controls were included in this study.

DNA Extraction

[0100] DNA was extracted from formalin-fixed, paraffin-embedded tissues using a modified Gentra (Gentra Systems Inc., Minneapolis, Minn.) protocol, and DNA was suspended in TE (10 mM Tris/0.1 mM EDTA, Integrated DNA Technologies, Coralville, Iowa). Quantification of total DNA was performed using the Picogreen assay (Invitrogen, Portland, Oreg.).

Genotyping

[0101] Samples are interrogated for the presence of additional SNP's in the nucleic acid sequences outlined in Table 13 below. If possible, testing is completed using Taqman genotyping kits (Applied Biosystems; ABI) after optimization for use with formalin fixed, paraffin embedded DNA samples. The 7900HT real-time PCR system is used for evaluating each sample. If a kit is not available for a given SNP, testing is then completed using traditional PCR followed by sequencing as described elsewhere (Garrity-Park et al., Am. J. Gastroenterol., 103(2):407-15 (2008)).

Statistical Analysis

[0102] The ability of a SNP to delineate a case from a control is determined using either a Fisher Exact test or chi-square, as appropriate. Any significant SNP is further interrogated using logistic regression with all other significant SNPs and previously known clinical risk factors (i.e., PSC). Modeling is then performed to determine the best diagnostic paradigm for predicting CRC.
[00013] [TABLE-US-00013]
  TABLE 13
 
  Analysis of nucleic acid polymorphisms in UC-CRC cases vs. UC controls
    SNP(s)    
  Target   Identified   p- value   Context sequence (ABI)
 
  IL-1      
 
  IL-6   −174G>C    
    (1800925)    
 
  IL-6   −6337T>C    
 
  IL-10   −1082G>A     TCCTCTTACCTATCCCTACTTCCCC[T/C]TCCC
    (1800896)     AAAGAAGCCTTAGTAGTGTTG
 
  IL-10   −819C.T     AGTGAGCAAACTGAGGCACAGAGAT[A/G]T
    (1800871)     TACATCACCTGTACAAGGGTACAC
 
  IL-10   −592 C>A     CTTTCCAGAGACTGGCTTCCTACAG[T/G]AC
    (1800872)     AGGCGGGGTCACAGGATGTGTTC
 
  IL-10   −627C>A    
 
  IL-15      
 
  IL-18      
 
  TGFB      
 
  TNF-α   -308G>A    
    (G19)    
 
  TNF-α   −238G>A    
    (673)    
 
  TNF-α   −863C>A    
    (1800630);    
 
  TNF-α   −857C>T    
 
  TNF-α   −301G>A    
 
  TNF-α   −293C>T    
 
  IL-12      
 
  IL-15      
 
  IL-23      
 
  IL-23R   2284C>A   >0.0001   TTTAATTTTAGCCATTCTTCTGCCT[A/C]AT
    (10889677)     TTCTTAAAATTAGAGAATTAAGG
 
  IL-23R   94G>T   =0.02   TTTTCCTTGCTTCCAGACATGAATCA[G/T]GT
    (1884444)     CACTATTCAATGGGATGCAGTAA
 
  IL-23R   1142G>A     ATTGGGATATTTAACAGATCATTCC[A/G]AA
    (11209026)     CTGGGTAGGTTTTTGCAGAATTT
 
  IL-7      
 
  NFKB   DelATTG    
 
  TLR1-10      
 
  IL-8   −251T>A     TTATCTAGAAATAAAAAAGCATACA[A/T]T
    (4073)     TGATAATTCACCAAATTGTGGAGC
 
  IL-8   2767A>T    
 
  IL-8   781C>T     AACTCTAACTCTTTATATAGGAAGT[C/T]G
    (2227306)     TTCAATGTTGTCAGTTATGACTGT
 
  IFNG      
 
  IL-4   −168C>T     TTAGCTTCTCCTGATAAACTAATTG[C/T]CT
    (2070874)     CACATTGTCACTGCAAATCGACA
 
  IL-4   −590C>T    
 
  IL-4   −34C>T    
 
  IL-4   −588C>T     ACACCTAAACTTGGGAGAACATTGT[C/T]C
    (2243250)     CCCAGTGCTGGGGTAGGAGAGTCT
 
  IL-1β   −31T>C   >0.0001   CCAGTTTCTCCCTCGCTGTTTTTAT[G/A]GC
    (1143627)     TTTCAAAAGCAGAAGTAGGAGGC
 
  IL-1β   −571C>T    
 
  IL-1β   3953C>T     CATAAGCCTCGTTATCCCATGTGTC[G/A]A
    (114634)     AGAAGATAGGTTCTGAAATGTGGA
 
  IL-1β   −511 C>T    
    (3087258)    
 
  IL-21      
 
  IL-17   −197G>A     TGCCCTTCCCATTTTCCTTCAGAAG[A/G]A
    (2275913)     GAGATTCTTCTATGACCTCATTGG
 
  TREM1      
 
  MPO      
 
  MIP-1α      
 
  MDR1      
 
  P16      
 
  RUNX3      
 
  COX2      
 
  MINT1      
 
  HPP1      
 
  MINT31      
 
  PPARγ   34C>G    
 
  PPARγ   161C>T    
 
  IL-1RA   86 bp repeat    
    (Intron 2)    
 
  IL-13   2044 G>A     TTAAAGAAACTTTTTCGCGAGGGAC[A/G]GT
    (20541)     TCAACTGAAACTTCGAAAGCATC
 
  IL-13   −1112 C>T     GGTTTCTGGAGGACTTCTAGGAAAA[C/T]GA
    (1800925);     GGGAAGAGCAGGAAAAGGCGACA
 
  IL-13   −1512 A>C    
 
  TLR1   R80T     AACACTGATATCAAGATACTGGATT[C/G]TA
    (5743611)     TTATGAGAAATTATCAAAATCCT
 
  TLR1   I602S    
    (5743618)    
 
  TLR2   R753Q    
    (5743708);    
 
  TLR2   GT repeat    
    (Intron2);    
 
  TLR2   P631H     GCCTGGCTCCAGGCCAAAAGGAAGC[A/C]C
    (5743704)     AGGAAAGCTCCCAGCAGGAACATC
 
  TLR3   N284I     CTCACTATGCTCGATCTTTCCTACA[A/T]CA
    (5743316);     ACTTAAATGTGGTTGGTAACGAT
 
  TLR3   L412F     ACTTGCTCATTCTCCCTTACACATA[T/C]TC
    (3775291);     AACCTAACCAAGAATAAAATCTC
 
  TLR3   908 T>C    
 
  TLR4   D299G     GCATACTTAGACTACTACCTCGATG[A/G]TA
    (4986790)     TTATTGACTTATTTAATTGTTTG
 
  TLR4   T399I     TGTTCTCAAAGTGATTTTGGGACAA[C/T]C
    (4986791)     AGCCTAAAGTATTTAGATCTGAGC
 
  TLR5   R392*    
 
  TLR6   S249P     TTGAGGGTAAAATTCAGTAAGGTTG[A/G]A
    (5743810)     CCTCTGGTGAGTTCTGATAAAAAT
 
  TLR6   −1401 A>G    
    (5743795)    
 
  TLR7   Q11L     TTTCCAATGTGGACACTGAAGAGAC[A/T]A
    (179008);     ATTCTTATCCTTTTTAACATAATC
 
  TLR7   A448V     AGTGAAGTTGGCTTCTGCTCAAATG[C/T]C
    (5743781)     AGAACTTCTGTAGAAAGTTATGAA
 
  TLR7   T801T     GGTTTGTCTGGTGGGTTAACCATAC[A/G]G
    (864058)     AGGTGACTATTCCTTACCTGGCCA
 
  TLR8   M1V     AATGAAAAATTAGAACAACAGAAAC[A/G]TG
    (3764880);     GTAAGCCACTTCTATTTCTTTAG
 
  TLR8   D118D     AATCAAATGGCTTGAATATCACAGA[C/T]G
    (2159377)     GGGCATTCCTCAACCTAAAAAACC
 
  TLR8   L651L     GTCTGGATTTATCCCTTAATAGGCT[C/G]A
    (2407992)     AGCACATCCCAAATGAAGCATTCC
 
  TLR9   1174 G>A     TGTGTGAGTGGCCGGCCCCCAGCTC[C/T]A
    (352139);     CCTCCACCCACTCCACTTCATGGG
 
  TLR9   1635 G>A     AGCTGAGGTCCAGGGCCTCCAGTCG[C/T]G
    (352140);     GTAGCTCCGTGAATGAGTGCTCGT
 
  TLR9   −1237 T>C    
    (5743836)    
 
  TLR10   N241H     AGCAATAGAACCGATGTCTTAGCAT[T/G]TT
    (11096957);     CTAAACTAAGATTTCGTTGCATT
 
  TLR10   I369L     AGAGTTTTCAAGTGAGGCAGTTGGA[T/G]A
    (11096955)     GTTCTTTTAAACAACTCGTCTGTT
 
  TLR10   I473T     GAGATCAGTTAGAAAATTAAATGCA[A/G]TA
    (11466657);     TTTAGTTCTCGTAAGGCCATCAG
 
  TLR10   R525W     AAATTTTTTAATTCACAGGTACACC[A/G]GA
    (11466658)     ATGGATTTCTTCCCGCATTTAGA
 

Example 4

Early Appearance of Nucleic Acid Markers in UC-CRC Pateints

[0103] Biopsies from 10 UC-CRC cases and 10 UC-controls obtained between 10 and 24 months prior to the index date analyzed in the work described in the above Examples were tested for methylation of RUNX3 and MINT1. RUNX3 was more frequently methylated in UC-CRC cases than controls (80% versus 10%). MINT1 was also more frequently methylated in UC-CRC cases than controls (60% versus 0%). These results demonstrate that the methylation changes apparent at the time of CRC (index date) actually occurred prior to overt neoplasm.

Other Embodiments

[0104] It is to be understood that while the invention has been described in conjunction with the detailed description thereof, the foregoing description is intended to illustrate and not limit the scope of the invention, which is defined by the scope of the appended claims. Other aspects, advantages, and modifications are within the scope of the following claims.
(57)

Claim

1. A method for assessing a mammal diagnosed with inflammatory bowel disease for the presence of or an increased risk of developing colorectal cancer, wherein said method comprises determining whether or not said mammal comprises at least two markers from the group consisting of elevated MPO polypeptide levels, elevated RUNX3 methylation status, elevated MINT1 methylation status, and reduced COX-2 methylation status as compared to a normal control, wherein the presence of said at least two markers is indicative of an increased risk of developing colorectal cancer.
2. The method of claim 1, further comprising determining whether or not said mammal comprises at least one polymorphism in a TNF alpha nucleic acid, wherein the presence of said polymorphism is indicative of an increased risk of developing colorectal cancer.
3. The method of claim 1, wherein said inflammatory bowel disease is ulcerative colitis.
4. The method of claim 1, wherein said determining step comprises performing an immunoassay.
5. The method of claim 1, wherein said determining step comprises performing methylation-specific PCR.
6. The method of claim 1, wherein said mammal is a human.
7. A method for assessing a mammal with histologically inactive inflammatory bowel disease for the presence of colorectal cancer or an increased risk of developing colorectal cancer, said method comprising determining whether or not said mammal comprises the presence of at least two markers selected from a group consisting of the presence of at least one polymorphism in a TNF alpha nucleic acid, an elevated MPO polypeptide level, an elevated methylation level of a RUNX3 nucleic acid, an elevated methylation level of a MINT1 nucleic acid, and a reduced methylation level of a COX-2 nucleic acid as compared to a normal control, wherein the presence of said at least two markers is indicative of the presence of colorectal cancer or an increased risk of developing said colorectal cancer.
8. The method of claim 7, wherein said mammal is assessed as having said increased risk of developing colorectal cancer and is categorized as a mammal needing more frequent monitoring than a mammal assessed as not having said increased risk of developing colorectal cancer.
9. A method for assessing a mammal diagnosed with inflammatory bowel disease for the presence of or an increased risk of developing colorectal cancer, wherein said method comprises:
(a) determining the methylation status in a RUNX3 nucleic acid and a COX-2 nucleic acid in said human,
(b) classifying said mammal as having or as having an increased risk of developing said colorectal cancer if said RUNX3 nucleic acid methylation status is elevated and said COX-2 nucleic acid methylation status is reduced as compared to a normal control, and
(c) classifying said mammal as not having or as not having at an increased risk of developing said colorectal cancer if said RUNX3 nucleic acid methylation status is not elevated and said COX-2 nucleic acid methylation status is not reduced.
10. The method of claim 9, wherein said inflammatory bowel disease is histologically inactive.
11. The method of claim 9, further comprising determining the level of an MPO polypeptide in said mammal, wherein an elevated level of said MPO polypeptide is indicative of the presence of or an increased risk of developing colorectal cancer.
12. The method of claim 9, further comprising determining the methylation status of a MINT1 nucleic acid in said mammal, wherein an elevated level of said MINT1 nucleic acid methylation status is indicative of the presence of or an increased risk of developing said colorectal cancer.
13. The method of claim 9, further comprising determining whether or not said mammal contains a polymorphism in a nucleic acid encoding a TNF-alpha protein, wherein the presence of said polymorphism is associated with the presence of or an increased risk of said colorectal cancer.
14. The method of claim 13, wherein said polymorphism is rs1800629.
15. A method for assessing a biopsy sample from an ulcerative colitis patient having the presence of a polymorphism in a TNF-alpha nucleic acid, said method comprising analyzing said biopsy sample for at least two markers selected from the group consisting of an MPO polypeptide level, methylation status of a RUNX3 nucleic acid, methylation status of a MINT1 nucleic acid, and methylation status of a COX-2 nucleic acid.
16. A method for assessing a mammal diagnosed with inflammatory bowel disease for the presence of or an increased risk of developing colorectal cancer, wherein said method comprises:
(a) detecting the presence of at least two markers in said mammal, wherein said at least two markers are selected from the group consisting of elevated MPO polypeptide levels, elevated RUNX3 methylation status, elevated MINT1 methylation status, and reduced COX-2 methylation status, and
(b) classifying said mammal as having colorectal cancer or as being at an increased risk of developing colorectal cancer based at least in part on said presence of said at least two markers.
17. A method for assessing a mammal with histologically inactive inflammatory bowel disease for the presence of colorectal cancer or an increased risk of developing colorectal cancer, where said method comprises:
(a) detecting the presence of at least two markers in said mammal, wherein said at least two markers are selected from a group consisting of the presence of at least one polymorphism in a TNF alpha nucleic acid, an elevated MPO polypeptide level, an elevated methylation level of a RUNX3 nucleic acid, an elevated methylation level of a MINT1 nucleic acid, and a reduced methylation level of a COX-2 nucleic acid, and
(b) classifying said mammal as having colorectal cancer or as being at an increased risk of developing colorectal cancer based at least in part on said presence of said at least two markers.
18. A method for assessing a mammal diagnosed with inflammatory bowel disease for the presence of or an increased risk of developing colorectal cancer, wherein said method comprises:
(a) detecting the presence of an elevated level of RUNX3 nucleic acid methylation and a reduced level of COX-2 nucleic acid methylation in said mammal, and
(b) classifying said mammal as having or as having an increased risk of developing said colorectal cancer based at least in part on said presence.
*****

Sign in to the Lens

Feedback