(19) United States
`(12) Patent Application Publication (10) Pub. No.: US 2012/0164238A1
`J00 st
`(43) Pub. Date:
`Jun. 28, 2012
`
`US 2012O164238A1
`
`(54) METHODS OF DETECTING COLORECTAL
`CANCER
`
`(75) Inventor:
`
`Louwagie Joost, Dornach (CH)
`
`(73) Assignee:
`
`MDXHEALTH SA
`
`(21) Appl. No.:
`
`13/147,570
`
`(22) PCT Filed:
`
`Feb. 3, 2010
`
`(86). PCT No.:
`
`PCT/GB2O1 O/OOO18O
`
`S371 (c)(1),
`(2), (4) Date:
`
`Mar. 12, 2012
`
`Related U.S. Application Data
`(62) Division of application No. 61/149,581, filed on Feb.
`3, 2009.
`
`Publication Classification
`
`(51) Int. Cl.
`(2006.01)
`A633/36
`(2006.01)
`A6IP35/00
`(2006.01)
`CI2O I/68
`(52) U.S. Cl. ........................ 424/620; 435/6.1 1; 435/6.12
`(57)
`ABSTRACT
`A method of detecting a predisposition to, or the incidence of
`colorectal cancer in a faecal sample comprises, in a first step
`(a), detecting the presence of blood in the faecal sample,
`wherein detection of the presence of blood is indicative of a
`predisposition to, or the incidence of colorectal cancer. The
`method additionally comprises, in second step (b), detecting
`an epi-genetic modification in the DNA contained within the
`faecal sample, wherein detection of the epigenetic modifica
`tion is indicative of a predisposition to, or the incidence of
`colorectal cancer. Based upon a positive result obtained in
`either (a) or (b) or in both (a) and (b) a predisposition to, or the
`incidence of colorectal cancer is detected. Related methods
`and kits involve detecting an epigenetic modification in a
`number of specific genes.
`
`Geneoscopy Exhibit 1018, Page 1
`
`

`

`US 2012/0164238 A1
`
`Jun. 28, 2012
`
`METHODS OF DETECTING COLORECTAL
`CANCER
`
`FIELD OF THE INVENTION
`0001. The present invention is concerned with the diagno
`sis, staging and treatment of disease, in particular cancer and
`more specifically colorectal cancer. The invention relates to
`methods and kits for diagnosing colorectal cancer based upon
`detecting epigenetic modifications, typically in specific
`genes. The methods and kits may also permit the detection of
`blood in a fecal sample, with the combined tests proving
`particularly advantageous.
`
`BACKGROUND OF THE INVENTION
`0002 Colorectal cancer (CRC) is a leading cause of can
`cer-related deaths worldwide, and is the second leading cause
`of cancer-related deaths in the United States. A patient's
`prognosis is good if the cancer is caught early, when the site
`of the cancer is confined to its site of origin. However, the cure
`rates fall once the cancer has spread. Most colon cancers arise
`from conventional adenomatous polyps (conventional
`adenoma-to-carcinoma sequence), while some colon cancers
`appear to arise from the recently recognized serrated
`adenomatous polyp (Serrated adenoma-to-carcinoma
`theory). Because conventional adenomas and Serrated
`adenomas are usually asymptomatic, mass screening of
`asymptomatic patients has become the cornerstone for detect
`ing and eliminating these precursor lesions to reduce the risk
`of colon cancer.
`0003) A number of different screening methods for CRC
`are available. Procedures such as digital rectal examination
`(DRE), colonoscopy or sigmoidoscopy are highly invasive,
`painful and can cause a great deal of patient discomfort. Other
`less invasive screening tests include fecal occult blood test
`(FOBT); fecal immunochemical test (FIT); barium enema
`with air contrast, virtual colonoscopy; biopsy (e.g., CT
`guided needle biopsy); and imaging techniques (e.g., ultra
`sound, CT scan, PET scan, and MRI).
`0004 Colonoscopy has become the primary screening test
`for CRC because of its high sensitivity and specificity, and the
`ability to perform polypectomy. While sensitive and specific,
`the procedure is invasive, costly, has limited availability and
`includes certain risks Such as induction of infection and per
`foration of the bowel.
`0005. A commonly used and less expensive way of screen
`ing for CRC is a fecal occult blood test (FOBT), which tests
`for the presence of blood in faeces. The presence of haemo
`globin as a representative blood protein in faeces is an indi
`cator of intestinal bleeding, which is frequently associated
`with CRC. However, since occult in a fecal sample could be
`indicative of a variety of gastrointestinal disorders, further
`medical testing Such as colonoscopy remains necessary to
`identify colorectal cancer.
`0006 Fecal occult blood tests fall primarily into two cat
`egories, tests based on the use of chromogenic chemical
`reagents such as gum guaiac and immunochemical tests. The
`chemically based guaiac methods determine the presence of
`occult blood by the detection of the perioxidase activity of the
`hemoglobin in the blood present in the faecal sample. They
`require catalysis of peroxide into oxygen and water, and the
`Subsequent oxidation of a colorless dye (most often into a
`colored form). However, peroxidase activity is also found in
`meats and vegetables. In order to produce accurate results,
`
`these tests require restriction of the intake of certain foods,
`drugs, vitamins, and other Substances prior to and during the
`sample collection period. The sensitivity of the most com
`monly used guaiac FOBT (Hemoccult) is approximately
`50%. Despite a specificity of 98%, the positive predictive
`value for FOBT is low. Methods of detecting occult blood
`based on porphyrin (heme and protpoporphyrin IX) analysis
`or immunologic tests using anti-hemoglobin antibodies
`improve on these results. Immunochemical tests (FIT or
`iFOBT) that use anti-hemoglobin antibodies specific for
`human blood in extracts from stool do not require dietary
`restrictions; however, they are more complicated and more
`expensive than peroxidase-based tests. In addition, human
`hemoglobin in fecal samples degrades with time, resulting in
`a loss of antigenicity which can produce false negative
`results. Reported sensitivity of these immunologic tests var
`ies widely but is typically 60-80% depending on the popula
`tion tested. Specificity is estimated to be -98%. Because of
`the intermittent nature of colorectal bleeding, the sensitivity
`of FOBT and FIT is directly proportional to the number of
`samples taken and the frequency of testing.
`0007 Recent developments in testing look specifically for
`mutations in DNA characteristic of colorectal neoplasia that
`are detectable in exfoliated epithelial cells in the stool (Pi
`gnone, et al., 2002; Ahlquist, et al., 2002). While neoplastic
`bleeding is intermittent, epithelial shedding is continual,
`potentially making stool-based DNA testing (i.e., also known
`as fecal DNA f-DNA and stool DNA sDNA) testing more
`sensitive than other methods. Early studies of molecular
`feacal screening primarily focused on single mutations. Gene
`mutations in P53, K-ras, and BAT 26, for instance, have been
`linked to colorectal cancer and remain detectable in feacal
`samples. Colorectal neoplasms are varied in nature and no
`single mutation has been identified as being expressed uni
`versally. For this reason, multiple target assay panels (MTAP)
`are preferably used. PreCien-PlusTM (EXACT Sciences Cor
`poration, Maynard, Mass.; Laboratory Corporation of
`America, Burlington, N.C.) is a single test that identifies the
`presence of 23 different microsatellite (MSI) mutations
`known to be associated with CRC, including mutations in
`BAT-26. Additionally, 21 other point mutations in other genes
`associated with CRC are included in this test: APC, K-ras, and
`p53. This test is further designed to detect long DNA frag
`ments, which have been specifically associated with cells
`called non-apoptotic colonocytes, which are common in
`CRC. While this test is more sensitive than fecal occult blood
`testing, it is not as sensitive as colonoscopy and will miss
`about half of cancers in an average risk group of people
`without symptoms.
`0008 Increased DNA methylation is an epigenetic alter
`ation that is common inhuman cancers and is often associated
`with transcriptional silencing. Aberrantly methylated DNA
`has also been proposed as a potential tumor marker for CRC
`detection. Genes such as Vimentin, which are transcription
`ally silent in normal epithelium, have been considered as
`targets for cancer-associated aberrant methylation and for use
`as cancer markers (JNCI Journal of the National Cancer Insti
`tute 2005 97(15): 1124-1132). A combined assay utilizing
`hypermethylated vimentin gene (hV) and a two site DNA
`integrity assay (DY), demonstrated a sensitivity of 88% for
`CRC with a specificity of 82% (Am J Gastroenterol. 2008
`November; 103(11):2862-70). Further, ColoSure R is a single
`marker laboratory developed, stool based DNA test. This
`method examines DNA in exfoliated colon cells for cancer
`
`Geneoscopy Exhibit 1018, Page 2
`
`

`

`US 2012/0164238 A1
`
`Jun. 28, 2012
`
`associated aberrant methylation of the vimentin gene and
`reaches a sensitivity range of 72-77% and a specificity range
`of 83-94% in average risk individuals.
`0009 Protein tests provide an alternative method for
`detecting CRC. Tests assessing the presence of tumor-derived
`enzymes such as M2 pyruvate kinase (M2-PK), and/or pro
`teins such as calprotectin, carcinoembryonic antigen (CEA),
`tissue inhibitor of metalloproteinase-1 (TIMP-1) and S100
`calcium binding protein A12 (S100A12) have been
`described. A diagnosis of colorectal cancer using a combina
`tion of fecal occult blood and novel fecal protein markers
`S100A12 and TIMP-1 has been described in Clin Gastroen
`terol Hepatol. 2008 October; 6(10): 1122-8. Dimeric isoen
`Zyme of pyruvate kinase, M2-PK, expressed by tumor cells,
`has as well been proposed as a screening tool for CRC. The
`performance of fecal M2-PK has been evaluated with IFOBT
`and colonoscopy in Am J Gastroenterol. 2008 June; 103(6):
`1496-504. Compared to immunochemical FOBTs, TuM2-PK
`does not have supplemental value for screening for CRC
`because of a lower sensitivity and specificity (Eur J. Gastro
`enterol Hepatol. 2007 October 19(10):878-82)
`0010 Although combined assays for detecting CRC have
`been described, their approach targets either multiple protein
`markers or either multiple DNA alterations. To date, immu
`nochemical tests and DNA tests for CRC detection have been
`evaluated and compared on a separate basis only.
`0011 EP0308227 describes a chemical fecal occult blood
`test employing a guaiac matrix.
`0012 EP0032782 describes a method for the detection of
`haemoglobin or decomposition products of haemoglobin in
`feces by means of an immunological reaction by using an
`antibody specific for human haemoglobin.
`0013 U.S. Pat. No. 7,288,413 describes methods that
`combine a chemical fecal occult blood test and an immu
`nochemical fecal occult blood test.
`0014 WO 04/092709 concerns a fecal blood test involv
`ing the dispersement of a dye in toilet water.
`00.15 EP0817968 describes several suitable stool collect
`ing and testing methods and devices.
`0016 WO 05/017207 discloses that the vimentingene can
`be a common target for methylation and epigenetic gene
`silencing in colon neoplasia, and may function as a candidate
`tumor Suppressor gene.
`0017 WO 2008/084219 relates to detection of colorectal
`cancer based upon determining methylation of a number of
`different genes, including panels of genes.
`0018 WO 2006/113671 and WO 2008/010975 describe
`methylation markers relevant to colorectal cancer.
`
`SUMMARY OF THE INVENTION
`0019. The invention provides a method of detecting a pre
`disposition to, or the incidence of colorectal cancer in a
`faecal sample comprising:
`0020 (a) detecting the presence of blood in the faecal
`sample, wherein detection of the presence of blood is
`indicative of a predisposition to, or the incidence of
`colorectal cancer,
`0021 (b) detecting an epigenetic modification in the
`DNA contained within the faecal sample, wherein detec
`tion of the epigenetic modification is indicative of a
`predisposition to, or the incidence of colorectal cancer
`and based upon a positive result obtained in either (a) or (b) or
`in both (a) and (b) detecting a predisposition to, or the inci
`dence of colorectal cancer.
`
`0022. Also described herein is a method of sample pro
`cessing, prior to carrying out a method of the invention,
`comprising removing a portion of a collected faecal sample
`and adding the removed portion of the sample to a buffer
`which prevents denaturation or degradation of blood proteins
`found in the sample.
`0023 The invention also provides a method of detecting a
`predisposition to, or the incidence of colorectal cancer in a
`sample comprising detecting an epigenetic modification in a
`panel of at least two genes selected from PHACTR3, NDRG4
`and FOXE1, wherein detection of the epigenetic modification
`in at least one of the genes in the panel is indicative of a
`predisposition to, or the incidence of colorectal cancer.
`0024. The invention also provides a method of detecting a
`predisposition to, or the incidence of cancer (and in particu
`lar colorectal cancer) in a sample comprising detecting an
`epigenetic modification in at least one gene selected from
`LAMA1 and CDO1, wherein detection of the epigenetic
`modification in the at least one gene is indicative of a predis
`position to, or the incidence of cancer (and in particular
`colorectal cancer).
`0025. The invention also relates to a method of detecting a
`predisposition to, or the incidence of colorectal cancer (in
`particular in a faecal sample) comprising detecting an epige
`netic modification in at least one gene selected from GPNMB
`and MMP2, wherein detection of the epigenetic modification
`in the at least one gene is indicative of a predisposition to, or
`the incidence of colorectal cancer.
`In related aspects, the invention provides
`0026
`0027 a method for predicting the likelihood of success
`ful treatment of colorectal cancer with a DNA demethy
`lating agent and/or a DNA methyltransferase inhibitor
`and/or HDAC inhibitor comprising detecting an epige
`netic modification in:
`(a) a panel of at least two genes selected from PHACTR3,
`NDRG4 and FOXE1,
`(b) at least one gene selected from LAMA1 and CDO1; or
`(c) at least one gene selected from GPNMB and MMP2 (in a
`faecal sample)
`wherein detection of the epigenetic modification in at least
`one of the genes in the panel or in the at least one gene is
`indicative that the likelihood of successful treatment is higher
`than if the epigenetic modification is not detected.
`0028 a method for predicting the likelihood of resis
`tance to treatment of colorectal cancer with a DNA dem
`ethylating agent and/or DNA methyltransferase inhibi
`tor and/or HDAC inhibitor comprising detecting an
`epigenetic modification in
`(a) a panel of at least two genes selected from PHACTR3,
`NDRG4 and FOXE1,
`(b) at least one gene selected from LAMA1 and CDO1; or
`(c) at least one gene selected from GPNMB and MMP2 (in a
`faecal sample)
`wherein detection of the epigenetic modification in at least
`one of the genes in the panel or in the at least one gene is
`indicative that the likelihood of resistance to treatment is
`lower than if the epigenetic modification is not detected.
`0029 a method of selecting a suitable treatment regi
`men for colorectal cancer comprising detecting an epi
`genetic modification in
`
`Geneoscopy Exhibit 1018, Page 3
`
`

`

`US 2012/0164238 A1
`
`Jun. 28, 2012
`
`(a) a panel of at least two genes selected from PHACTR3,
`NDRG4 and FOXE1,
`(b) at least one gene selected from LAMA1 and CDO1; or
`(c) at least one gene selected from GPNMB and MMP2 (in a
`faecal sample)
`wherein detection of the epigenetic modification in at least
`one of the genes in the panel or in the at least one gene results
`in selection of a DNA demethylating agent and/or a DNA
`methyltransferase inhibitor and/or a HDAC inhibitor for
`treatment and wherein if the epigenetic modification is not
`detected, a DNA demethylating agent and/or a DNA methyl
`transferase inhibitor and/or a HDAC inhibitor is not selected
`for treatment.
`0030 a method for monitoring treatment of colorectal
`cancer with a DNA demethylating agent and/or a DNA
`methyltransferase inhibitor and/or HDAC inhibitor
`comprising detecting an epigenetic modification in
`(a) a panel of at least two genes selected from PHACTR3,
`NDRG4 and FOXE1,
`(b) at least one gene selected from LAMA1 and CDO1; or
`(c) at least one gene selected from GPNMB and MMP2 (in a
`faecal sample)
`wherein detection of a reduction in the epigenetic modifica
`tion in at least one of the genes in the panel or in the at least
`one gene as treatment progresses is indicative of Successful
`treatment. Thus, the epigenetic modification may be mea
`sured at the start of the treatment and then once or more
`following treatment, or as treatment progresses, in order to
`determine if the treatment is achieving the desired effect. A
`return to lower levels of methylation of the genes is consid
`ered indicative of effective treatment.
`0031. The invention also relates to a kit for detecting a
`predisposition to, or the incidence of colorectal cancer in a
`faecal sample comprising:
`0032 (a) means for detecting an epigenetic modifica
`tion in the DNA contained within the faecal sample,
`wherein detection of the epigenetic modification is
`indicative of a predisposition to, or the incidence of
`colorectal cancer, and
`0033 (b) means for detecting the presence of blood in
`the faecal sample, wherein detection of the presence of
`blood is indicative of a predisposition to, or the inci
`dence of colorectal cancer.
`0034. Also provided is a kit for any of
`(a) detecting a predisposition to, or the incidence of colorec
`tal cancer in a sample
`(b) monitoring treatment of colorectal cancer with a DNA
`demethylating agent and/or a DNA methyltransferase inhibi
`tor and/or HDAC inhibitor
`(c) predicting the likelihood of successful treatment of col
`orectal cancer with a DNA demethylating agent and/or a
`DNA methyltransferase inhibitor and/or HDAC inhibitor
`(d) predicting the likelihood of resistance to treatment of
`colorectal cancer with a DNA demethylating agent and/or
`DNA methyltransferase inhibitor and/or HDAC inhibitor; or
`(e) selecting a suitable treatment regimen for colorectal can
`cer comprising means for detecting an epigenetic modifica
`tion in a panel of at least two genes selected from PHACTR3,
`NDRG4 and FOXE1.
`0035) Similarly, the invention also provides a kit for any
`of:
`(a) detecting a predisposition to, or the incidence of colorec
`tal cancer in a sample
`
`(b) predicting the likelihood of successful treatment of col
`orectal cancer with a DNA demethylating agent and/or a
`DNA methyltransferase inhibitor and/or HDAC inhibitor
`(c) predicting the likelihood of resistance to treatment of
`colorectal cancer with a DNA demethylating agent and/or
`DNA methyltransferase inhibitor and/or HDAC inhibitor; or
`(d) selecting a suitable treatment regimen for colorectal can
`cer comprising means for detecting an epigenetic modifica
`tion in at least one gene selected from LAMA1 and CDO 1.
`0036. The invention also provides a kit for any of
`(a) detecting a predisposition to, or the incidence of colorec
`tal cancer in a sample
`(b) predicting the likelihood of successful treatment of col
`orectal cancer with a DNA demethylating agent and/or a
`DNA methyltransferase inhibitor and/or HDAC inhibitor
`(c) predicting the likelihood of resistance to treatment of
`colorectal cancer with a DNA demethylating agent and/or
`DNA methyltransferase inhibitor and/or HDAC inhibitor; or
`(d) selecting a suitable treatment regimen for colorectal can
`cer comprising means for detecting an epigenetic modifica
`tion in at least one gene selected from GPNMB and MMP2
`and means for processing a faecal sample.
`0037. The invention also provides a method of detecting a
`predisposition to, or the incidence of colorectal cancer in a
`faecal sample comprising detecting an epigenetic modifica
`tion in the DNA contained within the faecal sample, wherein
`detection of the epigenetic modification is indicative of a
`predisposition to, or the incidence of colorectal cancer, char
`acterised in that the faecal sample has previously been stored
`for at least approximately 6 months, 1, 2, 3, 4, 5, 6 or more
`years and/or is less than approximately 4, 3, 2, or 1 g in
`weight.
`
`DETAILED DESCRIPTION OF THE INVENTION
`0038. The invention, as set out in the claims, is based upon
`Successful attempts to improve the detection of colorectal
`cancer. In particular, the invention aims to improve the posi
`tive and negative predictive value and also the sensitivity and
`specificity of detection of colorectal cancer through non
`invasive means. The methods of the invention may permit
`effective detection of colorectal cancer without the require
`ment for relatively expensive, highly invasive and painful
`procedures such as digital rectal examination, colonoscopy
`and sigmoidoscopy to be performed. The invention is based
`upon a combination of tests for detecting proteins and epige
`netic modification markers respectively in the same faecal
`sample, shown for the first time herein to provide a particu
`larly useful overall test.
`0039 Thus, according to a first aspect, the invention pro
`vides a method of detecting a predisposition to, or the inci
`dence of colorectal cancer in a faecal sample comprising,
`consisting essentially of or consisting of
`0040 (a) detecting the presence of blood in the faecal
`sample, wherein detection of the presence of blood is
`indicative of a predisposition to, or the incidence of
`colorectal cancer,
`0041 (b) detecting an epigenetic modification in the
`DNA contained within the faecal sample, wherein detec
`tion of the epigenetic modification is indicative of a
`predisposition to, or the incidence of colorectal cancer
`and based upon a positive result obtained in either (a) or (b) or
`in both (a) and (b) detecting a predisposition to, or the inci
`dence of colorectal cancer.
`
`Geneoscopy Exhibit 1018, Page 4
`
`

`

`US 2012/0164238 A1
`
`Jun. 28, 2012
`
`0042. As shown herein, the combination of methylation
`marker assay and fecal occult blood test (FOBT) gives very
`specific and sensitive results.
`0043. The combined methods of the invention improve the
`negative predictive value of the existing single tests. By
`improving sensitivity, the number of false negative results is
`decreased and this improves negative predictive value.
`0044 Step (a) of the methods involves detecting the pres
`ence of blood in the faecal sample, wherein detection of the
`presence of blood is indicative of a predisposition to, or the
`incidence of colorectal cancer. Blood in the faeces is an
`indicator of intestinal bleeding, which is frequently associ
`ated with colorectal cancer. Thus, detection of blood in the
`faecal sample is considered a "positive' result in step (a). Any
`suitable method for detecting the presence of blood in the
`sample may be employed. Often, the methods of detecting
`blood will rely upon detecting a representative blood protein
`in the faecal sample. In certain embodiments, detecting the
`presence of blood in the faecal sample comprises, consists
`essentially of or consists of detection of haemoglobin in the
`faecal sample. Detection may be through any Suitable means,
`and includes all variants of fecal occult blood tests. The test
`may be chromogenic or immunological in certain embodi
`ments. The test may rely upon peroxidase activity of hemo
`globin. Chromogenic tests are well known and commercially
`available and may rely upon chemical reagents such as gum
`guaiac. In specific embodiments, haemoglobin in the faecal
`sample is detected through immunochemical means. This
`may involve anti-hemoglobin antibodies in certain embodi
`ments. The term “antibody' or “antibodies' herein refers to
`an antibody or antibodies, or a derivative thereofthat retains
`specific binding activity. By specific binding activity is meant
`the ability to specifically bind to hemoglobin. Thus, such a
`reagent does not bind, or does not bind to a significant degree,
`to unrelated proteins found in the faecal sample. Any anti
`body or derivative may be employed. Thus, the antibody may
`be a monoclonal or polyclonal antibody. The derivative of the
`antibody that retains specific binding activity may comprise,
`consist essentially of or consist of a humanized version of a
`non-human antibody, a heavy chainantibody, a single domain
`antibody, a nanobody, a Fab fragment or Schv etc. in certain
`embodiments. Numerous techniques are available for pro
`ducing antibodies and their derivatized forms, as would be
`well known to one skilled in the art.
`0045. As mentioned above, the combination of techniques
`maximises sensitivity of detection, without significantly
`compromising specificity. Thus, the threshold detection con
`centrations for detection of blood/hemoglobin in step (a) may
`be those typically employed in fecal occult blood tests. Add
`ing in the step (b) test improves overall sensitivity by picking
`up additional positive samples. For example, in some
`embodiments, the result in step (a) is considered positive if
`the concentration of hemoglobin detected is more than
`between (about) 50 to (about) 150 ng/ml. In more specific
`embodiments, the result in step (a) is considered positive if
`the concentration of hemoglobin detected is more than
`(about) 100 ng/ml.
`0046. However, in other embodiments, the methods of the
`invention may be employed to improve the sensitivity of the
`step (a) method, whilst preventing a resultant loss in speci
`ficity. By lowering the threshold concentration of blood to be
`detected in the faecal sample to give a positive result in step
`(a), the sensitivity of the step (a) method is increased. In order
`to retain specificity, the step (b) method is employed on those
`
`samples in which low levels, that is to say lower than the
`typically used threshold, of blood were detected in step (a). A
`positive result from the step (b) method is required to confirm
`the positive result in step (a) for the “low level samples. For
`those samples having blood (especially hemaglobin) concen
`trations above the typically employed threshold in step (a), it
`is not necessary to perform the method of step (b), since for
`these samples the step (a) method is sufficiently specific for
`this not to be necessary. This has the advantage that the step
`(b) test is not required for all samples, thus reducing costs and
`increasing throughput. Thus, in certain embodiments, the
`result in step (a) is considered positive if the concentration of
`hemoglobin detected is lower than is typically employed as
`the threshold concentration of hemoglobin in hemoglobin
`detection tests, but for those samples in which a “lower than
`typical threshold’ concentration of hemoglobin is detected,
`step (b) is performed on these samples. The detection of the
`epigenetic modification in step (b) is then used to confirm the
`positive result in step (a). The step (b) test is not employed for
`those samples in which the concentration of hemoglobin
`detected is higher than the threshold typically employed in
`hemoglobin detection tests.
`0047. In specific embodiments, the result in step (a) is
`considered positive if the concentration of hemoglobin
`detected is more than or at least (about) 5 to (about) 50 ng/ml,
`more specifically more than or at least (about) 5 to (about) 20
`ng/ml and more particularly more than or at least (about) 10
`ng/ml. By lowering the threshold, the sensitivity of the test is
`increased. In Such embodiments, step (b) is performed only in
`the event that the concentration of hemoglobin detected is
`between (about) 5 ng/ml and (about) 250 ng/ml, more spe
`cifically between (about) 10 ng/ml and (about) 200 ng/ml.
`The detection of the epigenetic modification in step (b) is then
`used to confirm the positive result in step (a). Thus, a positive
`result in step (b) confirms the result in step (a) as positive. If
`no epigenetic modification of the DNA is detected, the result
`of step (a) is considered negative. For samples in which the
`concentration of hemoglobin detected is more than or at least
`(about) 200 ng/ml (or (about) 250 ng/ml), it is not necessary
`to perform step (b), since the result in step (a) will be of high
`specificity (i.e. is unlikely to be a false positive).
`0048 Step (b) involves detecting an epigenetic modifica
`tion in the DNA contained within the faecal sample, wherein
`detection of the epigenetic modification is indicative of a
`predisposition to, or the incidence of colorectal cancer. Thus,
`detection of the epigenetic modification is considered a "posi
`tive' result in step (b).
`0049. In some embodiments, the epigenetic modification
`is detected in at least one gene selected from PHACTR3,
`NDRG4, FOXE1, GATA4, GPNMB, TFPI2, SOX17,
`SYNE1, LAMA1, MMP2, OSMR, SFRP2 and CDO1, with
`detection of the epigenetic modification in at least one of the
`genes providing an indication of a predisposition to, or inci
`dence of colorectal cancer.
`0050. In certain embodiments, the epigenetic modification
`is detected in at least one gene selected from PHACTR3,
`NDRG4 and FOXE1, with detection of the epigenetic modi
`fication in at least one of the genes providing an indication of
`a predisposition to, or incidence of colorectal cancer.
`0051 PHACTR3 is the gene symbol approved by the
`HUGO Gene Nomenclature Committee. The gene is located
`on chromosome 20 (location 20q13.32-q13.33) and the gene
`
`Geneoscopy Exhibit 1018, Page 5
`
`

`

`US 2012/0164238 A1
`
`Jun. 28, 2012
`
`sequence is listed under the accession numbers AJ311 122 and
`NM 080672. The gene encodes the phosphatase and actin
`regulator 3.
`0052. NDRG4 is the gene symbol approved by the HUGO
`Gene Nomenclature Committee. The gene is located on chro
`mosome 16 (location q21-q22.3) and the gene sequence is
`listed under the accession number AB044947. The gene
`encodes NDRG family member 4.
`0053 FOXE1 is the gene symbol approved by the HUGO
`Gene Nomenclature Committee. The gene is located on chro
`mosome 9 (location 9q22) and the gene sequence is listed
`under the accession number U89995. The gene encodes the
`forkhead box E1 (thyroid transcription factor 2).
`0054 GATA4 is the gene symbol approved by the HUGO
`Gene Nomenclature Committee. The gene is located on chro
`mosome 8 (location 8p23.1-p22) and the gene sequence is
`listed under the accession numbers AK097060 and
`NM 002052. The gene encodes the GATA binding protein 4.
`0055 GPNMB is the gene symbol approved by the HUGO
`Gene Nomenclature Committee. The gene is located on chro
`mosome 7 (location 7p) and the gene sequence is listed under
`the accession numbers X76534 and NM 001005340. The
`gene encodes the glycoprotein (transmembrane) nmb.
`0056 TFPI2 is the gene symbol approved by the HUGO
`Gene Nomenclature Committee. The gene is located on chro
`mosome 7 (location 7q) and the gene sequence is listed under
`the accession numbers L27624 and NM 006528. The gene
`encodes tissue factor pathway inhibitor 2.
`0057. SOX17 is the gene symbol approved by the HUGO
`Gene Nomenclature Committee. The gene is located on chro
`mosome 8 (location 8q11.23) and the gene sequence is listed
`under the accession number AB073988. The gene encodes
`SRY (sex determining region Y)-box 17.
`0058 SYNE1 is the gene symbol approved by the HUGO
`Gene Nomenclature Committee. The gene is located on chro
`mosome 6 (location 6q24.2-q25.3) and the gene sequence is
`listed under the accession number AB018339. The gene
`encodes spectrin repeat containing, nuclear envelope 1.
`0059 LAMA1 is the gene symbol approved by the HUGO
`Gene Nomenclature Committee. The gene is located on chro
`mosome 18 (location 18p11.3) and the gene sequence is listed
`under the accession numbers X58531 and NM 00:5559. The
`gene encodes laminin, alpha 1.
`0060 MMP2 is the gene symbol approved by the HUGO
`Gene Nomenclature Committee. The gene is located on chro
`mosome 16 (location 16q13-q21) and the gene sequence is
`listed under the accession number NM 001 127891. The
`gene encodes matrix metallopeptidase 2 (gelatinase A, 72
`kDa gelatinase, 72 kDa type IV collagenase).
`0061 OSMR is the gene symbol approved by the HUGO
`Gene Nomenclature Committee. The gene is located on chro
`mosome 5 (location 5p13.2) and the gene sequence is listed
`under the accession number U60805 and NM 003999. The
`gene encodes the oncostatin M receptor.
`0062 SFRP2 is the gene symbol approved by the HUGO
`Gene Nomenclature Committee. The gene is located on chro
`mosome 4 (location 4q31.3) and the gene sequence is listed
`under the accession number AF017986. The gene encodes the
`secreted frizzl