WO2005062055A2

WO2005062055A2 - Methods for detecting neoplasia and markers thereof

Info

Publication number: WO2005062055A2
Application number: PCT/GB2004/005430
Authority: WO
Inventors: Tatjana Crnogorac-Jurcevic; Nick Lemoine
Original assignee: Cancer Research Technology Ltd
Current assignee: Cancer Research Technology Ltd
Priority date: 2003-12-24
Filing date: 2004-12-23
Publication date: 2005-07-07
Anticipated expiration: 2006-06-24
Also published as: WO2005062055A3; GB0329997D0

Abstract

A method of detecting and/or diagnosing a cancerous condition or predisposition to a cancerous condition in a patient is disclosed, the method comprising the step of detecting, in a sample taken from the patient, the presence of a marker selected from the group consisting of (i) SPAG1; (ii) MNK; (iii) Periostin; (iv) AGR2; or (v) UHRF1, wherein said marker is a marker polypeptide, nucleic acid or anti-marker antibody, or a fragment of said marker polypeptide, nucleic acid or anti-marker antibody.

Description

Methods for Detecting Neoplasia and Markers thereof

Field of the Invention

The present invention relates to the detection, diagnosis and treatment of cancerous conditions (e.g. neoplasia) and particularly, although not exclusively, to markers for use in the detection, diagnosis and treatment of cancerous conditions of the pancreas.

Background

Pancreatic cancer is a prevalent form of disease for which efficacious diagnosis and treatments are desired.

In the United States alone, there are up to 30,300 newly diagnosed cases of pancreatic cancer per year and an almost equal number of deaths per year from pancreatic cancers³⁶. In fact pancreatic cancer (PC) is now the fourth leading cause of cancer-related death in the USA for both men and women³⁷. The survival prognosis is low with only 19% of patients with cancer of the exocrine pancreas surviving more than one year after diagnosis.

Problems associated with pancreatic cancer include the inaccessibility of the organ, a lack of effective diagnostic tools and a poor response to currently available therapies³⁸. The late stage at which the disease is usually diagnosed is a reason that surgical resection is possible in less than 20% of patients, moreover, a surgical approach is itself inherently dangerous with survival rates following surgical intervention only about 40%³⁶. Furthermore, current chemotherapeutics are beneficial to less than 20% of patients with pancreatic adenocarcinoma .

Problems associated with the treatment of pancreatic cancer include the inaccessibility of the organ, a lack of effective diagnostic tools and a poor response to currently available therapies³⁸. Another problem is the late stage at which detection of the disease is normally made meaning that surgical resection is often the only option for treatment. Not only is a surgical approach inherently dangerous but survival rates following surgical intervention are only about 40%³⁶.

Early stage detection of PC is rarely made and the most frequently recognised clinical symptoms of PC - unexplained weight loss, nausea, diarrhoea, weakness, jaundice and upper abdominal and back pain - are indicators of an advanced stage of the disease. PC detection often relies on imaging techniques, e.g. ultrasonography, but visualisation is often impaired by the presence of intestinal gas. Computer Tomography can be used to detect lesions of about 1cm diameter. Owing to the fact that PC is characterised by early vascular dissemination and spread to regional lymph nodes, early diagnosis of PC will be of benefit in providing a successful treatment.

To date, the only well characterised biomarker expressed by cancer cells of the pancreas is the oncofetal antigen, CA19-9 (Carbohydrate Antigen 19.9). This marker is found to be elevated in the blood of 50-75% of PC patients, but is not detectable until a late stage of tumour progression. Additionally, CA19-9 is found in patients with non-neoplastic disease, inflammatory disease of the bowel, cirrhosis and autoimmune conditions and is thus not a specific marker of PC.

SPAGl

Sperm fertility related protein 1, SPAGl, (also called infertility-related sperm protein, Sperm-associated Antigen-1, SPAGl (SP75, TPIS, HSD-3.8, FLJ32920)) was identified in a rare form of infertility where anti-SPAGl antibodies derived from an infertile woman were reported to cause sperm agglutination and/or immobilisation. Expression of SPAGl is limited to the testis and particularly to germ cells at the early stage of spermatogenesis³⁹' ⁰. SPAGl has been reported to be located at the cell surface.

SPAGl protein localizes to the plasma membrane of germ cells in the testis and to the post-acrosomal plasma membrane of mature spermatozoa. Recombinant polypeptide binds GTP and exhibits GTPase activity. Thus, this protein may regulate GTP signal transduction pathways involved in spermatogenesis and fertilization. Two transcript variants of this gene encode the same protein.

Sequence data for SPAGl can be found in the NCBI database (http://www.ncbi.nlm.nih.gov) under accession number AF311312. The SPAGl mRNA is 3818bp in length and encodes a protein of 926 amino acids containing 3 tetratricopeptide (TPR) motifs that direct protein- protein interactions, an ATP/GTP binding site and putatative phosphorylation sites for PKC, CK2 and cA P/cGMP-dependent kinases. To date, SPAGl has not been described in pancreatic adenocarcinoma .

Menkes Disease protein, MNK (also called ATP7A)

The human X-linked recessive copper deficiency disorder, Menkes disease, is caused by mutations in the ATP7A (MNK) gene, which encodes a transmembrane copper-transporting P-type ATPase (MNK) . The MNK protein is localised to the Golgi apparatus and relocalises to the plasma membrane when copper levels are elevated.

Menkes disease is a fatal neurodegenerative disorder resulting in growth retardation in the first few weeks after birth which leads to death in early childhood, commonly in the first year.

The disease is the result of mutations in the ATP7A gene (herein referred to as Menkes disease protein or MNK) encoding a P-type transmembrane ATPase that translocates copper ions across infracellular membranes of compartments along the secretory pathway. ATP7A moves from the trans-Golgi network (TGN) to the cell surface in response to exogenously added copper ions and recycles back to the TGN upon copper removal³⁵. The protein contains a C-terminal di-leucine motif necessary for internalisation from the cell surface. The gene encoding the human Menkes disease protein is found at chromosome position Xq21.1 and encodes an mRNA of 8478bp which in turn encodes a 1500 amino acid full-length polypeptide. The human MNK sequence data can be found in the NCBI database under accession numbers NM_000052 and Q04656. Six isoforms of human MNK have been reported. To date, MNK protein has not been described in pancreatic cancer. Mutations in the MNK gene are known to cause growth retardation. Rats and hamsters provided with a copper deficient diet develop pancreatic atrophy after four weeks (mainly due to acinar cell depletion which is up to 90% after eight weeks) with an increase in the number of adipocytes, interstitial and ductular cells. The islets and ducts remained intact. This resembles the putative early events in the evolution of pancreatic cancer. Eight weeks after reintroduction of copper to the diet hepatocyte differentiation began within the ducts, periductally and around the islets. Occasional endocrine cells were also found interspersed within the ducts suggesting that they can act as a stem-cell equivalent .

Copper uptake is normally via the intestine and interaction with albumin in the blood. Storage or biliary excretion may then occur in the liver. Copper is also found bound to ceruloplasmin in the blood and bound to metallothioneins in the tissues.

MNK mutations have an effect on some copper dependent enzymes such as cytochrome oxidase (involved in electron transport) , lysis oxidase (involved in crosslinking collagen and elastin in the skin and vasculature) and super oxide dismutase (involved in detoxification of free-radicals) .

Periostin (also called OSF-2) The periostin gene encodes an 811-amino acid, glycosylated, secreted protein that was originally isolated from osteoblasts and functions as a cell adhesion molecule for preosteoblasts and is thought to. play a role in osteoblast recruitment, attachment and spreading .

Expression of periostin is a prognostic marker in non- small cell lung cancers (NSCLC) . Periostin is highly expressed at the tumour periphery of lung cancer tissue but not within the tumour by in situ RNA hybridization, suggesting that expression of periostin may be involved in the process of tumour invasion. NSCLC patients with periostin expression had significantly poorer survival than the patients without periostin expression¹⁰⁷.

Periostin is overexpressed in both ovarian tumours and cultured ovarian cell lines¹⁰⁸. Furthermore, periostin secreted by epithelial ovarian carcinoma is a ligand for alpha (V) beta (3 ) and alpha (V) beta (5) integrins and promotes cell motility¹⁰⁹.

Undetectable in normal human breast tissues, periostin was found to be overexpressed by the vast majority of human primary breast cancers examined. Tumour cell lines engineered to overexpress periostin showed a phenotype of accelerated growth and angiogenesis as xenografts in immunocompromised animals. The underlying mechanism of periostin-mediated induction of angiogenesis was found to derive in part from the up-regulation of the vascular endothelial growth factor receptor Flk-1/KDR by endothelial cells through an integrin alpha (v) beta (3) - focal adhesion kinase-mediated signaling pathway¹¹⁰. Acquired expression of periostin by colon cancer cells greatly promoted metastatic development of colon tumours. Periostin is overexpressed in more than 80% of human colon cancers examined with highest expression in metastatic tumours. Periostin expression dramatically enhanced metastatic growth of colon cancer by both preventing stress-induced apoptosis in the cancer cells and augmenting endothelial cell survival to promote angiogenesis¹¹¹. O03016471A3 relates to human periostin polypeptides and DNA sequences encoding them, to human periostin specific antibodies, diagnostic assays for metastasis of breast cancer to bone and preeclempsia .

Human anterior gradient-2, AGR2 (also called AG2; hAG-2; XAG-2) hAG-2 and hAG-3 are human homologues of the secreted Xenopus laevis proteins XAG-1/2 (AGR-1/2) that are expressed in the cement gland, an ectodermal organ in the head associated with anteroposterior fate determination during early development¹¹⁵. The roles of hAG-2 and hAG-3 in mammalian cells are unknown at present.

Coexpression of AGR2 with the estrogen receptor has been observed in breast cancer cell lines¹¹². hAG-2 and hAG-3, are associated with estrogen receptor-positive breast tumours and interact with metastasis gene C4.4a and dystroglycan (DAG-1)¹¹³.

Inverted CCAAT box Binding Protein of 90 kDa, ICBP90 (also called UHRFl; Np95; RNF106; FLJ21925) UHRF1 (also called ICBP90) is a nuclear protein that binds to one of the inverted CCAAT boxes of the topoisomerase II alpha gene promoter.

UHRF1 expression is altered in cancer cell lines and is upregulated by E2F-1 overexpression with an efficiency depending on the cancer status of the cell line¹¹⁴.

ICBP90 has not been described in pancreatic cancer before .

The problem of inability to detect pancreatic cancers at an early stage means that treatment of the disease begins when the cancer is well established thus reducing the likelihood -of successful treatment.

Summary of the Invention

The inventors have found that the SPAGl gene is upregulated in the development of pancreatic cancer and is present in cancers of the breast, uterus, stomach, lung and kidney. The inventors have confirmed the upregulation of SPAGl by quantitative RT-PCR and immunohistochemistry in early pancreatic lesions called PanlNs and pancreatic cancer tissues. A proportion of pancreatic cancer cell lines were also found to overexpress SPAGl by analysis of cDNA microarrays. The inventors also confirmed, by Northern blot and RT-PCR analysis, that normal pancreatic tissue does not express SPAGl.

The lack of expression of SPAGl protein or mRNA or upregulation of the SPAGl gene or mRNA in non-germ line cells, coupled with the finding that the SPAGl gene is not upregulated or expressed in late stage, chronic, pancreatitis, means that detection of the SPAGl gene, mRNA transcript, polypeptide, protein or protein precursor can be used to provide a specific marker for any stage, e.g. the early or late stages, of pancreatic cancer .

SPAGl can therefore be used to assist in the diagnosis of cancer in several tissues including the diagnosis of early stage pancreatic cancer. Localisation of SPAGl nucleic acid or polypeptide to cancer cells also provides a selectable marker for the targeting of chemotherapeutic agents to tumour cells.

The inventors have also found that MNK protein is upregulated in pancreatic cancer but is not present in normal pancreatic cells or chronic pancreatic tumour cells. Thus, detection of the MNK gene, mRNA transcript, polypeptide, protein or protein precursor can be used to provide a specific marker for any stage, e.g. the early or late stages, of pancreatic cancer.

MNK can therefore also be used to assist in the diagnosis of pancreatic cancer including the diagnosis of early stage pancreatic cancer. Local upregulation of MNK nucleic acid or polypeptide in cancer cells also provides a selectable marker for the targeting of chemotherapeutic agents to tumour cells.

In addition to SPAGl and MNK, the inventors have also found Periostin, AGR2 and UHRF1 to be upregulated in pancreatic cancer and in the development of pancreatic cancer, but not upregulated in normal pancreatic tissue. This has been confirmed through analysis of oligonucleotide and tissue arrays, through QRT-PCR, SAGE and immunohistochemical analyses .

Accordingly, in addition to SPAGl and MNK, periostin, AGR2 and UHRF1 have also been identified as markers useful in the detection and diagnosis of cancerous conditions of any kind at an early or late stage. In particular, these five markers are indicated to have utility in the detection and diagnosis of cancerous conditions of the pancreas and may be useful in detecting early or late stage pancreatic cancerous conditions .

Detection of one of the perisotin, AGR2 or UHRF1 gene, mRNA transcript, polypeptide, protein or protein precursor can be used to provide a specific marker for any stage, e.g. the early or late stages, of pancreatic cancer .

As periostin and AGR2 are normally secreted proteins, these markers may have particular utility as serum or blood-borne markers which may be detected in a patient blood sample or blood-derived sample. That detection may provide an indication of the presence of a cancerous condition .

At its most general, the present invention relates to the detection of a cancerous condition or predisposition to a cancerous condition.

This may comprise a method of detecting and/or diagnosing a cancerous condition (e.g. neoplasm) or predisposition to a cancerous condition in an individual comprising detecting a marker in a sample taken from said individual .

The present invention also encompasses the identification of a SPAGl and/or MNK and/or periostin and/or AGR2 and/or UHRF1 nucleic acid (i.e. DNA or RNA), peptide, polypeptide or protein as markers of a cancerous condition (e.g. neoplasia).

According to one aspect of the present invention there is provided a method of detecting and/or diagnosing a cancerous condition, or predisposition to a cancerous condition, in an individual, said method comprising the step of detecting the presence of a marker of said cancerous condition in a sample taken from said individual .

Preferably, the method is a diagnostic method or assay performed in vitro. The method may detect the expression or upregulation of expression of a marker polypeptide or nucleic acid (e.g. mRNA).

The method of detecting a cancerous condition may further comprise the step of obtaining a sample from said individual .

Preferably, the detecting step comprises detecting an increase in the amount of said marker in the sample relative to a control sample, e.g. from a corresponding individual not having a cancerous condition. The increased amount of the marker in the sample may reflect an increase or up-regulation of expression of the marker. The sample may comprise or may be derived from: a quantity of blood; a quantity of serum derived from the individual's blood which may comprise the fluid portion of the blood obtained after removal of the fibrin clot and blood cells; a quantity of pancreatic juice; a tissue sample or biopsy; or cells isolated from said individual.

The marker is preferably one of: a polypeptide; a polypeptide fragment (e.g. a peptide); a nucleic acid which may be a DNA or RNA; an mRNA; the amplification of genomic DNA; an antibody; or an antibody fragment. The antibody or antibody fragment may comprise a secondary marker in that the antibody or antibody fragment has been generated by the immune response of the individual to expression of a primary marker, e.g. polypeptide or polypeptide fragment, at the site of the cancerous condition .

In aspects of the present invention the marker is more preferably selected from the group consisting of:

(I) a SPAGl polypeptide or fragment thereof; an mRNA encoding a SPAGl polypeptide or fragment thereof; the amplification of all or a portion of the genomic DNA encoding a SPAGl polypeptide; and an antibody or antibody fragment recognising: (i) a SPAGl polypeptide; and/or (ii) a fragment of a SPAGl polypeptide;

(II) an MNK polypeptide or fragment thereof; an mRNA encoding an MNK polypeptide or fragment thereof; the amplification of all or a portion of the genomic DNA encoding an MNK polypeptide; and an antibody or antibody fragment recognising: (i) an MNK polypeptide; and/or (ii) a fragment of an MNK polypeptide; (III) a periostin polypeptide or fragment thereof; an mRNA encoding a periostin polypeptide or fragment thereof; the amplification of all or a portion of the genomic DNA encoding a periostin polypeptide; and an antibody or antibody fragment recognising: (i) a periostin polypeptide; and/or (ii) a fragment of a periostin polypeptide; (IV) an AGR2 polypeptide or fragment thereof; an mRNA encoding an AGR2 polypeptide or fragment thereof; the amplification of all or a portion of the genomic DNA encoding an AGR2 polypeptide; and an antibody or antibody fragment recognising: (i) an AGR2 polypeptide; and/or (ii) a fragment of an AGR2 polypeptide; or

(V) an UHRFl polypeptide or fragment thereof; an mRNA encoding an UHRFl polypeptide or fragment thereof; the amplification of all or a portion of the genomic DNA encoding an UHRFl polypeptide; and an antibody or antibody fragment recognising: (i) an UHRFl polypeptide; and/or (ii) a fragment of an UHRFl polypeptide.

Preferably, the method comprises detecting the presence at least one of a SPAGl, MNK, periostin, AGR2 or UHRFl polypeptide, or fragment thereof. Such detection may involve the step of contacting an antibody or antibody fragment capable of recognising said polypeptide, or fragment thereof, with said sample.

The sample may comprise a tissue sample or cells isolated from said individual and said detecting step may then comprise contacting an antibody or antibody fragment capable of recognising a SPAGl, MNK, periostin, AGR2 or UHRFl polypeptide, or fragment thereof, with said sample to detect the presence of SPAGl, MNK, periostin, AGR2 or UHRFl polypeptide, or fragment thereof. The SPAGl, MNK, periostin, AGR2 or UHRFl detected may be present on the surface of a cell of said sample

The detection step may comprise an immunological detection, e.g. western blot or immuno-blot, ELISA or RT- PCR, which may be quantitative RT-PCR.

In another aspect of the present invention there is provided a method of assessing the effectiveness of treatment of a cancerous condition by analysing the presence of a marker in a patient sample. The marker may be a polypeptide, a nucleic acid encoding a marker polypeptide, an antibody or a fragment of said polypeptide, nucleic acid or antibody.

The analysis of the presence of the marker may comprise monitoring the presence or absence of a marker polypeptide or nucleic acid in a sample taken from an individual. This may involve monitoring the expression of a marker polypeptide or nucleic acid in the sample or in specific components of the sample. The analysis may comprise measuring the amount of marker polypeptide or nucleic acid in the sample. This may comprise quantitative measurement of the amount of marker polypeptide or nucleic acid which is present or is expressed, e.g. using quantitative techniques such as quantitative western blot, QRT-PCR or quantitative northern blot. Alternatively the analysis may comprise a qualitative analysis, e.g. by monitoring the continued presence or absence of marker polypeptide or nucleic acid by microscopy, e.g. using immunohistochemical staining.

The assessment of effectiveness of treatment may be made by monitoring the change in the presence of marker polypeptide or nucleic acid, e.g. by monitoring a decrease or increase in marker polypeptide or nucleic acid presence or expression.

In other aspects of the present invention methods are provided for: (i) monitoring a cancerous condition in a patient; (ii) predicting a patient's response to therapeutic treatment of a cancerous condition; (iii) assessing the prognosis of a patient having a cancerous condition; (iv) screening a patient for the presence of a cancerous condition; and/or (v) determining the susceptibility of a patient to developing or contracting a cancerous condition; (vi) monitoring the response of a patient to therapeutic treatment for the cancerous condition, e.g. monitoring response to treatment with a selected drug, wherein said method may comprise detecting the presence of a marker polypeptide or a nucleic acid encoding a marker polypeptide, or a fragment of said polypeptide or nucleic acid in a sample taken from the patient.

In aspects of the present invention detection of a marker polypeptide or nucleic acid may be performed by any appropriate means available to the person skilled in the art .

One way of detecting the presence of a marker polypeptide or nucleic acid is by using marker specific antibodies which may be monoclonal or polyclonal antibodies. These may be used to detect the presence or expression of a marker polypeptide,- or marker epitope forming part of a marker polypeptide by assays involving binding between a marker polypeptide or nucleic acid and the anti-marker antibody. Suitable antibodies may be labelled with other compounds to aid visualisation, e.g. using fluorescent or radio- labels or secondary antibodies linked to similar labelling compounds. In one arrangement, detection of marker polypeptide may be by western blotting techniques using anti-marker antibodies . Quantitative western blot using radio-labelled antibodies may be used to determine a relative amount of a marker polypeptide or nucleic acid in a sample.

One way of detecting a marker nucleic acid is by hybridisation of a nucleic acid (DNA or RNA) probe to a marker nucleic acid under conditions of selected stringency. In one arrangement nucleic acid probes may be constructed to comprise nucleic acid sequences predetermined to have a high degree of sequence complementarity with a selected portion of a marker nucleic acid such that the probe will hybridise with the marker nucleic acid under high or very high stringency conditions, preferably under very high stringency conditions .

Accordingly, probe hybridisation may be used to detect a marker nucleic acid in a sample, e.g. by northern blot (hybridising to mRNA) or southern blot (hybridising to DNA) . Quantitative blotting techniques may be used to provide a relative assessment of the amount of marker nucleic acid detected. The quantitative assessment may be used to assess the change in marker presence or expression and/or to indicate a change in the state of the cancerous condition, e.g. an increase or decrease in the malignancy of the cancer.

Methods according to the present invention may be performed in vitro or in vivo. The term "in vitro" is intended to encompass experiments with cells in culture or with ex vivo cell or tissue samples whereas the term "in vivo" is intended to encompass experiments with intact multi-cellular organisms. Methods of detection, diagnosis and assessment of treatment effectiveness are preferably performed in vitro.

In another aspect of the present invention there is provided an assay kit for use in detecting and/or diagnosing a cancerous condition, or predisposition to a cancerous condition, in an individual, the kit comprising (i) an antibody capable of recognising a SPAGl, MNK, periostin, AGR2 or UHRFl polypeptide, or fragment thereof; and/or (ii) a nucleic acid capable of binding to a SPAGl, MNK, periostin, AGR2 or UHRFl nucleic acid under conditions of high or very high stringency. A SPAGl, MNK, periostin, AGR2 or UHRFl nucleic acid may be a nucleic acid encoding one of SPAGl, MNK, periostin, AGR2 or UHRFl polypeptides, a sequence complementary thereto or a fragment of such sequences.

In another aspect of the present invention there is provided a substance which binds to a SPAGl, MNK, periostin, AGR2 or UHRFl polypeptide, or fragment thereof, for use in the treatment of a cancerous condition .

According to another aspect of the present invention there is provided an antibody or antibody fragment which recognises at least one of a SPAGl, MNK, periostin, AGR2 and/or UHRFl polypeptide, or fragment thereof, for use in the treatment of a cancerous condition.

In another aspect of the present invention there is provided an anti-sense nucleic acid which hybridises to a DNA or RNA encoding a SPAGl, MNK, periostin, AGR2 or UHRFl polypeptide, or fragment thereof, under intermediate, high or very high stringency conditions, the anti-sense nucleic acid provided for use in the treatment of a cancerous condition.

In another aspect of the present invention there is provided a modulator of an activity of a SPAGl, MNK, periostin, AGR2 or UHRFl protein/polypeptide, or fragment thereof, for use in the treatment of a cancerous condition . In another aspect of the present invention there is provided the use of an antibody or antibody fragment capable of recognising at least one of a SPAGl, MNK, periostin, AGR2 and/or UHRFl polypeptide, or fragment thereof, in the manufacture of a medicament for the treatment of a cancerous condition.

In one preferred arrangement, the antibody or antibody fragment recognises a SPAGl polypeptide, or fragment thereof, and may be provided for use in the treatment of a cancerous condition (e.g. neoplasm) in male or female individuals .

The substance, antibody, antibody fragment, anti-sense nucleic acid or modulator may be linked to an anti-tumour compound. The linkage may be by conjugation.

Pharmaceutical compositions or medicaments for use in the treatment of a cancerous condition comprising one or more of said substance, antibody, antibody fragment, anti- sense nucleic acid or modulator may be provided and may further comprise a pharmaceutically acceptable carrier or adjuvant .

In another aspect of the present invention there is provided a method of treating a cancerous condition in an individual in need of such treatment comprising the step of administering to the individual a quantity of: (a) a substance that binds to a SPAGl, MNK, periostin, AGR2 or UHRFl polypeptide, or fragment thereof; or (b) an antibody or antibody fragment which recognises at least one of a SPAGl, MNK, periostin, AGR2 or UHRFl polypeptide, or fragment thereof; or (c) a SPAGl, MNK, periostin, AGR2 or UHRFl anti- sense nucleic acid; or (d) a modulator of an activity of a SPAGl, MNK, periostin, AGR2 or UHRFl protein/polypeptide; said quantity being effective to treat the cancerous condition or symptoms associated with the cancerous condition .

Inidividual /patient

In aspects of the present invention the individual may be any animal (e.g. non-human mammal) or human and is preferably a human patient.

The patient may be male or female.

As SPAGl is not normally expressed in female patients and detection of SPAGl in non-testis tissue is indicative of the presence of, or predisposition to, a cancerous condition, methods according to the present invention may be directed to detection of SPAGl in female patients.

Cancerous condition

The cancerous condition may be any unwanted cell proliferation (or any disease manifesting itself by unwanted cell proliferation) , neoplasm or tumour or increased risk of or predisposition to the unwanted cell proliferation, neoplasm or tumour. The cancerous condition may be a cancer and may be a benign or malignant cancer and may be primary or secondary (metastatic) . A neoplasm or tumour may be any abnormal growth or proliferation of cells and may be located in any tissue. Examples of tissues include the colon, pancreas, lung, breast, uterus, stomach, kidney, testis, central nervous system (including the brain), peripheral nervous system, skin, blood or lymph.

The cancerous condition is preferably an abnormal growth or tumour and is preferably located in the pancreas, lung, breast, uterus, stomach or kidney. Most preferably the cancerous condition is a pancreatic neoplasm or tumour being a form of, or involved in the development of, a pancreatic cancer.

The pancreatic neoplasm or tumour may be selected from the group consisting of: a Pancreatic Intraepithelial Neoplasia (PanIN) of any of type I to III; a Ductal Adenocarcinoma (DA); a Mucinous Cystic Neoplasm (MCN) ; an Intraductal Papillary Mucinous Neoplasm (IPMN); a Pancreacticoblastoma (PB) ; an Acinar Cell Carcinoma (ACC) ; a Solid Pseudopapillary Neoplasm (SPN); and a pancreatic adenocarcinoma (PDAC) .

The individual may be an animal or human, preferably a human patient in need of diagnosis or treatment. The individual may be a male or female.

Diagnosis

Detection of marker polypeptides or nucleic acids in accordance with the methods of the present invention may be used for the purpose of diagnosis of a cancerous condition in the patient, diagnosis of a predisposition to a cancerous condition or for providing a prognosis (prognosticating) of a cancerous condition. The diagnosis or prognosis may relate to an existing (previously diagnosed) cancerous condition, which may be benign or malignant, may relate to a suspected cancerous condition or may relate to the screening for cancerous conditions in the patient (which may be previously undiagnosed) .

Other diagnostic tests may be used in conjunction with those described here to enhance the accuracy of diagnosis or prognosis of a canσerous condition or to confirm a result obtained by using the tests described here.

The method of diagnosis may be an in vitro method performed on the patient sample, or following processing of the patient sample. Once the sample is collected, the patient is not required to be present for the in vitro method of diagnosis to be performed and therefore the method may be one which is not practised on the human or animal body.

Other diagnostic tests may be used in conjunction with those described here to enhance the accuracy of the diagnosis or prognosis or to confirm a result obtained by using the tests described here.

Sample types

A sample may be taken from any tissue or bodily fluid. The sample may comprise or may be derived from a tissue sample, biopsy or isolated cells from said individual.

The sample may comprise or may be derived from: a quantity of blood; a quantity of serum derived from the individual's blood which may comprise the fluid portion of the blood obtained after removal of the fibrin clot and blood cells; a quantity of pancreatic juice; a tissue sample or biopsy; or cells isolated from said individual.

The sample may comprise tissue or bodily fluid taken from the patients pancreas or pancreatic ducts.

In certain arrangements the sample may be a blood sample or blood-derived sample. The blood derived sample may be a selected fraction of a patient's blood, e.g. a selected cell-containing fraction or a plasma or serum fraction.

A selected cell-containing fraction of the blood may contain call types of interest which may include white blood cells ( BC) , e.g. peripheral blood mononuclear cells (PBC) and/or granulocytes, and/or red blood cells (RBC) .

A selected serum fraction may comprise the fluid portion of the blood obtained after removal of the fibrin clot and blood cells .

Antibodies and modulators

Antibodies and antibody fragments recognising a polypeptide or polypeptide fragment may bind to the polypeptide or polypeptide fragment. The bound complex may then be separately detectable by a further antibody or antibody fragment. Visualisation of binding reactions in vitro may be performed by linking a labelling compound to the antibody or antibody fragment, e.g. radio- or fluorescent label. A modulator of activity may be an inhibitor or an activator of a given polypeptide or protein activity, e.g. phosphorylation or dephosphorylation, binding or change of conformation.

Upregulation

Upregulation may relate to gene amplification, increased levels of a particular mRNA, polypeptide or protein.

Anti-tumour compounds

Anti-tumour compounds are well known to the person skilled in the art and include chemo-therapeutic compounds, e.g. sulphur based compounds such as vincristine or vinblastine. Linking an anti-tumour compound to a substance, antibody, antibody fragment, nucleic acid or modulator that is preferentially or selectively expressed in a cancerous tissue provides a means of targeting the anti-tumour compound to the cancerous tissue to provide effective therapy. In one arrangement, the anti-tumour compound may be a prodrug which is activated upon delivery to the neoplasm to an active form effective to treat the neoplasm.

Polypeptides, nucleic acids and fragments A polypeptide may be a protein in active form and conformation or may be a subunit of a protein made up of two or more subunits of the same or different type.

A fragment may comprise a nucleotide or amino acid sequence encoding a portion of the corresponding full length sequence. In this specification the corresponding full length sequence may be one of SEQ ID No.s 1 to 10. Said portion may be of defined length and may have a defined minimum and/or maximum length.

Accordingly, the fragment may comprise at least, i.e. have a minimum length of, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 25, 30, 40, 50, 60, 70, 80, 85, 90, 95, 96, 97, 98 or 99% of the corresponding full length sequence. The fragment may have a maximum length, i.e. be no longer than, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 25, 30, 40, 50, 60, 70, 80, 85, 90, 95, 96, 97, 98 or 99% of the corresponding full length sequence.

The fragment may comprise at least, i.e. have a minimum length of, 10 nucleotides or amino acids, more preferably at least 15, 20, 25, 30, 40, 50, 100, 150, 200, 300, 400, 500, 600, 700, 800, 900, 1000, 1100, 1200, 1300, 1400, 1500, 1600, 1700, 1800, 1900 2000, 2100, 2200, 2300, 2400, 2500, 2600, 2700, 2800, 2900, 3000, 3100, 3200, 3300, 3400, 3500, 3600, 3700, 3800, 3900 or 4000 nucleotides or amino acids.

The fragment may have a maximum length of, i.e. be no longer than, 10 nucleotides or amino acids, more preferably no longer than 15, 20, 25, 30, 40, 50, 100, 150, 200, 300, 400, 500, 600, 700, 800, 900, 1000, 1100, 1200, 1300, 1400, 1500, 1600, 1700, 1800, 1900 2000, 2100, 2200, 2300, 2400, 2500, 2600, 2700, 2800, 2900, 3000, 3100, 3200, 3300, 3400, 3500, 3600, 3700, 3800, 3900 or 4000 nucleotides or amino acids.

The fragment length may be anywhere between the said minimum and maximum length. Nucleic acids may include single or double-stranded DNA or RNA molecules, which may be siRNA' s, or chemically modified single or double-stranded DNA or RNA derivatives .

Anti-sense nucleic acid

By anti-sense nucleic acid is meant a nucleic acid having substantial sequence identity to the nucleic acid formed by the sequence of complementary bases to the single strand of a target nucleic acid. Thus, the anti-sense nucleic acid is useful in binding the target nucleic acid and may be used as an inhibitor to prevent or disrupt the normal activity or function, e.g. folding, binding, replication or transcription, of the target nucleic acid. The substantial sequence identity is preferably at least 50% sequence identity, more preferably at least 60, 70, 75, 80, 85, 90, 92, 94, 95, 96, 97, 98, 99 or 100% identity.

The anti-sense nucleic acid may be shorter or longer than the target nucleic acid. Identity of sequences is determined between the entire length of the shorter sequence when aligned with the complementary nucleic acid to obtain optimum base pairing.

The target nucleic acid may be a DNA or RNA encoding (i) a SPAGl polypeptide or fragment thereof; (ii) an MNK polypeptide or fragment thereof; (iii) a periostin polypeptide or fragment thereof; (iv) an AGR2 polypeptide or fragment thereof; or an UHRFl polypeptide or fragment thereof. The anti-sense nucleic acid may be a full-length nucleic acid, i.e. corresponding to the full-length of the target nucleic acid, or may be a fragment of the full-length nucleic acid.

For example, the human SPAGl mRNA has a full-length of 3818bp, and a corresponding full-length anti-sense nucleic acid would also comprise at least 3818bp. The human MNK mRNA comprises 8478bp and a corresponding full- length anti-sense nucleic acid would also comprise at least 8478bp. Fragments may comprise a nucleic acid having a given percentage length of the full-length target nucleic acid. Fragments may comprise a nucleic acid having a length which is any of 10, 20, 30, 40, 50, 60, 70, 80, 85, 90 or 95% of the full-length of the target nucleic acid sequence.

The anti-sense nucleic acid may be an siRNA of between 19 and 25 bp in length for use in RNAi based methods.

Hybridisation stringency

In accordance with the present invention, nucleic acid sequences may be identified by using hybridization and washing conditions of appropriate stringency.

Complementary nucleic acid sequences will hybridise to one another through Watson-Crick binding interactions. Sequences which are not 100% complementary may also hybridise but the strength of the hybridisation usually decreases with the decrease in complementarity. The strength of hybridisation can therefore be used to distinguish the degree of complementarity of sequences capable of binding to each other. The "stringency" of a hybridization reaction can be readily determined by a person skilled in the art.

The stringency of a given reaction may depend upon factors such as probe length, washing temperature, and salt concentration. Higher temperatures are generally required for proper annealing of long probes, while shorter probes may be annealed at lower temperatures. The higher the degree of desired complementarity between the probe and hybridisable sequence, the higher the relative temperature which can be used. As a result, it follows that higher relative temperatures would tend to make the reaction conditions more stringent, while lower temperatures less so.

For example, hybridizations may be performed, according to the method of Sambrook et al . , ("Molecular Cloning, A Laboratory Manual, Cold Spring Harbor Laboratory Press, 1989) using a hybridization solution comprising: 5X SSC, 5X Denhardt's reagent, 0.5-1.0% SDS, 100 μg/ml denatured, fragmented salmon sperm DNA, 0.05% sodium pyrophosphate and up to 50% formamide. Hybridization is carried out at 37-42°C for at least six hours. Following hybridization, filters are washed as follows: (1) 5 minutes at room temperature in 2X SSC and 1% SDS; (2) 15 minutes at room temperature in 2X SSC and 0.1% SDS; (3) 30 minutes-1 hour at 37°C in IX SSC and 1% SDS; (4) 2 hours at 42- 65°C in IX SSC and 1% SDS, changing the solution every 30 minutes .

One common formula for calculating the stringency conditions required to achieve hybridization between nucleic acid molecules is to calculate the melting temperature T_m (Sambrook et al . , 1989):

T_m = 81 . 5 ° C + 1 6. 6Log [Na + J + 0. 41 (% G+C) - 0. 63 (% formamide) - 600/n

where n is the number of bases in the oligonucleotide.

As an illustration of the above formula, using [Na+] = [0.368] and 50% formamide, with GC content of 42% and an average probe size of 200 bases, the T^ is 57°C. The T_m of a DNA duplex decreases by 1 - 1.5°C with every 1% decrease in sequence complementarity.

Accordingly, nucleotide sequences can be categorised by an ability to hybridise to a target sequence under different hybridisation and washing stringency conditions which can be selected by using the above equation. The T_π may be used to provide an indicator of the strength of the hybridisation.

The concept of distinguishing sequences based on the stringency of the conditions is well understood by the person skilled in the art and may be readily applied.

Sequences exhibiting 95-100% sequence complementarity are considered to hybridise under very high stringency conditions, sequences exhibiting 85-95% complementarity are considered to hybridise under high stringency conditions, sequences exhibiting 70-85% complementarity are considered to hybridise under intermediate stringency conditions, sequences exhibiting 60-70% complementarity are considered to hybridise under low stringency conditions and sequences exhibiting 50-60% complementarity are considered to hybridise under very low stringency conditions.

SPAGl

In this specification SPAG or SPAGl may relate to a SPAGl nucleic acid or polypeptide.

In this specification the terms ^ΛSPAG' or SPAGl' may relate to any SPAGl nucleic acid or polypeptide. Preferably SPAGl relates to human SPAGl or to an animal SPAGl homologue of human SPAGl, which may be a non-human mammalian homologue of human SPAGl.

T„he amino acid sequence (SEQ ID No.l) and nucleotide sequence (SEQ ID No.2) for the human SPAGl polypeptide and gene sequence are deposited in the NCBI database (http : //www. ncbi . nlm. nih . gov/) under accession number AF311312 [gi:10863767] (see Figure 25).

In this specification, a SPAGl nucleic acid may be any nucleic acid (DNA or RNA) having a nucleotide sequence having a specified degree of sequence identity to SEQ ID No.2, to an RNA transcript of SEQ ID No.2, to a fragment of SEQ ID No.2 or said RNA, or to the complementary sequence of any one of these sequences or fragments . Alternatively a SPAGl nucleic acid may be one that hybridises to one of these sequences under intermediate, high or very high stringency conditions.

In this specification, a SPAGl polypeptide may be any peptide, polypeptide or protein having an amino acid sequence having a specified degree of sequence identity to SEQ ID No.l or to a fragment of SEQ ID No .1.

The specified degree of sequence identity may be from at least 60% to 100% sequence identity. More preferably, the specified degree of sequence identity may be one of at least 65%, 70%, 75%, 80%, 85%, 86%, 87%, 88%, 89%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98% or 99% identity.

MNK

In this specification MNK may relate to a MNK nucleic acid or polypeptide.

In this specification the term ^ΛMNK' may relate to any MNK nucleic acid or polypeptide. Preferably MNK relates to human MNK or to an animal MNK homologue of human MNK, which may be a non-human mammalian homologue of human MNK.

The amino acid sequence (SEQ ID No.3) and nucleotide sequence (SEQ ID No .4 ) for the human MNK polypeptide and gene sequence are deposited in the NCBI database (http : //www. ncbi . nlm.nih . gov/) under accession number NM_000052 [gi:53986562] (see Figure 26).

In this specification, a MNK nucleic acid may be any nucleic acid (DNA or RNA) having a nucleotide sequence having a specified degree of sequence identity to SEQ ID No.4, to an RNA transcript of SEQ ID No.4, to a fragment of SEQ ID No.4 or said RNA, or to the complementary sequence of any one of these sequences or fragments. Alternatively a MNK nucleic acid may be one that hybridises to one of these sequence under intermediate, high or very high stringency conditions.

In this specification, a MNK polypeptide may be any peptide, polypeptide or protein having an amino acid sequence having a specified degree of sequence identity to SEQ ID No.3 or to a fragment of SEQ ID No .3.

Periostin

In this specification periostin may relate to a periostin nucleic acid or polypeptide.

In this specification the term ^periostin' may relate to any periostin nucleic acid or polypeptide. Preferably periostin relates to human periostin or to an animal periostin homologue of human periostin, which may be a non-human mammalian homologue of human periostin.

The amino acid sequence (SEQ ID No.5) and nucleotide sequence (SEQ ID No.6) for the human periostin polypeptide and gene sequence are deposited in the NCBI database (http : //www. ncbi .nlm.nih. gov/) under accession number NP_006466 (NM_006475 [gi : 5453833] ) (see Figure 27) . In this specification, a periostin nucleic acid may be any nucleic acid (DNA or RNA) having a nucleotide sequence having a specified degree of sequence identity •to SEQ ID No.6, to an RNA transcript of SEQ ID No.6, to a fragment of SEQ ID No .6 or said RNA, or to the complementary sequence of any one of these sequences or fragments. Alternatively a periostin nucleic acid may be one that hybridises to one of these sequence under intermediate, high or very high stringency conditions.

In this specification, a periostin polypeptide may be any peptide, polypeptide or protein having an amino acid sequence having a specified degree of sequence identity to SEQ ID No.5 or to a fragment of SEQ ID No.5.

AGR2

In this specification AGR2 may relate to an AGR2 nucleic acid or polypeptide.

In this specification the term AGR2' may relate to any AGR2 nucleic acid or polypeptide. Preferably AGR2 relates to human AGR2 or to an animal AGR2 homologue of human AGR2 , which may be a non-human mammalian homologue of human AGR2. The amino acid sequence (SEQ ID No.7) and nucleotide sequence (SEQ ID No .8 ) for the human AGR2 polypeptide and gene sequence are deposited in the NCBI database (http: //www.ncbi .nlm.nih.gov/) under accession number NP_006399 (NM_006408 [gi : 20070225] ) (see Figure 27).

In this specification, an AGR2 nucleic acid may be any nucleic acid (DNA or RNA) having a nucleotide sequence having a specified degree of sequence identity to SEQ ID No.8, to an RNA transcript of SEQ ID No.8, to a fragment of SEQ ID No .8 or said RNA, or to the complementary sequence of any one of these sequences or fragments . Alternatively an AGR2 nucleic acid may be one that hybridises to one of these sequence under intermediate, high or very high stringency conditions.

In this specification, an AGR2 polypeptide may be any peptide, polypeptide or protein having an amino acid sequence having a specified degree of sequence identity to SEQ ID No.7 or to a fragment of SEQ ID No .7.

UHRFl

In this specification UHRFl may relate to an UHRFl nucleic acid or polypeptide. In this specification the term ^ΛUHRF1' may relate to any UHRFl nucleic acid or polypeptide. Preferably UHRFl relates to human UHRFl or to an animal UHRFl homologue of human UHRFl, which may be a non-human mammalian homologue of human UHRFl .

The amino acid sequence (SEQ ID No.9) and nucleotide sequence (SEQ ID No.10) for the human UHRFl polypeptide and gene sequence are deposited in the NCBI database (http : //www.ncbi . nlm.nih . gov/) under accession number AAF28469 (NM_013282 [gil6507203] ) (see Figure 29).

In this specification, an UHRFl nucleic acid may be any nucleic acid (DNA or RNA) having a nucleotide sequence having a specified degree of sequence identity to SEQ ID No.10, to an RNA transcript of SEQ ID No.10, to a fragment of SEQ ID No.10 or said RNA, or to the complementary sequence of any one of these sequences or fragments. Alternatively an UHRFl nucleic acid may be one that hybridises to one of these sequence under intermediate, high or very high stringency conditions.

In this specification, an UHRFl polypeptide may be any peptide, polypeptide or protein having an amino acid sequence having a specified degree of sequence identity to SEQ ID No.9 or to a fragment of SEQ ID No .9.

The specified degree of sequence identity may be from at least 60% to 100% sequence identity. More preferably, the specified degree of sequence identity may be one of at least 65%, 70%, 75%, 80%, 85%, 86%, 87%, 88%, 89%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98% or 99% identity. Sequence identity

Percentage (%) sequence identity is defined as the percentage of amino acid residues in a candidate sequence that are identical with residues in the given listed sequence (referred to by the SEQ ID No.) after aligning the sequences and introducing gaps if necessary, to achieve the maximum sequence identity, and not considering any conservative substitutions as part of the sequence identity. Sequence identity is preferably calculated over the entire length of the respective sequences .

Where the aligned sequences are of different length, sequence identity of the shorter sequence is determined over the entire length of the longer sequence. For example, where a given sequence comprises 100 amino acids and the candidate sequence comprises 10 amino acids, the candidate sequence can only have a maximum identity of 10% to the entire length of the given sequence. This is further illustrated in the following examples:

(A)

Given seq: XXXXXXXXXXXXXXX (15 amino acids)

Comparison seq: XXXXXYYYYYYY (12 amino acids)

% sequence identity = the number of identically matching amino acid residues after alignment divided by the total number of amino acid residues in the longer sequence, i.e. (5 divided by 15) x 100 = 33.3%

(B)

Given seq: XXXXXXXXXX (10 amino acids) Comparison seq: XXXXXYYYYYYZZYZZZZZZ (20 amino acids)

% sequence identity = number of identical amino acids after alignment divided by total number of amino acid residues in the longer sequence, i.e. (5 divided by 20) x 100 = 25%.

Alignment for purposes of determining percent amino acid sequence identity can be achieved in various ways known to a person of skill in the art, for instance, using publicly available computer software such as BLAST, BLAST-2, ALIGN or Megalign (DNASTAR) software.

Identity of nucleic acid sequences may be determined in a similar manner involving aligning the sequences and introducing gaps if necessary, to achieve the maximum sequence identity, and calculating sequence identity over the entire length of the respective sequences . Where the aligned sequences are of different length, sequence identity of the shorter sequence is determined over the entire length of the longer sequence.

The invention includes the combination of the aspects and preferred features described except where such a combination is clearly impermissible or expressly avoided.

Aspects and embodiments of the present invention will now be illustrated, by way of example, with reference to the accompanying figures. Further aspects and embodiments will be apparent to those skilled in the art. All documents mentioned in this text are incorporated herein by reference. Brief Description of the Figures

Figure 1. SPAG expression in pancrea tic tumours . SPAGl was upregulated in all experiments analysed by GeneSpring. X-co-ordinate represents tumour samples interrogated (Tml5, 17, 20, 26, 39, 40, 42 and 9) .

Figure 2. Expression of SPAGl in pancrea tic cell lines . Northern blots were constructed using 3μg of mRNA per lane. Two different transcripts are identified: in addition to the larger transcript which corresponds to the expected size (approx 3.8kb), the smaller splicing variant of around 2.8kb was also noted.

Figure 3. SPAGl RT-PCR on pancreatic cell lines . RT- PCR confirmed cDNA microarray analysis data for expression of SPAGl in pancreatic cell lines.

Figure 4. SPAGl expression . (A) Expression of SPAGl in pancreatic cancer showed SPAGl to be upregulated in 70% of the specimens tested (T - Tumour; TM - Tumour Metastasis); (B) SPAGl was upregulated in 4 out of 8 Pancreatic Intraepithelial Neoplasias (PanlNs) ; (C) SPAGl was not upregulated in chronic pancreatitis (CP) .

Figure 5. SPAGl cancer profiling micro array. Increased SPAGl expression was evident in 8 out of 21 (38%) lung cancer specimens (N-normal; Ca-Cancer) . SPAGl was weakly expressed in other tumour types.

Figure 6. Western blot data from PowerBlot™ proteomic analysis . MNK and UHRF1/ICBP90 were found to be expressed in pancreatic cancer and not in normal or chronic pancreatitis specimens (CP-Chronic Pancreatiitis; PACA-Pancreatic Cancer) .

Figure 7. MNK expression analysed by QRT-PCR . (A) Expression of MNK in pancreatic cancer showed MNK expression to be upregulated in 16 out of 18 (85%) of pancreatic cancer specimens tested (T-Tumour; TM-Tumour Metastasis); (B) MNK expression was upregulated in two out of three type I, and one out of two type II, Pancreatic Intraepithelial Neoplasias (PanlNs ) ; (C) MNK was not upregulated in chronic pancreatitis (CP) specimens .

Figure 8. MNK expression in pancrea tic cell lines .

Figure 9. SPAGl monoclonal an tibody staining of pancrea ti c cancer cells . Light micrograph of SPAGl monoclonal antibody stained pancreatic cancer specimen showing SPAGl upregulation and localisation in ductal cells.

Figure 10. SPAGl monoclonal antibody staining of pancrea ti c cancer cells . Light micrograph of SPAGl monoclonal antibody stained pancreatic cancer (pancreatic adenocarcinoma) specimen showing SPAGl upregulation and localisation in ductal cells.

Figure 11. Immunohistochemical analysis on pancreatic cancer tissue arrays (PTMA) with commercially available AGR2 polyclonal antibodies (Imgenex, USA) developed against a synthetic peptide corresponding to amino acids 55-72 of AGR2 (TQTYEEALYKSKTSNKPL) . Figure 12. UHRFl immunohistochemical analysis using pancrea tic cancer and chronic pancrea ti tis tissue array. Representative images from normal (A, B) and chronic pancreatitis (C) displaying an absence of immunoreactivity and three pancreatic adenocarcinoma cases (D,E,F) with nuclear UHRFl expression are shown. Original magnification for A,C and D are x 100; for B, E and F x 200.

Figure 13. QRT-PCR analysis of (A) UHRFl and (B) MNK. More than three-fold upregulation of genes in PanIN and pancreatic adenocarcinoma samples is present, while this was not evident in chronic pancreatitis samples. The upregulation of UHRFl genes was retained in metastatic specimens as well. "T" stands for pancreatic adenocarcinoma .

Figure 14. Periostin immunohistochemical analysis using a commercially available periostin antibody . (A) Normal pancreas (negative result); (B,C,D) pancreatic cancer samples showing strong staining in the stroma.

Figure 15. Graph showing results of ORT-PCR for Periostin pancrea tic tumour samples . N=normal pancreas, CP=chronic pancreatitis, T=pancreatic adenocarcinoma.

Figure 16. Numerical data showing resul ts of Affymetrix™ U133A/B oligonucleotide array analysis for UHRFl, AGR2 and Periostin protein expression in normal and pancrea tic cancer samples . Figure 17. Average of numerical da ta from Figure 1 6.

Figure 18. Average of numerical da ta from Figure 16 wi th combined average for Panln samples .

Figure 19. Graph plotted using average val ues illustra ted in Figure 19 and showing expression of UHRFl , AGR2 and periostin across PanIN^', PDAC and metastasis (metastasis of pancrea tic cancer to l iver) samples . Data for UHRFl is represented by the lower line on the graph, data for AGR2 by the left hand peak and data for periostin by the right hand peak. It is evident that AGR2 and UHRFl are expressed early, in PanlNs, while periostin is expressed when pancreatic cancer has already developed. (Note the scale employed owing to large upregulation of both AGR2 and periostin) . M=metastasis of pancreatic adenocarcinoma to liver.

Figure 20. SAGE da ta for AGR2 .

Figure 21. SAGE da ta for UHRFl .

Figure 22. SAGE da ta for perios tin .

Figure 23. Table 1 - List of genes at least four-fold upregulated in four or more pancreatic adenocarcinoma specimens (identified by investiga tion of Sanger cDNA microarray) identifying SPAGl (3^rd entry) , periostin (osf- 2; 8^th entry) and AGR2 (14^th entry) .

Figure 24. Table 2 - Differentially expressed proteins between pancreatic adenocarcinoma (PDAC) and normal pancreas, and PDAC and CP . Figure 25. Sequence informa tion for SPAGl . Extract from NCBI accession number AF311312 showing amino acid sequence of SPAGl polypeptide (SEQ ID No.l) and SPAGl nucleotide sequence (SEQ ID No.2) encoding SPAGl polypeptide .

Figure 26. Sequence information for MNK. Extract from NCBI accession number NM_000052 showing amino acid sequence of MNK polypeptide (SEQ ID No.3) and MNK nucleotide sequence (SEQ ID No .4 ) encoding MNK polypeptide .

Figure 27. Seguence informa tion for periostin . Extract from NCBI accession number NM_006475 showing amino acid sequence of periostin polypeptide (SEQ ID No.5) and periostin nucleotide sequence (SEQ ID No.6) encoding periostin polypeptide.

Figure 28. Sequence informa tion for AGR2. Extract from NCBI accession number NM_006408 showing amino acid sequence of AGR2 polypeptide (SEQ ID No.7) and AGR2 nucleotide sequence (SEQ ID No .8 ) encoding AGR2 polypeptide .

Figure 29. Sequence informa tion for UHRFl . Extract from NCBI accession number NM_013282 showing amino acid sequence of UHRFl polypeptide (SEQ ID No.9) and UHRFl nucleotide sequence (SEQ ID No.10) encoding UHRFl polypeptide. Figure 30. SPAGl immunohistochemistry . Immunohistochemical staining for SPAGl in one normal, four PanIN and one PDAC specimen.

Detailed Description of the Best Mode of the Invention

Specific details of the best mode contemplated by the inventors for carrying out the invention are set forth below, by way of example. It will be apparent to one skilled in the art that the present invention may be practiced without limitation to these specific details.

Materials and Methods

Experimental procedures A

General approach

In order to understand the molecular changes underlying the complex pathology of pancreatic malignancy, global gene expression profiling of pancreatic adenocarcinoma compared with normal pancreatic tissue was performed.

Human cDNA microarrays comprising 9932 elements were interrogated with fluorescence-labelled normal and adenocarcinoma samples (nine tumours, three normal pancreata, and three cell lines) . The data were analysed for differential gene expression, which was confirmed by serial analysis of gene expression (SAGE) , digital differential display (DDD) analysis, and immunohistochemistry (IHC) for selected cases. The array data were filtered to produce a list of genes significantly up-regulated or down-regulated in pancreatic adenocarcinoma. See Table 1 (Figure 23).

Tissues and cell lines

Nine cases of human pancreatic cancer in the form of freshly frozen tissue blocks were obtained from the Department of Pathology, Hospital Beaujon, Clichy, Paris, France, with full ethical approval from the host institution. The histological characteristics of each case were as follows : Tm9 (moderately differentiated) , Tml5 (moderately differentiated) , Tml7 (moderately differentiated) , Tm20 (poorly differentiated) , Tm26 (moderately differentiated) , Tm27 (poorly differentiated) , Tm29 (moderately differentiated) , Tm40 (moderately differentiated) , and Tm 42 (well differentiated) . All specimens were enriched for tumour cell population by evaluating haematoxylin and eosin- stained sections of each individual tissue block and macrodissection by trimming in a cryostat. Thus, all specimens contained 70-80% of malignant cells and this was confirmed by repeated stained sections during cryosectioning . Out of four normal samples used in a control study, two originated from donor organs (coded ^ΛD' and ^λZ' ) and two were histologically normal tissues with no visible dysplastic changes in the ducts taken from the distal parts of pancreata harbouring ampullary tumours (coded H and H2) .

A panel of 20 pancreatic cancer cell lines was used: BxPC3, Capan2, FA6, IMMPC2, Mia-Paca, PT45, PaTul, PaTu2 , Paca3, Panel, A818.4, AsPCl, CFPAC, Colo357, HPAF, Hs766T, MDA, RWP, SUIT2, and T3M . They were obtained from Cancer Research UK Cell Services and were cultured in E4 medium (Cancer Research UK Media Production, Clare Hall, Middlesex, UK) supplemented with 10% heat-inactivated fetal calf serum (GibcoBRL, Life Technologies, Paisley, UK) .

The human pancreatic duct epithelial cell line HPDE was a kind gift from Dr Ming-Sound Tsao, University of Toronto, Canada and was grown in keratinocyte medium as described by Furukawa et al⁹ .

cDNA microarray experiments

Custom 10 K cDNA microarrays were obtained from the Sanger Centre, Cambridge, UK through the Cancer Research UK/Ludwig Institute/Wellcome Trust consortium. The version 1.2.1 human 10 K arrays contain 9932 elements: 817 cDNAs derived from the HGMP IMAGE collection; 647 cDNAs derived from the Research Genetics IMAGE collection; and 468 chromosome 22 gene-specific PCR products from the Sanger Centre.

The details of array fabrication and a complete annotated list of all the clones spotted are available on the Sanger Centre Microarray website

(www. sanger . ac . uk/Projects/Microarrays/informatics/datafi les . shtml) .

Experimental design was based on a reference model, where tumour cell RNA was compared with a single reference probe, providing normalized measures of the expression of each gene in each sample relative to the normal reference pool. The reference pool was constructed from all four normal pancreatic tissues. Three of those normal specimens ( ^ΛD' , ^ΛZ' , and H2') for which sufficient RNA was available were also used for individual comparison with the pool, so that the biological variability of the normal samples could be assessed.

All the RNAs were isolated using TRIZOL reagent (Gibco BRL, Life Technologies Inc, Frederick, MA, USA) . Throughout the study, 20μg of total RNA from the normal reference pool was labelled with Cy3-dCTP, while the same amount of RNA from each of the tumour samples and three normal specimens was labelled with Cy5 fluorophore (Amersham Pharmacia Biotech, Amersham, UK) . Direct labelling was performed using an anchored oligo- (dT) primer (Oligonucleotide Service, Cancer Research UK Clare Hall

Laboratories, Potters Bar, UK) during reverse transcription with Superscript II RT enzyme (Gibco BRL, Life Technologies Inc, Frederick, MA, USA) . Detailed descriptions of hybridization and washing procedures are available on the Sanger Centre Microarray website (www. sanger . ac . uk/proj ects/microarrays/arraylab/methods .h tml) .

Data analysis

Following hybridization, arrays were scanned using an Affymetrix™ 428 dual-laser scanner (Affymetrix™, High Wycombe, UK) and two acquired images (for Cy3 and Cy5, respectively) were analysed by Imagene 5.0 (BioDiscovery Inc, Los Angeles, CA, USA) . After grid assignment, the average pixel intensity within each spot was determined and a local background lying outside a three-pixel buffer range was deducted from each spot intensity value. Poor and negative spots were flagged automatically and omitted from further analysis. Signal intensities between the two fluorescent images were normalized using the robust linear smoother lowess¹⁰ in the Stata version 7 statistical software package (Stata Corporation, TX, USA). All spots with a signal intensity lower than 1.5 times the median background were considered as low signal values and excluded from further analysis.

Inter-experiment comparison and gene filtering were performed using GeneSpring 5.0 (Silicon Genetics, CA, ^' USA) and the cut-off values used were four-fold for both up- and down-regulated genes. The stringency of this selection was confirmed using a normal t -test, which is part of the filtering and statistical analysis in the GeneSpring software (the significance level used here was p < 0.05) .

Clustering analysis was performed with Cluster' software and displayed in ^ΛTree View' , both written by Michael B Eisen (Stanford University, Stanford, CA, USA)¹¹. Only genes present in 80% of the cases were selected for cluster analysis. A measure of similarity of logarithmically transformed data (log 2) was calculated using the Pearson correlation and the average linkage clustering algorithm was used to perform hierarchical clustering .

SAGE da tabase analysis and digital differential display (DDD)

SAGE is a public gene expression data repository and online data analysis for serial analysis of gene expression (SAGE) data. The SAGE technique measures not the expression level of a gene, but quantifies a "tag" which represents a gene transcript. A tag, for the purposes of SAGE, is a nucleotide sequence of a defined length, directly 3' adjacent to the 3' most restriction site for a particular restriction enzyme.

The publicly available SAGEmap (http://www.ncbi.nlm.nih.gov/SAGE) database and digital differential display (DDD) (http://www.ncbi.nlm.nih.gov/UniGene) were used to confirm the differential expression of genes obtained by our cDNA array study. In the SAGE analysis (performed on 1 November 2002 download) , two libraries of normal pancreatic ductal cells (HX and H126, with 32 226 and 32 512 total tags, respectively), two bulk pancreatic cancer tissues (Panc91-16 113 and Panc96-6252, with 33 957 a d 35 750 total tags, respectively), and four pancreatic cancer cell lines (CAPANl, CAPAN2, HS766T, and PANC1, with 37 962, 38 354, 40 461, and 24 924 total tags, respectively) were examined. In order to identify tags that were significantly overexpressed in the pancreatic cancer versus the normal ductal cells, we used a significance test specifically designed for the analysis of tag sampling data as described by Audic and Claverie¹².

In DDD, three pooled pancreatic cancer cDNA libraries from bulk samples (705, 395, and 306) were compared with three pooled cDNA libraries from normal pancreata (5012, 3612, and 3624) .

RT-PCR on pancrea tic cell lines cDNA was synthesized using the Superscript II reverse transcription kit (GibcoBTRL, Life Technologies, Paisley, Scotland) . SPAGl primers used were: forward- gctagctgatgggaacgtgaa; reverse- ggcttggctatcggaagtttc . The product size was 333bp.

PCR conditions were: 94 °C 2 mins 1 cycle; 94°C 20s 56°C 20s 30 cycles; 72°C 30s; final extension: 10 min at 72°C

lμl of the CDNA was used for each PCR in 20μl reaction volume containing 200μM dNTPs, lμM sense and anti-sense primers, 1.25mM MgCl₂, 2μl of lOx PCR buffer and 0.5 unit of Taq DNA polymerase (Roche Diagnostics Ltd, Bell Lane, Lewes, East Sussex, UK) . PCR was performed on a DNA thermal cycler (MJ Research, Inc., Waltham, MA, USA).

Quantita tive RT-PCR (QRT-PCR) cDNA was synthesized from lμg of total mRNA using an oligo (dT) primer and the Superscript II reverse transcription kit (GibcoBTRL, Life Technologies, Paisley, Scotland) . The cDNAs were column purified (Qiagen, Crawley, West Sussex, UK) , and diluted to a concentration of lOng/μl.

The primers used to amplify a 75bp amplicon were: forward- aaattgagattcaagaggtgaatgaa; reverse- ggcatcccgtggagacc

Reactions containing lOng cDNA, SYBR Green sequence detection reagents (Oswel Research Products Ltd, Southampton UK) and primers were assayed on an ABI7700 sequence detection system (Applied Biosystems) . The accumulation of PCR products was measured in real time as the increase in SYBR Green Fluorescence, and the data analysed using the Sequence Detector program vl .6.3 (Applied Biosystems) . Assays were performed in triplicate. Quantitative gene expression in tumour samples was compared to the average value of all normal samples arbitrarily set at 1.

SPAGl Cancer profiling in different tissues A cancer profiling array (BD Clontech) was used to hybridise the radiolabelled RT-PCR probe for SPAGl. The array comprises the tumour and matched normal tissues.

SPAGl Monoclonal Antibody Staining of Pancrea tic Cancer Specimens

Sections of two paraffin-embedded pancreatic cancer specimens were obtained and stained using a SPAGl monoclonal antibody (raised at the Monoclonal Research Services, LIF) . Dilution was 1:200 and the immunohistochemistry was performed on a Ventana™ machine using Hematoxylin as a counterstain.

Experimental Procedures B

MNK dysregula tion in pancrea tic cancer

Protein screening, using a proteomic analysis technique (BD Powerblot™, Fig.6), of pancreatic cancer specimens in comparison with normal pancreatic tissues was used to identify proteins having differentially upregulated expression in pancreatic cancer.

Confirma tion of MNK dysregula tion by QRT-PCR cDNA was synthesised from lμg of total RNA using an oligo (dT) primer and the Supersript II reverse transcription kit (GibcoBTRL, Life technologies, Paisley, Scotland) . The cDNAs were column-purified (Qiagen, Crawley, West Sussex, UK) , and diluted to a concentration of lOng/μl. The primers used to amplify a 128bp amplicon were: forward- gatgatgagctgtgtggcttga; reverse- getgttttactgttgtctccagtca

Reactions containing lOng cDNA, SYBR Green sequence detection reagents (Oswel Research Products Ltd, Southampton UK) and primers were assayed on an ABI7700 sequence detection system (Applied Biosystems). The accumulation of PCR products was measured in real time as the increase in SYBR Green fluorescence, and the data analysed using the Sequence Detector program vl .6.3 (Applied Biosystems) . Assays were performed in triplicate. Quantitative gene expression in tumour samples was compared to the average value of all normal samples arbitrarily set at 1.0.

Experimental procedures C

Tissue samples

All pancreatic tissue specimens (normal, ductal adenocarcinoma and chronic pancreatitis) were obtained from the Human Biomaterials Resource Centre (HBRC) , Hammersmith Hospital Trust, London and Cancer Tissue Bank Research Centre (CTBRC) in Liverpool. Eight PanIN specimens (three PanIN-1, two PanIN-2 and three PanIN-3) and four pancreatic liver metastases were kindly provided by Drs Teresa A. Brentnall, University of Washington, Seattle, USA and Makoto Sunamura, Tohoku University, Japan, respectively. All specimens were obtained with patients' written consent and full ethical approval from the host institutions.

Prior to RNA isolation or preparation of the lysates, each freshly-frozen tissue was sectioned and stained with Haematoxylin and Eosin in order to confirm the diagnosis and assess tumour cellularity.

Tissue microarrays were obtained from CTBRC, Liverpool.

BD PowerBlot™ analysis

BD PowerBlot™ is a protein screening method (BD Biosciences Pharmingen) in which samples are loaded for gel elecfrophoresis; proteins are transferred to PVDF membranes; membranes are probed with over 900 antibodies; blots are run in triplicate and electronic images of blots are captured using the Odyssey™ Infrared Imaging System; Images are subjected to automatic spot finding and spot matching using PDQuest™ software.

Three groups of tissue lysates were prepared, each from a total of 5 mg of tissue, and each containing five pooled clinical specimens from either normal pancreas (NP) , chronic pancreatitis (CP) or ductal adenocarcinoma (PDAC) , according to the protocol supplied by BD (http : //www.bdbiosciences . com/pharmingen/products/display _product .php?keyID=26) . Tissues from each of the disease groups were of confirmed clinical diagnosis and histologically closely resembled each other, with the proportion of neoplastic cellularity in cancer cases varying from 50 to 80%. Each lysate was interrogated with 900 antibodies, and the data for NP compared to both CP and PDAC; in addition, CP data were compared to PDAC. As the analysis was performed in triplicate, cross comparisons between the groups resulted in nine data points for each protein (antibody) . The data were organized into levels of confidence with scores of 1-10, depending on the expression level, reproducibility, and the intensity and the quality of the signal. A score of 10' represented the most stringent criteria (2-fold or higher over/under-expression in all nine comparisons from good quality signals that pass visual inspection) and was used in the selection of differentially expressed proteins .

QRT-PCR

Total RNA from all the tissues was extracted following the TriZQl protocol (Gibco BRL) . The cDNA was synthesized from lμg of total RNA using random primers and the MultiScribe™ Reverse Transcription kit (Applied Biosystems, Warrington, Cheshire). QRT-PCR primers used were as follows: UHRFl (145 bp amplicon) sense 5' gcccgttccagttgttcct3' ; antisense

5' aacacctgtgcccgaaagg3' ; MNK/ATP7A (128 bp amplicon): sense 5' gatgatgagctgtgtggcttga3' ; antisense 5' gctgttttactgttgtctccagtca 3'. Reactions containing 10 ng cDNA, primers and SYBR Green sequence detection reagents (Applied Biosystems, Warrington, Cheshire. UK) were assayed on an ABI7700 sequence detection system (Applied Biosystems). The accumulation of PCR product was measured in real time as an increase in SYBR Green fluorescence, and the data analysed using the Sequence Detector program vl .9.1. (Applied Biosystems) . Assays were performed in triplicate utilizing 5 normal pancreatic specimens (N) , 8 PanIN lesions, 18 primary adenocarcinomas (T1-T18), 4 metastases to liver (TM1-TM4) and 9 chronic pancreatitis specimens (CP1-CP9) . Quantitative gene expression of experimental samples was compared to the average value of all normal samples set arbitrarily at 1.

Immunohistochemical analysis

This analysis was performed using a pancreatic cancer- specific tissue array (PTMA) that comprises specimens with full clinical history including age at presentation, gender, tumour size, resection margins, grade, TNM stage, presence of lymph node, perineural or vascular invasion, and survival data. It contains 180 core biopsies, 110 of which are PDAC cores corresponding to 55 cases spotted in duplicate, the remaining being either normal pancreatic or other tissue control cores, such as kidney or lung, again represented by duplicate cores. Immunohistochemical analysis was also performed on a chronic pancreatitis tissue array (CPTMA) containing cores representing 24 normal ducts and 24 chronic pancreatitis specimens, each arrayed in duplicate (96 cores in total) with 5 cores each of normal colonic, liver and kidney samples.

In addition, a third set of tissue arrays, LandMark High Density Cancer Tissue MicroArrays from Ambion™ was used to obtain a comprehensive analysis of expression of selected proteins across a variety of cancer and matched normal tissues. The High Density Cancer Tissue MicroArray comprised 280 tumour cores from 24 different tissues including matched normal cores for the majority of cancer types presented. Further details on these arrays can be obtained from the Ambion™ web site (www.ambion.com) . Immunostaining was performed using an anti-UHRFl monoclonal antibody at 1:200 dilution. All antibodies were obtained from BD Transduction Laboratory (Cowley, Oxford, UK) .

The tissue arrays were stained using the Ventana Discovery™ automated stainer, with antigen retrieval using Tris Borate EDTA Cell Conditioning buffer (pH 8) for 50 min at 95°C and the DAB Detection Kit according to the protocols provided by the Ventana Discovery™ System, Illkirch, France (www.ventanadiscovery.com). The sections were counterstained with Haematoxylin.

UHRFl was considered positive if immunoreactivity was present in at least 5% or more of the cell nuclei. Due to the relatively small number of individual cores representing each of 24 different tissue types, High Density Cancer Tissue MicroArrays were scored using a simple scoring system - as presence or absence of immunoreactivity .

Sta tistical Analysis

A Fisher's exact test was used to test the independence of rows and columns in the 2x2 contingency table of staining patterns between the tumour and normal tissue, with p-values taken directly from the hypergeometric distribution. For each protein, the contingency tables were constructed using the number of positive outcomes in the first column and the number of negative outcomes in the second for both the tumour and normal samples. Logistic regression was used to assess the associations between the staining pattern within the tumour samples and the available clinical data. It was performed in the R statistical environment on a Linux platform, using the glm (generalized linear model) function. In all cases, the staining pattern was transformed to l's and 0's, representing present and absent staining, respectively.

Experimental procedures D

Analysis of oligonucleotide arrays

Large-scale gene expression profiling to identify genes involved in both the early and late phases of pancreatic tumourigenesis was performed using Affymetrix™ Human Gene U133A/B oligonucleotide arrays comprising 47 000 gene elements. After hybridisation, scanning was performed in GeneChip Scanner 3000. After scanning, the analysis of the data began with the *.CEL file, allowing one to access the perfect match (PM) and mismatch (MM) probe intensities, which together constitute a probe set. Using the Bioconductor packages within the R statistical environment, in particular the ^ΛAffy' package, background correction was performed using the ^VMAS' function, perfect matches were selected using ^λPMONLY' , the probe sets were summarised using median polishing ( MEDIANPOLISH' ) and normalised using 'QUANTILES' . The normalisation was carried out globally, to minimise chip to chip variation, and smooth the data across all the experiments. Differentially expressed genes were then identified by a Welch two sample t-test. In addition, a distribution free test (permutation test using the t- statistic) with subsequent FDR (false discovery rate) correction with a cut off p-value of less than 0.05 was also performed. For the proteins UHRFl, AGR2 and periostin, numerical results for protein expression are shown in Figures 16-18 and average data for normal, PanlNs, pancreatic ductal adenocarcinoma (PDAC) and. metastasis samples is graphically illustrated in Figure 19 (data for UHRFl is indicated by the lower line on the graph, periostin by the right hand peak and AGR2 by the left hand peak) . Notably periostin is highly elevated in PDAC.

The results can be used to assess whether upregulated candidate biomarkers are likely to be markers of early or late stage development of pancreatic tumourigenesis. The oligonucleotide array data for AGR2 (Figure 19) indicates that AGR2 is a potential marker of early stage tumourigenesis and the data for periostin (Figure 19) indicates that periostin is a potential marker of later stage tumourigenesis.

Immunohistochemical (IHC) analysis of AGR2 and periostin For AGR2, IHC analysis was performed on PTMA with commercially available anti-AGR2 polyclonal antibodies developed against a synthetic peptide corresponding to amino acids 55-72 (TQTYEEALYKSKTSNKPL) . The results are shown in Figure 11 and show that out of 29 PDAC specimens, 26 were positive (90%) and 3 were negative (10%) .

The results of IHC staining for periostin using a commercially available antibody are shown in Figure 14. Figure 14 B, C and D show strong periostin staining in the stroma of pancreatic cancer samples but not in normal pancreatic tissue. Periostin QRT-PCR

To determine relative periostin mRNA levels in situ, QRT- PCR was performed for periostin on a range of pancreatic cancer samples. The results are shown in Figure 15.

Results

Experimental Procedures A

Differen tial gene expression in pancrea tic adenocarcinoma

Using a very stringent four-fold cut-off, which was chosen to produce manageable lists of genes, we have identified 29 up-regulated and 46 down-regulated genes that showed the change in at least four out of nine cancer specimens. Table 1 (Figure 23) ¹³'¹⁴ displays a list of the up-regulated genes, together with a ^λconfirmation' column, which states the results of a search of the SAGEmap database and DDD that were employed to validate the differential expression of these genes. Increased expression of a variety of genes such as fibronectin, laminin, collagen, MMPll, galectin 1, and lumican, which were already associated with pancreatic adenocarcinoma, as well as several genes previously not well described in this cancer type, namely osf-2 (periostin) , AGR2, ATDC, and TEM8, were found to be up-regulated in this study.

SPAGl was shown to be upregulated at least four fold in 7/8 pancreatic cancer tissues and upregulated 1.5 fold in the eighth sample tested. 13 out of 20 pancreatic cell lines exhibited a four-fold or greater upregulation of SPAGl with 2 of the remaining 7 showing SPAGl upregulation of greater than 2 fold. The results of RT- PCR on pancreatic cell lines (Fig.3) are in complete agreement with the cDNA microarray data.

5 out of 15 IPMTs (intraductal papillary mucinous tumours) , which in a proportion of cases progress to "classical" adenocarcinoma and can therefore be regarded as one of the precursor lesions, showed a greater than 2- fold upregulation of SPAGl.

Analysis of SAGE showed 1-3 tags in 44/113 of the libraries examined with 5 tags in the gastric cancer SAGE library and 24 tags in the breast intraductal carcinoma library.

SAGE data for AGR2, periostin and UHRFl are shown in Figures 20-22.

QRT-PCR was employed to confirm dysregulation of SPAGl gene. The results (Fig.4) demonstrate that SPAGl is upregulated in around 70% of pancreatic carcinomas (Fig. 4A) and that this overexpression starts in a proportion of Pancreatic Intraepithelial Neoplasias (Fig. 4B) which represent the early phases of pancreatic cancer development. It is also of note that SPAGl was not upregulated in chronic pancreatitis specimens (Fig. 4C) . Thus SPAGl has the potential to be used as a specific marker of early stage pancreatic cancer.

Higher SPAGl expression was also present in 8 out of 21, i.e 38%, of lung cancer specimens (Fig.5) in comparison to normal lung. In efforts to generate a monoclonal antibody to SPAGl the supernatant obtained from three hybridomas have also been found, by Western blotting, to be positive for SPAGl present in human sperm extract and testis.

One monoclonal antibody to SPAGl obtained from the Monoclonal Research Services, LIF, was used to stain paraffin-embedded pancreatic cancer specimens. The results are shown in Figures 9 and 10. The cytoplasm of malignant ductal cells was stained and an increased stain intensity was evident towards the luminal end of the cells. This data confirms the upregulated expression of SPAGl protein in pancreatic cancer and provides evidence in support of SPAGl protein localisation to ductal cells.

SPAGl immunohistochemistry also confirmed upregulation of SPAGl in PanIN and PDAC specimens compared to normal (non-tumour) specimens (see Figure 30) .

The experiments performed have identified SPAGl to be upregulated in the development of pancreatic cancer. Detection of SPAGl nucleic acid or polypeptide can therefore provide a means of diagnosing pancreatic cancer at an early stage of development when treatment can be most effective. The specific localisation of SPAGl upregulation also provides a biochemical target. Antibodies binding specifically to SPAGl, (e.g. a monoclonal antibody) can be used to deliver anti-tumour compounds directly to the tumour cells by chemically linking the anti-tumour compound to the SPAGl antibody. The diagnosis and treatment of tumours, particularly pancreatic tumours/neoplasms can thus be significantly improved. Experimental Procedures B

MNK is upregula ted in pancrea tic cancer and not present in normal or chronic pancrea ti tis specimens Protein screening, using a proteomic analysis technique (BD powerblot™, Fig.6), of pancreatic cancer specimens in comparison with normal pancreatic tissues revealed MNK protein to be upregulated in pancreatic cancer specimens but not present in normal and chronic pancreatitis specimens .

MNK is upregula ted in 85% of pancreatic carcinomas QRT-PCR (Fig.7 and 8) demonstrated that MNK is upregulated in approximately 85% of pancreatic carcinomas (Fig. 7A) . This overexpression is already present in a proportion of Pancreatic Intraepithelial Neoplasias (Fig. 7B) , which represent the early phases of pancreatic cancer development.

Of particular note was that MNK is not upregulated in chronic pancreatitis specimens (Fig.7C).

The upregulation of MNK in pancreatic cancer is also observed to be coincident with an upregulation of ceruloplasmin and a down regulation of metallothioneins .

These results indicate that MNK can provide a marker for the specific identification and diagnosis of early phase pancreatic cancer. Detection of MNK nucleic acid or polypeptide can therefore provide a means of diagnosing pancreatic cancer at an early stage of development when therapeutic intervention can be most effective. The specific localisation of MNK upregulation also provides a biochemical target. Antibodies (e.g. a monoclonal antibody) specifically binding to MNK protein can thus be used to deliver anti-tumour compounds directly to the tumour cells by chemically linking the anti-tumour compound to the MNK antibody. The diagnosis and treatment of tumours, particularly pancreatic tumours/neoplasms can thus be significantly improved.

Experimental procedures C

33 proteins were up- or down-regulated more than two-fold in the comparison between the CP and NP, while 113 proteins (58 upregulated and 55 downregulated) were deregulated between PDAC and NP (Tajole 2 (Figure 24)). Table 2 also shows that only 18 of 58 upregulated proteins were previously disclosed in earlier studies on pancreatic adenocarcinoma ⁴²' ⁴⁶' ⁴⁸' ⁵²' ^61~74. Comparison of the tables showed that more than half of proteins are commonly upregulated in both CP and PDAC.

The inventors selected for more detailed study two proteins that appeared overexpressed only in the cancer cases and were not present in chronic pancreatitis or normal specimens, namely UHRFl and ATP7A/MNK.

To establish if UHRFl and MNK (ATP7A) protein overexpression was due to transcriptional regulation QRT- PCR was employed.

Figure 13 shows QRT-PCR data for UHRFl (A) , and MNK (ATP7A) (B) . Six out of eight (75%) PanIN lesions, 14 out of 18 (78%) tumour cases and all four metastatic samples (metastasis of pancreatic cancer to liver) showed more than a two-fold increase of UHRFl transcript; compared to normal pancreas. No upregulation in CP specimens was noted (Figure 13A) . For MNK (ATP7A) , six out of eight PanIN (75%) and 11 out of 18 (61%) tumour samples showed more than two-fold transcript increase, while this was not noted in any of the CP and the majority of metastatic (metastasis of pancreatic cancer to liver) specimens (Figure 13B) .

Tissue microarrays were used to confirm the deregulation of UHRFl protein by immunohistochemistry on a large number of clinical specimens to compensate for both pooling of the samples and a small number of cases initially analyzed. Although the inventors had successfully used anti-ATP7A (anti-MNK) antibody in the western blot analysis (data not shown) , even after multiple optimization trials, it failed to give consistent and reproducible results on formalin-fixed, paraffin-embedded tissue sections and was therefore excluded from further immunohistochemical analysis.

UHRFl immunoreactivity was detected only in the nuclei of malignant cells, and it was absent from the normal ductal, acinar and islet cell compartments. Out of 34 cores representing individual PDAC cases, 29 (85%) showed nuclear staining in at least 5% of the cells. Amongst 28 scoreable cores derived from normal pancreatic tissue, only 2 cases (7%) showed positive staining in several scattered acinar cells, while the remaining 26 (93%) specimens were completely negative. Using a Fisher's exact test for independence, this difference in staining between normal and cancer cases (85% versus 7%, respectively) was highly significant, with p = 4.576e-10. Due to the relative uniformity of staining, no significant correlation could be made between the protein expression and the clinicopathological data. Where both cores representing the same specimen were present, good concordance between the results was observed. Figure 12 shows representative images from normal (A,B), CP (C) and three PDAC cases (D,E,F). Of note, there was no UHRFl immunoreactivity evident on any of the CP cores in the CPTMA array.

High Density Cancer Tissue MicroArray analysis showed that UHRFl protein is expressed in 4/11 colonic cancer cores, 4/14 prostate cancers, 7/17 bladder cancer and 2/7 lymph node specimens. Brain, thyroid, breast, lung, liver, kidney and other remaining tissue represented on the array showed absence of immunoreactivity in both cancer and normal tissue cores (data not shown) .

Discusssion

Experimental procedures C

Although multiple gene expression profiling studies of pancreatic adenocarcinoma (several including CP) have been performed previously ⁴¹' ^{42, 4 ~53}' ⁷⁵, very few proteomic studies of pancreatic adenocarcinoma tissue specimens have been conducted to date ^{53, 54}. The inventors utilized a large-scale Western blot approach that has resulted in identification of a number of proteins with difference in the levels of expression between NP, CP and PDAC, with the additional benefit of the availability of antibody reagents for further studies . As can be seen in Table 2 (Figure 24), 65% (85/113) of differentially expressed proteins are commonly deregulated between CP and PDAC, similar to the findings of Logsdon et al ⁴². Amongst the commonly overexpressed proteins in CP and PDAC specimens are actin and several associated proteins, namely L-caldesmon, p21-Arc, α- actinin and fascin. Together they indicate common cytoskeletal modifications accompanying the development of both CP and PDAC. Several common signaling molecules were also identified, such as G protein alpha transducing activity polypeptide 2 (Gat) , tyrosine protein kinase HCK, src homology 2 domain containing transforming protein C3 ShCC (SHC3), c-src tyrosine kinase (CSK) and pathogen-recognition receptor Gb, as well as several small GTPases: Rho, Rap2 and Rab27. Two proteins involved in apoptotic pathways, namely Bid and Bad were found downregulated in both CP and PDAC specimens.

This significant overlap in deregulated proteins between CP and PDAC is probably due to shared morphological changes seen in CP and PDAC, which include acinar cell degeneration, fibrosis, immune cell infiltration and ductular hyperplasia. Moreover, both hereditary and sporadic forms of CP show an increased risk of developing PDAC, 67 and 17 times, respectively ⁶⁰' ^7δ.

It is well established that inflammation can create an environment that supports tumour formation in a variety of human cancers. At least in a proportion of PDAC cases, pancreatic inflammation mediated by cytokines, reactive oxygen species and increased pro-inflammatory pathways (including COX2 and Nfkb) ^37~39 may represent an early step in the development of the malignancy with increased genomic damage and stimulation of "reactive" cellular proliferation. Increased cell division is indeed a well- known feature of human cancer ⁷⁰.

UHRFl and ATP7A showed exclusive expression in PDAC and not in CP or normal pancreas .

The UHRFl (ubiquitin-like, containing PHD and RING finger domains, 1)/ICBP90 gene encodes a transcription factor of 793 amino acids. It is a novel CCAAT box-binding nuclear protein involved in the regulation of topoisomerase Ilα (TopoIIα) gene during the cell cycle ⁷¹. It belongs to the family of E3 ligases of the RING finger type and has been described as a target of protein kinase A. Phosphorylation of S298, situated just upstream of the DNA binding site causes UHRFl to acquire a conformational change that results in its increased transcriptional activity and enhances TopoIIα expression⁷². Higher levels of both UHRFl and TopoIIα transcripts were recently described in several cancer cell lines and breast carcinoma tissues when compared to normal breast⁴³. We have found upregulation of UHRFl in around 80% of PDAC cases at both RNA and protein levels; moreover, this expression was noted in the majority of PanlN lesions and was retained in all metastatic (metastasis of pancreatic cancer to liver) specimens. As TopoIIα is known to be upregulated in PDAC⁴⁴, it is likely that UHRFl is involved in TopoIIα regulation in pancreatic adenocarcinomas as well . MNK (also called ATP7A, Menkes disease gene) is a 160 kDa copper- transporting ATPase. Defects in this ATPase are associated with Menkes disease, an X-linked recessive disorder characterized by growth retardation, neurodegeneration, connective tissue disorders and death in early childhood. This was thought to be caused by dysfunction of copper-requiring enzymes such as cytochrome oxidase (electron transport), lysyl oxidase (crosslinking of collagen and elastin) and SOD (free radical detoxication) ⁵⁵. ATP7A is an integral membrane protein that cycles between the trans-Golgi network and relocalizes to the plasma membrane in response to elevated copper levels ⁵⁶. Copper is the third most abundant trace element after iron and zinc, and it needs to be tightly regulated due to its high toxicity. After intestinal absorption, copper is bound to albumin and transported to liver. It is then secreted from hepatocytes into the plasma where it is bound to ceruloplasmin. Ceruloplasmin is on the other hand, an acute phase-responsive oxidase enzyme that potentially reflects increased oxidant stress ⁷. Of note, increased levels of ceruloplasmin were also found in the present study (Table 2 (Figure 24)) . Association of changes in copper and ceruloplasmin levels and tumour growth have already been reported in other tumour types ⁹⁸' ⁹⁹.

In the free form, copper can catalyze the formation of highly reactive hydroxyl radicals, which are the most powerful oxidizing radicals that can induce DNA strand breaks and oxidation of bases ⁶⁰. MNK overexpression in pancreatic cancer appears to be an early event, as suggested by increased transcript levels in the majority of PanlN lesions. In summary, the inventors have described protein expression of normal, CP and pancreatic adenocarcinoma specimens and have shown involvement of multiple differentially expressed proteins involved in various aspects of cellular functions (signaling, metal ion binding, cytoskeletal proteins, apoptosis etc) . Interestingly, several oxidative stress-related proteins, namely superoxide dismutase 2 (SOD2) , gene associated with retinoic-interferon-induced mortality 19 protein (GRIM 19/NADH-ubiquinone oxidoreductase B16.6 subunit) and 5-lipoxygenase (ALOX5) (Table 2 (Figure 24)), were found commonly deregulated in both CP and PDAC. The contribution of early oxidative stress in the development of genomic instability in histologically normal-appearing ductal epithelial cells from the patients with both CP and PDAC has recently been described ⁶⁵. MNK also appeared to be directly or indirectly implicated in the metabolism of ROS. As their deregulation was absent in CP and exclusive for PDAC, this suggests the need for additional (genotoxic) events that would lead to the development of cancer. The involvement of several other proteins, namely SOD1, catalase and thioredoxin has already been described in PDAC previously ^{106, 44} and combined with our present findings strongly supports the importance of extensive oxidative stress in the pathogenesis of pancreatic adenocarcinoma.

Furthermore, the identification of UHRFl and MNK indicates their utility as diagnostic tools in distinguishing chronic pancreatitis from pancreatic adenocarcinoma . Summary

The inventors have used a range of techniques to identify protein markers upregulated in forms of pancreatic cancer or in the development of pancreatic cancer. In particular, five markers SPAGl, MNK, Periostin, AGR2 and UHRFl have been identified.

CDNA microarray studies clearly show a multiple fold increase in upregulation of SPAGl (7.2 fold increase), Periostin (6 fold increase) and AGR2 (20 fold increase) . Affymetrix™ oligonucleotide array analysis showed upregulation of all five markers, periostin (21 fold) and AGR2 (12 fold) showing high levels of upregulation. Results for marker upregulation in both cDNA microarray and Affymetrix™ arrays showed good correlation.

QRT-PCR has also been used to confirm overexpression of candidate markers. Overexpression of periostin (in 17 out of 18 pancreatic tumour tissues tested) , SPAGl (in 13 out of 18 pancreatic tumour tissues and 5 out of 9 Panln specimens tested) , MNK (in 85% of pancreatic tissues tested) and UHRFl (in 14 out of 18 pancreatic tumour tissues and 6 out of 8 Panln specimens tested) .

SAGE analysis confirmed expression of periostin, AGR2, and UHRFl expression in pancreatic cancer tissues.

Immunohistochemical analysis using marker specific antibodies showed positive staining for periostin, AGR2, SPAGl and UHRFl in over 90% of pancreatic tumour cores tested. Investigations of tissue arrays (Ambion™) confirmed marker expression in a range of normal and cancer tissues .

AGR2 was indicated to be weakly expressed in normal colon, stomach and kidney, and not expressed in normal breast, prostate, testis, skin, brain, oesophagus, larynx, mouth, tongue or lung. In cancer tissues, AGR2 expression was identified in breast, gastric and colon adenocarcinomas as well as in transitional cell carcinoma of kidney and bladder but was not identified in lymphomas, mesotheliomas, brain tumours, or squamous-cell carcinomas of skin, oesophagus, larynx, mouth, tongue or lung .

In normal tissues, SPAGl was indicated to be strongly expressed in sperm cells of testis tissue (as expected) . In addition, SPAGl immunoreactivity was detected in stomach, duodenum, colon, gallbladder and liver, as well as in the kidney. No immunoreactivity was detected in normal pancreas, brain, thyroid, lung, heart, lymphoid tissue, spleen, urinary bladder, skeletal muscle and ovary. As well as showing high level of expression in pancreatic cancer, SPAGl was found to be expressed in 3/11 prostate cancers, 2/12 breast cancers, 6/11 colonic, 1/9 liver and kidney cancers, 4/13 ovarian, 4/11 endometrial and 1/14 laryngeal cancers.

UHRFl expression was identified in normal stomach tissue and in cancer tissues of the colon, prostate and bladder.

Periostin and AGR2 are secreted proteins and are therefore proposed as candidate serum or blood markers. Expression data indicates AGR2, MNK, UHRFl and SPAGl show potential as early markers of neoplasia with periostin having potential as a slightly later stage marker.

References

1. Shoemaker DD, Schadt EE, Armour CD, et al . Experimental annotation of the human genome using microarray technology. Na ture 2001; 409: 922-927.

2. Golub TR, Slonim DK, Tamayo P, et al . Molecular classification of cancer: class discovery and class prediction by gene expression monitoring. Science 1999; 286: 531-537.

3. Perou CM, Sorlie T, Eisen MB, et al . Molecular portraits of human breast tumours. Na ture 2000; 406: 747- 752.

4. Alizadeh AA, Eisen MB, Davis RE, et al . Distinct types of diffuse large B-cell lymphoma identified by gene expression profiling. Na ture 2000; 403: 503-511.

5. Ross DT, Scherf U, Eisen MB, et al . Systematic variation in gene expression patterns in human cancer cell lines. Na ture Genet 2000; 24: 227-235.

6. Lemoine NR. Molecular advances in pancreatic cancer. Digestion 1997; 58: 550-556.

7. Crnogorac-Jurcevic T, Efthimiou E, Capelli P, et al . Gene expression profiles of pancreatic cancer and stromal desmoplasia. Oncogene 2001; 20: 7437-7446. 8. Crnogorac-Jurcevic T, Efthimiou E, Neilsen T, et al . Expression profiling in microdissected adenocarcinomas of the pancreas. Oncogene 2002; 21: 4587-4594.

9. Furukawa T, Duguid WP, Rosenberg L, et al . Long-term culture and immortalization of epithelial cells from normal adult human pancreatic ducts transfected by the E6/E7 gene of human papilloma virus 16. Am J Pa thol 1996; 148: 1763-1770.

10. Yang Y, Dudoit S, Luu P, et al . Normalization for cDNA microarray data: a robust composite method addressing single and multiple slide systematic variation. Nucl eic Acids Res 2002; 15: el5.

11. Eisen MB, Spellman PT, Brown PO, Botstein D. Cluster analysis and display of genome-wide expression patterns. PNAS 1998; 95: 14 863-14 868.

12. Audic S, Claverie JM. The significance of digital gene expression profiles. Genome Res 1997; 7: 986-995.

13. Maass N, Hojo T, Ueding M, et al . Expression of the tumour suppressor gene Maspin in human pancreatic cancers. Clin Cancer Res 2001; 7: 812-817.

14. Ebert M, Yokoyama M, Kobrin MS, et al . Induction and expression of amphiregulin in human pancreatic cancer. Cancer Res 1994; 54: 3959-3962.

15. Ryu B, Jones J, Hollingsworth MA, Hruban RH, Kern

SE . Invasion-specific genes in malignancy: serial analysis of gene expression comparisons of primary and passaged ancers. Cancer Res 2001; 61: 1833-1838.

16. Gress TM, Muller-Pillasch F, Geng M, et al . A pancreatic cancer specific expression profile. Oncogene 1996; 13: 1819-1830.

17. Iacobuzio-Donahue CA, Maitra A, Shen-Ong GL, et al . Discovery of novel tumour markers of pancreatic cancer using global gene expression technology. Am J Pa thol 2002; 160: 1239-1249.

18. Ryu B, Jones J, Blades NJ, et al . Relationships and differentially expressed genes among pancreatic cancers examined by large scale serial analysis of gene expression. Cancer Res 2002; 62:819-826.

19. Argani P, Rosty C, Reiter RE, et al . Discovery of new markers of cancer through serial analysis of gene expression: prostate stem cell antigen is overexpressed in pancreatic adenocarcinoma. Cancer Res 2001; 61: 4320- 4324.

20. Schafer BW, Wicki R, Engelkamp D, Mattei MG, Heizmann CW. Isolation of a YAC clone covering a cluster of nine S100 genes on human chromosome lq21: rationale for a new nomenclature of the S100 calcium-binding protein family. Genomics 1995; 25: 638-643.

21. Donato R. S100: a multigenic family of calcium- modulated proteins of the EF-hand type with infracellular and extracellular functional roles. Int J Biochem Cell Biol 2001; 33: 637-668. 22. Ilg EC, Schafer BW, Heizmann C . Expression pattern of S100 calcium-binding proteins in human tumours. In t J Cancer 1996; 68: 325-332.

23. Ljubimova JY, Khazenzon NM, Chen Z, et al . Gene expression abnormalities in human glial tumours identified by gene array. Int J Oncol 2001; 18: 287-295.

24. Hough CD, Cho KR, Zonderman AB, Schwartz DR, Morin PJ. Coordinately up-regulated genes in ovarian cancer. Cancer Res 2001; 61: 3869-3876.

25. Cochran AJ, Lu HF, Li PX, Saxton R, Wen DR. S-100 protein remains a practical marker for melanocytic and other tumours. Melanoma Res 1993; 3: 325-330.

26. Rosty C, Ueki T, Argani P, et al . Overexpression of S100A4 in pancreatic ductal adenocarcinomas is associated with poor differentiation and DNA hypomethylation. Am J Pa thol 2002; 160: 45-50.

27. Emoto Y, Kobayashi R, Akatsuka H, Hidaka H. Purification and characterization of a new member of the S-100 protein family from human placenta. Biochem Biophys Res Commun 1992; 182: 1246-1253.

28. Becker T, Gerke V, Kube E, Weber K. S100P, a novel Ca(2+)- binding protein from human placenta. cDNA cloning, recombinant protein expression and Ca (2+) binding properties. Eur J Biochem 1992; 207: 541-547.

29. Da Silva IDCG, Fu Hu Y, Russo IH, et al . S100P calcium binding protein overexpression is associated with immortalization of human breast epithelial cells in vi tro and early stages of breast cancer development in vivo . Int J Oncol 2000; 16: 231-240.

30. Amler LC, Agus DB, LeDuc C, et al . Dysregulated expression of androgen-responsive and nonresponsive genes in the androgen independent prostate cancer xenograft model CWR22-R1. Cancer Res 2000; 60: 6134-6141.

31. Mousses S, Bubendorf L, Wagner U, et al . Clinical validation of candidate genes associated with prostate cancer progression in the CWR22 model system using tissue microarrays . Cancer Res 2002; 62: 1256-1260.

32. Terris B, Blaveri E, Crnogorac-Jurcevic T, et al .

Characterization of gene expression profiles in intraductal papillarymucmous tumours of the pancreas. Am J Pathol 2002; 160: 1745-1754.

33. Cullen PJ, Lockyer PJ. Integration of calcium and Ras signalling. Na ture Rev Mol Cell Biol 2002; 3: 339- 348.

34. Chambers AF, Tuck AB . Ras-responsive genes and tumour metastasis. Crit .Rev Oncog 1993; 4: 95-114.

35. Cobbold et al . , The Menkes disease ATPase (ATP7A) is internalised via a Rac-lregulated, clathrin- and caveolae-independent pathway. Human Molecular Genetics, Vol. 12 No. 13 (2003) . 36. Curr Probl Cancer 2002:26:276-275 (July/August 2002) .

37. WHO, 2003 World Cancer Report.

38. Molecular alterations in pancreatic carcinoma: expression profiling shows that dysregulated expression of S100 genes is highly prevalent. Crnogorac-Jurecevic et al., Journal of Pathology 2003; 201: 63-74.

39. Isolation and sequencing of the cDNA encoding the 75-kD human sperm protein related to infertility. Zhang et al., Chin. Med. J. 105 (12), 998-1003 (1992).

40. Expression and function of the HSD-3.8 gene encoding a testis-specific protein. Lin et al . , Mol. Hum. Reprod. 7 (9) , 811-818 (2001) .

41. Jemal A, Murray T, Samuels A, Ghafoor A, Ward E, Thun MJ. Cancer statistics, 2003. CA Cancer J Clin 2003;53:5-26.

42. Logsdon CD, Simeone DM, Binkley C, Arumugam T, Greenson JK, Giordano TJ, Misek DE, Kuick R, Hanash S. Molecular profiling of pancreatic adenocarcinoma and chronic pancreatitis identifies multiple genes differentially regulated in pancreatic cancer. Cancer Res 2003;63:2649-57.

43. Crnogorac-Jurcevic T, Efthimiou E, Capelli P, Blaveri E, Baron A, Terris B, Jones M, Tyson K, Bassi C, Scarpa A, Lemoine NR. Gene expression profiles of pancreatic cancer and stromal desmoplasia. Oncogene 2001;20:7437-46. 44. Crnogorac-Jurcevic T, Missiaglia E, Blaveri E, Gangeswaran R, Jones M, Terris B, Costello E, Neoptolemos JP, Lemoine NR. Molecular alterations in pancreatic carcinoma: expression profiling shows that dysregulated expression of SlOO genes is highly prevalent. J Pathol 2003;201:63-74.

45. Nakamura T, Furukawa Y, Nakagawa H, Tsunoda T, Ohigashi H, Murata K, Ishikawa 0, Ohgaki K, Kashimura N, Miyamoto M, Hirano S, Kondo S, Katoh H, Nakamura Y, Katagiri T. Genome-wide cDNA microarray analysis of gene expression profiles in pancreatic cancers using populations of tumor cells and normal ductal epithelial cells selected for purity by laser microdissection . Oncogene 2004;23:2385-400.

46. Friess H, Ding J, Kleeff^'J, Fenkell L, Rosinski JA, Guweidhi A, Reidhaar-Olson JF, Korc M, Hammer J, Buchler M . Microarray-based identification of differentially expressed growth- and metastasis- associated genes in pancreatic cancer. Cell Mol Life Sci 2003;60:1180-99.

47. Iacobuzio-Donahue CA, Ashfaq R, Maitra A, Adsay NV, Shen-Ong GL, Berg K, Hollingsworth MA, Cameron JL, Yeo CJ, Kern SE, Goggins M, Hruban RH . Highly expressed genes in pancreatic ductal adenocarcinomas: a comprehensive characterization and comparison of the transcription profiles obtained from three major technologies. Cancer Res 2003;63:8614-22.

48. Ryu B, Jones J, Blades NJ, Parmigiani G, Hollingsworth MA, Hruban RH, Kern SE . Relationships and differentially expressed genes among pancreatic cancers examined by large-scale serial analysis of gene expression. Cancer Res 2002;62:819-26. 49. Tanaka H, Hata F, Nishimori H, Honmou 0, Yasoshima T, Nomura H, Ohno K, Hirai I, Kamiguchi K, Isomura H, Hirohashi Y, Denno R, Sato N, Hirata K. Differential gene expression screening between parental and highly metastatic pancreatic cancer variants using a DNA microarray. J Exp Clin Cancer Res 2003;22:307-13.

50. Yu XJ, Long J, Fu DL, Zhang QH, Ni QX . Analysis of gene expression profiles in pancreatic carcinoma by using cDNA microarray. Hepatobiliary Pancreat Dis Int 2003;2:467-70.

51. Grutzmann R, Pilarsky C, Staub E, Schmitt AO, Foerder M, Specht T, Hinzmann B, Dahl E, Alldinger I, Rosenthal A, Ockert D, Saeger HD. Systematic isolation of genes differentially expressed in normal and cancerous tissue of the pancreas. Pancreatology 2003;3:169-78.

52. Tan ZJ, Hu XG, Cao GS, Tang Y. Analysis of gene expression profile of pancreatic carcinoma using cDNA microarray. World J Gastroenterol 2003; 9:818- 23.

53. Shekouh AR, Thompson CC, Prime W, Campbell F, Hamlett J, Herrington CS, Lemoine NR, Crnogorac- Jurcevic T, Buechler MW, Friess H, Neoptolemos JP, Pennington SR, Costello E. Application of laser capture microdissection combined with two- dimensional elecfrophoresis for the discovery of differentially regulated proteins in pancreatic ductal adenocarcinoma. Proteomics 2003;3:1988-2001.

54. Koopmann J, Zhang Z, White N, Rosenzweig J, Fedarko N, Jagannath S, Canto MI, Yeo CJ, Chan DW, Goggins M. Serum diagnosis of pancreatic adenocarcinoma using surface-enhanced laser desorption and ionization mass specfromefry. Clin Cancer Res 2004;10:860-8.

55. Gygi SP, Rochon Y, Franza BR, Aebersold R. Correlation between protein and mRNA abundance in yeast. Mol Cell Biol 1999;19:1720-30.

56. Kurosaki M, Demontis S, Barzago MM, Garattini E, Terao M. Molecular cloning of the cDNA coding for mouse aldehyde oxidase: tissue distribution and regulation in vivo by testosterone. Biochem J 1999;341 ( Pt 1) :71-80.

57. Kim HJ, Lotan R. Identification of retinoid- modulated proteins in squamous carcinoma cells using high-throughput immunoblotting . Cancer Res 2004;64:2439-48.

58. Yoo GH, Piechocki MP, Ensley JF, Nguyen T, Oliver J, Meng H, Kewson D, Shibuya TY, Lonardo F,, Tainsky MA. Docetaxel induced gene expression patterns in head and neck squamous cell carcinoma using cDNA microarray and PowerBlot. Clin Cancer Res 2002;8:3910-21.

59. Malka D, Hammel P, Maire F, Rufat P, Madeira I, Pessione F, Levy P, Ruszniewski P. Risk of pancreatic adenocarcinoma in chronic pancreatitis. Gut 2002;51:849-52.

60. Lowenfels AB, Maisonneuve P, Cavallini G, Ammann RW, Lankisch PG, Andersen JR, Dimagno EP, Andren- Sandberg A, Domellof L. Pancreatitis and the risk of pancreatic cancer. International Pancreatitis Study Group. N Engl J Med 1993;328:1433-7.

61. Iacobuzio-Donahue CA, Maitra A, Shen-Ong GL, van Heek T, Ashfaq R, Meyer R, Walter K, Berg K, Hollingsworth MA, Cameron JL, Yeo CJ, Kern SE, Goggins M, Hruban RH . Discovery of novel tumor markers of pancreatic cancer using global gene expression technology. Am J Pathol 2002;160:1239-49.

62. Han H, Bearss DJ, Browne LW, Calaluce R, Nagle RB, Von Hoff DD. Identification of differentially expressed genes in pancreatic cancer cells using cDNA microarray. Cancer Res 2002;62:2890-6.

63. Iacobuzio-Donahue CA, Maitra A, Olsen M, Lowe AW, van Heek NT, Rosty C, Walter K, Sato N, Parker A, Ashfaq R, Jaffee E, Ryu B, Jones J, Eshleman JR, Yeo CJ, Cameron JL, Kern SE, Hruban RH, Brown PO, Goggins M. Exploration of global gene expression patterns in pancreatic adenocarcinoma using cDNA microarrays . Am J Pathol 2003;162:1151-62.

64. Izumo A, Yamaguchi K, Eguchi T, Nishiyama K, Yamamoto H, Yonemasu H, Yao T, Tanaka M, Tsuneyoshi M. Mucinous cystic tumor of the pancreas: immunohistochemical assessment of "ovarian-type stroma". Oncol Rep 2003;10:515-25.

65. Vishwanatha JK, Chiang Y, Kumble KD, Hollingsworth MA, Pour PM. Enhanced expression of annexin II in human pancreatic carcinoma cells and primary pancreatic cancers. Carcinogenesis 1993;14:2575-9.

66. Yen TW, Aardal NP, Bronner MP, Thorning DR, Savard CE, Lee SP, Bell RH, Jr. Myofibroblasts are responsible for the desmoplastic reaction surrounding human pancreatic carcinomas. Surgery 2002;131:129-34.

67. Tani T, Lumme A, Linnala A, Kivilaakso E, Kiviluoto T, Burgeson RE, Kangas L, Leivo I, Virtanen I. Pancreatic carcinomas deposit laminin-5, preferably adhere to laminin-5, and migrate on the newly deposited basement membrane. Am J Pathol 1997;151:1289-302. 68. Tanno S, Mitsuuchi Y, Altomare DA, Xiao GH, Testa JR. AKT activation up-regulates insulin-like growth factor I receptor expression and promotes invasiveness of human pancreatic cancer cells. Cancer Res 2001;61:589-93.

69. Ng SSW, Tsao MS, Chow S, Hedley DW. Inhibition of phosphatidylinositide 3-kinase enhances gemcitabine- induced apoptosis in human pancreatic cancer cells. Cancer Res 2000;60:5451-5.

70. Kusama T, Mukai M, Iwasaki T, Tatsuta M, Matsumoto Y, Akedo H, Inoue M, Nakamura H. 3-hydroxy-3- methylglutaryl-coenzyme a reductase inhibitors reduce human pancreatic cancer cell invasion and metastasis. Gastroenterology 2002;122:308-17.

71. Hennig R, Ding XZ, Tong WG, Schneider MB, Standop J, Friess H, Buchler MW, Pour PM, Adrian TE . 5- Lipoxygenase and leukotriene B(4) receptor are expressed in human pancreatic cancers but not in pancreatic ducts in normal tissue. Am J Pathol 2002;161:421-8.

72. Maitra A, Iacobuzio-Donahue C, Rahman A, Sohn TA, Argani P, Meyer R, Yeo CJ, Cameron JL, Goggins M, Kern SE, Ashfaq R, Hruban RH, Wilentz RE. Immunohistochemical validation of a novel epithelial and a novel stromal marker of pancreatic ductal adenocarcinoma identified by global expression microarrays: sea urchin fascin homolog and heat shock protein 47. Am J Clin Pathol 2002;118:52-9.

73. Hall PA, Coates P, Lemoine NR, Horton MA. Characterization of integrin chains in normal and neoplastic human pancreas. J Pathol 1991;165:33-41. 74. Goto M, Shinno H, Ichihara A. Isozyme patterns of branched-chain amino acid transaminase in human tissues and tumors. Gann 1977;68:663-7.

75. Friess H, Ding J, Kleeff J, Liao Q, Berberat PO, Hammer J, Buchler MW. Identification of disease- specific genes in chronic pancreatitis using DNA array technology. Ann Surg 2001;234:769-78; discussion 778-9.

76. Lowenfels AB, Maisonneuve P, DiMagno EP, Elitsur Y, Gates LK, Jr., Perrault J, Whitcomb DC. Hereditary pancreatitis and the risk of pancreatic cancer. International Hereditary Pancreatitis Study Group. J Natl Cancer Inst 1997;89:442-6.

77. Tucker ON, Dannenberg AJ, Yang EK, Zhang F, Teng L, Daly JM, Soslow RA, Masferrer JL, Woerner BM, Koki AT, Fahey TJ, 3rd. Cyclooxygenase-2 expression is up-regulated in human pancreatic cancer. Cancer Res 1999;59:987-90.

78. Farrow B, Sugiyama Y, Chen A, Uffort E, Nealon W, Mark Evers B. Inflammatory mechanisms contributing to pancreatic cancer development. Ann Surg 2004;239:763-9; discussion 769-71.

79. Chandler NM, Canete JJ, Callery MP . Increased expression of NF-kappa B subunits in human pancreatic cancer cells. J Surg Res 2004;118:9-14.

80. Preston-Martin S, Pike MC, Ross RK, Jones PA, Henderson BE. Increased cell division as a cause of human cancer. Cancer Res 1990;50:7415-21.

81. Hopfner R, Mousli M, Jeltsch JM, Voulgaris A, Lutz Y, Marin C, Bellocq JP, Oudet P, Bronner C. ICBP90, a novel human CCAAT binding protein, involved in the regulation of topoisomerase Ilalpha expression. Cancer Res 2000;60:121-8. 82. Trotzier MA, Bronner C, Bathami K, Mathieu E, Abbady AQ, Jeanblanc M, Muller CD, Rochette-Egly C, Mousli M. Phosphorylation of ICBP90 by protein kinase A enhances topoisomerase Ilalpha expression. Biochem Biophys Res Commun 2004;319:590-5.

83. Mousli M, Hopfner R, Abbady AQ, Monte D, Jeanblanc M, Oudet P, Louis B, Bronner C. ICBP90 belongs to a new family of proteins with an expression that is deregulated in cancer cells . Br J Cancer 2003;89:120-7.

84. Maitra A, Adsay NV, Argani P, Iacobuzio-Donahue C, De Marzo A, Cameron JL, Yeo CJ, Hruban RH . Multicomponent analysis of the pancreatic adenocarcinoma progression model using a pancreatic intraepithelial neoplasia tissue microarray. Mod Pathol 2003;16:902-12.

85. de Fraipont F, Nicholson AC, Feige JJ, Van Meir EG. Thrombospondins and tumor angiogenesis . Trends Mol Med 2001;7:401-7.

86. Bornstein P. Thrombospondins as matricellular modulators of cell function. J Clin Invest 2001;107:929-34.

87. Kamochi J, Tokunaga T, Tomii Y, Abe Y, Hatanaka H, Kijima H, Yamazaki H, Watanabe N, Matsuzaki S, Ueyama Y, Nakamura M. Overexpression of the thrombospondin 2 (TSP2) gene modulated by the matrix metalloproteinase family expression and production in human colon carcinoma cell line. Oncol Rep 2003;10:881-4.

88. Grutz ann R, Saeger HD, Luttges J, Schackert HK, Kalthoff H, Kloppel G, Pilarsky C. Microarray-based gene expression profiling in pancreatic ductal carcinoma: status quo and perspectives. Int J Colorectal Dis 2004;19:401-13.

89. Kunz M, Koczan D, Ibrahim SM, Gillitzer R, Gross G, Thiesen HJ. Differential expression of thrombospondin 2 in primary and metastatic malignant melanoma. Acta Derm Venereol 2002;82:163-9.

90. Koda a J, Hashimoto I, Seki N, Hongo A, Yoshinouchi M, Okuda H, Kudo T. Thrombospondin-1 and -2 messenger RNA expression in epithelial ovarian tumor. Anticancer Res 2001;21:2983-7.

91. Kyriakides TR, Zhu YH, Yang Z, Bornstein P. The distribution of the matricellular protein thrombospondin 2 in tissues of embryonic and adult mice. J Histochem Cytochem 1998;46:1007-15.

92. Lopes N, Gregg D, Vasudevan S, Hassanain H, Goldschmidt-Clermont P, Kovacic H. Thrombospondin 2 regulates cell proliferation induced by Racl redox- dependent signaling. Mol Cell Biol 2003;23:5401-8.

93. Irani K, Xia Y, Zweier JL, Sollott SJ, Der CJ, Fearon ER, Sundaresan M, Finkel T, Goldschmidt- Clermont PJ. Mitogenic signaling mediated by oxidants in Ras-transformed fibroblasts. Science 1997;275:1649-52.

94. Joneson T, Bar-Sagi D. A Racl effector site controlling mitogenesis through superoxide production. J Biol Chem 1998;273:17991-4.

95. Turner Z, Lund C, Tolshave J, Vural B, Tonnesen T, Horn N. Identification of point mutations in 41 unrelated patients affected with Menkes disease. Am J Hum Genet 1997;60:63-71.

96. Petris MJ, Mercer JF. The Menkes protein (ATP7A; MNK) cycles via the plasma membrane both in basal and elevated extracellular copper using a C-terminal di-leucine endocytic signal. Hum Mol Genet 1999;8:2107-15.

97. Cunningham J, Leffell M, Mearkle P, Harmatz P. Elevated plasma ceruloplasmin in insulin-dependent diabetes mellitus : evidence for increased oxidative stress as a variable complication. Metabolism 1995;44:996-9.

98. Nayak SB, Bhat VR, Upadhyay D, Udupa SL. Copper and ceruloplasmin status in serum of prostate and colon cancer patients . Indian J Physiol Pharmacol 2003;47:108-10.

99. Senra Varela A, Lopez Saez JJ, Quintela Senra D. Serum ceruloplasmin as a diagnostic marker of cancer. Cancer Lett 1997;121:139-45.

100. Gaatke LM, Chow CK. Copper toxicity, oxidative stress, and antioxidant nutrients. Toxicology 2003;189:147-63.

101. Garattini E, Mendel R, Romao MJ, Wright R, Terao M. Mammalian molybdo-flavoenzymes, an expanding family of proteins: structure, genetics, regulation, function and pathophysiology . Biochem J 2003; 372: 15- 32.

102. Obach RS . Potent inhibition of human liver aldehyde oxidase by raloxifene. Drug Metab Dispos 2004; 32:89- 97.

103. Huang DY, Furukawa A, Ichikawa Y. Molecular cloning of retinal oxidase/aldehyde oxidase cDNAs from rabbit and mouse livers and functional expression of recombinant mouse retinal oxidase cDNA in Escherichia coli. Arch Biochem Biophys 1999; 36 : 264- 72.

104. Tulachan SS, Doi R, Kawaguchi Y, Tsuji S, Nakajima S, Masui T, Koizumi M, Toyoda E, Mori T, Ito D, Kami K, Fujimoto K, Imamura M. All-trans retinoic acid induces differentiation of ducts and endocrine cells by mesenchymal/epithelial interactions in embryonic pancreas. Diabetes 2003;52:76-84.

105. Moskovitz AH, Linford NJ, Brentnall TA, Bronner MP, Storer BE, Potter JD, Bell RH, Jr., Rabinovitch PS. Chromosomal instability in pancreatic ductal cells from patients with chronic pancreatitis and pancreatic adenocarcinoma. Genes Chromosomes Cancer 2003;37:201-6.

106. Crnogorac-Jurcevic T, Efthimiou E, Nielsen T, Loader J, Terris B, Stamp G, Baron A, Scarpa A, Lemoine NR. Expression profiling of microdissected pancreatic adenocarcinomas. Oncogene 2002;21:4587-94.

107. Sasaki H et al . .Expression of periostin, homologous with an insect cell adhesion molecule, as a prognostic marker in non-small cell lung cancer. Jpn J Cancer Res.2001 Aug 92 ( 8 ) : 869-73.

108. Ismail eta 1. Differential gene expression between normal and tumor derived ovarian epithelial cells. Cancer Res. 2000 Dec 1; 60 (23) : 6744-9.

109. Gillan et al . Periostin secreted by epithelial ovarian carcinoma is a ligand alpha (V) beta (3) and alpha (V) beta (5) integrins and promotes motility. Cancer Res. 2002 Sep 15; 62 ( 18 ) : 5538.

110. Shao et al . Acquired expression of periostin by human breast cancers promotes tumor anigigenesis through up-regulation of vascular endothelial growth factor receptor 2 expression. Mol Cell Biol. 2004 May;24 (9) :3992-4003.

111. Bao et al. Periostin potently promotes metastatic growth of colon cancer augmenting cell survival via the Akt/PKB pathway. Cancer Cell. 2004 Apr;4 (4) :329-39.

112. Thompson and Weigel . hAG-2, the human homologue of the Xenopus laevis cement gland gene XAG-2, is co- expressed with estrogen receptor in breast cancer cell lines. Biochem Biophys res Commun. 1998 Oct 9;25(1) :lll-6.

113. Fletcher et al . hAG-2 and hAG-3, human homologues of genes involved in differentiation, are associated with oestrogen receptor-positive breast tumours and interact with metastasis gene C4.4a and dystroglycan. Br J Cancer. 2003 Feb 24 ; 88 (4 ) : 579- 85.

114. Mousli et al . ICBP90 belongs to a new family of proteins with an expression that is deregulated in cancer cells. Br J Cancer. 2003 Jul 7 : 89 (1) : 120-7.

115. Aberger F, Weidinger G, Grunz H, Richter K. Anterior specification of embryonic ectoderm: the role of the Xenopus cement gland-specific gene XAG-2. Mech Dev 1998;72:115-30.

Claims

Claims :

1. A method of detecting a cancerous condition of the pancreas or predisposition to a cancerous condition of the pancreas in a patient comprising the step of detecting, in a sample taken from the patient, the presence of a marker selected from the group consisting of: (i) SPAGl; (ii) MNK; (iii) UHRFl (iv) Periostin; or (v) AGR2, wherein said marker is a marker polypeptide, nucleic acid or anti-marker antibody, or a fragment of said marker polypeptide, nucleic acid or anti-marker antibody.

2. A method as claimed in claim 1 wherein said method comprises detecting the upregulation of the marker polypeptide, nucleic acid, anti-marker antibody or fragment .

3. A method as claimed in claim 2 wherein detection of upregulation of the marker comprises detecting an increase in the level of the marker relative to a control or standard.

4. A method as claimed in claim 2 or 3 wherein detection of upregulation of the marker comprises determining the level of expression of marker polypeptide or polypeptide fragment (s) in the sample and comparing said level with a control or standard.

5. A method as claimed in claim 2 or 3 wherein detection of upregulation of the marker comprises determining the level of mRNA, or mRNA fragment (s), encoding the marker in the sample and comparing said level with a control or standard.

6. A method as claimed in claim 2 or 3 wherein the detection of upregulation comprises determining the level of amplification of genomic DNA encoding said marker in the sample and comparing the level with a control or standard.

7. A method as claimed in claim 2 or 3 wherein the detection of upregulation comprises determining the level of anti-marker antibody, or antibody fragment (s), in the sample and comparing said level with a control or standard.

8. A method as claimed in any one of claims 3 to 7 wherein the standard comprises a corresponding value for the respective marker level which is indicative of the absence of a cancerous condition of the pancreas or predisposition to a cancerous condition of the pancreas.

9. A method of assessing the effectiveness of treatment of a cancerous condition of the pancreas comprising analysing, in a sample taken from a patient, the presence of a marker selected from the group consisting of: (i) SPAGl; (ii) MNK; (iii) UHRFl (iv) Periostin; or (v) AGR2, wherein said marker is a marker polypeptide, nucleic acid or anti-marker antibody, or a fragment of said marker polypeptide, nucleic acid or anti-marker antibody.

10. A method as claimed in claim 9 wherein the method comprises detecting the presence or absence of the marker in the sample.

11. A method as claimed in claim 9 or 10 wherein the method comprises: (a) at a first point in time, detecting the level of the marker in a patient sample; (b) at a second point in time, detecting the level of the marker in a patient sample; and (c) monitoring a change in said level.

12. A method as claimed in any one of the preceding claims wherein said method forms part of a method of diagnosis or prognosis of the cancerous condition.

13. A method as claimed in any one of the preceding claims wherein said method forms part of a method of monitoring the cancerous condition in the patient.

14. A method as claimed in any one of the preceding claims wherein said method forms part of a method of screening a patient for the presence of the cancerous condition .

15. A method as claimed in any one of the preceding claims wherein said method is performed in vitro.

16. A method as claimed in any one of the preceding claims wherein detection of the marker comprises immunological detection.

17. A method as claimed in any one of the preceding claims comprising detecting or analysing the presence of any two, three, four or five of said markers.

18. A method according to any one of the preceding claims wherein a said step of detecting the presence of a marker polypeptide, or fragment thereof, comprises contacting an antibody to said marker polypeptide or fragment with said sample and detecting bound complex of marker polypeptide and antibody.

19. A method according to claim 18 wherein said detection involves western blot, ELISA or immunohistochemical staining performed on the sample or a portion of the sample.

20. A method as claimed in any one of the preceding claims wherein the marker is a SPAGl nucleic acid and said method comprises detecting the presence of a SPAGl nucleic acid, or a fragment thereof, said method comprising the step of contacting said sample with a nucleic acid probe and detecting hybridisation of said probe to said marker, wherein said probe is: (a) a nucleic acid having substantial sequence identity to SEQ ID No .2 or to a fragment thereof; and/or (b) a nucleic acid which hybridises to SEQ ID No.2 under high or very high stringency conditions.

21. A method as claimed in any one of claims 1 to 19 wherein the marker is a MNK nucleic acid and said method comprises detecting the presence of a MNK nucleic acid, or a fragment thereof, said method comprising the step of contacting said sample with a nucleic acid probe and detecting hybridisation of said probe to said marker, wherein said probe is: (a) a nucleic acid having substantial sequence identity to SEQ ID No .4 or to a fragment thereof; and/or (b) a nucleic acid which hybridises to SEQ ID No.4 under high or very high stringency conditions.

22. A method as claimed in any one of claims 1 to 19 wherein the marker is an UHRFl nucleic acid and said method comprises detecting the presence of an UHRFl nucleic acid, or a fragment thereof, said method comprising the step of contacting said sample with a nucleic acid probe and detecting hybridisation of said probe to said marker, wherein said probe is: (a) a nucleic acid having substantial sequence identity to SEQ ID No.10 or to a fragment thereof; and/or (b) a nucleic acid which hybridises to SEQ ID No.10 under high or very high stringency conditions.

23. A method as claimed in any one of claims 1 to 19 wherein the marker is a periostin nucleic acid and said method comprises detecting the presence of a periostin nucleic acid, or a fragment thereof, said method comprising the step of contacting said sample with a nucleic acid probe and detecting hybridisation of said probe to said marker, wherein said probe is: (a) a nucleic acid having substantial sequence identity to SEQ ID No .6 or to a fragment thereof; and/or (b) a nucleic acid which hybridises to SEQ ID No .6 ^' under high or very high stringency conditions.

24. A method as claimed in any one of claims 1 to 19 wherein the marker is an AGR2 nucleic acid and said method comprises detecting the presence of an AGR2 nucleic acid, or a fragment thereof, said method comprising the step of contacting said sample with a nucleic acid probe and detecting hybridisation of said probe to said marker, wherein said probe is: (a) a nucleic acid having substantial sequence identity to SEQ ID No .8 or to a fragment thereof; and/or (b) a nucleic acid which hybridises to SEQ ID No.8 under high or very high stringency conditions.

25. A method as claimed in any preceding claim wherein said cancerous condition is selected from the group consisting of: (a) pancreatic intraepithelial neoplasia (PanlN) ; (b) ductal adenocarcinoma (DA) ; (c) mucinous cystic neoplasm (MCN) ; (d) intraductal papillary mucinous neoplasm ( I PMN ) ; (e) pancreacticoblastoma (PB) ; (f) acinar cell carcinoma (ACC) ; (g) solid pseudopapillary neoplasm (SPN) ; and (h) pancreatic adenocarcinoma (PDAC) .

26. A method as claimed in any one of the preceding claims wherein said sample is selected from the group consisting of: (I) a quantity of blood; (II) a quantity of serum; (III) a quantity of pancreatic juice; (IV) a tissue sample or biopsy; or (V) cells isolated from the patient.

27. A method as claimed in claim 26 wherein said tissue sample, biopsy or cells are from the pancreas of the patient .

28. An assay kit for use in diagnosing a cancerous condition of the pancreas or predisposition to a cancerous condition of the pancreas or for assessing the effectiveness of a treatment for a cancerous condition of the pancreas, in a patient, the kit comprising an antibody to SPAGl, MNK, UHRFl, periostin or AGR2 polypeptide, or to a fragment thereof.

29. An assay kit for use in diagnosing a cancerous condition of the pancreas or predisposition to a cancerous condition of the pancreas or for assessing the effectiveness of a treatment for a cancerous condition of the pancreas, in a patient, the kit comprising a nucleic acid capable of hybridising, under conditions of high or very high stringency, to one of: (a) SEQ ID No.2 (SPAGl) ; (b) SEQ ID No.4 (MNK) ; (c) SEQ ID No.10 (UHRFl); (d) SEQ ID No.6 (periostin); (e) SEQ ID No.8 (AGR2) ; (f ) a fragment of one of (a) to (e) ; (g) an RNA transcript of one of (a) to (e) or a fragment thereof; or (h) to the complementary sequence of a sequence of any one of (a) to (g) .

30. An antibody or antibody fragment which binds to one of (a) SPAGl; (b) MNK; (c) UHRFl (d) Periostin; ( e ) AGR2 ; polypeptide, or fragment thereof, for use in the treatment of a cancerous condition of the pancreas.

31. An antibody or antibody fragment as claimed in claim 30 which is conjugated to an anti-tumour compound.

32. An antibody or antibody fragment as claimed in claim 30 or 31 which binds to SPAGl polypeptide, or fragment thereof, for use in the treatment of a cancerous condition in a female patient.