WO2016198807A1

WO2016198807A1 - Method for predicting the sensitivity of a gastrointestinal tumor to an inhibitor treatment

Info

Publication number: WO2016198807A1
Application number: PCT/FR2016/051419
Authority: WO
Inventors: Stéphane ANSIEAU; Alain Puisieux; Guillaume Collin; Pierre SAINTIGNY
Original assignee: Centre Leon Berard
Priority date: 2015-06-12
Filing date: 2016-06-13
Publication date: 2016-12-15
Also published as: FR3037403A1

Abstract

The invention primarily relates to an in vitro method for selecting a gastrointestinal cancer patient who is likely to respond to anticancer treatment involving the administration of a quinolinate phosphoribosyltransferase (QPRT) inhibitor. A method of the invention involves in particular a step of determining QPRT expression in a test sample of the gastrointestinal tumor, QPRT expression in cells of the test sample being an indicator that the patient is sensitive to a QPRT inhibitor treatment. The invention also relates to the use of QPRT as a biomarker for predicting the sensitivity of a gastrointestinal tumor patient to a QPRT inhibitor treatment.

Description

METHOD FOR PREDICTING THE SENSITIVITY OF A TUMOR

GASTROINTESTINAL TO INHIBITORY TREATMENT

The present invention is in the medical field and more specifically in the field of anti-cancer therapy. The main subject of the invention is an in vitro or ex vivo method of selecting patients with gastrointestinal tumors that may benefit from anti-tumor therapy comprising the administration of an inhibitor of the kynurenine pathway. A method according to the invention comprises in particular a step of determining the expression of the QPRT protein in a test sample of said gastrointestinal tumor, the expression of QPRT in epithelial cells of the test sample being indicative of the sensitivity said patient to an inhibitory treatment of the kynurenine pathway.

The present invention also relates to the use of the protein QPRT as a biomarker for predicting the sensitivity of a patient suffering from a gastrointestinal tumor to an inhibitory treatment of the kynurenine pathway.

In order to optimize therapeutic treatments and the relevance of clinical trials in the field of anti-cancer therapy, the identification of new therapeutic targets and biomarkers predictive of the patients' sensitivity to this treatment, be it used alone. one or in combination with other therapeutic approaches, remains an important goal for the future development of these treatments. It is therefore necessary to have new treatments and effective methods to predict the sensitivity of a patient to these heavy treatments and associated with significant side effects.

The inventors have now developed a new method for the selection of patients with a gastrointestinal tumor likely to benefit from an inhibitory treatment of the kynurenine pathway. This method comprises determining the expression of the QPRT protein in the epithelial cells of a sample of the patient's tumor prior to any treatment, a higher expression of the QPRT protein in the test sample compared to a reference sample being indicative of the patient's sensitivity to an inhibitory treatment of the kynurenine pathway.

Several gene expression signatures ("classifiers") to monitor the evolution of Crohn's disease and colitis have been described (Montero-Melendez et al., 2013).

The publication by Watanabe (2013) describes a gene expression signature and a method for predicting the progression from ulcerative colitis to colorectal cancer. The signature described in this document comprises 40 differentially expressed genes according to whether patients with neoplastic lesions, or not affected by such lesions.

The publication of Malik et al. (2014) describes the crystallographic structure of human QPRT in a form complexed with its inhibitor, phthalic acid. This inhibitor is however not very specific, the concentration at which it is effective in vitro is of the order of millimolar (10 ^"3 M). The publication of Sahm et al. (2013) teaches that quinolic acid, an endogenous metabolite of tryptophan and a precursor of NAD ⁺ , confers resistance to oxidative stress on gliomas.

Chronic inflammation is fertile ground for cancer development. In inflamed mucosa, however, oxidative stress induces erosion of telomeres and signs of replicative senescence in intestinal stem cells.

The inventors have shown that the comparison of the gene expression profile of human colon epithelial cells that has escaped the senescence induced by oxidative stress with the gene profile of the parental line highlights the induction of a specific gene program, characterized by the activation of the expression of about thirty genes and the repression of about thirty others. This gene program is orchestrated by ZEB transcription factors and is differentiated from the epithelio-mesenchymal transition (EMT) program commonly induced by these factors. This new gene program is named ALEP for Alternative to EMT Program. The inventors have shown in vitro that the expression of QPRT confers a survival advantage to cells having escaped senescence, under conditions of chronic inflammation or when they are subjected to an oxidizing agent. The inventors have also shown that QPRT depletion of cells that have escaped senescence induces the death of these cells by apoptosis under conditions of chronic inflammation, and that the inhibition of QPRT expression prevents these cells from tolerating oncogenic activations. They also demonstrated that intestinal cancer cell lines with an active ALEP program (high signature score) are dependent on the QPRT protein for proliferation. Beyond the QPRT protein, the inhibition of another key enzyme of the kynurenine pathway located upstream of QPRT, the TD02 protein, very similarly inhibits the proliferation of these lines. The activity of the kynurenine pathway as a whole would therefore be necessary for the proliferation of colorectal lines in which the ALEP program is activated.

Moreover, the inventors have shown that the QPRT protein is absent in the enterocytes of normal intestinal tissue, whereas an increase in the number of labeled epithelial cells during the detection of QPRT is observed in a significant proportion of dysplasias of low- or high-grade and cancers developed in an inflammatory context, as well as in epithelial cells with a significant proportion of adenomas and sporadic cancers. In addition, a proportion of oesophageal adenocarcinomas express the QPRT protein whereas no labeling is observed in the squamous epithelium, used as internal control.

The QPRT protein, and beyond it the other enzymes of the kynurenine pathway, are therefore particularly interesting therapeutic targets and their inactivation therefore constitutes a promising anti-tumor therapeutic approach, if appropriate in combination with another anti-tumor treatment, in particular with a therapy capable of inducing oxidative stress. The invention will now be described in detail with the aid of examples intended to illustrate it without limiting it, various modifications that may be made without departing from the general scope of the invention.

The present invention relates to an ex vivo method of selecting a patient suffering from a gastrointestinal tumor, likely to benefit from an inhibitory treatment of the kynurenine pathway, from a test sample of said tumor before treatment of said patient, said method comprising the following steps:

a) determination of the expression of the QPRT protein in the epithelial cells of said test sample,

b) The comparison of the expression of the QPRT protein in the epithelial cells of the said test sample with the expression of QPRT in a reference sample, c) The selection of a patient likely to benefit from an inhibitor treatment of the pathway of kynurenine from the comparison of step b), a higher expression level of QPRT protein in the test sample epithelial cells than that of the reference sample being indicative of the sensitivity of said test sample. tumor to an inhibitory treatment of the kynurenine pathway.

The present invention also relates to an ex vivo method for predicting the sensitivity of a patient suffering from a gastrointestinal tumor (before treatment) to an inhibitory treatment of the kynurenine pathway, said method comprising the following steps: i) obtaining a test sample of said tumor, ii) determining the expression of QPRT protein in the epithelial cells of said test sample, iii) comparing the expression of QPRT protein in said test sample with the expression of QPRT in a reference sample, iv) predicting the sensitivity of said patient to an inhibitory treatment of the kynurenine pathway from the comparison of step iii), a higher expression level of QPRT protein epithelial cells of the test sample relative to that of the reference sample being indicative of the sensitivity of said patient to an inhibitory treatment of the kynurenine pathway.

By "sensitivity of a patient to a treatment" is meant the observation of a therapeutic effect such as improving the survival of a patient having received said therapeutic treatment. The absence of a patient's response, or the patient's lack of sensitivity to a therapeutic treatment, is instead defined by the term "resistance" of a patient to said therapeutic treatment, in which no therapeutic effect is observed. An ex vivo method of selecting a patient likely to benefit from an inhibitory treatment of the kynurenine pathway makes it possible to predict the sensitivity of a patient suffering from a gastrointestinal tumor to an inhibitor treatment of the gastrointestinal tract. kynurenine.

The expression "before treatment" or "untreated" means before or after a surgical procedure, such as an excision, performed on the tumor but before any therapeutic treatment associated but non-surgical, such as chemotherapy, radiotherapy, hormone therapy or immunotherapy.

Quinolinate phosphoribosyltransferase or QPRT (UniProtKB Q 15274 updated on May 18, 2010, SEQ ID NO: 1), also known as Nicotinate-nucleotide pyrophosphorylase catalyzes the conversion of quinolinic acid (QA) to nicotinamide mononucleotide (NAMN) by transfer. phosphoribosyl group of 5-phospho-D-ribose-1-diphosphate (PRPP), during the first stage of NAD + biosynthesis.

By "kynurenine pathway" is meant the metabolic pathway leading to the de novo production of NAD ⁺ from tryptophan, an essential amino acid. The enzymes involved in this pathway have been described in particular by Vécsei et al., 2013.

By "inhibitory treatment of the kynurenine pathway" is meant the application of a treatment intended to inhibit the activity and / or the effects of at least one of the enzymes which contribute to this metabolic pathway, namely tryptophan 2 , 3-dioxygenase (TD02), kynurenine formylase, kynurenine 3-monooxygenase (KMO), kynureninase (KYNU), 3-hydroxyanthranilic acid oxygenase (3HAO) or QPRT. This treatment includes, for example, the use of inhibitors of the tryptophan protein 2,3-dioxygenase or TD02 such as the molecule 690C91 ((E) -6-fluoro-3- [2- (3-pyridyl) vinyl] -1H indole, Sigma), inhibitors of kynurenine 3-monooxygenase or KMO such as 2-3 teriflunomide and Ro-61-8048 (Selleckchem), PDE10A (Pfizer), JM6 (3,4-dimethoxy-N- (4) - (3-nitrophenyl) -5- (piperidin-1-ylmethyl) thiazol-2-yl) benzenesulfonamide), inhibitors of the enzyme 3HAO as described by Vallerini et al., 2013 or QPRT inhibitors. A known example of an antagonist of the enzymatic activity of QPRT is phthalic acid (Malik et al., 2013).

By "inhibitor of the kynurenine pathway" is meant any substance, molecule, compound or complex capable of specifically inhibiting, in particular by competition or conformation, the activity and / or the effects of at least one enzyme of this type. metabolic pathway. For example, these inhibitors may be of the allosteric type, substrate analog (s), product analogue (s) or enzymatic reaction intermediate (s), especially non-catabolizable, specific for the enzyme concerned.

In all aspects of the invention, the inhibitory treatment of the kynyrenine pathway is preferably an inhibitor treatment of QPRT or TD02, and more particularly QPRT.

Therefore, preferably, the subject of the invention is an ex vivo method of selecting a patient suffering from a gastrointestinal tumor, likely to benefit from an inhibitory treatment of the kynurenine pathway, from a test sample of said tumor prior to treatment of said patient, said method comprising the following steps:

b) comparing the expression of the QPRT protein in the epithelial cells of said test sample with the expression of QPRT in a reference sample, c) The selection of a patient likely to benefit from an inhibitory treatment of the kynurenine pathway from the comparison of step b), a higher expression level of QPRT protein in the epithelial cells of the sample test relative to that of the reference sample being indicative of the sensitivity of said tumor to an inhibitory treatment of the kynurenine pathway.

According to a particular aspect, the subject of the invention is an ex vivo method of selecting a patient suffering from a gastrointestinal tumor, before treatment, likely to benefit from an inhibitory treatment of the kynurenine pathway, comprising the determining the level of expression of QPRT protein in the epithelial cells of a tumor test sample, wherein the tumor test sample is selected from: tumor tissue, an operative specimen and a biopsy.

The selection method according to the invention relates to an ex vivo method of a patient suffering from a gastrointestinal tumor likely to benefit from an inhibitory treatment of the kynurenine pathway, before said patient has been treated for said tumor.

In general, in the context of the present invention, the sample from the patient with a gastrointestinal tumor is obtained before treatment; said patient being considered untreated for said tumor.

The sample of the gastrointestinal tumor may be derived from the primary tumor or from metastasis (s). According to a particular aspect of the ex vivo selection process according to the invention, the test sample of said tumor is derived from the primary tumor.

The expression "higher expression level" indicates that the level of expression observed in said test sample is greater than the level of expression observed in the reference sample, in terms of the number of cells in which the Expression is detected and / or the intensity of the detected signal and / or the location of the cells for which the protein is detected.

Indeed, the inventors have shown that the QPRT protein is absent in the enterocytes of normal intestinal tissue, whereas an increase in the number of labeled epithelial cells during the detection of QPRT is observed in a significant proportion of dysplasias of low- or high-grade and cancers developed in an inflammatory context, as well as in epithelial cells with a significant proportion of adenomas and sporadic cancers.

In addition, a proportion of oesophageal adenocarcinomas express the QPRT protein whereas no labeling is observed in the squamous epithelium, used as internal control.

Advantageously, in a method according to the invention, the expression of the QPRT protein is determined by comparison with the expression of the QPRT protein in a reference sample, the expression of the QPRT protein in a reference sample being especially chosen from: the expression of QPRT in a test sample of healthy tissue of the same nature as the tissue of the test sample, the expression of QPRT in a tissue sample of a patient reaches inflammatory gastrointestinal disease not associated with dysplastic (ie, non-tumor) lesions, and / or expression of QPRT in a tissue sample of the subject patient taken from the patient did not have a gastrointestinal tumor.

The known non-tumorous inflammatory gastrointestinal diseases are: Crohn's disease, ulcerative colitis; chronic gastritis and chronic gastroesophageal reflux (as part of adenocarcinoma).

The determination of the level of expression of the QPRT protein is carried out quantitatively or semi-quantitatively by any method known to those skilled in the art allowing the specific detection and / or quantification of a particular protein in a sample. Such a determination method is chosen in particular from methods using QPRT-binding antibodies, including in particular immunohistochemistry (IHC) and immunofluorescence (IF) detection.

The expression of QPRT can be defined by means of a histological score taking into account, on the one hand, the observed marking intensity and, on the other hand, the percentage of labeled cells, or by combining the percentage labeled cells and the intensity of labeling. According to this method, tumor sample sections are analyzed by microscopy after bringing said sections into contact with an antibody specifically binding QPRT.

By "labeled cells" is meant cells for which a signal is detected when looking for expression of the QPRT protein.

According to a more particular embodiment, the subject of the invention is an ex vivo method of selecting a patient suffering from a gastrointestinal tumor likely to benefit from an inhibitory treatment of the kynurenine pathway, in which the Determination of QPRT protein expression comprises determining the presence of QPRT comprising at least one immunohistochemistry step.

By "immunohistochemistry" is meant a method of locating proteins in the cells of a tissue section, by detecting a particular antigen by means of polyclonal or monoclonal antibodies binding that antigen. The antibody-antigen pair can be visualized for example by conjugation to an enzyme which can catalyze a color production reaction, by the use of labeled antibodies, in particular by a fluorophore, or by the use of antibodies known as " secondary antibodies capable of binding antigen-binding antibodies.

In an immunohistochemical method, the tissue is placed on a solid support. A primary antibody bound to a detection system (or a primary antibody followed by a secondary antibody) is added. The signal is observed by microscopy techniques.

In a particular embodiment of a method according to the invention for selecting a patient suffering from a gastrointestinal tumor likely to benefit from an inhibitory treatment of the kynurenine pathway, the test sample of said tumor is contacted with a QPRT-binding antibody, said antibody being selected from polyclonal anti-QPRT antibodies and anti-QPRT monoclonal antibodies. According to a particular embodiment of a method according to the invention, said antibody binds QPRT under native conditions, one skilled in the art can easily choose an antibody meeting this criterion, there may be mentioned in particular the Sigma HPA011887 antibody.

According to an even more particular embodiment, the subject of the invention is an in vitro method for selecting a patient suffering from a gastrointestinal tumor that may benefit from an inhibitory treatment of the kynurenine pathway, in which the determination of the presence of the QPRT protein in the test sample of said tumor is carried out by means of at least one immunohistochemistry step comprising the localization of labeled cells within the epithelial tissue of the test sample.

According to an even more particular embodiment, an ex vivo method for predicting the sensitivity of a patient suffering from a gastrointestinal tumor to an inhibitory treatment of the kynurenine pathway, according to the invention, comprises the determination of the percentage of labeled epithelial cells within the epithelial tissue of the test sample.

According to a particular aspect, the subject of the invention is an ex vivo method of selecting a patient suffering from a gastrointestinal tumor likely to benefit from an inhibitory treatment of the kynurenine pathway, in which the determination of the QPRT protein expression comprises determining the percentage of labeled tumor cells.

According to a more particular aspect, in a method according to the invention, a level of expression of QPRT protein in the test sample corresponding to at least 10% of the labeled epithelial cells is indicative of the sensitivity of said tumor to a treatment inhibitor of the way of kynurenine.

In the context of the present invention, there is also disclosed an ex vivo method for predicting the sensitivity of a tumor patient to an inhibitory treatment of the kynurenine pathway, wherein the determination of the level of QPRT expression includes the determination of the level of expression of an mRNA encoding the QPRT protein. More particularly, a method according to the invention comprises determining the amount of an mRNA encoding the QPRT protein, particularly in the epithelial cells of the tumor. The determination of the amount of the mRNA encoding the QPRT protein can be carried out by any method known to those skilled in the art allowing the specific detection of an mRNA, for example after microdissection of the epithelial cells of a tumor. Such a method notably comprises the specific detection of an mRNA using nucleic acid probes complementary to the sequence of said mRNA, in particular by RT-PCR, by a Northern-Blot type method. It can also be obtained by hybridization on an immobilized probe (in situ hybridization).

Quantitative PCR is used to measure the initial amount of nucleic acid. The amplified nucleic acids, and in particular the mRNAs, can be detected and quantified by any method known to a person skilled in the art, such as in particular by 260 nm spectrometry or by Northern blotting using a hybridization method for radioactive or fluorescent probes. amplified nucleic acid. The nucleotide sequence of the mRNA corresponding to quinolinate phosphoribosyltransferase is known (NCBI reference NM 014298, updated on March 15, 2015, SEQ ID NO: 2). A person skilled in the technical field will readily conceive, from this nucleotide sequence, primers capable of specifically hybridizing to the nucleic acid encoding QPRT, thus enabling its detection. By way of example, the primers described in the present patent application (SEQ ID NO: 3 and SEQ ID NO: 4) can be used in a process according to the invention.

According to another particular aspect, the subject of the invention is an ex vivo method of selecting a patient suffering from a gastrointestinal tumor likely to benefit from an inhibitory treatment of the kynurenine pathway, starting from a sample of said tumor before any treatment, said method comprising the following steps:

a) Determining, in said test sample, the level of expression of the QPRT protein and at least one other protein of the ALEP program, in the epithelial cells of the tumor

b) comparing the protein expression, in said test sample, of QPRT and at least one other protein of the ALEP program, with the protein expression of QPRT and at least the other protein of the ALEP program, in the epithelial cells of a reference sample,

c) The selection of a patient likely to benefit from an inhibitory treatment of the kynurenine pathway from the comparison of step b),

i. a higher expression level of QPRT protein in the test sample epithelial cells and a higher level of expression of at least one other ALEP program protein in the test sample epithelial cells than in the test sample reference sample, and / or

ii. a higher expression level of QPRT protein in the test sample epithelial cells and a lower level of expression of at least one ALEP program protein in the test sample epithelial cells relative to that of the test sample. reference sample being indicative of the sensitivity of said tumor to an inhibitory treatment of the kynurenine pathway.

ALEP program proteins refer to the 59 proteins encoded by the genes shown in Figure 13, which the inventors have demonstrated that the expression is either higher or lower in epithelial cells resistant to oxidative stress.

The genes encoding the ALEP program proteins whose expression is greater are:

APOE, SPARC, C6ORF160, SNHG5, SLC2A3, GALC, MT1G, SH3GL2, GNE, SUNC1, EXOSC8, LCP1, KCNJ8, NDRG1, SNCG, FGD3, LGALS1, TXNIP, SUM03, EMILIN2, TNNC1, TNNT1, TGFB1, LOC728031, KRT19 and GABARAPL1. The genes coding for the ALEP program proteins whose expression is lower are: IL32, ETV5, S100A3, SGK1, PRSS1, IRAK2, S100A9, GPR56, TNFAIP3, PIM2, PRRX2, BIRC3, CXCL5, DIAPH1, AD AMI9, LRIG1, SLC7A11, EVL, WNT5A, TNFAIP2, EN02, TNFRSF1B, PROM1, TNF, TMPRSS2, LOC100134294, NFIX, RIN2, CXCL1, CXCL2, CARD9, CCL20.

Preferably, step b) consists in comparing the expression of the QPRT protein and at least the expression of the APOE protein in the epithelial cells of said test sample, it being understood that a higher expression level of QPRT and of APOE is indicative of the sensitivity of said tumor to an inhibitory treatment of the kynurenine pathway, in particular to a QPRT or TD02 inhibitor, in particular to a QPRT inhibitor.

According to a particular embodiment, in an ex vivo method for predicting the sensitivity of a patient suffering from a gastrointestinal tumor to an inhibitory treatment of the kynurenine pathway, according to the invention, said gastrointestinal tumor is selected from: colorectal cancer, gastric cancer and adenocarcinoma of the esophagus.

The term "tumor" refers to a malignant tumor or cancer.

The term "colorectal cancer" refers to a tumor resulting from the neoplastic transformation of the epithelial cells of the mucosa of the colon or rectum. Colorectal cancer can develop in an inflammatory context (ulcerative colitis or Crohn's disease) or not and may result from malignant progression of dysplasia or polyps. Depending on the size and depth of the tumor, ganglion infiltration and number, the tumors are classified as O to 4 (TNM classification).

Gastric adenocarcinoma refers to a tumor resulting from the neoplastic transformation of the stomach lining and results from the progression of chronic gastritis in atropic chronic gastritis, metaplasia, dysplasia before possible progression to adenocarcinoma. Localized, locally advanced and metastatic stages are distinguished.

Esophageal adenocarcinoma refers to a tumor derived from the epithelium of the

Barrett and results from an abnormal healing of lesions related to chronic aggression of the lower esophageal lining by chronic gastroesophageal reflux. Tumors are classified according to the TNM classification).

According to a more particular embodiment, in an ex vivo method for predicting the sensitivity of a patient suffering from a gastrointestinal tumor to an inhibitory treatment of the kynurenine pathway, according to the invention, said gastrointestinal tumor is intestinal is developed in a chronic inflammatory context.

The term "chronic inflammatory context" refers to the repetition of stages of inflammation and remission, this period can extend over years, as for patients with Crohn's disease or ulcerative colitis, leading progressively to an alteration of the epithelium, for example by inducing genetic instability in the epithelial cells. This chronic inflammation increases the incidence of cancer in the long term. According to an even more specific embodiment, the subject of the invention is an ex vivo method for predicting the sensitivity of a patient suffering from a gastrointestinal tumor developed on a chronic inflammatory context to an inhibitor treatment of the gastrointestinal tract. kynurenine, according to the invention, said gastrointestinal tumor being a colorectal cancer.

According to another particular embodiment, in an ex vivo method for predicting the sensitivity of a patient suffering from a gastrointestinal tumor to an inhibitory treatment of the kynurenine pathway, according to the invention, said gastrointestinal tumor intestinal is a sporadic colorectal cancer.

Tumors that do not develop in an inflammatory context are said to be sporadic; they generally result from the mutation of the APC gene.

According to another particular embodiment, in an ex vivo method for predicting the sensitivity of a patient suffering from a gastrointestinal tumor to an inhibitory treatment of the kynurenine pathway, according to the invention, said gastrointestinal tumor intestinal is a colorectal cancer that affects patients with germline mutations on the APC gene. These patients develop juvenile adenomatous polyposes (PAF), which can evolve into CRC.

According to another particular aspect, the subject of the invention is an ex vivo method for predicting the sensitivity of a patient suffering from a tumor to an inhibitory treatment of the kynurenine pathway, combined with another antitumor treatment. More particularly, in a method according to the invention, said anti-tumor treatment is chosen from: surgery, chemotherapy and radiotherapy.

According to a still more particular aspect, the subject of the invention is an ex vivo method for predicting the sensitivity of a patient suffering from a gastrointestinal tumor to an inhibitory treatment of the kynurenine pathway, said patient having made the subject of a surgical step for removing said tumor (or excision).

The subject of the present invention is also an ex vivo method for identifying a tumor capable of responding to a treatment comprising the administration of a kynurenine inhibiting agent, from a test sample of said tumor before treatment of said tumor. patient, said method comprising the steps of:

at. Determination of QPRT expression in the epithelial cells of said test sample,

b. Comparing QPRT expression in epithelial cells of said test sample with expression of QPRT in epithelial cells of a reference sample,

vs. The identification of the tumor likely to respond to a treatment comprising the administration of an inhibitory agent of the kynurenine pathway, from the comparison of step b), a higher expression level of QPRT in the epithelial cells of the test sample compared to that of the epithelial cells of a reference sample being indicative of the sensitivity of said tumor to an inhibitory treatment of the kynurenine pathway.

In the context of this subject of the invention, the expression of QPRT can be determined by the protein expression of QPRT or the transcriptional expression of QPRT, in particular the expression of the QPRT mRNA. According to a particularly preferred embodiment of this subject of the invention, it is the expression of the QPRT protein which is determined.

In another aspect, the invention relates to the use of QPRT as a protein biomarker for predicting the sensitivity of a patient with a gastrointestinal tumor to an inhibitory treatment of the kynurenine pathway.

The invention also relates to the use of QPRT as a biomarker for predicting the sensitivity of a patient suffering from a gastrointestinal tumor to an inhibitory treatment of the kynurenine pathway, preferably chosen from a QPRT inhibitor and a TD02 inhibitor, even more preferably QPRT inhibitor treatment, said patient being untreated for said tumor and QPRT protein expression being measured in epithelial cells of a sample of said tumor.

In a particular aspect, the invention also relates to the use of QPRT as a biomarker for predicting the sensitivity of a patient suffering from a gastrointestinal tumor to an inhibitory treatment of the kynurenine pathway, preferably selected from a QPRT inhibitor and a TD02 inhibitor, combined with an anti-tumor treatment, such as oxaliplatin, 5-FU or irrinotecan, and even more preferably the sensitivity of a patient with a gastrointestinal tumor -intestinal to an inhibitory treatment of QPRT.

Other features and advantages of the invention appear in the following examples and figures.

BRIEF DESCRIPTION OF THE FIGURES

Figures 1A and 1B: The escape to senescence induced in response to chronic inflammation allows the emergence of risk cells.

Fig.lA: Submission of colon epithelial cells to chronic inflammation forces them to adapt to oxidative stress. Immortalized human colon epithelial cells (HCEC-hTERT) were subjected every two days to activated macrophage supernatants (AMS). Cells rapidly stop proliferating (as demonstrated by H3-phosphorylated histone level, pS10-H3), enter senescence as demonstrated by detection of SA-β-galactosidase activity, accumulation of acetylated p53 protein ( Ac-p53) and the induction of its target gene CIP1 (p21). After a month, the cells begin to proliferate. Fig. 1B: Emergent cells are resistant to oxidative stress and tolerate mitogenic activations. The cells that have escaped senescence (Ech-Inf A and Ech-Inf B from two separate experiments) and the Parental HCEC cells (Par) were either treated with tert-butyl-hydoxy-peroxide (TBHP) (panel left) or transduced by an activated version of the RAS protein (HRASG12V, right panel) and stained with 12-day crystal violet later. The level of RAS protein is analyzed by Western blot.

Figures 2A and 2B: ZEB transcription factors orchestrate the mechanism of adaptation to oxidative stress. Fig. 2A: Escaping to induced senescence in response to chronic inflammation or chemical peroxide results from induction of ZEB embryonic genes. Analysis of SNAI1, TWIST2, ZEB1 and ZEB2 gene expression by q-RT-PCR in cells escaping chronic inflammation (left panel) or oxidative stress induced by a chemical peroxide (right panel). Relative expression is defined in comparison to the HPRT1 household gene and parental HCEC cells. The expression of the SNAI2 and TWIST1 genes is below the detection threshold. Fig. 2B: The ectopic expression of one of the two ZEB transcription factors is sufficient to allow cells to adapt to oxidative stress. HCEC-hTERT cells were transduced by Zebl or Zeb2, submitted to either activated macrophage supernatants (AMS) or TBHP. After 12 days of treatment, the cells were stained with crystal violet.

Figures 3A and 3B: The resistance to oxidative stress is related to the induction of a specific gene program. Fig. 3A: Comparison of the HCEC-hTERT gene profiles that escaped chronic inflammation (from two independent experiments, Ech-Inf A and B), chemical oxidative stress induced (Ech-TBHP) or transduced by Zebl (HCEC-ZeW) or Zeb2 (HCEC-Zeb2) highlights a set of 27 commonly induced genes and 32 commonly repressed genes (see Figure 13). Among this list, the two most induced genes are QPRT and APOE. Fig. 3B: The ectopic expression of either of these two proteins is sufficient to protect cells from oxidative stress. HCEC cells were transduced by QPRT or APOE and subjected to either activated macrophage supernatants (AMS) or TBHP treatment. After 12 days of treatment, the cells were stained with crystal violet. FIGS. 4A and 4B: The activity of the QPRT protein contributes to resistance to oxidative stress. Fig. 4A: Survival of cells that have escaped senescence induced in response to chronic inflammation is dependent on QPRT expression. Panel at the top: HCEC cells that escaped senescence induced in response to chronic inflammation (Ech-Inf) were infected with lentiviral vectors expressing two distinct shRNAs directed against QPRT RNA (shRNA QPRT A and B) and subjected activated macrophage supernatants (SMA). After 12 days of treatment, the cells are stained with crystal violet. The parental HCEC (Par) and Ech-Inf cells infected with a control shRNA were used respectively as negative and positive controls. Bottom panel: analysis of QPRT and cleaved caspase 3 (Cl, case 3) by Western blot.

Fig. 4B: QPRT protein activity is essential to allow cells that have escaped senescence induced in response to chronic inflammation to tolerate activations oncogenic. Panel at the top: HCEC cells that escaped induced senescence in response to chronic inflammation (Ech-Inf) were disrupted in QPRT by RNA interference and infected with a retroviral vector expressing the activated version of the HRAS protein (HRASG12V). 12 days after infection, the cells were stained with crystal violet. Bottom panel: QPRT and RAS analysis by Western blot.

Figure 5: Classification of human lines of colorectal carcinoma based on the ALEP score. The lines used in the following tests and in which the program is induced are OUMS-23, C2BBel, CL-14; the lines in which the program is not induced are SW116, HCT116, HT-29, SW403, COLO-320.

Figure 6: The proliferation of colorectal cancer lines in which the ALEP program is induced is dependent on the expression of QPRT. Top panels: different colorectal lines in which the alternative program is induced (ALEP active) or not (ALEP inactive) were infected with lentiviral vectors expressing shRNAs directed against QPRT RNA (shRNA QPRT A and B) and stained with purple crystal twelve days after infection. Bottom panel: QPRT analysis in the different lines by Western blot.

Figure 7: Panels A and B. The inhibition of the kynurenine pathway in ALEP -positive ORMS-23 (panel A) or C2BBel (panel B) colorectal lines is accompanied by a proliferation arrest, a induction of autophagy and DNA damage. Panels from the top. Inhibition of QPRT expression via shRNA (A or B) induces cell proliferation arrest as judged by proliferation curves (panels on the upper left) and histone H3 phospho accumulation Serine 10 (pS10-H3, panels top right). Inhibition of QPRTcw shRNA or inactivation of the enzyme TD02 by chemical compound 680C91 induces autophagy, judging by the accumulation of vacuoles (panels on the lower left) and the LC3 protein (panels central and right) and the accumulation of DNA damage, judged by the accumulation of the γ-Η2ΑΧ protein (central bottom and right panels).

Figure 8: QPRT analysis on human samples by immunohistochemistry (I). QPRT marking examples by immunohistochemistry on healthy colon samples (upper left) of inflamed mucous membranes (top right), dysplasias of low- or high-grade (2 ^nd and 3 ^rd row from the high) or carcinomas developed on an inflammatory context. Scale: 100 μιη.

Figure 9: QPRT analysis on human samples by immunohistochemistry (II). Examples of QPRT staining by immunohistochemistry in samples of weakly or strongly dysplastic adenomas and in sporadic cancers. Scale 100 μιη.

Figures 10A and 10B: The ALEP program is induced in the inflamed mucosa of patients with ulcerative colitis or Crohn's disease. The comparison of the ALEP program signature with the gene profiles of patients with Crohn's disease (CD) or ulcerative colitis (UC) (series GSE36807 and GSE37283) makes it possible to segregate a certain number of patients. Note in the series of patients with ulcerative colitis GSE37283 the ability of the signature to segregate those with or without cancerous colorectal lesions.

Figures 11A to 11C: Analysis of the biological significance of the alternative program on a large cohort of patients with sporadic colorectal cancers (primary resected tumors of patients who have not yet undergone any non-surgical associated therapeutic treatment). The analysis is conducted on a cohort of 566 colorectal carcinomas (GSE39582) classified according to their gene profile into six molecular types (from C1 to C6, Marisa et al., 2013). Fig. 11A. Analysis of the signature score of the alternative program in the six molecular types. Fig. 11B. Analysis of the correlation between the scores of the alternative program signatures and intestinal stem cells Lgr5 and EphB2. Fig. 11C. Kaplan-Meier representation of the recurrence-free survival of colorectal cancer patients by high, intermediate, or low ALEP score (with thresholds taken at 4000 and 6000 by combining the scores of each of the overexpressed or under-expressed genes even extinct).

Figure 12: QPRT analysis by immunohistochemistry in oesophageal adenocarcinoma. Examples of oesophageal adenocarcinoma expressing (QPRT +) or non - QPRT - protein QPRT. The squamous epithelium of the lower third of the esophagus is used as an internal control (panel at the bottom right). Scale 100 μιη.

Figure 13: List of genes commonly induced or repressed in HCEC cells that have escaped senescence induced in response to chronic inflammation (Ech-Inf A and B) to a chemical peroxide (Ech-TBHP A and B) or transduced by Zebl or Zeb2.

EXAMPLES

Example 1: Material and methods

Plasmids and Lineages: The HRAS ^G12V , ZEB1 and ZEB2 pBabe Puro constructs have been previously described (Morel et al., 2012). The expression vectors ΑΡΟΕε3 and QPRT in Pbabe Puro were generated from the vectors pLenti-GIII-hAPOE-GFP-2A-Puro constructs (abm) and QPRT pCMV6 (Origene). The QPRT RK138 / 139 cDNA was synthesized by the Genscript company and subcloned into the Puro pBabe retro viral vector (Addgene). QPRT A shRNAs (5'-TCGCTCTGAAGGTGGAGT-3 ', SEQ ID NO: 5) and shRNA QPRT B (5'-AGTCCTAAACCGGAAGAGG-3', SEQ ID NO: 6) in pLKO.I were obtained from Sigma . HT-29, HCT116, SW403, COLO320HSR and C2BBel were obtained from ATCC. The lines OUMS23 and CL-14 were provided by the JCRB cellular bank (Japan) and the DSMZ GmbH (Germany), respectively. The lines OUMS23 and c2BBel were cultured in DMEM medium (Life Technologies) supplemented with 10% FCS (Cambrex) and 100 U / ml Penicillin / Streptomycin (PS, Life Technologies), lines HT-29 and HCT116 in McCoy's 5A medium ( Life Technologies) supplemented with 10% FCS + 100 U / ml PS, lines SW116, SW403 and COLO320 in RPMI medium (Life Technologies) supplemented with 10% FBS + 100 U / ml PS, and the line CL-14 in DMEM / HAMF 12 1: 1 medium (Invitrogen) supplemented with 20% FBS and 100 U / ml PS. The primary epithelial colon cells were obtained from the abm company (T4056), immortalized by transduction at HTERT and cultured in Prigrow III medium (TM003, abm) supplemented with 5% FBS and 100 U / ml PS on collagen plates. (BD BioCoat Collagen Type I, Dutscher).

Induction of oxidative stress in HCEC cells: The monocyte line THP-1 was obtained from ATCC and cultured in RPMI medium (Life Technologies) supplemented with 10% FBS, 100 U / ml PS, 1 mM Sodium pyruvate (Life Technologies) and non-essential amino acids (Life Technologies). Cell differentiation (5.10 ⁷ cells) was induced by a treatment with PMA (0.162 μΜ, Sigma) for one day and the activation of the macrophages thus obtained then induced by LPS (ultra-pure LPS-EB of the E line. coli 0111: B4, InvivoGen) for one day. The supernatant was collected, filtered over 0.45 μl, diluted half in the HCEC culture medium and placed on the HCEC cells (5 × 10 ⁵ in a 6-well dish). The supernatant of activated macrophages was renewed every other day. Similarly, the treatment of HCEC cells with tert-Butyl hydroperoxide (30 μl, Sigma) was repeated every other day. The cells were stained with crystal violet after ten days of treatment. SA-β-galactosidase assays were performed as previously described (Dimri et al., 1995).

Retroviral and lentiviral infections: Ectopic production of proteins in HCEC-hTERT cells was performed through retroviral infections as previously described (Gras et al., 2014). Briefly, the cells are "murine" by the ectopic expression of an ecotropic receptor (Baker et al., 1992; Boehm et al., 2005) before being infected with the retroviral expression vectors. Sequential infections are spaced 48 hours apart and selection is initiated 24 hours after the second infection with puromycin (1.5 μg / ml). The lentiviral particles of the QPRT shRNAs were generated by transfection of the HEK293T line with the pLKO.l vectors, pCMV AR8.91 (gag-pop-Tat-Rev) (Zufferey et al., 1997) and phCMVG-VSVG (env). (Burns et al., 1993) by the calcium phosphate technique. Depletion in QPRT was obtained by three successive infections, each separated by 12h. We confirmed the infection of more than 90%> cells.

Western-bot protein analysis: The cell extracts were carried out in RIPA medium (100 mM NaCl, 1% NP40, 0.1% SDS, 50 mM Tris pH8) supplemented with protease inhibitor (Roche) and phosphatase cocktails. (Sigma) and cell debris removed by centrifugation. The proteins were analyzed by Western blot using anti-TWIST1 2Cla monoclonal antibodies (Abcam), anti-p21CIP1 / WAF1 clone SX118 (Dako) and anti-QPRT (ab57125, Abcam), and polyclonal anti-H-antibodies. RAS N-18 (Santa Cruz), anti-ZEB1 H102 (Santa Cruz), anti-ZEB2 (Sayan et al., 2009), p-histone H3 (Ser10) -R (Santa Cruz), and histone H2A.X Ser139 (Cell Signaling) and secondary antibodies coupled to peroxidase (Dako). The antigen-antibody complexes were revealed by the Luminol reagent (Santa Cruz). Immunohistochemistry: Patient samples were obtained from the Biological Resources Center of Lyon Est and Saint-Antoine Hospital (Paris) with the agreement of local ethics committees. The samples were used with the informed written agreement of the patients. This study has been approved by the ethical committees of both institutions. The markings were performed on sections of 40 μιη of formalin fixed tissues and incubated in paraffin. The slides were successively deparaffinized by three successive baths of xylene and rehydrated with baths of pure ethanol, 90% and 80%. The antigen was then unmasked by heating the slides in the microwave (5 and 10 min) in a PH6 (Dako) unmasking solution. After blocking the endogenous peroxidase activity by a 0.3% hydrogen peroxide bath for 10 min, the slides were incubated in the presence of QPRT antibodies (Atlas antibody, HPA011887, Sigma) at room temperature for 1 h at a dilution at l / 100th. The labeling is then revealed using the Novolink Polymer Kit (Leica).

Analysis of the transcriptional expression of the genes by q-RT-PCR: The preparation of the DNAs and the reverse transcription were performed as previously described (Gras et al., 2014). The PCR primers were selected using the primer3 software. The household gene HPRT1 was used for normalization. The following primer pairs have been selected:

Table 1 Gene Expression Profiles: Gene expression profiles and data analysis were performed through the ProfileXpert platform (Lyon, France). The expression profiles were analyzed on the basis of chips containing 47231 probes (HumanHT-12 v4 Expression BeadChip, Illumina Inc., USA). Total RNA (500 ng) was amplified and labeled with biotin using the Illumina TotalPrep ™ RNA Amplification Kit kit (Ambion Inc., USA). Hybridization was performed with 750 ng of biotin labeled cRNA. The chips were scanned using Illumina's standard protocol and theiScan scanner (Illumina Inc., USA) and the data was standardized using Genome Studio 2010 software (Illumina Inc., USA).

In silico analyzes of the relevance of the alternative program: The analysis of the data was carried out through the software Array Studio (Omicsoft Corporation). The raw data from the analyzed series (CEL) were processed using quantile normalization and the Robust multi-array average algorithm (PMA and transformed into Log2 (Irizarry et al., 2003). Various computer tools including the "Hierarchical clustering analysis The Principal Component Analysis (PCA) and ssGSEA projection methodology were used.The signature score was determined using the "Empirical cumulative Distribution Functions" (ECDF) genes and correlations between signatures determined by a Pearson correlation.

Example 2: The escape of senescence in response to chronic inflammation leads to the emergence of risk cells and results from reprogramming of the cells, orchestrated by the transcription factor ZEB1.

The model established in vitro is based on the use of immortalized colon epithelial cells (HCEC-hTERT) cultured in the presence of activated macrophage supernatant (SMA, conditions mimicking chronic inflammation). These experimental conditions lead to the induction of single-stranded DNA damage that evolves into double-stranded damage (confirmed by γΗ2ΑΧ protein accumulation). These damages induce a stop of the cellular proliferation (demonstrated by the reduction of the phosphorylated H3 histone level on the residue Ser10, p-S10-H3) and the engagement of the epithelial cells in a program of senescence (evidenced by the acetylation of the p53 protein (Ac-p53), induction of its CIP1 target gene (p21) and detection of an SA-galactosidase activity): a sequence of events summarizing the observations made in vivo. However, after a period of one month, the cells invariably begin to proliferate (Figure 1A). Emergent cells (later referred to as "Ech-Inf" for "inflammation escape") are not only able to proliferate under conditions of chronic inflammation but also to withstand oxidative stress that is induced in response to a peroxide chemical (such as tert-butyl-hydroxy-peroxide or TBHP) or in response to oncogenic activation (transduction of an activated version of the RAS protein, H-RAS ^G12V ) (Figure 1B). Emerging cells have therefore acquired intrinsic resistance to oxidative stress and tolerant oncogenic activation is likely to evolve into a malignant phenotype. Submission of cells to a chemical peroxide (TBHP) similarly generates a halt to proliferation of cells. After three weeks, cells again begin to proliferate (noted after Ech-TBHP).

The status of the TP53 gene is wild in the Ech-Inf cells and the signaling pathway remains inducible in response to genotoxic stress. The analysis of the expression of the members of the three main families Snail, Twist and Zeb was analyzed. Invariably, on three independent experiments, the oxidative stress escape induced by inflammation or chemical peroxide is associated with an induction of the ZEB1 or ZEB2 gene (Figure 2A), suggesting that ZEB transcription factors orchestrate the mechanism of adaptation to oxidative stress. Consistent with this hypothesis, ectopic production of murine ZEB 1 protein or the related murine ZEB2 protein (HCEC-ZeW and HCEC-Zeb2 lines) gives HCEC cells a proliferative advantage under oxidative stress conditions (induced either by chronic inflammation or TBHP treatment) (Figure 2B).

Example 3 The ZEB1 and ZEB2 proteins induce a specific gene program that gives the cells resistance to oxidative stress.

The comparison of the gene expression profiles of the Ech-Inf, Ech-TBHP, HCEC-Zebl and HCEC-Zeb2 lines to the parental line (HCEC-Par) made it possible to highlight the activation of a common gene program constituted by the activation of about 30 genes and the repression of about thirty others (Figure 3A, activated genes listed in Figure 13). This program called ALEP (Alternative to EMT Program, an alternative to the EMT program commonly induced by these transcription factors) does not include any of the known detoxification genes. The implication in granting this survival advantage of the two most induced genes of this program, namely QPRT and APOE, was tested. The ectopic expression of either, with superior QPRT efficiency, effectively confers survival benefit to cells under conditions of chronic inflammation or when subjected to chemical peroxide (Figure 3B). . This observation confirms that several components of this program contribute to providing survival and / or proliferation benefit to colon epithelial cells under these stress conditions. Example 4 Activation of the alternative program results in a dependence of the cells on the QPRT protein for their proliferation or survival.

QPRT protein (for quinolinate phosphoribosyl transferase) is an enzyme originally described as being involved in the kynurenine pathway (Liu et al., 2007). Although its contribution to the de novo production of NAD ⁺ is effective in the liver and kidney, the QPRT protein is also present in other cell types, presumably providing other functions, including inhibition of the activation of NAD ^+. caspase 3 as recently described in HeLa cells (Ishidoh et al., 2010). Depletion of HCEC Esc cells into QPRT induces apoptotic cell death under conditions of chronic inflammation (Figure 4A). Interestingly, inhibition of QPRT expression also prevents these cells from tolerating oncogenic activations (Figure 4B). The comparison of the alternative program with the expression profiles of the 55 colorectal cancer lines of the Cancer Cell Line Encyclopedia (CCLE) database (data GSE36133) (Barretina et al., 2012) highlights in a certain number of lines a joint expression several of the genes induced by the ALEP program (no joint expression of the repressed genes). The set of activated genes from the ALEP program was used as a signature, the lines were ranked according to their score and their QPRT addiction was evaluated (Figure 5). Of all the lines analyzed, only the cells having an activated alternative program (high signature score) seem to be dependent on the QPRT protein for their proliferation, judging by a crystal violet coloration (FIG. proliferation curve analysis (results not shown). All of this data demonstrates the biological relevance of the ALEP signature and highlights its interest in predicting QPRT dependence.

The ALEP signature score is parallel to the level of QPRT protein detected in Western blot (Figure 6). However, although the program is not active in the COLO-320 line, the QPRT protein is detectable at a lower rate. In immunohistochemistry (IHC), the signal detected in this line is sufficiently high to be considered as weakly positive.

Inhibition of the QPRT protein induces proliferation arrest of the ALEP-positive colorectal lines as judged by the accumulation of phosphorylated H3 histone on the Ser10 residue. This arrest is accompanied by an increase in autophagy (visualized by the accumulation of LC3 protein) and the accumulation of damage to DNA (visualized by the accumulation of γΗ2ΑΧ). Inactivation of the TD02 enzyme upstream of the QPRT protein in the kynurenine pathway has similar consequences (Figure 7). ALEP-positive colorectal cells are therefore dependent on the activity of the kynurenine pathway to maintain their proliferation.

EXAMPLE 5 The expression of QPRT is induced in the early phases of the development of colorectal cancers, developed on an inflammatory context or not.

The analysis of IHC QPRT expression in a cohort of patients with chronic bowel disease shows an absence of the protein in the enterocytes of normal tissue (unlike lamina propria and cell fibroblasts). infiltrated immune cells that strongly express the protein), an induction of its expression limited to a few epithelial cells located at the bottom of crypts in inflamed mucosa (a place where intestinal stem cells are located) and an increase in the number of labeled epithelial cells in dysplasias low-grade (45.4%, n = 22), high-grade (18.2%, n = 22) and cancers (38.1%>, n = 21) developed in an inflammatory context (Figure 8). Similarly, the expression of QPRT is induced in epithelial cells adenomas (weakly or strongly dysplastic, 50%, n = 20) and sporadic cancers (30%, n = 10) (Figure 9).

Example 6: The ALEP program is induced in intestinal stem cells and is a factor of poor prognosis.

The gene signature of the ALEP program was established on an epithelial cell model. To refine this signature so that it can be used on tumors (taking into account their cellular heterogeneity), it has been compared to two cohorts of patients suffering from ulcerative colitis (UC) or Crohn's disease (CD) since over seven years (datasets GSE36807 and GSE37283, (Montero-Melendez et al., 2013, Pekow et al., 2013)). These two series were selected because the gene profiles were established on macroscopically non-inflamed mucosa so as to limit the direct impact of inflammation on transcript levels. The analysis of the score of this signature allows in both cohorts of patients to clearly segregate two subgroups (Figure 10 A-B). In particular, in the second series of patients, it makes it possible to distinguish, from non-neoplastic lesions, those who have developed dysplasias or colorectal carcinomas (FIG. 10B). The signature score of the ALEP program thus reflects the progress of the disease. QPRT analyzes by IHC, however, reveal a number of patients with positive low-grade dysplasias and high-grade negative dysplasias. The detection of QPRT does not seem to be a factor of poor prognosis in itself.

The distribution of the ALEP signature was analyzed in a third cohort of patients (n = 466) who exclusively developed colorectal carcinomas in a non-inflammatory setting (data GSE39582) (Marisa et al., 2013). These tumors have recently been classified on the basis of their gene profile into 6 categories (from C1 to C6). The analysis of the ALEP signature highlights activation of the program in a number of tumors (18%>, n = 566), regardless of their molecular subtype (Figure 11A). There is a higher proportion of positive tumors in the C4 subgroup, described to be enriched in stem cells (Marisa et al., 2013). In this cohort, the ALEP signature score and that of the Lgr5 and EphB2 intestinal stem cell signatures (Merlos-Suarez et al., 2011) correlate significantly, supporting the hypothesis that ALEP is induced in stem cells. intestinal precursors (Figure 11B). As previously demonstrated for the signatures EphB2 and Lgr5 (Merlos-Suarez et al., 2011), and demonstrated for the C4 subgroup (Marisa et al., 2013), the ALEP signature proves to be predictive of a risk of high recurrence (Figure 11 C).

In conclusion, this analysis highlights a group of distinct patients, with a high risk of recurrence characterized by intestinal strains signatures and high ALEP program scores. Example 7: Extension of the observations to a second model of tumors developed on an inflammatory context.

Esophageal adenocarcinomas originate from Barrett's esophagus (located in the lower third of the esophagus) and chronic inflammation related to the rise of gastric juices plays a key role. QPRT analysis by immunohistochemistry shows that a proportion of the adenocarcinomas tested express the protein QPRT (20%, n = 10) suggesting that the induction of the program can in this context also play a key role in tumor initiation. No labeling is observed in the squamous epithelium, used as an internal control (Figure 11). Conclusions: These findings highlight the potential value of QPRT as a therapeutic target. These results also highlight a new program induced in gastrointestinal cancers, most likely within stem cells. Beyond its predictive value in terms of risk of recurrence, the interest of this signature is to constitute a new tool to identify tumors that could be targeted by QPRT inhibitors.

BIBLIOGRAPHIC REFERENCES

Baker, B.W., et al. (1992). Nucleic Acids Res. 20, 5234.

Barretina, J., Et al. (2012). Nature 483, 603-607.

Boehm, J.S., et al. (2005). Mol. Cell Biol. 25, 6464-6474.

Burns, J.C. et al. (1993). Proc. Natl. Acad. Sci. U.S.A. 90, 8033-8037.

Dimri, G. P., et al. (1995). Proc. Natl. Acad. Sci. U.S. 92, 9363-9367.

Fat, B, et al. (2014). PLoS. One. 9, e92254.

Irizarry, R.A., et al. (2003). Biostatistics. 4, 249-264.

Ishidoh, K., Et al. (2010). Biochim. Biophys. Acta 1803, 527-533.

Liu, H., et al. (2007). J. Mol. Biol. 373, 755-763.

Malik, S.S., et al. (2014) Proteins 82, 405-14.

Marisa, L., et al. (2013). PLoS. Med. 10, el001453.

Merlos-Suarez, A., et al. (2011). Cell Stem Cell 8, 511-524.

Montero-Melendez, T. et al. (2013). PLoS. One. 8, e76235.

MoreLA.P., Et al. (2012). PLoS. Broom. 8, e 1002723.

Pekow, J. et al. (2013) Inflamm. Bowel. Dis. 19, 461-470.

Sahm, F., et al. (2013) Cancer Res. 3225-34.

Sayan, A.E., et al. (2009). Proc. Natl. Acad. Sci. U.S. A 106, 14884-14889.

Vallerini, G.P., et al. (2013). J. Med. Chem. 56, 9482-95.

Vecsei, L., et al (2013). Nat. Rev. Drud Discov. 12, 64-82.

Watanabe, T., et al (2013) Clin Can. Res. 13, 415-420.

Zufferey, R., et al. (1997). Nat. Biotechnol. 15, 871-875.

Claims

An ex vivo method for selecting a patient suffering from a gastrointestinal tumor that may benefit from an inhibitory treatment of the kynurenine pathway from a test sample of said tumor prior to treatment of said patient, said method comprising the following steps :

The method of claim 1, wherein said test sample of said tumor is derived from the primary tumor.

The method according to one of claims 1 to 2, wherein the determination of QPRT expression comprises at least one immunohistochemistry step.

The method according to one of claims 2 to 3, wherein determining QPRT expression comprises locating labeled cells within the epithelial tissue of the test sample.

The method of one of claims 2 to 4 wherein the determination of QPRT expression comprises determining the percentage of labeled epithelial cells.

The method according to one of claims 1 to 5, wherein said gastrointestinal tumor is selected from colorectal cancer, gastric cancer and adenocarcinoma of the esophagus.

Method according to one of claims 1 to 6, wherein said gastrointestinal tumor is developed in a chronic inflammatory context.

The method of claim 7, wherein said gastrointestinal tumor is colorectal cancer developed in a chronic inflammatory context.

The method according to one of claims 1 to 6, wherein said gastrointestinal tumor is sporadic colorectal cancer.

The method according to one of claims 1 to 9, wherein said inhibitory treatment of the kynurenine pathway is selected from a QPRT inhibitor treatment and a TD02 inhibitor treatment.

11. Use of QPRT as a biomarker for predicting the sensitivity of a gastrointestinal tumor patient to an inhibitory treatment of the kynurenine pathway, said patient being untreated for said tumor and the expressing the QPRT protein being measured in the epithelial cells of a sample of said tumor.

The use of claim 11, wherein said kynurenine pathway inhibitory treatment is selected from a QPRT inhibitor treatment and a TD02 inhibitor treatment.