WO2013092960A1 - Methods and kits for the diagnosis of colorectal cancer - Google Patents

Abstract

The invention relates to methods for the diagnosis of colorectal cancer based on the determination of the expression levels of a set of genes, the altered expression of which in relation to a reference value allows diagnosing said cancer with a high reliability.

Description

METHODS AND KITS FOR THE DIAGNOSIS OF COLORECTAL CANCER

TECHNICAL FIELD OF THE INVENTION The invention is within the field of diagnosis, more specifically, in the field of diagnosis of colorectal cancer by means of using panels of genes the expression of which is altered in CRC samples compared to samples from a healthy subject.

BACKGROUND OF THE FNVENTION

Colorectal cancer (CRC) is one of the most frequent neoplasias in the western world, it is the third cause of death in men, after lung cancer and prostate cancer and it is the second in frequency among women, after breast cancer. In Spain, approximately 25,000 new cases are diagnosed every year, which constitutes an incidence of 50 new cases for every 100,000 inhabitants and year (www.aecc.org). However, in a European context, the incidence in Spain can be considered as medium-low, in part due to the Mediterranean diet (FFNI, L., et al., 2007). Several studies have shown that diets with a high caloric dose and high cholesterol levels are a risk factor for the development of colorectal cancer (BOYLE, P., et al, 2000, BMJ 321: 805-808; KEY, T. J., et al, 2002, Lancet 360: 861-868.; FERNANDEZ, E., et al, 2006, J Br Menopause Soc 12 (4): 139- 142.), specially the intake of fats, proteins and refined carbohydrates, as well as the deficiency in fiber.

Traditionally, the diagnosis of CRC is carried out by means of the detection of blood concealed in feces, flexible sigmoidoscopy and colonoscopy. However, these methods are carried out once the cancer has already begun to invade tissues, due to the symptoms that it causes (changes of intestinal habits or rectal bleeding, among others). This drawback can be solved by means of using molecular markers, i.e., molecules the expression of which is altered in tumor tissue compared to healthy tissue or which show mutations in tumor tissue with respect to the healthy tissue. Different molecular CRC diagnostic markers such as genes APC, K-RAS and TP53 have been described up until now. However, given the diversity of types of tumors, it is unlikely that the use of a single marker can be suitable for the diagnosis of all the possible types of tumors. For this reason, several CRC diagnosis methods based on the determination of the expression levels of a set of genes have been proposed. Thus, WO07032631 describes a method for the diagnosis of a colorectal tumor based on the determination of alterations in the expression levels of genes CTHRC1, CANP and KIAA0101. WO06015742 describes a method for the diagnosis of a colorectal carcinoma based on the determination of the expression levels of a panel of 120 genes. WO05044990 describes a method for the diagnosis of a colorectal carcinoma based on the determination of the expression levels of a panel of 93 genes. EP1439393 describes a method for the diagnosis of colorectal carcinoma based on the determination of the expression levels of TIMP1. WO03097872 describes a series of genes the expression levels of which are altered in colorectal carcinoma samples as well as methods for the diagnosis of colorectal carcinoma based on the determination of the expression levels of said genes. WO04001072 describes methods for the diagnosis of colorectal cancer based on the determination of the expression levels of 52 genes the expression of which is high in colorectal carcinomas and in 376 genes the expression of which is repressed in colorectal carcinomas with respect to normal colorectal epithelium. This document describes an array comprising probes suitable for the detection of 4 of the 50 genes the expression of which increases in tumors and 24 of the 50 genes the expression of which increases in tumors. EP1355151 describes a method for the diagnosis of colorectal cancer based on the determination of the expression levels of a set of 39 genes.

Nevertheless, despite the research carried out in this topic, today there are very few tumor markers which are useful from the clinical point of view both for the diagnosis of CRC. Therefore, there is a need in the art for markers or panels of markers which allow the diagnosis of CRC with a high reliability.

SUMMARY OF THE INVENTION

In a first aspect, the invention relates to a method for diagnosing colorectal cancer comprising determining the level of the expression of one or more genes selected from the group consisting of ENC1, MADCAM1, CMTM7, FAM60A, AC ATI and DDX55 in a biofluid of a subject, wherein a change in the level of expression of said genes compared to the level in a reference value is indicative of colorectal cancer. In a second aspect, the invention relates to a method for monitoring the effect of a therapy in a patient suffering from colorectal cancer and being treated with said therapy, comprising determining the levels of the expression of one or more genes selected from the group consisting of ENC1, MADCAM1, CMTM7, FAM60A, AC ATI and DDX55 in a biofluid of said patient, wherein a change of the level of expression of said genes compared to the level in a reference value indicates a positive effect of said therapy.

In a third aspect, invention relates to a kit comprising a set of reagents which allow determining the expression levels of genes one or more genes selected from the group consisting of ENC1, MADCAM1, CMTM7, FAM60A, AC ATI and DDX55.

DETAILED DESCRIPTION OF THE INVENTION Diagnostic method The authors of the present invention, using real-time PCR have identified a series of genes the expression levels of which in blood samples allowed determining if a subject suffers from colorectal cancer with an accuracy rate higher than 90%, being said accuracy, when some combinations of genes are used, of 94.64 %.

Thus, in a first aspect, the invention relates to a method for diagnosing colorectal cancer comprising determining the level of the expression of one or more genes selected from the group consisting ENC1, MADCAM1, CMTM7, FAM60A, AC ATI and DDX55 in a biofluid of a subject, wherein a change in the level of expression of said genes compared to the level in a reference value is indicative of colorectal cancer.

As used in the present invention, the expression "method for the diagnosis" relates to a method that may essentially consist of the previously mentioned steps or may include additional steps. However, it must be understood that the method, in a preferred embodiment, is a method that is carried out in vitro, i.e., it is not carried out in the human or animal body. Diagnosing as used herein relates to evaluating the probability according to which a subject suffers from a disease. As will be understood by persons skilled in the art, such evaluation, although it is preferred that it is, normally may not be correct for 100% of the subjects to be diagnosed. The term, however, requires that a statistically significant part of the subjects can be identified as suffering from the disease or having a predisposition for it. The person skilled in the art can determine if a part is statistically significant without further delay by using several well known statistic evaluation tools, for example, determination of confidence intervals, determination of the p value, Student's t-test, Mann- Whitney test, etc. The details are in Dowdy and Wearden, Statistics for Research, John Wiley & Sons, New York 1983. The preferred confidence intervals are of at least 50%, at least 60%, at least 70%>, at least 80%>, at least 90%>, at least 95%. The p values are preferably p < 0.05, 0.01, etc.

As used herein, the term "colorectal cancer" includes any type of colon, rectal and appendix neoplasia and refers both to early and late adenomas and to carcinomas as well as to the hereditary, familial or sporadic cancer. Hereditary CRC includes those syndromes which include the presence of polyps, such as the hamartomatous polyposis syndromes and the most known, familial adenomatous polyposis (FAP) as well as nonpolyposis syndromes such as hereditary nonpolyposis colorectal cancer (HNPCC) or Lynch syndrome I.

In a preferred embodiment of the invention, said colorectal cancer (CRC) is colon cancer, rectal cancer and/or vermiform appendix cancer. In another embodiment of the invention, said colorectal cancer is stage 0, stage I [please let us know whether you can provide any evidence for stages 0 and I], stage II, stage III and/or stage IV colorectal cancer. The stages of CRC referred to herein correspond to the American Joint Committee on Cancer (AJCC) CRC staging, although other staging methods, such as Dukes and Astler-Coller staging, can equally be used.

"Stage 0" refers to the cancer in the earliest stage. It has not grown beyond the inner layer (mucosa) of the colon or rectum. This stage is also known as carcinoma in situ or intramucosal carcinoma.

"Stage I" refers to the cancer having grown through the muscularis mucosa into the submucosa or also having grown into the muscularis propria. It has not spread to nearby lymph nodes or distant sites.

"Stage Π" including sub-stages A, B and C refers to the cancer (A) grown into the outermost layers of the colon or rectum but has not gone through them; it has not reached nearby organs and it has not yet spread to the nearby lymph nodes or distant sites; (B) grown through the wall of the colon or rectum but has not grown into other nearby tissues or organs; it has not yet spread to the nearby lymph nodes or distant sites; and/or (C) grown through the wall of the colon or rectum and is attached to or has grown into other nearby tissues or organs; it has not yet spread to the nearby lymph nodes or distant sites.

"Stage III" including sub-stages A, B and C refers to the following: (A) At least one of the following applies: (i) The cancer has grown through the mucosa into the submucosa and it may also have grown into the muscularis propria. It has spread to 1 to 3 nearby lymph nodes or into areas of fat near the lymph nodes but not the nodes themselves. It has not spread to distant sites, (ii) The cancer has grown through the mucosa into the submucosa. It has spread to 4 to 6 nearby lymph nodes. It has not spread to distant sites. (B) At least one of the following applies: (i) The cancer has grown into the outermost layers of the colon or rectum or through the visceral peritoneum but has not reached nearby organs. It has spread to 1 to 3 nearby lymph nodes or into areas of fat near the lymph nodes but not the nodes themselves. It has not spread to distant sites, (ii) The cancer has grown into the muscularis propria or into the outermost layers of the colon or rectum. It has spread to 4 to 6 nearby lymph nodes. It has not spread to distant sites, (iii) The cancer has grown through the mucosa into the submucosa or it may also have grown into the muscularis propria. It has spread to 7 or more nearby lymph nodes. It has not spread to distant sites. (C) At least one of the following applies: (i) The cancer has grown through the wall of the colon or rectum (including the visceral peritoneum) but has not reached nearby organs. It has spread to 4 to 6 nearby lymph nodes. It has not spread to distant sites, (ii) The cancer has grown into the outermost layers of the colon or rectum or through the visceral peritoneum but has not reached nearby organs. It has spread to 7 or more nearby lymph nodes. It has not spread to distant sites, (iii) The cancer has grown through the wall of the colon or rectum and is attached to or has grown into other nearby tissues or organs. It has spread to at least one nearby lymph node or into areas of fat near the lymph nodes. It has not spread to distant sites.

"Stage IV" including sub-stages A and B refers to the following: (A) The cancer may or may not have grown through the wall of the colon or rectum, and it may or may not have spread to nearby lymph nodes. It has spread to 1 distant organ (such as the liver or lung) or set of lymph nodes. (B) The cancer may or may not have grown through the wall of the colon or rectum, and it may or may not have spread to nearby lymph nodes. It has spread to more than 1 distant organ (such as the liver or lung) or set of lymph nodes, or it has spread to distant parts of the peritoneum (the lining of the abdominal cavity). The term "bio fluid" is a biological fluid and refers to an aqueous fluid of biological origin. It may be obtained from any location (such as blood, plasma, serum, urine, bile, cerebrospinal fluid, aqueous or vitreous humor, or any bodily secretion).

In a preferred embodiment of the invention, the biofluid is blood, plasma or serum. In a more preferred embodiment, it is blood.

The term "subject", as used herein, refers to all animals classified as mammals and includes, but is not restricted to, domestic and farm animals, primates and humans, e.g., human beings, non-human primates, cows, horses, pigs, sheep, goats, dogs, cats, or rodents. Preferably, the patient is a male or female human of any age or race. A subject may or may not have or suffer from CRC. In other words, it is not known whether the subject has or suffers from CRC.

Marker genes in the diagnostic method of the invention include gene CMTM7 (CKLF-like MARVEL transmembrane domain containing 7 NM 138410)), FAM60A (family with sequence similarity 60, member A, NM_021238), ENC1 (ectodermal neural cortex- 1, NM_003633), AC ATI (acetyl-coenzyme A acetyltransferase 1, NM_000019), MADCAM1 (mucosal vascular addressing cell adhesion molecule 1, NM 130760) and gene DDX55 (DEAD (Asp-Glu-Ala-Asp) box polypeptide 55, NM 020936).

The methods of the invention preferably comprise the simultaneous determination of the expression levels of genes ENC1, MADCAM1, CMTM7, FAM60A, AC ATI and DDX55. However, the invention contemplates the use of a limited number of said genes, including cancer diagnosis methods in which the expression of genes ENC1, MADCAM1, CMTM7, FAM60A, AC ATI and DDX55 is individually determined as well as diagnosis methods based on the determination of each of said genes individually as well as of subsets of markers formed by a combination of any two genes, any three genes, any four genes, any five genes or the six genes.

In a particular embodiment the methods of the invention consist of the determination of the expression level of ENC1.

In a particular embodiment, the gene or genes used in the method of the invention are those which can diagnose colorectal cancer with an accuracy rate of at least 90%, at least 91%, at least at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98%, at least 99% or 100%. The term "accuracy rate", as used herein is defined as the condition of being true, exact or correct, that is the degree of agreement between the classification of a subject as healthy or illness and the real condition of said subject.

In a particular embodiment, the invention contemplates methods based on the determination of the level of a set of genes selected from the group consisting of ENCl and DDX55; ENCl and FAM60A; ENCl, FAM60A and DDX55; ENCl and CMTM7; ENC1,CMTM7 and DDX55; ENCl, CMTM7 and MADCAM1 ; ENCl and MADCAM1; ENCl, MADCAM1 and DDX55; ENCl, ACATl and MADCAM1; ENCl, ACATl, MADCAM1 and DDX55; ENCl, ACATl, CMTM7 and MADCAM1; ENCl, ACATl, CMTM7, MADCAM1 and DDX55; ENCl, ACATl, CMTM7, FAM60A and MADCAM1; ENCl, CMTM7, MADCAM1 and DDX55; ENCl, CMTM7 and FAM60A; ENCl, CMTM7, FAM60A and DDX55; ENCl, CMTM7, FAM60A and MADCAM1 ; ENCl, CMTM7, FAM60A, MADCAM1 and DDX55;ENC1 and ACATl ; ENCl, ACATl and FAM60; ENCl, ACATl, FAM60 and DDX55; ENCl, ACATl and DDX55; ENCl, ACATl, FAM60A and MADCAM1 ; ENC, ACATl, FAM60A, MADCAM1 and DDX55.

In another particular embodiment, the invention contemplates methods based on the determination of the level of a set of genes selected from the group consisting of ACATl, MADCAM1 and DDX55; ACATl and MADCAM1; CMTM7 and MADCAM1; CMTM7, MAD CAM 1 and DDX55; CMTM7, FAM60A and MADCAM1; CMTM7, FAM60A, MADCAM1 and DDX55, ACATl, CMTM7, MADCAM1 and DDX55; ACATl, CMTM7 and MADCAM1; MADCAM and DDX55; FAM60A and MADCAM1; FAM60A, MADCAM 1 and DDX55.

A "change" as used herein, means an alteration of the expression level of the genes, for example an increase or a decrease. The expression of a gene is considered increased in a sample of the subject under study when the levels increase with respect to the reference value by at least 5%, by at least 10%, by at least 15%, by at least 20%, by at least 25%, by at least 30%, by at least 35%, by at least 40%, by at least 45%, by at least 50%, by at least 55%, by at least 60%, by at least 65%, by at least 70%, by at least 75%, by at least 80%, by at least 85%, by at least 90%, by at least 95%, by at least 100%, by at least 110%, by at least 120%, by at least 130%, by at least 140%, by at least 150%, or more. Similarly, the expression of a gene is considered decreased when its levels decrease with respect to the reference value by at least 5%, by at least 10%>, by at least 15%, by at least 20%, by at least 25%, by at least 30%, by at least 35%, by at least 40%, by at least 45%, by at least 50%, by at least 55%, by at least 60%, by at least 65%, by at least 70%, by at least 75%, by at least 80%, by at least 85%, by at least 90%, by at least 95%, by at least 100% (i.e., absent).

A "reference value", as used herein, means a value determined in a sample obtained from a pool of healthy subjects which does not have a disease state or particular phenotype or from a bio fluid of said patient determined at the start of the therapy and/or determined previously during said therapy. The suitable reference expression levels of genes can be determined by measuring the expression levels of said genes in several suitable subjects, and such reference levels can be adjusted to specific subject populations (for example, a reference level can be linked to the age so that comparisons can be made between expression levels in samples of subjects of a certain age and reference levels for a particular disease state, phenotype, or lack thereof in a certain age group). In a preferred embodiment, the reference value is obtained from several healthy subjects or from subjects without prior history of colorectal cancer. The person skilled in the art will appreciate that the type of reference value can vary depending on the specific method to be performed

The expression profile of the genes in the reference value can preferably, be generated from a population of two or more individuals. The population, for example, can comprise 3, 4, 5, 10, 15, 20, 30, 40, 50 or more individuals. Furthermore, the expression profile of the genes in the reference value and in the sample of the individual that is going to be diagnosed according to the methods of the present invention can be generated from the same individual, provided that the profiles to be assayed and the reference profile are generated from bio fluids taken at different times and are compared to one another. For example, a sample of an individual can be obtained at the beginning of a study period. A reference biomarker profile from this sample can then be compared with the biomarker profiles generated from subsequent samples of the same individual. In a preferred embodiment, the reference value is from a pool of samples from several individuals.

The methods of the invention comprise the step of determining the expression levels of the marker genes. Virtually any conventional method can be used within the frame of the invention to detect and quantify the levels of mRNA encoded by the marker genes. By way of a non-limiting illustration, the expression levels are determined by means of the quantification of the levels of mRNA encoded by said genes. The latter can be quantified by means of using conventional methods, for example, methods comprising the amplification of mRNA and the quantification of the amplification product of said mRNA, such as electrophoresis and staining, or alternatively, by means of Northern blot and the use of suitable probes, Northern blot and use of specific probes of the mRNA of the genes of interest or of their corresponding cDNA/cRNA, mapping with the SI nuclease, RT-PCR, hybridization, microarrays, etc. Similarly, the levels of the cDNA/cRNA corresponding to said mRNA encoded by the marker genes can also be quantified by means of using conventional techniques; in this event, the method of the invention includes a step of synthesis of the corresponding cDNA by means of reverse transcription (RT) of the corresponding mRNA followed by the synthesis (RNA polymerase) and amplification of the cRNA complementary to said cDNA. Conventional methods of quantifying the expression levels can be found, for example, in Sambrook et al, 2001 "Molecular cloning: to Laboratory Manual", 3^rd ed., Cold Spring Harbor Laboratory Press, N.Y., Vol. 1-3.

In order to normalize the values of mRNA expression among the different samples, it is possible to compare the expression levels of the mRNA of interest in the test samples with the expression of a control RNA. A "control RNA" as used herein, relates to RNA whose expression levels do not change or change only in limited amounts in tumour cells with respect to non-tumorigenic cells. Preferably, the control RNA is mRNA derived from housekeeping genes and which code for proteins which are constitutively expressed and carry out essential cellular functions. Preferred housekeeping genes for use in the present invention include 18-S ribosomal protein, β-2- microglobulin, ubiquitin, cyclophilin, GAPDH, PSMB4, tubulin and β-actin.

Alternatively, it is also possible to determine the expression levels of the marker genes by means of the determination of the expression levels of the proteins encoded by said genes, since if the expression of genes is increased, an increase of the amount of corresponding protein should occur and if the expression of genes is decreased, a decrease of the amount of corresponding protein should occur. The determination of the expression levels of the different proteins can be carried out using any conventional method. By way of a non-limiting example, said determination can be carried out using antibodies with the capacity for binding specifically to the protein to be determined (or to fragments thereof containing the antigenic determinants) and subsequent quantification of the resulting antigen-antibody complexes. The antibodies that are going to be used in this type of assay can be, for example polyclonal sera, hybridoma supernatants or monoclonal antibodies, antibody fragments, Fv, Fab, Fab' and F(ab')2, scFv, diabodies, triabodies, tetrabodies and humanized antibodies. At the same time, the antibodies may or may not be labeled. Illustrative, but non-exclusive, examples of markers that can be used include radioactive isotopes, enzymes, fluorophores, chemo luminescent reagents, enzyme cofactors or substrates, enzyme inhibitors, particles, dyes, etc. There is a wide variety of well known assays that can be used in the present invention, using non-labeled antibodies (primary antibody) and labeled antibodies (secondary antibodies); these techniques include Western-blot or immunoblot, ELISA (enzyme- linked immunosorbent assay), RIA (radioimmunoassay), competitive EIA (enzyme immunoassay), DAS-ELISA (double antibody sandwich ELISA), immunocytochemical and immunohistochemical techniques, techniques based on the use of biochips or protein microarrays including specific antibodies or assays based on the colloidal precipitation in formats such as reagent strips. Other forms of detecting and quantifying the protein include affinity chromatography techniques, ligand-binding assays, etc.

Once the expression levels of the marker genes in relation to the expression levels of said genes in a reference value have been determined, it is necessary to identify if there are changes in the expression of said genes (increase or decrease of the expression). The gene expression profiles of the invention can be used in combination with other genetic and non-genetic markers already known for the diagnosis of CRC. Thus, the invention contemplates the use of the expression profiles previously mentioned together with the determination of the expression levels of one or several additional markers such as serum markers (for example, carcinoembrionic antigen) or analytes such as CA19-9, CA 125, CK-BB and the guanylyl cyclase C or together with methods based on the detection of blood in feces or the detection in feces of polymorphisms in genes K-ras, APC, p53 and BAT-26.

Method for determining the effect of a therapy

In a second aspect, the invention relates to a method for monitoring the effect of a therapy in a patient suffering from colorectal cancer and being treated with said therapy, comprising determining the levels of the expression of ENC1, MADCAM1, CMTM7, FAM60A, AC ATI and DDX55 in a bio fluid of said patient, wherein a change of the level of expression of said genes compared to the level in a reference value indicates a positive effect of said therapy

The term "therapy" or "treatment", as used herein, refers to the attempted remediation of a health problem, usually following a diagnosis. As such, it is not necessarily a cure, i.e. a complete reversion of a disease. Said therapy may or may not be known to have a positive effect on CRC or, in other words, to be useful to treat CRC. Therapies include, without limitation, surgery, radiotherapy, chemotherapy, or targeted therapies (e.g. using antibodies) and the choice of a therapy or a combination of therapies depends, for example, on the stage of the cancer, whether the cancer has recurred and the patient's general health.

The term "surgery", as used herein, means any therapeutic procedure that involves methodical action of the hand or of the hand with an instrument, on the body of a human or other mammal, to produce a curative or remedial.

The term "radiotherapy" refers to multiple types of radiation therapy including internal and external radiation therapies or radio immunotherapy, and the use of various types of radiations including X-rays, gamma rays, alpha particles, beta particles, photons, electrons, neutrons, radioisotopes, and other forms of ionizing radiations. In a preferred embodiment, the therapy is neoadjuvant or adjuvant chemotherapy. The term "neoadjuvant therapy", as used herein, refers to any type of treatment of cancer given prior to surgical resection of the primary tumor, in a patient affected.

As used herein, the term "chemotherapy" refers to the use of a chemical drug or a combination thereof for the treatment of cancer, tumors or malign neoplasia, including both cytotoxic or cytostatic drugs. Examples of chemotherapy agents which may be in accordance to the present invention include: alkylating agents (for example mechlorethamine, chlorambucil, cyclophosphamide, ifosfamide, streptozocin, carmustine, lomustine, melphalan, busulfan, dacarbazine, temozolomide, thiotepa or altretamine); platinum drugs (for example cisplatin, carboplatin or oxaliplatin); antimetabolite drugs (for example 5-fluorouracil, capecitabine, 6-mercaptopurine, methotrexate, gemcitabine, cytarabine, fludarabine or pemetrexed); anti-tumor antibiotics (for example daunorubicin, doxorubicin, epirubicin, idarubicin, actinomycin-D, bleomycin, mitomycin-C or mitoxantrone); mitotic inhibitors (for example paclitaxel, docetaxel, ixabepilone, vinblastine, vincristine, vinorelbine, vindesine or estramustine); and topoisomerase inhibitors (for example etoposide, teniposide, topotecan, irinotecan, diflomotecan or elomotecan).

Suitable chemotherapy agents include but are not limited to alkylating agents such as for example, cyclophosphamide, carmustine, daunorubicin, mechlorethamine, chlorambucil, nimustine, melphalan and the like; anthracyclines, such as for example, daunorubicin, doxorubicin, epirubicin, idarubicin, mitoxantrone, valrubicin and the like; taxane compounds, such as, for example, paclitaxel, docetaxel and the like; topoisomerase inhibitors such as for example, etoposide, teniposide, tuliposide, irinotecan and the like; nucleotide analogs such as for example, azacitidine, azathioprine, capecitabine, cytarabine, doxifluridine, fluorouracil, gemcitabine, mercaptopurine, methotrexate, thioguanine, ftorafur and the like; platinum-based agents such as for example, carboplatin, cisplatin, oxaliplatin and the like; antineoplastic agents such as for example, vincristine, leucovorin, lomustine, procarbazine and the like; hormone modulators such as for example, tamoxifen, finasteride, 5-□ -reductase inhibitors and the like; vinca alkaloids such as, for example, vinblastine, vincristine, vindesine, vinorelbine and the like. Suitable chemotherapy agents are described with more detail in the literature, such as in The Merck Index in CD-ROM, 13^th edition.

In the present invention, "antitumor agent" is understood as that chemical, physical or biological agent or compound with antiproliferative, antioncogenic and/or carcinostatic properties which can be used to inhibit tumor growth, proliferation and/or development. Examples of antitumor agents which can be used in the present invention are (i) antimetabolites, such as antifolates and purine analogs; (ii) natural products, such as antitumor antibiotics and mitotic inhibitors; (iii) hormones and antagonist thereof, such as androgens and corticosteroids; and (iv) biological agents, such as viral vectors. A list of compounds that can be used as antitumor agents is described in patent application WO2005/112973.

In a particular embodiment of the invention, the antitumor agent comprises antiangiogenic agents and signaling pathway inhibitors that, in another more particular embodiment, the antiangiogenic agent is Bevacizumab and the signaling pathway inhibitor is Cetuximab, Panitumumab or Erlotinib.

In a particular aspect of the invention, the reference value is from a biofluid of said patient determined at the start of the therapy and/or determined previously during said therapy. Marker genes in the diagnostic method of the invention include gene CMTM7 (CKLF-like MARVEL transmembrane domain containing 7 NM 138410)), FAM60A (family with sequence similarity 60, member A, NM_021238), ENCl (ectodermal neural cortex- 1, NM_003633), ACATl (acetyl-coenzyme A acetyltransferase 1, NM_000019), MADCAMl (mucosal vascular addressing cell adhesion molecule 1, NM 130760) and gene DDX55 (DEAD (Asp-Glu-Ala-Asp) box polypeptide 55, NM 020936).

In preferred embodiment, the method is carried out by determining a set of genes wherein said set of genes is selected from the group consisting of ENCl and DDX55; ENCl and FAM60A; ENCl, FAM60A and DDX55;ENC1 and CMTM7; ENC1,CMTM7 and DDX55; ENC1,CMTM7 and MADCAMl; ENCl and MADCAMl; ENCl, MADCAMl and DDX55; ENCl, ACATl and MADCAMl; ENCl, ACATl, MADCAMl and DDX55; ENCl, ACATl, CMTM7 and MADCAMl; ENCl, ACATl, CMTM7, MADCAMl and DDX55; ENCl, ACATl, CMTM7, FAM60A and MADCAMl; ENCl, CMTM7, MADCAMl and DDX55; ENCl, CMTM7 and FAM60A; ENCl, CMTM7, FAM60A and DDX55; ENCl, CMTM7, FAM60A and MADCAMl; ENCl, CMTM7, FAM60A, MADCAMl and DDX55;ENC1 and ACATl; ENCl, ACATl and FAM60; ENCl, ACATl, FAM60 and DDX55; ENCl, ACATl and DDX55; ENCl, ACATl, FAM60A and MADCAMl; ENC, ACATl, FAM60A, MADCAMl and DDX55.

In another particular embodiment, the method is carried out by determining a set of genes selected from the group consisting of ACATl, MADCAMl and DDX55;ACAT1 and MADCAMl; CMTM7 and MADCAMl; CMTM7, MADCAMl and DDX55; CMTM7, FAM60A and MADCAMl; CMTM7, FAM60A,MADCAM 1 and DDX55, ACAT1,CMTM7,MADCAM1 and DDX55; ACATl, CMTM7 and MADCAMl; MADCAM and DDX55; FAM60A and MADCAMl; FAM60A, MADCAMl and DDX55.

The term "targeted therapy" refers to a type of medication that blocks the growth of or kills cancer cells by interfering with specific targeted molecules in or on tumour cells rather than by simply interfering with rapidly dividing cells (e.g. with traditional chemotherapy). The main agent types used for targeted therapies are small molecules and antibodies.

The term "patient", as used herein, refers to a subject as defined above which suffers and is known to suffer from CRC. "Start of said therapy" means a time point just before the therapy is applied, preferably immediately before, e.g. before and on the same or the previous day. The term "previously", as used herein, refers to any prior time point. Preferably though it means at least 1 day, at least 2 days, at least 4 days, at least 5 days, at least 6 days, at least 7 days, at least 10 days, at least 2 weeks, at least 3 weeks, at least 4 weeks, at least 6 weeks, at least 2 months or at least 3 months before. It is emphasised that the level of step (i) can be compared to more than one level of step (ii), wherein the more than one levels of step (ii) are preferably determined at different time points, e.g. any of the time points recited above.

"Stagnation" preferably means a level within a 3%, a 5%, an 8%, a 10%, a 15%, or a 20%) range of a level determined before, e.g. previously as defined above. In a preferred embodiment, this "level determined before" is the value determined at the start of the therapy.

The term "change" has been described in detail in relation to the first method of the invention and is equally applicable to the second method of the invention. The change in the level of the expression of the genes is compared to the level determined before, e.g. previously as defined above. In a preferred embodiment, this "level determined before" is the value determined at the start of the therapy.

A "positive effect of a therapy" means that said therapy slows down disease progression, prevents disease progression, reduces or eliminates tumour burden (i.e. size and/or number of tumours), and/or cures the patient. It may also mean an increase in the duration of Recurrence-Free interval (RFI), an increase in the time of survival or Overall Survival (OS), an increase in the time of Disease-Free Survival (DFS), an increase in the duration of Distant Recurrence-Free Interval (DRFI), and/or the like.

The terms "colorectal cancer", "biofluid" and "subject" have been defined above and are used with the same meaning in the context of the present method.

In a preferred embodiment, the expression level of the genes is determined by determining the levels of the corresponding mRNAs. In a still more preferred embodiment, the determination of the levels of the mRNAs is determined by means of RT-PCR.

In a preferred embodiment, the biofluid is blood.

The reference value is preferably obtained from a sample of said patient before starting the treatment. Other terms used to define the method of the second aspect have the meaning as defined with respect to the method of the first aspect. Equally, embodiments described to further characterize the method of the first aspect also further characterize the method of the second aspect, if applicable.

Kit of the invention

In another aspect, the invention relates to a kit for diagnosis of colorectal cancer comprising a set of reagents which allow determining the expression levels of genes ENC1, MADCAM1, CMTM7, FAM60A, AC ATI and DDX55.

In the context of the present invention, "kit" is understood as a product containing the different reagents necessary for carrying out the methods of the invention packed so as to allow their transport and storage. Materials suitable for packing the components of the kit include crystal, plastic (polyethylene, polypropylene, polycarbonate and the like), bottles, vials, paper, envelopes and the like. Additionally, the kits of the invention can contain instructions for the simultaneous, sequential or separate use of the different components which are in the kit. Said instructions can be in the form of printed material or in the form of an electronic support capable of storing instructions such that they can be read by a subject, such as electronic storage media (magnetic disks, tapes and the like), optical media (CD-ROM, DVD) and the like. Additionally or alternatively, the media can contain Internet addresses that provide said instructions.

"Reagent which allows determining the expression levels of a gene" means a compound or set of compounds that allows determining the expression levels of a gene both by means of the determination of the levels of mRNA and by means of the determination of the levels of protein. Thus, reagents of the first type include probes capable of specifically hybridizing with the mRNAs encoded by said genes. Reagents of the second type include compounds that bind specifically with the proteins encoded by the marker genes and preferably include antibodies, although they can be specific ap tamers.

In a preferred embodiment, the first component of the kit of the invention consists of a specific probe for each of genes ENC1, MADCAM1, CMTM7, FAM60A, ACAT1, and DDX55. In a preferred embodiment, the reagents of the kit allow determining the expression levels of a set of genes and wherein said set is selected from the group consisting of ENCl and DDX55; ENCl and FAM60A; ENCl, FAM60A and DDX55;ENC1 and CMTM7; ENC1,CMTM7 and DDX55; ENC1,CMTM7 and MADCAM1 ; ENCl and MADCAM1 ; ENCl, M ADC AMI and DDX55; ENCl, ACATl and MADCAM1; ENCl, ACATl, MADCAM1 and DDX55; ENCl, ACATl, CMTM7 and MADCAM1; ENCl, ACATl, CMTM7, MADCAM1 and DDX55; ENCl, ACATl, CMTM7, FAM60A and MADCAM1; ENCl, CMTM7, MADCAM1 and DDX55; ENCl, CMTM7 and FAM60A; ENCl, CMTM7, FAM60A and DDX55; ENCl, CMTM7, FAM60A and MADCAM1; ENCl, CMTM7, FAM60A, M ADC AMI and DDX55;ENC1 and ACATl; ENCl, ACATl and FAM60; ENCl, ACATl, FAM60 and DDX55; ENCl, ACATl and DDX55; ENCl, ACATl, FAM60A and MADCAM1; ENC, ACATl, FAM60A, M ADC AMI and DDX55.

In another particular embodiment, the reagents of the kit allow determining the expression levels of a set of genes and wherein said set is selected from the group consisting of ACATl, MADCAM1 and DDX55; ACATl and MADCAM1; CMTM7 and MADCAM1; CMTM7, MADCAM1 and DDX55; CMTM7, FAM60A and MADCAM1; CMTM7, FAM60A, MADCAM1 and DDX55, AC AT 1 ,CMTM7, M ADC AMI and DDX55; ACATl, CMTM7 and MADCAM1; MADCAM and DDX55; FAM60A and MADCAM 1; FAM60A, MADCAM 1 and DDX55.In a preferred embodiment the set of reagents in the kit of the invention, which allow determining the expression levels of genes ENCl, MADCAM 1, CMTM7, FAM60A, ACATl, and DDX55 represents the 10%, 20%, 30%, 40%, 50%, 60%, 70%, 80%, 90%, 95% or 99% of total reagent in the kit.

In another preferred embodiment, the probes or antibodies forming the kit of the invention are coupled to an array.

In the event that the expression levels of several of the genes identified in the present invention are to be simultaneously determined, it is useful to include probes for all the genes the expression of which is to be determined in a microarray hybridization. The microarrays comprise a plurality of nucleic acids that are spatially distributed and stably associated to a support (for example, a biochip). The nucleic acids have a sequence complementary to particular subsequences of genes the expression of which is to be detected, therefore are capable of hybridizing with said nucleic acids. In the methods of the invention, a microarray comprising an array of nucleic acids is put into contact with a preparation of nucleic acids isolated from the patient object of the study. The incubation of the microarray with the preparation of nucleic acids is carried out in conditions suitable for the hybridization. Subsequently, after the elimination of the nucleic acids which have not been retained in the support, the hybridization pattern is detected, which provides information on the genetic profile of the sample analyzed. Although the microarrays are capable of providing both qualitative and quantitative information of the nucleic acids present in a sample, the invention requires the use of arrays and methodologies capable of providing quantitative information.

The invention contemplates a variety of arrays with regard to the type of probes and with regard to the type of support used. The probes included in the arrays that are capable of hybridizing with the nucleic acids can be nucleic acids or analogs thereof which maintain the hybridization capacity such as for example, nucleic acids in which the phosphodiester bond has been substituted with a phosphorothioate, methylimine, methylphosphonate, phosphoramidate, guanidine bond and the like, nucleic acids in which the ribose of the nucleotides is substituted with another hexose, peptide nucleic acids (PNA). The length of the probes can be of 7, 10, 15, 20, 25, 30, 35, 40, 45, 50, 55, 60, 65, 70, 75, 100 nucleotides and vary in the range of 10 to 1000 nucleotides, preferably in the range of 15 to 150 nucleotides, more preferably in the range of 15 to 100 nucleotides and can be single-strand or double-strand nucleic acids.

The selection of the specific probes for the different target genes is carried out such that they bind specifically to the target nucleic acid with a minimum hybridization to non-related genes. However, there are probes of 20 nucleotides which are not unique for a certain mR A. Therefore, probes directed to said sequences will show a cross- hybridization with identical sequences that appear in mRNA of non-related genes. In addition, there are probes that do not specifically hybridize with the target genes in the conditions used (because of secondary structures or of interactions with the substrate of the array). This type of probe must not be included in the array. Therefore, the person skilled in the art will observe that the probes that are going to be incorporated in a certain array must be optimized before their incorporation to the array. The optimization of the probes is generally carried out by generating an array containing a plurality of probes directed to the different regions of a certain target polynucleotide. This array is put into contact firstly with a sample containing the target nucleic acid in an isolated form and, secondly, with a complex mixture of nucleic acids. Probes which show a highly specific hybridization with the target nucleic acid but low or no hybridization with the complex sample are thus selected for their incorporation to the arrays of the invention. Additionally, it is possible to include in the array hybridization controls for each of the probes that is going to be studied. In a preferred embodiment, the hybridization controls contain an altered position in the central region of the probe. In the event that high levels of hybridization are observed between the studied probe and its hybridization control, the probe is not included in the array.

In a preferred embodiment, the array contains a plurality of probes complementary to subsequences of the target nucleic acid of a constant length or of variable length in a range of 5 to 50 nucleotides. The array can contain all the specific probes of a certain mRNA of a certain length or can contain probes selected from different regions of an mRNA. Each probe is assayed in parallel with a probe with a changed base, preferably in a central position of the probe. The array is put into contact with a sample containing nucleic acids with sequences complementary to the probes of the array and the signal of hybridization with each of the probes and with the corresponding hybridization controls is determined. Those probes in which a higher difference is observed between the signal of hybridization with the probe and its hybridization control are selected. The optimization process can include a second round of optimization in which the hybridization array is hybridized with a sample that does not contain sequences complementary to the probes of the array. After the second round of selection, those probes having signals of hybridization lower than a threshold level will be selected. Thus, probes which pass both controls, i.e., which show a minimum level of unspecific hybridization and a maximum level of specific hybridization with the target nucleic acid are selected.

The microarrays of the invention contain not only specific probes for the polynucleotides indicating a determined pathophysiological situation, but also containing a series of control probes, which can be of three types: normalization controls, expression level controls and hybridization controls.

Normalization controls are oligonucleotides that are perfectly complementary to labeled reference sequences which are added to the preparation of nucleic acids to be analyzed. The signals derived from the normalization controls after the hybridization provide an indication of the variations in the hybridization conditions, intensity of the marker, efficiency of the detection and another series of factors that can result in a variation of the signal of hybridization between different microarrays. The signals detected from the rest of probes of the array are preferably divided by the signal emitted by the control probes, thus normalizing the measurements. Virtually any probe can be used as normalization control. However, it is known that the efficiency of the hybridization varies according to the composition of nucleotides and the length of the probe. Therefore, preferred normalization probes are those which represent the mean length of the probes present in the array, although they can be selected such that they include a range of lengths that reflect the rest of probes present in the array. The normalization probes can be designed such that they reflect the mean composition of nucleotides of the rest of probes present in the array. A limited number of normalization probes is preferably selected such that they hybridize suitably, i.e., they do not have a secondary structure and do not show sequence similarity with any of the probes of the array is used. The normalization probes can be located in any position in the array or in multiple positions in the array to efficiently control variations in hybridization efficiency related to the structure of the array. The normalization controls are preferably located in the corners of the array and/or in the center thereof.

The expression controls levels are probes which hybridize specifically with genes which are expressed constitutively in the sample which is analyzed. The expression level controls are designed to control the physiological state and the metabolic activity of the cell. The examination of the covariance of the expression level control with the expression level of the target nucleic acid indicates if the variations in the expression levels are due to changes in the expression levels or are due to changes in the overall transcriptional rate in the cell or in its general metabolic activity. Thus, in the case of cells which have deficiencies in a certain metabolite essential for cell viability, the observation of a decrease both in the expression levels of the target gene as in the expression levels of the control is expected. On the other hand, if an increase in the expression of the expression of the target gene and of the control gene is observed, it probably due to an increase of the metabolic activity of the cell and not to a differential increase in the expression of the target gene. Any probe corresponding to a gene expressed constitutively, such as genes encoding proteins which exert essential cell functions such as β-2-microglobulin, ubiquitin, ribosomal protein 18S, cyclophilin A, transferrin receptor, actin, GAPDH and the like, can be used. In a preferred embodiment, the expression levels controls are GAPDH, tyrosine 3-monooxygenase/tryptophan 5- monooxygenase activation protein (YWHAZ), ubiquitin, beta-actin and β-2- microglobulin.

Hybridization controls can be included both for the probes directed to target genes and for the probes directed to the expression level or to the normalization controls. Error controls are probes of oligonucleotides identical to the probes directed to target genes but which contain mutations in one or several nucleotides, i.e., which contain nucleotides in certain positions which do not hybridize with the corresponding nucleotide in the target gene. The hybridization controls are selected such that, applying the suitable hybridization conditions, the target gene should hybridize with the specific probe but not with the hybridization control or with a reduced efficiency. The hybridization controls preferably contain one or several modified positions in the center of the probe. The hybridization controls therefore provide an indication of the degree of unspecific hybridization or of cross-hybridization to a nucleic acid in the sample to a probe different from that containing the exactly complementary sequence.

The arrays of the invention can also contain amplification and sample preparation controls which are probes complementary to subsequences of selected control genes because they normally do not appear in the biological sample object of the study, such as probes for bacterial genes. The R A sample is supplemented with a known amount of a nucleic acid which hybridizes with the selected control probe. The determination of the hybridization to said probe indicates the degree of recovery of the nucleic acids during their preparation as well as an estimation of the alteration caused in the nucleic acids during the processing of the sample.

Once a set of probes showing the suitable specificity and a set of control probes are provided, the latter are arranged in the array in a known position such that, after the steps of hybridization and of detection, it is possible to establish a correlation between a positive signal of hybridization and the particular gene from the coordinates of the array in which the positive signal of hybridization is detected.

The microarrays can be high density arrays with thousands of oligonucleotides by means of photolithographic in situ synthesis methods (Fodor et al, 1991, Science, 767- 773). This type of probe is usually redundant, i.e., they include several probes for each mRNA which is to be detected. In a preferred embodiment, the arrays are low density arrays or LDA containing less than 10000 probes per square centimeter. In said low density arrays, the different probes are manually applied with the aid of a pipette in different locations of a solid support (for example, a crystal surface, a membrane). The supports used to fix the probes can be obtained from a large variety of materials, including plastic, ceramics, metals, gels, membranes, crystals and the like. The microarrays can be obtained using any methodology known for the person skilled in the art.

After the hybridization, in the cases in which the non-hybridized nucleic acid is capable of emitting a signal in step of detection, a step of washing is necessary to eliminate said non-hybridized nucleic acid. The step of washing is carried out using methods and solutions known by the person skilled in the art.

In the event that the labeling in the nucleic acid is not directly detectable, it is possible to connect the microarray comprising the target nucleic acids bound to the array with the other components of the system necessary to cause the reaction giving rise to a detectable signal. For example, if the target nucleic acids are labeled with biotin, the array is put into contact with conjugated streptavidin with a fluorescent reagent in suitable conditions so that the binding between biotin and streptavidin occurs. After the incubation of the microarray with the system generating the detectable signal, it is necessary to carry out a step of washing to eliminate all the molecules which have bound non- specifically to the array. The washing conditions will be determined by the person skilled in the art using suitable conditions according to the system generating the detectable signal and which are well known for the person skilled in the art.

The resulting hybridization pattern can be viewed or detected in several different ways, said detection being determined by the type of system used in the microarray. Thus, the detection of the hybridization pattern can be carried out by means of scintillation counting, autoradiography, determination of a fluorescent signal, calorimetric determinations, detection of a light signal and the like.

Prior to the step of detection, it is possible to treat the microarrays with an specific endonuclease for single-strand DNA, such that the DNA that has bound non- specifically to the array is eliminated whereas the double-strand DNA resulting from the hybridization of the probes of the array with the nucleic acids of the sample object of study remains unchanged. Endonucleases suitable for this treatment include the SI nuclease, mung bean nuclease and the like. In the event that the treatment with endonuclease is carried out in an assay in which the target nucleic acid is not labeled with a directly detectable molecule (for example, in an assay in which the target nucleic acid is biotinylated), the treatment with endonuclease will be carried out before putting the microarray into contact with the other members of the system producing the detectable signal.

After the hybridization and the possible subsequent washing and treatment processes, the hybridization pattern is detected and quantified, for which the signal corresponding to each point of hybridization in the array is compared to a reference value corresponding to the signal emitted by a known number of terminally labeled nucleic acids in order to thus obtain an absolute value of the number of copies of each nucleic acid which is hybridized in a certain point of the microarray.

In the event that the expression levels of several of the proteins identified in the present invention is to be simultaneously determined, compositions containing at least one specific antibody for each of the genes are useful. For this purpose, the arrays of antibodies such as those described by De Wildt et al. (2000) Nat. Biotechnol. 18:989- 994; Lueking et al. (1999) Anal. Biochem. 270: 103-111; Ge et al. (2000) Nucleic Acids Res. 28, e3, I- VII; MacBeath and Schreiber (2000) Science 289: 1760-1763; WO 01/40803 and WO 99/51773 Al are useful. The antibodies of the array include any immunological agent capable of binding to a ligand with high affinity, including IgG, IgM, IgA, IgD and IgE, as well as molecules similar to antibodies which have an antigen binding site, such as Fab', Fab, F(ab')2, single domain antibodies or DABS, Fv, scFv and the like. The techniques for preparing said antibodies are very well known for the person skilled in the art and include the methods described by Ausubel et al. ((Current Protocols in Molecular Biology, eds. Ausubel et al, John Wiley & Sons (1992)).

The antibodies of the array can be applied at high speed, for example, using commercially available robotic systems (for example, those produced by Genetic Microsystems or Biorobotics). The substrate of the array can be nitrocellulose, plastic, crystal or can be of a porous material as for example, acrylamide, agarose or another polymer. In another embodiment, it is possible to use cells producing the specific antibodies for detecting the proteins of the invention by means of their culture in array filters. After the induction of the expression of the antibodies, the latter are immobilized in the filter in the position of the array where the producing cell was located.

An array of antibodies can be put into contact with a labeled target and the binding level of the target to the immobilized antibodies can be determined. If the target is not labeled, a sandwich type assay can be used in which a second labeled antibody specific for the polypeptide which binds to the polypeptide which is immobilized in the support is used. The quantification of the amount of polypeptide present in the sample in each point of the array can be stored in a database as an expression profile. The array of antibodies can be produced in duplicate and can be used to compare the binding profiles of two different samples.

In another aspect, the invention relates to the use of a kit of the invention for the diagnosis of colorectal cancer or for determining the effect of a treatment in a colorectal cancer patient.

The invention is detailed below by means of the following examples which are merely illustrative and by no means limiting for the scope of the invention.

EXAMPLES Blood samples

47 blood samples from patients with colorectal cancer (6% stage I, 33% stage II, 44% stage III and 17%> stage IV) were processed at Hospital de Cruces and 10 blood samples from healthy patients were processed at Gaiker. The RNA was obtained using Qiamp Blood RNA kit (Qiagen) according to the manufacturer's instructions.

RT-PCR

Starting from 500 ng of total RNA in a volume of 10 μΐ, the retrotranscription was carried out by means of Superscript III First-Strand synthesis kit (Invitrogen) according to the manufacturer's instructions.

Once the complementary DNA was obtained, real-time PCR was carried out using Platinum Quantitative PCR Supermix-UDG with ROX procedure (Invitrogen). The reaction mixture was made by adding 2.8 μΐ of water, 10 μΐ of the Supermix mixture, 1.2 μΐ of MgCl₂ and 1 μΐ of the probe with the primers, either for the gene to be studied or for the gene used as a normalizing or housekeeping gene (18S). 5 ml of cDNA (2 ng/μΐ) were added to 15 μΐ of the reaction mixture. The process was carried out in 96-well plates. A blank well was added in which cDNA was not included. The program used in the MyiQ thermal cycler (Biorad) consisted of 2 cycles at 95°C for 2 minutes, 1 cycle at 95°C for 15 seconds, 40 cy cycles at 60°C for 1 minute and 1 cycle at 25°C for 8 minutes.

The essential parameter in a real-time PCR is the threshold cycle (or Ct). It is defined as the number of PCR cycles from which the amplified product is detected (HU, N., et al, 2006). It is inversely proportional to the number of copies of the gene of interest.

Two different methods are used when quantifying the results. On one hand, there is the use of the calibration line, which is constructed from known DNA concentrations and, once made, the results of the samples are extrapolated. However, in this case, the quantification was carried out by means of the comparative Ct method, also known as the 2^"AACt method, using the 18S rRNA gene as the reference (housekeeping) gene.

Values of a gene above 1 indicate that that gene is expressed that number of time more (fold-change) in the condition tested with respect to the control. Values below 1 indicate that that gene is more expressed in the control.

RESULTS

After RT-PCR we concluded that the level of expression of ENC1, CMTM7, FAM60A, DDX55, AC ATI and MADCAMl were increased in blood samples from patients with colorectal cancer.

Table 1: Expression levels ofENCl, ACT1, CMTM7, FAM60A, MADCAMl and DDX5 genes in blood samples from patients with colorectal cancer.

Bioinformatics and data mining analysis Sample number 'tumoral 44' was removed from the analysis because there was only RT-PCR expression for 1 gene (ENCl gene). To sum up, our final dataset is formed by 46 tumoral and 10 control RT-PCR expression samples, measured in 6 genes (ENCl, CMTM7, FAM60A, DDX55, ACAT1, MADCAM1).

The main objective of the following bioinformatics and data mining analysis is to gain insight in the discrimination-diagnosis of both phenotypes (tumoral-T vs control- NT), making use of the exposed RT-PCR dataset.

Univariate feature ranking: t-test and non-parametric test.

In the following lines, the univariate discriminative power (statistical significance) of the presented genes is tested.

When a parametric t-test is performed for each of the 6 genes, comparing their tumoral-samples versus their non-tumoral values, all the tests show significant values, denoting statistically significant differences in the RT-PCR values of both phenotypes. All p-values of the t-tests are close to zero (when equal and non-equal variances are assumed). The use of this parametric test, which assumes a normal distribution in the data samples, is justified: a Kolmogorov-Smirnov test over the data shapes of all genes (conditioned to their specific phenotype), does not reject the Gaussian assumption performed in it. As an alternative and complementary way to study the same task, the individual relevance of each gene is measured by means of a well-known non-parametric (not assuming any specific data shape distribution) test called symmetrical uncertainty (SU):

where X is the predictive variable (in our case, each gene), C is the class label to be predicted (it takes both phenotypic values: non-tumoral and tumoral), H(X) denotes the entropy of X (level of uncertainty in its values) and H(X|C) is the conditional entropy of X given C. All the genes show notable values for this non-parametric SU metric, denoting significant differences RT-PCR values in both phenotypes. While genes show almost the same p-values in the previous t-test, genes' relevance can be ranked in the following way by means of SU (for each gene, in brackets, the value of the SU correlation degree), denoting a different degree of relevance among them: CMTM7 (0.487), ENC1 (0.431), DDX55 (0.423), FAM60A (0.395), AC ATI (0.39), MADCAM1 (0.286). (Note that the correlation coefficients range between 1 to 0 for SU metric: a value of 1 corresponds to a perfect correlation of the values of the gene with the class phenotype).

Based on these results on the statistical significance of their expressions, all six genes are included in the following multivariate analysis stage. Multivariate diagnostic model.

Different classification models to discriminate between both phenotypes (non- tumoral and tumoral), based on the RT-PCR values of the presented 6 genes, are learned. The performance of each model (accuracy, confusion matrix, ROC AUC values, sensitivity, etc.) is estimated by means of a leaving-one-out technique, which works as follows. Fifty-six classification models are considered, each of them built with 55 training samples (all except one sample): the accuracy of each of these models is tested in the remainder sample, which has not been taken into account to build the model. The accuracy rates of these models (in their respectives remainder instance) have been pooled in order to estimate the exposed accuracy metrics. In this way, samples used for testing do not participate in the modelization-phase, assuring an honest validation of the models. Leavingone- out is considered as the most powerful accuracy estimation technique when few samples are available, and it is frequently used in bio informatics studies.

Although a large set of models is learned, only the most remarkable results are shown. There exists a large set of classification techniques within the data mining literature. Nevertheless, we focus our research on models with a transparent output which can be interpreted and understood by biologists. Therefore, the main part of our results is based on Bayesian network classifiers. In order to learn this type of models, gene values need to be discretized: as it is common in gene expression literature, when the opposite is not noted, gene RT-PCR values are discretized in 3-equal-width intervals, mimicking the classic 'underexpression'- 'baseline'-'overexpression' window of range- values.

The simplest Bayesian network structure, known as naive Bayes, assumes that all predictive genes are independent when the phenotype is known. It obtains 83.92% accuracy rate. The associated confusion matrix is shown below (rows correspond to real- observed phenotype values, whereas columns collect the predictions of the model):

Although global recognition rates are high, the large number (8) of 'false negative' samples (assuming the 'tumoral' phenotype as 'positive') seems a non- desirable effect in a clinical study.

It is possible to induce another Bayesian network model which allows dependencies among predictive genes, namely k-dependence Bayesian classifiers. For this model, a 92.85% accuracy rate is obtained, with the following associated confusion matrix:

The number of 'false negative' samples is notably reduced, while the number of 'false positive' cases is maintained. The 'sensitivity' value of the model is 0.943, while 'specificity' is 0.9.

In an attempt to continue reducing the number of 'false negative' predictions, the results of the same type of k-dependence Bayesian classifier can be improved if the gene values are discretized using the supervised multi-interval discretization technique. This discretization technique tends to fix the binning intervals of the variable in mid-points that induce notably different distributions of the class phenotype.

In our case, this technique discretizes all the variables in two intervals.

Estimated accuracy rate is 94.64%>, with the following associated confusion matrix: tumoral non-tumoral

tumoral 45 1

non-tumoral 2 8

While the number of 'false negative' samples is reduced, the number of 'false positive' cases increases in one. The 'sensitivity' value of the model is 0.978, while 'specificity' is 0.8.

For the sake of simplicity, the associated conditional probability tables of the Bayesian network structures, which show the strength of modeled 'parent-child' relationships, are not displayed. The following items can be highlighted:

• A strong correlation relationship between ENCl and AC ATI genes, the former conditioning the latter values. The biological insights of this relationship should be studied.

• The same holds for the ACAT1→ CMTM7 gene relationship.

• CMTM7 and ACAT1 genes strongly condition the values of the FAM60A gene.

The biological insights of this relationship should be studied.

Different combination of markers were analyzed and the accuracy rates obtained were between 91.0 % and 96.43%.

Table 2 shows the accuracy rate, sensitivity and specificity of different combinations of markers.

Subset of Markers

Accuracy rate Sensitivity Specificity

ENC1-MADCAM1 96.43% 1.000 0.200

ENC1-MADCAM1-DDX55 96.43% 1.000 0.200

ENC1-ACAT1-MADCAM1 96.43% 1.000 0.200

ENC1-ACAT1-MADCAM1-DDX55 96.43% 1.000 0.200

ENC1-ACAT1-CMTM7-MADCAM1 94.64% 0.978 0.200

ENC1-ACAT1-CMTM7-MADCAM1-DDX55 94.64% 0.978 0.200

ENC1-ACAT1-CMTM7-FAM60A-MADCAM1 94.64% 0.978 0.200

ACAT1-MADCAM1-DDX55 94.64% 0.957 0.100

ACAT1-MADCAM1 94.64% 0.957 0.100

CMTM7-MADCAM1 94.64% 0.935 0,000

CMTM7-MADCAM1-DDX55 94.64% 0.935 0,000

CMTM7-FAM60A-MADCAM1 94.64% 0.935 0,000 TPR FPR

Subset of Markers

Accuracy rate Sensitivity Specificity

CMTM7-FAM60A-MADCAM1-DDX55 94.64% 0.935 0,000

ACAT1-CMTM7-MADCAM1-DDX55 92.86% 0.935 0.100

ACAT1-CMTM7-MADCAM1 92.86% 0.935 0.100

ENC1 91.07% 0.935 0.200

MADCAM-DDX55 91.07% 0.890 0,000

FAM60A-MADCAM1 91.07% 0.891 0,000

FAM60A-MADCAM1-DDX55 91.07% 0.891 0,000

ENC1-DDX55 91.07% 0.935 0.200

ENC1-FAM60A 91.07% 0.935 0.200

ENC1-FAM60A-DDX55 91.07% 0.935 0.200

ENC-1-CMTM7 91.07% 0.935 0.200

ENC1-CMTM7-DDX55 91.07% 0.935 0.200

ENC1-CMTM7-MADCAM1 91.07% 0.935 0.200

ENC1-CMTM7-MADCAM1-DDX55 91.07% 0.935 0.200

ENC1-CMTM7-FAM60A 91.07% 0.935 0.200

ENC1-CMTM7-FAM60A-DDX55 91.07% 0.935 0.200

ENC1-CMTM7-FAM60A-MADCAM1 91.07% 0.935 0.200

ENC1-CMTM7-FAM60A-MADCAM1-DDX55 91.07% 0.935 0.200

ENC1-ACAT1 91.07% 0.935 0.200

ENC1-ACAT1-FAM60 91.07% 0.935 0.200

ENC1-ACAT1-FAM60-DDX55 91.07% 0.935 0.200

ENC1-ACAT1-DDX55 91.07% 0.935 0.200

ENC1-ACAT1-FAM60A-MADCAM1 91.07% 0.935 0.200

ENC-ACAT1-FAM60A-MADCAM1-DDX55 91.07% 0.935 0.200

ENC-1-ACAT1-CMTM7-FAM60A-MADCAM1-

94.64% 0.978 0.2

DDX55

Table 2 Accuracy rate, sensitivity and specificity of different combinations of markers

Other kind of classification algorithms were also tested (e.g. random forest, classification trees, neural networks, logistic regression, etc.), obtaining similar or lower accuracy values. However, as it was previously stated, Bayesian network models allow a more clear interpretation and comprehension of the underlying modeled relationships.

Claims

1. A method for diagnosing colorectal cancer, comprising determining the level of the expression of one or more genes selected from the group consisting of ENCl, MADCAM1, CMTM7, FAM60A, AC ATI and DDX55 in a bio fluid of said subject, wherein a change in the expression of said one or more genes compared to the level in a reference value is indicative of colorectal cancer.

2. A method according to claim 1 wherein the expression levels of a set of genes is determined wherein said set is selected from the group consisting of

ENCl and DDX55;

ENCl and FAM60A;

ENCl, FAM60A and DDX55;

ENCl, CMTM7;

ENCl, CMTM7 and DDX55;

ENCl, CMTM7 and MADCAM1 ;

ENCl, MADCAM1;

ENCl, MADCAM1 and DDX55;

ENCl, AC ATI and MADCAM1 ;

ENCl, ACAT1, MADCAM1 and DDX55;

ENCl, ACAT1, CMTM7 and MADCAM1;

ENCl, ACAT1, CMTM7, MADCAM1 and DDX55;

ENCl, ACAT1, CMTM7, FAM60A and MADCAM1;

ENCl, CMTM7, MADCAM1, and DDX55;

ENC 1 , CMTM7 and FAM60A;

ENCl, CMTM7, FAM60A and DDX55;

ENCl, CMTM7, FAM60A and MADCAM1 ;

ENCl, CMTM7, FAM60A, MADCAM1 and DDX55;

ENCl and AC ATI;

ENCl, AC ATI and FAM60;

ENCl, ACAT1, FAM60 and DDX55;

ENCl, AC ATI and DDX55;

ENCl, ACAT1, FAM60A andMADCAMl ; and

ENC, ACAT1, FAM60A, MADCAM1 and DDX55.

3. A method according to claim 1 wherein the expression levels of a set of genes is determined wherein said set is selected from the group consisting of

ACAT1, MADCAM1 and DDX55;

AC ATI and MADCAM1 ;

CMTM7 and MADCAM1;

CMTM7, MADCAM1 and DDX55;

CMTM7, FAM60A and MADCAM1 ;

CMTM7, FAM60A, MADCAM1 and DDX55;

ACAT1, CMTM7 , MADCAM1 and DDX55;

ACAT1, CMTM7 and MADCAM1;

MADCAM and DDX55;

FAM60A, MADCAM 1 ;

and

FAM60A, MADCAM 1 and DDX55.

4. The method according to any of claims 1 to 3 wherein the reference value is from a bio fluid of a subject not having colorectal cancer.

A method for monitoring the effect of a therapy in a patient suffering from colorectal cancer and being treated with said therapy, comprising determining the levels of the expression of ENC1, MADCAM 1, CMTM7, FAM60A, ACAT1, and DDX55 in a bio fluid of said subject, wherein a change of the level of expression of said genes compared to the level in a reference value indicates a positive effect of said therapy.

A method according to claim 5 wherein the expression levels of a set of genes is determined wherein said set is selected from the group consisting of

ENC1 and DDX55;

ENC1 and FAM60A;

ENC1, FAM60A and DDX55; ENCl CMTM7;

CMTM7 and DDX55;

CMTM7 and MADCAMl ;

MADCAMl ;

MADCAMl and DDX55;

ACATl and MADCAMl ;

ACATl , MADCAMl and DDX55;

ACATl, CMTM7 and MADCAMl;

ACATl, CMTM7, MADCAMl and DDX55;

ACATl, CMTM7, FAM60A and MADCAMl;

CMTM7, MADCAMl, and DDX55;

CMTM7 and FAM60A;

CMTM7, F AM 60 A and DDX55;

CMTM7, F AM 60 A and MADCAMl ;

CMTM7, F AM 60 A, MADCAMl and DDX55;

and ACATl;

ACATl and FAM60;

ACATl, FAM60 and DDX55;

ACATl and DDX55;

ACATl , F AM 60 A andMADCAMl ;

7. A method according to claim 5 wherein the expression levels of a set of genes is determined wherein said set is selected from the group consisting of

ACATl, MADCAMl and DDX55;

ACATl and MADCAMl ;

CMTM7 and MADCAMl ;

CMTM7, MADCAMl and DDX55;

CMTM7, FAM60A and MADCAMl ;

CMTM7, FAM60A, MADCAMl and DDX55;

ACATl, CMTM7 , MADCAMl and DDX55;

ACATl, CMTM7 and MADCAMl; MADCAM and DDX55;

FAM60A, MADCAM 1 ;

and

FAM60A, MADCAM 1 and DDX55.

8. The method according to any of claims 5 to 7 wherein the reference value is from a biofluid of said patient determined at the start of the therapy and/or determined previously during said therapy.

A method according to any of the previous claims, wherein the determination the expression levels of the genes is determined by means of RT-PCR.

A kit for diagnosis of colorectal cancer comprising a set of reagents which allow determining the expression levels of one or more genes selected from the group consisting of ENC1, MADCAM 1, CMTM7, FAM60A, ACAT1, and DDX55.

11. A kit according to claim 10 wherein the reagents allow determining the expression levels of a set of genes and wherein said set is selected from the group consisting of

ENC1 and DDX55;

ENC1 and FAM60A;

ENC1, FAM60A and DDX55;

ENC1 and CMTM7;

ENC1, CMTM7 and DDX55;

ENC1, CMTM7 and MADCAM 1 ;

ENC 1 and MADCAM 1 ;

ENC1, MADCAM 1 and DDX55;

ENC1, AC ATI and MADCAM 1 ;

ENC1, ACAT1, MADCAM 1 and DDX55;

ENC1, ACAT1, CMTM7 and MADCAM 1;

ENC1, ACAT1, CMTM7, MADCAM 1 and DDX55;

ENC1, ACAT1, CMTM7, FAM60A and MADCAM 1;

ENC1, CMTM7, MADCAM 1 and DDX55;

ENC1, CMTM7 and FAM60A; ENC1, CMTM7, FAM60A and DDX55;

ENC1, CMTM7, FAM60A and MADCAM1 ;

ENC1, CMTM7, FAM60A, MADCAM1 and DDX55;

ENC1 and AC ATI;

ENC 1 , AC AT 1 and FAM60;

ENC1, ACAT1, FAM60 and DDX55;

ENC1, AC ATI and DDX55;

ENC1, ACAT1, FAM60A, MADCAM1 ;

and

ENC, ACAT1, FAM60A, MADCAM1 and DDX55.

12. A kit according to claim 10 wherein the reagents allow determining the expression levels of a set of genes and wherein said set is selected from the group consisting of ACAT1, MADCAM1 and DDX55

AC AT 1 and MADC AM 1

CMTM7 and MADC AMI

CMTM7, MADC AMI and DDX55

CMTM7, FAM60A and MADC AMI

CMTM7, FAM60A, MADC AMI and DDX55

ACAT1, CMTM7, MADC AMI and DDX55

ACAT1, CMTM7 and MADC AMI

MADCAM and DDX55

FAM60A and MADCAM 1

and

FAM60A and MADCAM 1 and DDX55.

13. A kit as defined in any of claim 10 to 12 wherein the reagents are immobilized in a support. 14. Use of a kit according to any of claims 10 to 13 for the diagnosis of colorectal cancer or for determining the effect of a treatment in a colorectal cancer patient.