WO2008031165A1

WO2008031165A1 - Methods and compositions for the diagnosis and treatment of tumours

Info

Publication number: WO2008031165A1
Application number: PCT/AU2007/001364
Authority: WO
Inventors: Kerrie Mcdonald; Bruce Robinson; Maree O'sullivan; Glenn Stone
Original assignee: Northern Sydney And Central Coast Area Health Service
Priority date: 2006-09-14
Filing date: 2007-09-14
Publication date: 2008-03-20

Abstract

The present invention relates to methods for the diagnosis, including prognosis, of tumours of the central nervous system, including of the brain, particularly tumours of neuroepithelial tissue (glioma(s)). The present invention also relates to novel markers for use in such methods, including the markers IQGAP1, Homer1, IGFBP2, and C1QL1. The present invention also relates to compositions and methods for the treatment of tumours, and to novel screening methods for the identification of agents for therapeutic use in the treatment of tumours.

Description

Methods and compositions for the diagnosis and treatment of tumours

Technical Field

The present invention relates to methods for the diagnosis and prognosis of tumours of the central nervous system, including of the brain, particularly tumours of neuroepithelial tissue (glioma(s)). The present invention also relates to novel markers for use in such methods. The present invention also relates to compositions and methods for the treatment of tumours, and to novel screening methods for the identification of agents for therapeutic use in the treatment of tumours.

Background

Gliomas are the most common primary tumour of the central nervous system (CNS) and are separated into groups based upon their presumed cell of origin. The two commonest groups are astrocytoma and oligodendroglioma. Glial tumours are generally graded on the basis of the most malignant area identified, generally according to the World Health Organisation (WHO) system (Kleihues et al., 2000). This system uses presence or absence of nuclear atypia, mitosis, microvascular proliferation and necrosis as indicators of increasing tumour aggressiveness. Tumour heterogeneity is a significant issue faced by pathologists. Patients diagnosed with gliomas demonstrate highly refractory responses to available therapies. Current treatment for glioma patients include surgical debulking followed by radiotherapy and adjuvant chemotherapy treatment with temozolomide (Stupp et al., 2005.). Oligodendrogliomas respond much more favourably to treatment which is reflected in long term survival. There are also large variations in survival, even amongst high grade tumours such as glioblastoma multiforme (GBM). Whilst the classification system is largely reproducible, it is acknowledged that some gliomas are refractory to treatment despite no discernible histological differences to sensitive tumours. Furthermore the current classification system does not offer mechanistic insights or suggest treatment strategies. The diagnosis of oligodendrogliomas and mixed oligoastrocytomas has increased dramatically in recent years raising concerns that some astrocytomas could be misidentified as oligodendrogliomas. In addition, the identification of GBMs with morphological features associated with oligodendroglial differentiation is also increasing in incidence. The histological criteria for the classification of these tumours are often poorly defined, making the differential diagnosis between anaplastic oligoastrocytoma and ordinary glioblastoma a difficult and very subjective issue. There is thus an increasing need to improve current diagnostic practices and to identify novel markers that could increase diagnostic accuracy.

Summary of the Invention The present invention aims to provide methods for improved diagnosis of tumours of the central nervous system, such as brain tumours and in particular tumours of neuroepithelial tissue (glioma(s)).

In a first aspect of the invention there is provided a method of diagnosing a tumour of the central nervous system (CNS) in an individual comprising: a) determining the expression of at least one gene selected from the group consisting of IQGAPl, Homer 1, and ClQLl or determining the expression of at least two genes selected from the group consisting of IQGAPl, Homerl, IGFBP2, and ClQLl in a biological sample from said individual; and b) comparing said expression with that of at least one reference sample; wherein a difference in said expression is diagnostic of a tumour of the CNS. In one embodiment the tumour of the CNS is a brain tumour, such as a glioma. In one embodiment the glioma is selected from the group consisting of glioblastoma multiforme, astrocytomas, oligodendrogliomas, oligoastrocytomas and ependymomas.

In one embodiment the method comprises discriminating high grade tumours and low grade tumours. In one embodiment the method is a method of diagnosing a biologically aggressive type or grade of tumour, such as an anaplastic oligodendroglioma associated with poor survival outcome. In one embodiment the method is a method of predicting responsiveness of the individual to therapy or predicting survival of the individual. In one embodiment the method is a method of grading a tumour.

In one embodiment the method comprises determining the level of expression of one or more gene sets selected from the group consisting of (i) Homerl and IQGAPl; (ii) IGFBP2 and ClQLl; (iii) IQGAPl and IGFBP2; (iv) IGFBP2 and Homerl; and (v) IQGAPl and ClQLl.

In one embodiment the method further comprises determining the expression of at least one gene selected from the group consisting of LGALSl, LRRC20, KIAA0599, KPNA5, NFYB, CARHSPl, COPZ2, ARS, HS7SLP, CH13L1, HSxS138, SERPINA3, SPPl, and RBPl.

In one embodiment the method is a method of diagnosing high grade glioma, wherein expression of IQGAPl in the biological sample is about 10-fold higher than IQGAPl expression in said reference sample. In one embodiment the method is a method of diagnosing high grade glioma wherein the expression of IGFBP2 in the biological sample is about 20-fold higher than IGFBP2 expression in said reference sample, and the expression of IQGAPl in the biological sample is about 10-fold higher than IQGAPl expression in said reference sample.

In one embodiment the method is a method of discriminating high grade glioma and low grade glioma, wherein expression of IQGAPl in high grade glioma is about 5-fold higher in high grade glioma compared to low grade glioma. In one embodiment the method is a method of discriminating high grade glioma and low grade glioma, wherein expression of IGFBP2 in high grade glioma is about 15-fold higher in high grade glioma compared to low grade glioma. In one embodiment of the method absence of expression of IQGAPl and IGFBP2 is diagnostic of long term survival in an individual having glioblastoma multiforme.

In one embodiment the expression of Homer 1 is reduced in said biological sample compared to said reference sample.

In one embodiment the expression of ClQLl is elevated in said biological sample compared to said reference sample. In one embodiment expression of ClQLl is about 10-fold higher in said biological sample compared to said reference sample.

In one embodiment the method comprises determining the expression of said at least one gene in a plurality of samples of said individual, wherein the samples are sourced from different states of said individual. The biological sample may be a known tumour sample or a sample suspected of comprising tumour cells.

In an embodiment the biological sample is brain tissue, such as may be obtained by biopsy.

In one embodiment determining the expression of a gene comprises determining the level of a nucleic acid sequence or fragment thereof corresponding to said gene or of a polypeptide or fragment thereof encoded by said gene.

In one embodiment the reference sample is normal brain tissue obtained from a similar or identical region of the brain of a second individual, such as a cadaver. In one embodiment the reference sample is a whole brain nucleic acid preparation or a brain tissue type specific nucleic acid preparation. In one embodiment the nucleic acid preparation is selected from the group consisting of total RNA, cDNA, poly A⁺ RNA.

In one embodiment the reference sample is a tumour sample, such as a known high grade glioma or a known low grade glioma.

In one embodiment the method comprises a method of discriminating high grade tumours and low grade tumours. In one embodiment determining expression comprises contacting said sample with at least one antibody specific to a polypeptide encoded by said gene or a fragment thereof.

In a second aspect of the invention there is provided a method of treating a tumour of the CNS in an individual comprising a) diagnosing a tumour of the central nervous system (CNS) in said individual by a method comprising determining the expression of at least one gene selected from the group consisting of IQGAPl, Homer 1, and ClQLl or determining the expression of at least two genes selected from the group consisting of

IQGAPl, Homer 1, IGFBP2, and ClQLl in a biological sample from said individual, and comparing said expression with that of a reference sample; wherein a difference in said expression is diagnostic of a tumour of the CNS; and b) formulating a therapeutic regime suitable for the treatment of an individual having said diagnosed tumour; and c) administering said therapeutic regime to said individual.

In a third aspect of the invention there is provided a kit for use in diagnosing a tumour of the CNS, the kit comprising at least one probe specific for a gene or gene product selected from the group consisting of IQGAP 1 , Homer 1 , and C 1 QL 1.

In one embodiment the kit further comprises at least one probe specific for IGFBP2 or a gene product thereof.

In one embodiment the kit further comprises at least one probe specific for a gene or gene product selected from the group consisting of LGALSl, LRRC20, KIAA0599, KPNA5, NFYB, CARHSPl, COPZ2, ARS, HS7SLP, CH13L1, HSxS138, SERPINA3, SPPl, and RBPl.

In one embodiment the probe is selected from the group consisting of a nucleic acid sequence and an antibody. In one embodiment the kit further comprises one or more additional components selected from the group consisting of (i) one or more reference probe(s); (ii) one or more detection reagent(s); (iii) one or more agent(s) for immobilising a polypeptide on a solid support; (iv) a solid support material; (v) instructions for use of the kit or a component(s) thereof in a method for diagnosing a tumour of the CNS. In one embodiment the kit comprises one or more probe(s) immobilised on a solid support, such as a biochip. In a fourth aspect of the invention there is provided a biochip comprising at least one probe specific for a gene or gene product selected from the group consisting of IQGAPl, Homerl, and ClQLl.

In one embodiment the biochip further comprises at least one probe specific for IGFBP2 or a gene product thereof. In one embodiment the biochip further comprises at least one probe specific for a gene or gene product selected from the group consisting of LGALSl, LRRC20, KIAA0599, KPNA5, NFYB, CARHSPl, COPZ2, ARS, HS7SLP, CH13L1, HSxS138, SERPINA3, SPPl, and RBPl. In one embodiment the probe(s) is at least one member selected from the group consisting of a nucleic acid sequence and an antibody.

In a fifth aspect of the invention there is provided a method for the treatment of a tumour of the CNS in a subject wherein said tumour is characterised by elevated expression of IQGAPl, the method comprising administering a therapeutically effective amount of an agent selected from the group consisting of (i) an inhibitor of expression of IQGAPl; and (ii) an antagonist of a biological activity of a polypeptide encoded by IQGAPl .

In a sixth aspect of the invention there is provided a method for the treatment of a tumour of the CNS in a subject wherein said tumour is characterised by decreased expression of Homer 1, the method comprising administering a therapeutically effective amount of an agent selected from the group consisting of (i) a stimulant of expression of

Homerl; and (ii) a stimulant of a biological activity of a polypeptide encoded by Homer 1.

In a seventh aspect of the invention there is provided a method for the treatment of a tumour of the CNS in a subject wherein said tumour is characterised by elevated expression of ClQLl, the method comprising administering a therapeutically effective amount of an agent selected from the group consisting of (i) an inhibitor of expression of ClQLl; and (ii) an antagonist of a biological activity of a polypeptide encoded by ClQLl.

In one embodiment the antagonist of a biological activity of a polypeptide encoded by IQGAPl or ClQLl, respectively, is an antibody specific for IQGAPl or ClQLl, respectively. In a preferred embodiment the antibody is a monoclonal antibody.

In an eighth aspect the invention provides a method for screening for an agent capable of modulating the expression of one or more genes selected from the group consisting of IQGAPl, Homerl, IGFBP2 and ClQLl, the method comprising exposing a cell to a candidate agent and comparing the expression of the gene in the presence of the agent to the expression of the gene in the absence of the agent, whereby a difference in expression of the gene in the presence compared to the absence of the agent is indicative of an agent capable of modulating the expression of the gene. In one embodiment, multiple candidate agents may be screened simultaneously, such as by exposing said cell to a composition comprising a plurality of discrete candidate agents.

In one embodiment the cell is an isolated cell or cells. In one embodiment the cell is comprised in an organism, such as a mouse, rat or primate.

In one embodiment the organism is a transgenic organism, comprising recombinant nucleic acid sequence encoding one or more of IQGAPl, Homer 1, IGFBP2 and ClQLl . In one embodiment the cell is a tumour cell, such as a glioma.

In one embodiment the agent is selected from the group consisting of antisense molecules and siRNA.

In a ninth aspect the invention provides a method for screening for an agent capable of modulating a biological activity of a polypeptide, or fragment thereof, encoded by a gene selected from the group consisting of IQGAPl, Homer 1, IGFBP2 and ClQLl, the method comprising exposing said polypeptide or said fragment to a candidate agent under conditions suitable for the expression of the biological activity and comparing the biological activity of the polypeptide or fragment in the presence of the agent to the biological activity in the absence of the agent, whereby a difference in biological activity of the polypeptide or fragment in the presence compared to the absence of the agent is indicative of an agent capable of modulating a biological activity of the polypeptide or fragment.

In one embodiment the polypeptide is an isolated polypeptide.

In one embodiment the polypeptide is expressed by a transgenic organism. In one embodiment the agent is an antagonist of a biological activity of a polypeptide encoded by a gene selected from the group consisting of IQGAPl, Homer 1, IGFBP2 and ClQLl.

In one embodiment the agent is a monoclonal antibody which binds to the polypeptide. In a tenth aspect of the invention there is provided a method for the treatment of a tumour of the CNS in a subject wherein said tumour is characterised by elevated expression of IGFBP2, the method comprising administering a therapeutically effective amount of an agent selected from the group consisting of (i) an inhibitor of expression of IGFBP2; and (ii) an antagonist of a biological activity of a polypeptide encoded by IGFBP2. In a thirteenth aspect of the present invention there is provided an isolated antibody or fragment thereof that specifically binds to a polypeptide encoded by IQGAPl or a fragment thereof. In one embodiment the antibody specifically binds to a polypeptide selected from the group consisting of (i) a polypeptide encoded by a nucleic acid sequence comprising the sequence of SEQ ID NO:1 or a sequence complementary thereto; (ii) a polypeptide comprising the amino acid sequence of SEQ ID NO:2; (iii) a polypeptide comprising an amino acid sequence of at least about 80% identity to the polypeptide of (i) or (ii); (iv) a polypeptide comprising an antigenic fragment of (i), (ii), or (iii). In a fourteenth aspect of the present invention there is provided an isolated antibody or fragment thereof that specifically binds to a polypeptide encoded by IGFBP2 or a fragment thereof. In one embodiment the antibody specifically binds to a polypeptide selected from the group consisting of (i) a polypeptide encoded by a nucleic acid sequence comprising the sequence of SEQ ID NO: 3 or a sequence complementary thereto; (ii) a polypeptide comprising the amino acid sequence of SEQ ID NO:4; (iii) a polypeptide comprising an amino acid sequence of at least about 80% identity to the polypeptide of (i) or (ii); (iv) a polypeptide comprising an antigenic fragment of (i), (ii), or (iii).

In a fifteenth aspect of the present invention there is provided an isolated antibody or fragment thereof that specifically binds to a polypeptide encoded by Homer 1 or a fragment thereof. In one embodiment the antibody specifically binds to a polypeptide selected from the group consisting of (i) a polypeptide encoded by a nucleic acid sequence comprising the sequence of SEQ ID NO: 5 or a sequence complementary thereto; (ii) a polypeptide comprising the amino acid sequence of SEQ ID NO:6; (iii) a polypeptide comprising an amino acid sequence of at least about 80% identity to the polypeptide of (i) or (ii); (iv) a polypeptide comprising an antigenic fragment of (i), (ii), or (iii).

In a sixteenth aspect of the present invention there is provided an isolated antibody or fragment thereof that specifically binds to a polypeptide encoded by ClQLl or a fragment thereof. In one embodiment the antibody specifically binds to a polypeptide selected from the group consisting of (i) a polypeptide encoded by a nucleic acid sequence comprising the sequence of SEQ ID NO:7 or a sequence complementary thereto; (ii) a polypeptide comprising the amino acid sequence of SEQ ID NO:8; (iii) a polypeptide comprising an amino acid sequence of at least about 80% identity to the polypeptide of (i) or (ii); (iv) a polypeptide comprising an antigenic fragment of (i), (ii), or (iii).

In one embodiment the antibody is a monoclonal antibody.

The summary of the invention described above is not limiting and other features and advantages of the invention will be apparent from the following detailed description of the preferred embodiments, as well as from the claims.

Definitions The term "therapeutically effective amount" as used herein includes within its meaning a non-toxic but sufficient amount of a compound or composition for use in the invention to provide the desired therapeutic effect. The exact amount required will vary from subject to subject depending on factors such as the species being treated, the age and general condition of the subject, co-morbidities, the severity of the condition being treated, the particular agent being administered and the mode of administration and so forth. Thus, it is not possible to specify an exact "effective amount". However, for any given case, an appropriate "effective amount" may be determined by one of ordinary skill in the art using only routine methods.

In the context of this specification, an "antagonist" is any substance, agent or drug that inhibits physiological activity of a polypeptide that is normally capable of being stimulated by a naturally occurring or endogenous regulatory substance. It is to be understood that, in the context of this specification, the term "antagonist" includes partial antagonists in which the substance, agent or drug may be only partly effective in inhibiting physiological activity, such as partly inhibiting or reducing one of a number of biological activities of the subject polypeptide or protein. It will be understood that, in the context of this specification, the term "antagonist" includes any substance, agent or drug that acts directly on a subject polypeptide, such as IQGAPl, Homerl, ClQLl or IGFBP2, including those which have affinity for the subject polypeptide, and any substance, agent or drug which may act indirectly on the subject polypeptide, such as through one or more intermediate substance(s) or pathways, to inhibit activity of the subject polypeptide. Furthermore, it will be understood that the term "antagonist" also includes any substance, agent or drug which acts to reduce the expression of the subject polypeptide or protein, such as through inhibition of transcription, translation, or post- translational modification. In the context of this specification, the term "patient" includes humans and individuals of any species of social, economic or research importance including but not limited to members of the genus ovine, bovine, equine, porcine, feline, canine, primates, rodents. In the context of this specification the term "subject individual" is used to refer to the individual who, for example, is being assessed by a practitioner through a method of the invention. The subject individual will thus be understood to be one who is suspected of, or known to, have a tumour of the CNS. The subject individual may also be referred to herein as the "patient". Unless indicated to the contrary in the specific use, the "biological sample" referred to herein means a sample taken from or derived therefrom the subject patient for which diagnosis is sought. The biological sample will thus generally refer to the sample which is a tumour sample or is suspected of comprising a tumour.

In the context of this specification, the term "combined with" and similar terms such as "in conjunction with" when used in relation to a therapeutic regime means that each of the drugs and other therapeutic agent(s), such as agonists and antagonists, is used in the treatment of an individual and that each of the drugs and other therapeutic agents in the "combined" therapeutic regime may be administered to the individual simultaneously with one or more of the other agents in the therapeutic regime, or may be administered to the individual at a different time to one or more of the other agents in the therapeutic regime. That is, the term "combined with" and similar terms such as "in conjunction with" when used in relation to a therapeutic regime may mean that any one or more of the drugs or other agents may be physically combined prior to administration to the patient, and it will be understood that the term also includes administration of the one or more drugs and other therapeutic agents as separate agents not in prior physical combination.

In the context of this specification the terms "brain tumour", "tumour of the brain" and "glioma" will be understood to have the same meaning. It will be understood that reference herein to a "tumour" is intended to also include cancer, hence for example, reference to a tumour sample or a sample suspected of comprising a tumour will be understood to also mean a cancer sample or a sample suspected of being cancerous.

In the context of this specification the term "polypeptide" means a polymer made up of amino acids linked together by peptide bonds. The polypeptide may be of any length. Except where the context indicates otherwise it will be understood that the term polypeptide also includes peptides and proteins. In the context of this specification, the term "comprising" means "including principally, but not necessarily solely". Furthermore, variations of the word "comprising", such as "comprise" and "comprises", have correspondingly varied meanings. In the context of this specification, the term "high grade" refers to a brain tumours with highly aggressive biological behaviour and the presence of the histological features necrosis and microvascular proliferation (NMVP). "High grade" brain tumours are associated with a median survival time of 337 days after diagnosis.

In the context of this specification, the term "low grade" refers to a brain tumour with temperate biological behaviour and generally the absence of the histological features necrosis and microvascular proliferation. "Low grade" brain tumours are associated with 80% of patients displaying greater than 5 years survival after diagnosis.

The term "at least one" when used in the context of a group of selectable elements includes any and all members of the group individually selected and includes any combination of the members of the group. Similarly, the term "at least two" when used in the context of a group of selectable elements includes any selection of two or more members of the group in any combination.

To the extent that it is permitted, all references cited herein are incorporated by reference in their entirety.

Abbreviations

For convenience, the following abbreviations used in this specification are listed below.

"WHO" is used herein as an abbreviation for World Health Organisation. "GBM" is used herein as an abbreviation for Glioblastoma multiforme.

"CNS" is used herein as an abbreviation for central nervous system.

"AIII" is used herein as an abbreviation for Anaplastic Astrocytoma Grade III.

"All" is used herein as an abbreviation for Astrocytoma Grade II.

"Odg III" is used herein as an abbreviation for Anaplastic Oligodendroglioma Grade III.

"Odg II" is used herein as an abbreviation for Oligodendroglioma Grade II.

"OAIII" is used herein as an abbreviation for Anaplastic Oligoastrocytoma Grade III.

"LOH" is used herein as an abbreviation for loss of heterozygosity. "SDDA" is used herein as an abbreviation for Stepwise Diagonal Discriminant Analysis.

"qPCR" is used herein as an abbreviation for quantitative polymerase chain reaction(s). "MAb" is used herein as an abbreviation for monoclonal antibody.

"DAB" is used herein as an abbreviation for 3,3'-diaminobenzidine.

"IHC" is used herein as an abbreviation for immunohistochemical.

"Homer 1" is used herein as an abbreviation for the gene Horner homolog I (Drosophila) or a gene product thereof such as included under Accession No. NM_004272.

"IQGAPl" is used herein as an abbreviation for the gene IQ motif containing GTPase activating protein I or a gene product thereof such as included under Accession No. NM_003870.

"LGALS" is used herein as an abbreviation for the gene Galectin 1 or a gene product thereof such as included under Accession No. NM_002305.

"LRRC20" is used herein as an abbreviation for the gene Leucine rich repeat containing 20 or a gene product thereof such as included under Accession No. NM_018239.

"IGFBP2" is used herein as an abbreviation for the gene Insulin-like growth factor binding protein 2 or a gene product thereof such as included under Accession No. M35410.

"ClQLl" is used herein as an abbreviation for the gene Complement component I, q subcomponent-like or a gene product thereof such as included under Accession No. NM_006688. "SPPl" is used herein as an abbreviation for the gene secreted phosphoprotein 1

(psteopontiή) or a gene product thereof such as included under Accession No. NM_000582.

"RBPl" is used herein as an abbreviation for the gene retinol binding protein 1, cellular or a gene product thereof such as included under Accession No. NM_002899. "NMVP" is used herein as an abbreviation for the term necrosis and microvascular proliferation.

"LTS" is used herein as an abbreviation for the term "long term survival" and "long term survivors", depending on context.

"STS" is used herein as an abbreviation for the term "short term survival" and "short term survivors", depending on context. "KPNA5" is used herein as an abbreviation for the gene Karyopherin alpha 5 or a gene product thereof such as included under Accession No. NM_002269.

"NFYB" is used herein as an abbreviation for the gene nuclear transcription factor Y beta or a gene product thereof such as included under Accession No. NM_006166. "CARHSPl" is used herein as an abbreviation for the gene calcium regulated heat stable protein 1 or a gene product thereof such as included under Accession No. NM_014316.

"COPZ2" is used herein as an abbreviation for the gene coatomer protein complex subunit zeta 2 or a gene product thereof such as included under Accession No. NM_016429.

"ARS" is used herein as an abbreviation for the gene human autonomously replicating sequence or a gene product thereof such as included under Accession No. L08441 .

"HS7SLP" is used herein as an abbreviation for the gene 7SL pseudogene or a gene product thereof such as included under Accession No. X02067.

"CH13L1" is used herein as an abbreviation for the gene Chitinase-3-like 1 or a gene product thereof such as included under Accession No. NM OO 1276.

"HSxS138" is used herein as an abbreviation for the gene (xsl38)mRNA or a gene product thereof such as included under Accession No. . "SERP IN A3" is used herein as an abbreviation for the gene serpin peptidase inhibitor clade A or a gene product thereof such as included under Accession No. NMJ)01085.

Figure Legends Figure 1. Survival analysis of histology based classification of gliomas.

(A) Disease-specific survival of the patients entered in this studies classified according to histologic grade (WHO 2000) where i) All (astrocytoma grade II), «=12; ii) AIII (astrocytoma grade III), «=20; iii) GBM (glioblastoma multiforme), «=52; iv) OdgII (oligodendroglioma grade II), «=20; v) OdgIII (oligodendroglioma grade III), «=26 and vi) OAIII (oligoastrocytoma grade III), «=12. (B) Disease-specific survival of patients entered in this study and designated as i) High grade gliomas, «=73 or ii) Low grade gliomas, «=70 (p<0.001, log-rank test).

Figure 2. Scatter graph of normalised log₂ gene expression results demonstrating separation of the 20 low grade gliomas from 17 high grade gliomas. (A) Separation of high and low grade gliomas (+ve and -ve) with the gene subsets Homer] and IQGAPl. (B) Separation of high and low grade gliomas (+ve and -ve) gliomas with the gene subsets, IGFBP2 and ClQLl. The gene subsets were validated by qPCR. (C) log transformed class median values for the qPCR expression (normalized to the ribosomal 18S endogenous control) for each of the 25 high grade gliomas and 23 low grade gliomas after comparison to a normal brain reference (Columns); Standard error of the expression across all examined samples (Bars). * denotes significance p O.005, ** significance p O.001 derived from Student /-test analysis.

Figure 3. Expression and localisation of IQGAPl and IGFBP2 in normal brain tissue and gliomas classified by histology and by the modified scheme based on a high or low grade designation.

Panel (A) normal brain and AIII, OdgIII and OAIII gliomas, alternatively classified as low grade gliomas with little or no IQGAPl immunostaining. Panel (B) GBMl, GBM2, OdgIII and OAIII, alternatively classified as high grade gliomas. Strong IQGAPl staining is visible. Panel (C) normal brain and AIII, OdgIII and OAIII gliomas, alternatively classified as low grade gliomas with little or no IGFBP2 immunostaining. Panel (D) GBMl, GBM2, OdgIII and OAIII, alternatively classified as high grade gliomas demonstrating moderate to strong cytoplasmic IGFBP2 staining. Magnification: *200.

Figure 4. Prognostic value of IQGAPl and IGFBP2 protein expression scores in 143 glioma patients.

(A) disease-specific survival of patients with increasing age: i) age group 20-39 years, «=36; ii) age group 40-49 years («=23); iii) age group 50-59 years («=35); iv) age group 60+ years («=49). Log Rank PO.001. (B) disease-specific survival of patients with i) astrocytic («=80) or ii) oligodendroglial tumours (n=57). Log Rank PO.001. (C) disease- specific survival of patients with IQGAPl protein expression scores: i) score 0 («=7); ii) score 1 («=18); iii) score 2 («=36); iv) score 3 («=17); v) score 4 («=65). Log Rank .PO.001. (D) disease-specific survival of patients with IGFBP2 protein expression scores: i) score 0 («=50); ii) score 1 («=18); iii) score 2 («=21); iv) score 3 («=54). Log Rank PO.001.

Figure 5. Prognostic value of IQGAPl and IGFBP2 protein expression when applied as an adjunct to the histologically classified glioma patients. (A) disease-specific survival of patients diagnosed with an AIII or GBM with positive (score>3) or negative (score<2) IQGAPl protein expression (IQGAP l+ve/-ve). i) AIII/IQGAPl-ve (n=16) and ii) AIII/IQGAPl+ve (n=4). Log Rank P=0.032. iii) GBM/IQGAPl-ve («=11) and iv) GBM/IQGAPl+ve (n=4\) and. Log Rank P=0.550. (B) disease-specific survival of patients diagnosed with OAIII or OdgIII with positive (score>3) or negative (score<2) IQGAPl protein expression (IQGAP l+ve/-ve). i) OAIII/IQGAPl-ve (H=6) and iv) OAIII/IQGAP+ve («=6) Log Rank P=0.336. ii) OdgIII/IQGAP-ve («=17) and iii) OdgIII/IQGAP+ve («=9) Log Rank P=0.038. (C) disease-specific survival of patients diagnosed with an AIII or a GBM with positive (score>2) or negative (score≤l) IGFBP2 protein expression (IGFBP2+ve/-ve). i) AHI/IGFBP2-ve (n=17) and ii) AIII/IGFBP2+ve («=3) Log Rank i^>=0.022. iii) GBM/IGFBP2-ve (n=4) and GBM/IGFBP2+ve («=48) Log Rank P=0.518. (D) disease- specific survival of patients diagnosed with OAIII or OdgIII with positive (score>2) or negative (score≤l) IGFBP2 protein expression (IGFBP2+ve/-ve). i) OAIII/IGFBP2-ve (κ=5) and iv) OAIII/IGFBP2+ve («=7) Log Rank iM).O86. ii) OdgIII/IGFBP2-ve («=17) and iii) OdgIII/IGFBP2+ve (n=9) Log Rank P=COl 9.

Figure 6. Percentage of patients diagnosed with a GBM and are devoid of IGFBP2 (p=0.001) or IQGAPl (p=0.012) when measured immunohistochemically. LTS (long term survivors); STS (Short term survivors) Figure 7: (A) nucleic acid sequence encoding IQGAPl and (B) amino acid sequence of IQGAPl, which are SEQ ID NO: 1 and 2, respectively.

Figure 8: (A) nucleic acid sequence encoding IGFBP2 and (B) amino acid sequence of IGFBP2, which are SEQ ID NO: 3 and 4, respectively.

Figure 9: (A) nucleic acid sequence encoding Homerl and (B) amino acid sequence of Homerl, which are SEQ ID NO: 5 and 6, respectively.

Figure 10: (A) nucleic acid sequence encoding ClQLl and (B) amino acid sequence of ClQLl, which are SEQ ID NO: 7 and 8, respectively.

Detailed Description and Best Mode

The invention will now be described in more detail, including, by way of illustration only, with respect to the examples which follow.

In the current study, the inventors used two-colour chip expression analysis to identify genes that could discriminate between different glioma types as well as grades. As described herein, the inventors used a technology called GeneRaVE® which assisted in the identification of small sets of genes which provide better predictive accuracy than the usually much larger sets found by methods and marker sets previously used. By pooling the tumours into a high grade class and a low grade class based on the presence or absence of necrosis and microvascular proliferation, the inventors surprisingly identified two new gene/protein sets, IQGAPl /Homer 1 and IGFBP 2/C IQLl that provide, inter alia, improved diagnosis and prognosis of tumours of the CNS. The members of the gene/protein sets may be used individually or as sets in the methods of the invention. The present invention provides, inter alia, nucleic acid and protein sequences that are differentially expressed in tumours of the central nervous system (CNS) when compared to normal samples. In preferred embodiments, the nucleic acid and protein sequences are selected from the group consisting of IQGAPl, Homerl, ClQLl and IGFBP2.

The present invention provides, inter alia, new markers and methods useful in the diagnosis and treatment of tumours of the central nervous system (CNS), in particular gliomas.

In one aspect of the invention there is provided a method of diagnosing a tumour of the central nervous system (CNS) in an individual comprising: a) determining the expression of at least one gene selected from the group consisting of IQGAPl, Homerl, IGFBP2 and ClQLl in a biological sample of said individual; and b) comparing said expression of said at least one gene with that of at least one reference sample; wherein a difference in said expression is an indication that the individual may have a tumour of the CNS.

It will be understood that "diagnosing" a tumour includes identification of the presence of tumour cells in a sample or individual, identifying a specific category or type of tumour cell, such as any tumour cell of the CNS, discriminating between different types of tumours, such as discriminating between different types of gliomas, for example discriminating between ependymomas, astrocytomas, mixed gliomas and oligodendrogliomas, investigating the severity of tumour presence, for example to assist in assessing prognosis, discriminating between high grade and low grade tumours, and assessing tumour progression or regression.

In one embodiment the method comprises determining the expression of IQGAPl and IGFBP2. The inventors have determined that such "dual" use of markers can accurately predict survival in glioma patients. As described herein the use of these two markers may permit identification of biologically aggressive grade III astrocytomas, oligodendrogliomas and oligoastrocytomas from their less aggressive counterparts. The use of the markers and methods described herein may be used to more accurately diagnose a biopsy sample, for example where the pathology is difficult to ascertain by current methods. s With particular reference to the markers IQGAPl and IGFBP2 the present inventors have identified that these markers may be used to predict poor overall survival. Similarly, the absence of these markers in particular the absence of both IQGAPl and IGFBP2 may be used to predict long term survival. o Methods of screening for expression

The practice of the present invention will employ, unless otherwise indicated, conventional techniques of molecular biology (including recombinant techniques), microbiology, cell biology, and biochemistry, which are within the skill of the art. Such techniques are explained fully in the literature, such as, "Molecular Cloning: As Laboratory Manual", 2^nd edition (Sambrook et al., 1989); "Oligonucleotide Synthesis" (M. J. Gait, ed., 1984); "Animal Cell Culture" (R. I. Freshney, ed., 1987); "Methods in Enzymology" (Academic Press, Inc.); "Handbook of Experimental Immunology", 4.sup.th edition (D. M. Weir & C. C. Blackwell, eds., Blackwell Science Inc., 1987); "Gene Transfer Vectors for Mammalian Cells" (J. M. Miller & M. P. Calos, eds., 1987);0 "Current Protocols in Molecular Biology" (F. M. Ausubel et al., eds., 1987); and "PCR:

The Polymerase Chain Reaction", (Mullis et al., eds., 1994).

In general, analysis of expression may be undertaken on the basis of nucleic acid analysis or polypeptide analysis. A combination of methods can of course also be used.

Methods of nucleic acid-based gene expression analysis can broadly be divided into5 two groups: methods based on hybridization analysis of polynucleotides, and methods based on sequencing of polynucleotides. The most commonly used methods known in the art for the quantification of mRNA expression in a sample include northern blotting and in situ hybridization (Parker & Barnes, Methods in Molecular Biology 106:247-283 (1999)); RNAse protection assays (Hod, Biotechniques 13:852-854 (1992)); and reverse0 transcription polymerase chain reaction (RT-PCR) (Weis et al., Trends in Genetics 8:263- 264 (1992)). Representative methods for sequencing-based gene expression analysis include Serial Analysis of Gene Expression (SAGE) which allows the simultaneous and quantitative analysis of a large number of gene transcripts, without the need of providing an individual hybridization probe for each transcript (eg., see Velculescu et al., Science5 270:484-487 (1995); and Velculescu et al., Cell 88:243-51 (1997), and gene expression analysis by massively parallel signature sequencing (MPSS) such as described by Brenner et al., Nature Biotechnology 18:630-634 (2000). Alternatively, antibodies may be employed that can recognize specific duplexes, including DNA duplexes, RNA duplexes, and DNA-RNA hybrid duplexes or DNA-protein duplexes. In the methods of the invention a polypeptide corresponding to a marker may be detected using any of a variety of techniques and binding agents. Any such technique and agent may be used according to the present invention. In certain preferred embodiments, the binding agent is an antibody that binds specifically to the polypeptide. The invention also encompasses the use of protein arrays, including antibody arrays, for detection of a polypeptide. The use of antibody arrays is described, for example, in Haab et al., "Protein microarrays for highly parallel detection and quantitation of specific proteins and antibodies in complex solutions", Genome Biol. 2(2):2001, 2001. Other types of protein arrays are known in the art. In general, antibodies that bind specifically to a polypeptide may be generated by methods well known in the art and described, for example, in Harlow, E, Lane, E, and Harlow, E, (eds.) Using Antibodies: A Laboratory Manual, Cold Spring Harbor Laboratory Press, Cold Spring Harbor, 1998. Details and references for the production of antibodies may also be found in U.S. Pat. No. 6,008,337. Antibodies may include, but are not limited to, polyclonal, monoclonal, chimeric (e.g., "humanized"), single chain antibodies, Fab fragments, antibodies generated using phage display technology, etc. The invention encompasses the use of "fully human" antibodies produced using the XENOMOUSE™ technology (AbGenix Corp., Fremont, Calif.) according to the techniques described in U.S. Pat. No. 6,075,181. The antibody may have a detectable label. The antibodies or functional antibody parts may be purchased, isolated, or produced using known methods. In addition, in certain embodiments of the invention the polypeptides are detected using other specific binding agents known in the art for the detection of polypeptides, such as aptamers (Aptamers, Molecular Diagnosis, Vol. 4, No. 4, 1999), reagents derived from combinatorial libraries for specific detection of proteins in complex mixtures, random peptide affinity reagents, etc. In general, any appropriate binding agent for detecting a polypeptide may be used in conjunction with the present invention, although antibodies may represent a particularly appropriate modality.

A variety of formats can be employed to determine whether a sample contains a protein that binds to a given antibody. Examples of such formats include, but are not limited to, enzyme immunoassay (EIA), radioimmunoassay (RIA), Western blot analysis and enzyme linked immunoabsorbant assay (ELISA). A skilled artisan can readily adapt known protein/antibody detection methods for use in determining whether ovarian cells express a marker of the present invention.

In one format, antibodies, or antibody fragments, can be used in methods such as Western blots or immunofluorescence techniques to detect the expressed proteins. In such uses, it is generally preferable to immobilize either the antibody or proteins on a solid support. Suitable solid phase supports or carriers include any support capable of binding an antigen or an antibody. Well-known supports or carriers include glass, polystyrene, polypropylene, polyethylene, dextran, nylon, amylases, natural and modified celluloses, polyacrylamides, gabbros, and magnetite. Expression analysis according to the present invention may be undertaken by proteomics techniques that are well known in the art. The proteome is the totality of the proteins present in a sample (e.g. tissue, organism, or cell culture) at a certain point of time. Proteomics includes, among other things, study of the global changes of protein expression in a sample (also referred to as "expression proteomics"). Proteomics typically includes the following steps: (1) separation of individual proteins in a sample by 2-D gel electrophoresis (2-D PAGE); (2) identification of the individual proteins recovered from the gel, e.g. by mass spectrometry and/or N-terminal sequencing, and (3) analysis of the data using bioinformatics. Proteomics methods are valuable supplements to other methods of gene expression profiling, and can be used, alone or in combination with other methods of the present invention, to detect the products of the gene markers of the present invention.

Tumours of the CNS

Gliomas are a type of primary tumour of the CNS that arise from neuroepithelial tissue in the brain. Gliomas are often classified on the basis of the specific cell type they most closely resemble, the main types of gliomas being ependymomas, astrocytomas, mixed gliomas and oligodendrogliomas. Briefly, ependymomas arise from ependymal cells which form the ventricles and the lining of the central canal of the spinal cord.

Specific types of ependymomas include myxopapillary ependymoma which is commonly found in the spine, papillary ependymoma, subependymoma, and anaplastic ependymoma which is generally a malignant form of ependymoma. As with many types of gliomas, ependymomas may metastasise to other locations within the CNS.

Astrocytomas arise from astrocyte cells and are one of the most common of the gliomas. Various specific forms of astrocytomas have been described, including juvenile pilocytic astrocytoma (such as cerebellar astrocytoma and optic nerve glioma), subependymal giant cell astrocytomas, infiltrating low grade astrocytomas, gemistocytic astrocytoma, anaplastic astrocytoma, and malignant astrocytoma.

Mixed gliomas are tumours that contain a high proportion of cells of more than one type, such as tumours containing both astrocytes and oligodendrocytes, which may be referred to as oligoastrocytomas, or astrocytes and ependymal cells. Glioblastoma multiforme (also referred to as glioblastoma multiforme and as grade IV astrocytoma; GBM) is the most common example of a high grade glioma and represents approximately 30% of all primary brain tumours.

Oligodendrogliomas are primary tumours that arise in oligodendrocytes. Oligodendrogliomas frequently also contain astrocytes and are thus termed oligoastrocytomas or mixed gliomas.

Tumour and reference samples

Practice of methods of the invention, such as methods of diagnosing a tumour in an individual, may involve comparison of various characteristics of the suspected tumour sample from the subject individual with characteristics of one or more reference samples. As used herein reference to a tumour sample includes a sample known to comprise or consist of one or more tumour cells as well as a sample suspected of comprising or consisting of tumour cells. Any of a variety of different tumour samples may be used in the practice of the invention and an appropriate tumour sample in a given situation will be apparent to the skilled addressee. By way of non-limiting example, the tumour sample may be a tissue sample obtained from an appropriate region of the CNS or brain, obtained as a freshly frozen sample and/or fixed and paraffin embedded. The tumour sample may be blood and/or serum obtained from the patient, for example at the time of surgery or at any other time.

It will be understood that a variety of reference samples may be used. An appropriate reference sample will be apparent to the skilled addressee in any given circumstance. By way of non-limiting example, a reference sample may be a tissue sample from the same subject individual or from a second or other individual. The reference sample may be a sample of normal brain tissue. In this context it will be understood that the "normal" brain tissue is tissue that the practitioner considers is not affected by a tumour. The normal brain tissue may be obtained from similar region of the brain or from a distinct region of the brain. When obtained from a similar region of the brain such tissue may be referred to as a 'matched' sample. The reference sample may be a tumour sample, such as a sample comprising a high grade or low grade glioma. Typically, when used as a reference sample, the diagnosis of the reference (tumour) sample will be known.

The reference sample may comprise material capable of acting as a positive control for assessing the validity of the expression assay. For example, as described herein the absence of detectable expression of IQGAPl and IGFBP2 in a sample taken from an individual having glioblastoma multiforme may be diagnostic of long term survival of such an individual. In such cases it may be advantageous to include a reference sample in the form of a positive control for the assay which permits the user to ensure that the assay is performing as expected. Similarly, a negative control as a reference sample may be used.

The reference sample may be a natural sample, derived from natural sample or a synthetic sample or composition. The reference sample may have a pre-determined amount of a gene or gene product, such as at least one selected from the group consisting of IQGAPl, IGFBP2, Homer 1 and ClQLl. Such a reference sample would be expected to provide a known strength of signal in a given assay method and hence may be advantageous for a user desirous of determining the level of expression of a gene.

Any number of reference sample(s) may be used. For example, a single reference may be used or a plurality of reference samples, which may be the same or different, may be used. In this regard, a "plurality" is any number greater than one. A tissue sample, such as a tumour sample or a reference sample, may be obtained by biopsy, or surgical debulking or resection. The biopsy may be an open or a closed biopsy, such as through the use of stereotaxic instrumentation. The reference sample may be obtained from a commercial source, such as a brain nucleic acid mix, for example whole brain total RNA mix or a whole brain poly A⁺ RNA. Sources of whole brain total RNA mix are known and include, for example Ambion, Inc. USA, and Clontech Laboratories, Inc. USA. Sources of whole brain polyA⁺ RNA are known and include Clontech Laboratories, Inc. USA. The reference sample may be an RNA mix or polyA⁺ RNA preparation of a specific region or regions of the brain, such as the temporal lobe. Where the reference sample comprises or is substantially of completely obtained from a specific region of the brain, the reference sample may be matched to the tumour sample such that the reference sample and the tumour sample comprise or are the same region of the brain in the respective subject individual and the source of the reference sample.

Preparation of Samples In methods of the invention a sample may be assayed to assist determination of the expression of a given gene or genes. The present invention describes genes and gene sets which are shown herein to have altered expression levels in tumours of the CNS compared to the absence of such tumours, for example elevated expression in tumour tissue sample compared to normal tissue sample. The expression level of a gene may be increased or may be decreased compared to the expression level of the same gene in a reference sample. In one embodiment the expression level of the gene in a tumour sample is increased.

It will be understood that preparation of samples for analysis in the methods of the invention may be performed by any suitable means. By way of non-limiting example, determination of the expression of a gene in a sample may be undertaken by analysis of the sample, or an extract thereof, at the nucleic acid level or at the protein or polypeptide level, or a combination thereof.

Where analysis at the nucleic acid level is desired, the sample may be prepared by any suitable means for analysis of one or more nucleic acid(s) in a sample, including extraction of total RNA from the sample, extraction of poly A⁺ RNA from the sample, and/or preparation of cDNA representative of expression of messenger RNA within the sample. The cDNA preparation may be a total cDNA preparation, such that it represents a library of all or substantially all mRNA species in the sample or the cDNA preparation may be prepared in such a way that it is enriched for the inclusion of particular species, such as one or more gene markers of interest for a given analysis.

Any suitable method for determination of expression levels in a sample may be used. In general, the method may involve assay of the sample material, or a product derived therefrom such as amplified nucleic acid, with one or more probes having selectivity for the gene or genes of interest. By way of non-limiting example, an RNA extract may be prepared from the sample, such as a tumour sample, and the RNA may be assayed by quantitative PCR (Q-PCR) analysis using TaqMan Probes (Applied Biosystems) (Pflaff et al. 2002)

In an embodiment one or more nucleic acid probe(s) selective for a target gene, such as a gene described herein as a marker of tumours of the CNS or a reference gene, or a complement thereof, may be attached to a biochip. In a preferred embodiment the degree of complementarity between a probe and its target sequence is sufficient to permit selective hybridisation of the probe to its target sequence in preference to hybridisation of the probe and other sequences which may be in the sample under investigation. In a preferred embodiment a probe will be substantially complementary to its target sequence. By "substantially complementary" it will be understood that the degree of complementarity is sufficient for the probe to hybridize with a target sequence under normal reaction conditions.

The skilled addressee will be able to determine an appropriate stringency of s hybridization for use in an assay according to the method of the invention, such as by an empirical calculation dependent upon probe length, washing temperature, and salt concentration. In general, longer probes require higher temperatures for proper annealing, while shorter probes need lower temperatures. Hybridization generally depends on the ability of denatured DNA to reanneal when complementary strands are

I₀ present in an environment below their melting temperature. The higher the degree of desired homology between the probe and hybridizable 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 additional details and explanation of stringency of hybridization reactions, see for is example Ausubel et al., Current Protocols in Molecular Biology, Wiley Interscience Publishers, (1995) and Sambrook et al., Molecular Cloning: A Laboratory Manual, New York: Cold Spring Harbor Press, 1989.

Examples of stringent conditions or high stringency conditions which may be used in the methods of the invention include (1) low ionic strength and high temperature for

20 washing, for example 0.015 M sodium chloride/0.0015 M sodium citrate/0.1% sodium dodecyl sulfate (SDS) at 5O⁰C; (2) a denaturing agent present during hybridization, such as formamide, for example, 50% (v/v) formamide with 0.1% bovine serum albumin/0.1% Ficoll/0.1% polyvinylpyrrolidone/50 mM sodium phosphate buffer at pH 6.5 with 750 mM sodium chloride, 75 mM sodium citrate at 42⁰C; or (3) 50% formamide, 5xSSC (0.75

2₅ M NaCl, 0.075 M sodium citrate), 50 mM sodium phosphate (pH 6.8), 0.1% sodium pyrophosphate, 5x Denhardt's solution, sonicated salmon sperm DNA (50 μg/ml), 0.1% SDS, and 10% dextran sulfate at 42°C, with washes at 42⁰C in 0.2xSSC (sodium chloride/sodium citrate) and 50% formamide at 55⁰C, followed by a high-stringency wash consisting of 0. IxSSC containing EDTA at 55⁰C.

30 Examples of moderately stringent conditions which may be used in the methods of the invention include the use of washing solution and hybridization conditions (e.g., temperature, ionic strength and % SDS) less stringent that those described above. An example of moderately stringent conditions is overnight incubation at 37⁰C in a solution comprising: 20% formamide, 5xSSC (150 mM NaCl, 15 mM trisodium citrate), 50 mM

3₅ sodium phosphate (pH 7.6), 5x Denhardt's solution, 10% dextran sulfate, and 20 mg/ml denatured sheared salmon sperm DNA, followed by washing the filters in IxSSC at about 37-5O⁰C. The skilled artisan will recognize how to adjust the temperature, ionic strength, and other conditions influential on stringency of hybridization as necessary to accommodate factors such as probe length and the like. In a preferred embodiment high stringency conditions are used for hybridisation to permit highly selective hybridisation of a probe and its target sequence. The complementarity of a probe to its target sequence does not need to be perfect; one or more mismatches between a probe and its target may be present. It will be understood that more than one probe selective for a target may be used. For example, two, three, four, five or more probes for any given target sequence may be used. The multiple probes for a given target may have partially overlapping sequences or may each be directed to discrete regions of a target sequence. An advantage offered by such a multiple probes approach is that an alteration in a region of the target gene or gene product that modulates binding or hybridisation of a probe, such decreasing the hybridisation of a given probe, will be less likely to result in the diagnostic method returning a false negative result.

The probe(s) may be attached to or immobilised on a supporting substrate, such as a solid support, usually by covalent or non-covalent binding. A plurality of probes may be attached to or immobilised on a supporting substrate, preferably in discrete locations. It

, will be understood that the attachment or immobilisation of the probe on the supporting substrate may be direct or indirect, such as through the inclusion of a cross-linker, linker, or reactive group on one or both of the probe and supporting substrate. The supporting substrate may comprise any suitable material or materials. It will be understood that the probe(s) may be synthesised according to methods known in the art and then attached to or immobilised on the biochip or the probe(s) may be synthesised on the surface of the biochip, such as through photolithographic techniques. Methods for preparation of biochips are known in the art and include methods described in WO/95/35505 and US Patent No. 5,700,637.

In a preferred embodiment the probe is attached to or immobilised on a biochip. The biochip may be of any suitable shape or configuration. For example the biochip may be tubular with probe(s) located on the inside surface to permit flow-through analysis of sample. In embodiments, the biochip is planar or substantially planar. It will be appreciated that a biochip, such as a planar or substantially planar biochip, may comprise a configuration that may assist in efficient exposure of relevant components to each other, such as probe sequence(s), target sequence(s), hybridisation material or solutions and wash material or solution. For example, a biochip may include pockets, pits, channels, indentations, depressions, protrusions, or the like. The biochip may be of any suitable material that can be modified to contain discrete sites for locating individual probe, such as a nucleic acid sequence or antibody. By way of non-limiting example, suitable materials include glass and modified or functionalized glass, plastics, such as acrylics, polystyrene and copolymers of styrene and other materials, polypropylene, polyethylene, polybutylene, polyurethanes, and TeflonJ, polysaccharides, nylon or nitrocellulose, resins, silica or silica-based materials including silicon and modified silicon, carbon, metals, inorganic glasses, plastics, etc. In general, the substrates allow optical detection and do not appreciably fluoresce. For a desired marker gene the skilled addressee will be able to select an appropriate probe(s) for the investigation of a sample. By way of non-limiting example, where the material to be investigated is a nucleic acid sample, such as an RNA or cDNA preparation derived from a tumour sample or normal sample, a nucleic acid probe may be employed. The nucleic acid probe for a desired gene may be a genomic probe, a cDNA probe, or a synthetic probe such as an oligonucleotide probe. The nucleic acid probe may be of any suitable length and sequence to provide an appropriate level of selectivity of hybridisation with the sample material. For example, the length of the nucleic acid probe(s) may be in the range of about 5 to about 100 bases, with preferred lengths of probes being from about 10 to about 80 bases, or from about 20 to about 70 bases, or from about 25 to about 50 bases or from about 30 to about 40 bases. Longer probes may be used, such as fragments or derivatives of the gene(s) of interest which may represent a substantial fraction of the length of the gene of interest or of a cDNA derived from the gene of interest. For example, a probe may be from about 100 to about 1000 bases long, such as about 200, about 300, about 400, about 500, about 600, about 700, about 800, or about 900 bases long. Methods for the preparation of probes are known in the art and include synthesis of probes through the use of selective primer oligonucleotides in the polymerase chain reaction (PCR) and automated synthesis, such as commercial phosphoramidite based synthesis.

Methods for the preparation of total RNA, polyA⁺ RNA, cDNA and nucleic acid sequences, such as probes and primers, are known to the skilled addressee, for example as described in Current Protocols in Molecular Biology (Eds Roger Brent, Robert E. Kingston, J. G. Seidman, Kevin Struhl, Frederick M. Ausubel, Virginia Benson Chanda, David D. Moore, J.G. Seidman, F.M. Ausubel; John Wiley & Sons Inc.) and Molecular Cloning: A Laboratory Manual (Third Edition, Sambrook, and Russell; Cold Spring Harbor Laboratory Press, New York), the contents of each of which are incorporated herein to the extent permissible.

In one embodiment, nucleic acids, proteins and/or antibodies of the invention are labelled. Such a "labelled" compound has at least one element, isotope or chemical entity attached to enable detection of the compound. Any suitable label may be used including, for example, a) an isotopic label, which may be radioactive or heavy isotopes; b) an immune label, which may be antibodies or antigens; and c) a coloured or fluorescent dye. A label may be incorporated into the compound, such as a nucleic acid or polypeptide or antibody at any appropriate position. For example, the label should be capable of producing, either directly or indirectly, a detectable signal. The detectable moiety may be a radioisotope, such as ³H, ¹⁴C, ³²P, ³⁵S, or ¹²⁵I, a fluorescent or chemiluminescent compound, such as fluorescein isothiocyanate, rhodamine, or luciferin, or an enzyme, such as alkaline phosphatase, beta-galactosidase or horseradish peroxidase. Any method known in the art for conjugating the antibody to the label may be employed, including those methods described by Hunter et al., Nature, 144:945 (1962); David et al., Biochemistry, 13:1014 (1974); Pain et al., J. Immunol. Meth., 40:219 (1981); and Nygren, J. Histochem. and Cytochem., 30:407 (1982).

As a further non-limiting example, a sample comprising nucleic acid sequences, such as RNA or cDNA, may be analysed by quantitative polymerase chain reaction (qPCR) using one or more oligonucleotide(s) selective for the gene or genes of interest as primer oligonucleotides. As the skilled addressee will be aware, the primer oligonucleotide(s) sequence need not be identical to the corresponding region of the sequence of the gene of interest to be selective. The primer oligonucleotide(s) used for qPCR may be of any appropriate length, for example the primer may be any length in the range of about 10 to about 100 bases long, typically such oligonucleotides may be in the range of about 15 to about 30 bases long, more typically about 18 to about 25 bases long. The primer oligonucleotide(s) may be labelled with a detectable label or may be unlabelled. Where oligonucleotide primer(s) corresponding to more than one specific gene or sequence are used in a single qPCR, for example by the inclusion in a single reaction of oligonucleotide primers specific for the amplification of gene products derived from expression of two or more distinct genes, such as IQGAP I, Homer 1, IGFBP2, and/or ClQlL, at least one oligonucleotide primer specific for each gene may be labelled with a detectable label which discriminates the products of amplification of that gene from the products of other genes amplified in the same reaction. Methods of labelling nucleic acids with detectable labels are known in the art and are described, for example, in Plaff et al. (2002). Where one or more labelled oligonucleotide primer(s) are used, the product of the qPCR will be labelled and methods of detection of the product(s) appropriate to the label or labels used may be used to assess the level of a specific PCR product in the reaction and hence relate that level to the level of specific RNA in the sample, such as a tumour sample as an indication of the expression of the gene marker of interest. Where the oligonucleotide primer(s) used in the qPCR are not labelled, an appropriate labelling entity, such as a radiolabel, for example one or more radiolabeled nucleotide precursors, may be included in the qPCR reaction mixture such that reaction products are labelled.

Nucleic acid probes for use in the methods of the invention, such as those described above, may be prepared with reference to the nucleic acid sequences described herein, namely SEQ ID NO:1 (IQGAPl), SEQ ID NO:3 (IGFBP2), SEQ ID NO:5 (Homerl) and SEQ ID NO:7 (ClQLl).

Antibodies

The present invention also provides antibodies which are capable of binding to an expression product of one or more genes associated with a tumour of the CNS. The expression product may be any expression product of an appropriate gene which is capable of being bound by an antibody. Typically the expression product is a polypeptide, or fragment thereof, encoded by a gene of interest. Typically, the gene of interest is a gene which is characterised by an altered level of expression in tumour tissue compared to normal tissue. In preferred embodiments the polypeptide, or fragment thereof, is an expression product encoded by IQGAPl, Homerl, IGFBP2 or ClQLl. Thus, in one embodiment, the antibody is an antibody capable of binding to a polypeptide, or fragment thereof, encoded by IQGAPl, Homerl, IGFBP2 or ClQLl.

In preferred embodiments the antibody is capable of binding specifically to an expression product of a gene of interest. Typically, the expression product is a polypeptide or fragment thereof. By "binding specifically" it will be understood that the antibody is capable of binding to the polypeptide or fragment thereof encoded by a given gene with a higher affinity than it binds to an unrelated polypeptide. For example, the antibody may bind to the polypeptide or fragment thereof with a binding constant in the range of at least 10^"4M to 10^"10M. Preferably the binding constant is at least about 10^"5M, or at least about 10^"6M, more preferably the binding constant of the antibody to the polypeptide encoded by a gene of interest, or fragment thereof, is at least about 10^"7M, at least about 10^" M, or at least about 10^" M or more.

In one embodiment the antibody is an antibody capable of specifically binding to IQGAPl or a fragment thereof. In one embodiment the antibody is an antibody capable of specifically binding to IGFBP2 or a fragment thereof. In one embodiment the antibody is a monoclonal antibody.

Antibodies of the present invention may exist in a variety of forms, including for example as a whole antibody, or as an antibody fragment, or other immunologically active fragment thereof, such as complementarity determining regions. Similarly, the antibody may exist as an antibody fragment having functional antigen-binding domains, that is, heavy and light chain variable domains. Also, the antibody fragment may exist in a form selected from the group consisting of, but not limited to: Fv, F_atø, F(ab)2, scFv

(single chain Fv), dAb (single domain antibody), chimeric antibodies, bi-specific antibodies, diabodies and triabodies. An antibody 'fragment' may be produced by modification of a whole antibody or by synthesis of the desired antibody fragment. Methods of generating antibodies, including antibody fragments, are known in the art and include, for example, synthesis by recombinant DNA technology. The skilled addressee will be aware of methods of synthesising antibodies, such as described in, for example, United States Patent No. 5,296,348, issued March 22, 1994 To Rakowicz-Szulczynska, et al. and entitled "Methods for screening monoclonal antibodies for therapeutic use."

Preferably antibodies are prepared from discrete regions or fragments of the polypeptide of interest, such as a polypeptide or fragment thereof encoded by IQGAPl, Homer, IGFBP2 or ClQLl. An antigenic portion of a polypeptide of interest may be of any appropriate length, such as from about 5 to about 15 amino acids. Preferably, an antigenic portion contains at least about 5, 6, 7, 8, 9, 10, 11, 12, 13 or 14 amino acids.

In the context of this specification reference to an antibody specific to a polypeptide of interest, such as a polypeptide encoded by IQGAPl, Homer, IGFBP2 or ClQLl, includes an antibody that is specific to a fragment of the polypeptide of interest. Methods for the generation of suitable antibodies will be readily appreciated by those skilled in the art. For example, a monoclonal antibody, typically containing Fab portions, may be prepared using the hybridoma technology described in Antibodies-A Laboratory Manual, Harlow and Lane, eds., Cold Spring Harbor Laboratory, N.Y. (1988). It will also be understood that antibodies of the invention include humanised antibodies, chimeric antibodies and fully human antibodies. An antibody of the invention may be a bi-specific antibody, having binding specificity to more than one antigen or epitope. For example, the antibody may have specificity for one or more of IQGAPl, or a fragment thereof, and IGFBP2, or a fragment thereof, and additionally have binding specificity for another antigen. In one embodiment the binding specificity for another antigen may be an antigen associated with presence of a tumour of the CNS. Methods for the preparation of humanised antibodies, chimeric antibodies, fully human antibodies, and bispecific antibodies are known in the art and include, for example as described in United States Patent No. 6,995,243 issued February 7, 2006 to Garabedian, et al. and entitled "Antibodies that recognize and bind phosphorylated human glucocorticoid receptor and methods of using same".

An antibody of the invention may be therapeutic or diagnostic or both.

The antibodies of the invention may be used in diagnostic applications, such as for diagnosing a tumour of the central nervous system (CNS) in an individual. It will be understood that "diagnosing" a tumour includes identification of the presence of tumour cells in a sample or individual, identifying a specific category or type of tumour cell, such as any tumour cell of the CNS, discriminating between different types of tumours, such as discriminating between different types of gliomas, for example discriminating between ependymomas, astrocytomas, mixed gliomas and oligodendrogliomas, investigating the severity of tumour presence, for example to assist in assessing prognosis, discriminating between high grade and low grade tumours, and assessing tumour progression or regression.

As described herein the methods of the invention may be undertaken on the basis of any appropriate expression product of a gene of interest. Various methods of the invention are described above, for example methods of diagnosing tumours on the basis of expression of gene products such as RNA corresponding to the gene(s) of interest.

Gene expression may also be determined by assaying a biological sample for the presence of a polypeptide, or fragment or variant thereof, encoded by a gene of interest. Such a polypeptide may be referred to as a "target polypeptide". In one embodiment, where diagnosis is on the basis of identification of a polypeptide expressed by a gene of interest, the diagnosis includes determining gene expression by contacting a suitable antibody with a sample in an in vitro assay or in vivo assay.

For example, in vitro detection of a polypeptide, or variants or fragments thereof, of a gene of interest in the diagnosis of a tumour, may be achieved using a variety of techniques including ELISA (enzyme linked immunosorbent assay), Western blotting, immunoprecipitation, immunofluorescence and "sandwich" assays. Such techniques are commonly used by those of skill in the art. Similarly, suitable techniques of the in vivo detection of the polypeptide, or fragments or analogues thereof, including immunohistochemistry using a labelled antibody will be readily understood by persons skilled in the art.

In one embodiment gene expression may be determined by contacting protein of a sample, such as a tissue sample or an isolated protein preparation of a tissue sample, with an antibody specific for the polypeptide of interest, such as an anti-IQGAPl, anti- Homer 1, anti-IGFBP2 or anti-ClQLl specific antibody and measuring the binding of the antibody. In the context of this specification an isolated protein preparation or isolating protein from a sample includes any method in which a sample is enriched for the presence of protein compared to the sample initially obtained, such as obtained through biopsy.

Expression of one or more genes of interest may be investigated simultaneously. For example, an ELISA may include multiple targets analysed at the same time by contacting a sample with a panel of antibodies to the polypeptides of interest, such as through the use of a multi-well assay or the use of a chip, as described above, wherein antibodies specific for a given set of gene products are affixed to discrete locations of the chip. In one embodiment, the chip or ELISA includes an anti-IQGAPl antibody, an anti- Homer 1 antibody, an anti-IGFBP2 antibody and an anti-ClQLl antibody in discrete locations on or in the multi-well plate or chip. The assay may include more than one antibody for a given polypeptide of interest. The assay may include additional antibodies, such as antibodies for additional tumour or cancer markers. The assay may include antibodies directed to a gene product of LGALSl, LRRC20, KIAA0599, KPNA5, NFYB, CARHSPl, COPZ2, ARS, HS7SLP, CH13L1, HSxS138, SERPINA3, SPPl, and RBPl. The assay may include antibodies for reference polypeptides. In this context a reference polypeptide is a polypeptide the expression of which is not associated with cancerous tissue. For example, the reference polypeptide may be a polypeptide the expression of which is indicative of tissue type, to assist the practitioner in determining that an appropriate tissue sample is being assayed. By way of further example the reference polypeptide may be a polypeptide the expression of which is constitutive, thereby providing for example an indication of the amount of sample analysed. In a preferred embodiment the assay for expression of a gene product of interest comprises an ELISA (Coligan, et al, Current Protocols in Immunology, 1(2), Chapter 6, 1991). Methods for performing ELISA are known to the skilled addressee and briefly stated may include the following steps. The method initially comprises preparing or obtaining an antibody specific to a gene product of interest, such as a polypeptide or fragment thereof of IQGAPl, Homer, IGFBP or ClQLl. Preferably, the antibody is a monoclonal antibody. A reporter antibody to the monoclonal antibody is prepared or obtained. To the reporter antibody is attached a suitably detectable reagent such as radioactivity, fluorescence, or enzymic reporter, for example a horseradish peroxidase enzyme. The sample to be analysed is incubated on a solid support, such as a polystyrene dish, that binds proteins in the sample. Free protein binding sites on the dish are covered by incubating with a non-specific protein, such as BSA. The monoclonal antibody specific for the target polypeptide is incubated in the dish during which time the monoclonal antibodies attach to the specific polypeptide attached to the polystyrene dish. Unbound monoclonal antibody is removed by washing with buffer. The reporter antibody, with attached detectable reagent, is added to the dish resulting in binding of the reporter antibody to the specific monoclonal antibody bound to the target gene polypeptide. Unattached reporter antibody is removed by washing with buffer. If necessary, suitable substrate(s) is added to promote detection of bound antibody and quantitation. For example, when utilising horse radish peroxidase reporter, peroxidase substrates are added to the dish and the amount of colour developed in a given time period is a measurement of the amount of the target polypeptide present in a given sample. The result may be compared to a reference sample(s) analysed concurrently or to a standard curve, such as may have been generated at another time.

A competition assay may be employed where antibodies specific to a target polypeptide are attached to a solid support. The sample comprising the target polypeptide is then labelled, for example with radioactive label and is then passed over the solid support and the amount of label detected, for example, by liquid scintillation chromatography, can be correlated to a quantity of the target polypeptide in the sample.

A "sandwich" assay is similar to an ELISA assay. In a "sandwich" assay, sample containing target polypeptides are passed over a solid support and bind to antibody attached to the solid support. A second antibody is then bound to the target polypeptide. A third antibody which is labelled and is specific to the second antibody, is then passed over the solid support and binds to the second antibody and an amount can then be quantified.

Determination of the expression level of one or more given genes may also be made by staining paraffin sections prepared from tissue samples with one or more antibodies specific to the polypeptides of interest. Methods for the preparation and staining of paraffin sections are known in the art for example as described in Trojanowski JQ, Obrocka MA, Lee VMY. (1983; A comparison of eight different chromogen protocols for the demonstration of immunoreactive neurofilaments or glial filaments in rat cerebellum using the peroxidase-antiperoxidase method and monoclonal antibodies. J. Histochem Cytochem 31, 1217-23)

In one embodiment the method determines the level of mRNA corresponding to the at least one gene, or determines the level of a polypeptide encoded by said at least one gene, or a fragment of said polypeptide. Where the method determines the expression of more than one gene, such as two, three, four or more genes, it will be understood that the method may include determination of different indicators of expression for different genes. For example, expression of one gene may be determined by levels of mRNA corresponding to that gene, whilst expression of another gene may be determined by levels of polypeptide encoded by the gene, whilst yet another may be determined by levels of a fragment of said polypeptide. Similarly, multiple indicators of the expression of a given gene may also be determined and are included within the scope of the invention. Therefore any suitable combination of indicators of the expression of a gene is included within the scope of the invention.

The methods of the invention include diagnosis and/or prognosis of cancers of the CNS, typically of gliomas, by determination of the level of expression of one or more genes selected from the group consisting of IQGAPl, Homer 1, IGFBP2 and ClQLl and comparison to the level in a reference sample. A difference in the level of expression of the gene in the test or biological sample which may also be referred to herein as a tumour sample (which includes a sample suspected of comprising a tumour) compared to the reference sample is indicative of the presence of a tumour and/or indicative of the type or grade of tumour. The difference may be an increase or a decrease in comparison to the reference sample. Typically, an increase in the level of expression of any one or more of IQGAPl, IGFBP2 and CIQLl is indicative of the presence of a glioma in the sample. Any level of increase may be indicative although typically an increase in the range of about 2-fold to about 25-fold, such as about 2-fold, about 5-fold, about 10-fold, about 20- fold or about 25-fold compared to reference sample would be indicative of the presence of a tumour. Typically, a decrease in the level of expression of Homer 1 is indicative of the presence and or type of a tumour. Any level of decrease in the level of expression of Homer 1 may be indicative, although the decrease is typically in the range of about 2-fold to about 20-fold, such as about 2-fold, about 5-fold, about 10-fold or about 15-fold.

In one embodiment the reference sample is a sample obtained form normal brain. In the diagnosis or prognosis of high grade versus low grade glioma, the reference sample may be derived from normal brain or may be derived from a tumour sample, such as a tumour that has previously been characterized as high or low grade glioma.

The level of expression in the said biological sample may be compared to one or more reference(s) that has been pre-determined, for example, a panel of known predetermined samples. For example the reference sample or panel may be a database of results of previously analysed reference samples. The one or more or panel of pre- determined reference samples may include known high grade, known low grade and or known normal samples.

In one embodiment a higher level of expression of one or both of IQGAPl and IGFBP2 in the biological sample compared to a known low grade sample may be indicative of a high grade glioma. In one embodiment expression of IQGAPl of about 5- fold higher in the biological test sample compared to the level of expression in a known low grade glioma reference sample may be indicative of a high grade glioma. Similarly, expression of IGFBP2 of about 15-fold higher in the biological test sample compared to the level of expression in a known low grade glioma reference sample may be indicative of a high grade glioma. The present invention also provides that high levels of expression of IQGAPl and or IGFBP2 can predict survival. The invention also provides that high levels of expression of IQGAPl and or IGFBP2 can predict tumour grade. The invention also provides that high levels of expression of IQGAPl and or IGFBP2 can predict astrocytic tumour type. Tumour grade and age predict survival. In one embodiment, the absent protein expression of IQGAPl and or IGFBP2 is predictive of long term survival which is defined as survival for greater than 3 years after initial diagnosis of a GBM. Treatment regime

The methods of the present invention provide an alternative or an adjunct to available methods for the diagnosis of tumours of the CNS, such as tumours of the brain. Hence the methods of the invention provide the practitioner with improved methods by which various types of tumours of the CNS may be diagnosed. The methods also allow the categorisation of various types of tumour, such as categorisation of high grade and low grade tumours. Accordingly the invention also provides methods for treatment of various types of tumours of the CNS.

For example, in one aspect, the invention permits the practitioner to accurately diagnose the tumour and to select a treatment regime appropriate to the type of tumour present in an individual. In one aspect, this may be accomplished by a method comprising a) diagnosing a tumour of the central nervous system (CNS) in an individual by a method comprising determining the expression of at least one gene selected from the group consisting of IQGAPl, Homerl, IGFBP2 and ClQLl in a tumour sample of said individual, and comparing said expression of said at least one gene with that of a reference sample; wherein a difference in said expression is an indication that the individual may have a tumour of the CNS; and b) formulating a therapeutic regime suitable for the treatment of an individual having said tumour; and c) administering said therapeutic regime to said individual. Current practice in Australia for high-grade gliomas is usually resection, followed by concomitant radiotherapy and chemotherapy (Temozolomide). Current practices for low grade tumours are typically at the discretion of the treating neuro-oncologist and neuro-surgeon. Some neuro-clinicians closely monitor patients with low grade tumours, without surgical, radiotherapy or chemotherapy recommendations. Others will call for surgery only, or surgery and radiotherapy. Markers of malignancy are needed in order to better guide neuro-clinicians.

These markers could also provide higher accuracy in grade prediction in small biopsies and gliomas that are surgically inoperable to determine their aggressive potential.

The present invention offers further advantages in the treatment of tumours of the CNS, for example gliomas. As described herein the inventors have identified differential expression of certain genes in cancerous tissue compared to normal tissue, particularly brain tissue. One such gene identified herein is IQGAPl. No previous studies have linked IQGAPl with gliomas. IQGAPl is thought to be involved in cellular motility and morphogenesis by interacting directly with cytoskeletal, cell adhesion and signal transduction proteins. This ability of IQGAPl to regulate cell migration and the identification herein of aberrant expression of IQGAPl indicates this gene and gene product as a candidate for therapeutic applications.

The present inventors have also demonstrated herein that Homer 1 is underexpressed in gliomas. No previous reports have described a link between expression of Homer 1 and cancer.

The present investigators have also described herein that ClQLl is on average about ten-fold higher in tumours compared to normal brain. No previous reports have described a link between the expression of ClQLl nor is its biological function documented.

The present inventors have also described herein that IGFBP2 acts as a discriminatory gene in gliomas.

According to the present invention any one or more of these genes may act as a marker for targeting a therapeutic agent to tumour cells. For example, as shown herein tumour cells have increased levels of expression of IQGAPl compared to normal cells of the CNS. This differential expression may act as a target by which to direct, preferentially, therapeutic agents to tumour cells, such as by way of an anti-IQGAPl antibody conjugated to a therapeutic moiety, such as a small molecule that modulates the activity of the IQGAPl, or a cytotoxic agent that reduces the number of tumour cells, or a prodrug. Small molecule drugs, cytotoxic agents and prodrugs are known in the art and include, for example, radiochemicals, toxins, such as diphtheria A chain, ricin A chain and the like. Anti-Homer 1 and/or anti-IGFBP2 and or anti-ClQLl antibodies may be used in a corresponding manner.

The invention described herein offers more direct methods for the treatment of tumours of the CNS, such as glioma. Accordingly, antibodies to any one or more of IQGAPl, Homer 1, ClQLl and IGFBP2 may be antagonists and may be used as a therapeutic agent. In one aspect, there is provided a method of treating a tumour of the CNS, comprising administering to an individual requiring said treatment an agent capable of binding to one or more of IQGAPl, Homer 1, IGFBP2 or ClQLl in an amount capable of reducing a biological function of an expression product of IQGAPl, Homer 1 or IGFBP2. In one embodiment the agent is an antibody. It will be understood that the method contemplates any combination of administration of the agent or agents, such as antibody or antibodies, including singular administration of one antibody, singular administration of multiple antibodies, either at the same time or separated over time, multiple administration of a single antibody and multiple administration of multiple antibodies. Additional therapeutic advantages also arise from the invention described herein. The identification by the inventors of the elevated levels of expression of IQGAPl, IGFBP2 and ClQLl in tumours of the CNS, such as gliomas, is consistent with a therapeutic approach in which agents capable of reducing the elevated levels of expression are utilised. Similarly, the identification herein of reduced levels of expression of Homer 1 in some gliomas is consistent with a therapeutic approach in which agent(s) capable of stimulating expression of Homer 1 or of activating a gene product of Homer 1 are utilised. By altering the level of expression of one or more of IQGAPl, Homer 1, IGFBP2 and ClQLl progression of the tumour may be diminished or inhibited. Methods for the suppression of expression of a given gene are known in the art, for example suppression of expression of a given gene may comprise RNA interference (RNAi), such as described in Bourguigon, L. Y. W., et al. (2005; Hyalururonan-CD44 Interaction with IQGAPl promotes Cdc42 and ERH signalling, leading to actin binding, EIk-I /Estrogen receptor Transcriptional Activation, and Ovarian cancer Progression. J. Biol. Chem. 280, 11961-11972)

A treatment regime described herein may be used in conjunction with known treatments for tumours of the CNS, such as surgical debulking, radiotherapy, chemotherapy, for example with temozolomide. It will be understood that use of therapy "in conjunction with" another includes simultaneous application of the therapies, as well as separate application of the therapies in any order and separated by an suitable time period.

The most appropriate treatment regime for any particular patient may be determined by the treating physician and will depend upon a variety of factors including: the particular type or types of tumour being treated and the severity of the disorder; activity of the compound or agent employed; the composition employed; the age, body weight, general health, sex and diet of the patient; the time of administration; the route of administration; the rate of sequestration of the agent or compound; the duration of the treatment; drugs used in combination or coincidental with the treatment, together with other related factors well known in medicine. The compounds and agents proposed for the present invention may be administered as compositions therapeutically. In a therapeutic application, compositions are administered, for example, to a patient already having a brain tumour whether symptomatic or not, in an amount sufficient to effectively treat the patient. The composition should thus provide a quantity of the compound or agent sufficient to effectively treat the patient. The therapeutically effective dose level for any particular patient will depend upon a variety of factors including: the disorder being treated and the severity of the disorder; activity of the compound or agent employed; the composition employed; the age, body weight, general health, sex and diet of the patient; the time of administration; the route of administration; the rate of sequestration of the agent or compound; the duration of the treatment; drugs used in combination or coincidental with the treatment, together with other related factors well known in medicine.

One skilled in the art would be able, by routine experimentation, to determine an effective, non-toxic amount of the therapeutic agent or agents where appropriate, which would be required to treat the condition.

Typically, in therapeutic applications, the treatment would be for the duration of the disease state.

Further, it will be apparent to one of ordinary skill in the art that the optimal quantity and spacing of individual dosages and, where combination therapy is used, optimal quantity and spacing of administration of the various agents of the combination therapy, will be determined by the nature and extent of the disease state being treated, the form, route and site of administration, and the nature of the particular individual being treated. Also, such optimum conditions can be determined by conventional techniques.

It will also be apparent to one of ordinary skill in the art that the optimal course of treatment, such as, the number of doses of the composition or compositions given per day for a defined number of days, can be ascertained by those skilled in the art using conventional course of treatment determination tests.

Screening Methods The identification herein of specific genes and gene products as having altered expression in tumours of the CNS, such as gliomas, permits the development of methods for the identification of agents capable of modulating the development of such tumours. These agents may also be referred to as candidate bioactive agents and offer the potential for use in methods of treating and/or preventing tumour development or progression. Accordingly, any one or more of the nucleic acid sequences, polypeptides and antibodies described herein may be used in screening for potential drugs for the treatment of tumours of the CNS. The nucleic acids, polypeptides and antibodies may be used to evaluate the effect of drug candidates on gene expression profiles, such as an expression profile characteristic of a particular tumour of the CNS. Similarly, nucleic acids, polypeptides and antibodies may be used to identify agents capable of interfering with a biological function or physiological activity of a polypeptide or protein encoded by a gene with aberrant expression in a tumour of the CNS, such as IQGAPl, Homer 1, IGFBP2 and ClQLl. The candidate agent, which includes candidate compounds and candidate drugs and similar terms, describes any molecule or class of molecules to be tested for bioactivity capable of directly or indirectly altering tumour development or progression, including transition from benign to malignant, and metastasis, or altering expression of one or more of the genes identified herein. Candidate agents include, for example, proteins, oligopeptides, small organic molecules, polysaccharides, polynucleotides. In preferred embodiments the candidate agent is capable of inhibiting tumour development or promoting tumour regression.

For example, candidate bioactive agents may be screened against cells expressing one or more of IQGAPl, Homer 1, IGFBP2 and ClQLl for the ability of the candidate agent to modulate the expression of one or more of IQGAPl, Homer 1 IGFBP2, and ClQLl. It will be understood that reference to a candidate agent having the ability to "modulate the expression of a gene includes an agent that has the ability to increase the expression of the gene and includes an agent that has the ability to decrease the expression of the gene. Preferably, the agent will decrease the expression of the gene. In development of therapeutic applications agents that decrease the expression of the over- expressed gene are preferred. In other circumstances, agents that increase the expression of a gene may be advantageous, such as in the investigation of tumour development and progression in vitro or in animal models.

The candidate agent may modulate expression either directly or indirectly, such as by triggering a response in the cell which in turn directly modulates the expression. The preferred degree of modulation may be determined by the skilled addressee, depending on the change in gene expression of the gene in tumour tissue compared to normal tissue. For example, if a tumour type exhibits an increase in expression of a given gene of about 25%, or about 50% or about 75%, or about 2-fold, about 3-fold, about 4-fold, about 5- fold, 10-fold or more compared to normal tissue, an agent capable of reducing expression of the gene by the respective amount may be suitable for therapy of the tumour. It will be understood that any degree of ability of a candidate agent to modulate, such as to decrease, the expression of a given gene will suffice to identify that agent as a suitable candidate. Hence, in a further aspect the invention provides a method for screening for an agent capable of modulating the expression of one or more genes selected from the group consisting of IQGAPl, Homer 1, IGFBP2 and ClQLl, the method comprising exposing a cell to a candidate agent and comparing the expression of the gene in the presence of the agent to the expression of the gene in the absence of the agent, whereby a difference in expression of the gene in the presence compared to the absence of the agent is indicative of an agent capable of modulating the expression of the gene. In one embodiment multiple candidate agents may be screened simultaneously, such as by exposing said cell to a composition comprising a plurality of discrete candidate agents. In one embodiment the cell is an isolated cell or cells. In one embodiment the cell is comprised in an organism, such as a mouse, rat or primate. In one embodiment the organism is a transgenic organism, comprising recombinant nucleic acid sequence encoding one or more of IQGAPl, Homer 1, IGFBP2 and ClQLl. In one embodiment the cell is a tumour cell, such as a glioma. Another potential antagonist is an antisense construct prepared using antisense technology. Antisense technology can be used to control gene expression through triple- helix formation or antisense DNA or RNA, both of which methods are based on binding of a polynucleotide to DNA or RNA. Various methods of antisense technology are known in the art and include, for example those described in Lee et al., Nucl. Acids Res. 6:3073 (1979); Cooney et al, Science 241 :456 (1988); and Dervan et al., Science 251:1360 (1991); Okano, J., Neurochem. 56:560 (1991); Oligodeoxynucleotides as Antisense Inhibitors of Gene Expression, CRC Press, Boca Raton, FIa. (1988)). In effect, the methods prevent or reduce transcription and the production of IQGAPl, Homer 1 or IGFBP2. The invention also provides a method for screening for an agent capable of modulating a biological activity of a polypeptide, or fragment thereof, encoded by a gene selected from the group consisting of IQGAPl, Homer 1, IGFBP2 and ClQLl, the method comprising exposing said polypeptide or said fragment to a candidate agent under conditions suitable for the expression of the biological activity and comparing the biological activity of the polypeptide or fragment in the presence of the agent to the biological activity in the absence of the agent, whereby a difference in biological activity of the polypeptide or fragment in the presence compared to the absence of the agent is indicative of an agent capable of modulating a biological activity of the polypeptide or fragment. In the performance of the method of screening the polypeptide may be exposed as an isolated or substantially purified polypeptide, or it may be exposed in the form of an expression product encoded by a cell, such as a recombinant cell engineered to express the polypeptide. For example, the polypeptide may be expressed by a transgenic organism, such as a mouse, rat or primate.

To determine the effect of an agent or agents on a biological activity of the polypeptide, it will be understood that the screening method will include at least one step in which the conditions are suitable for the subject biological activity to be expressed. The skilled addressee is capable of determining the appropriate conditions, taking into accounts factors such as the particular biological activity of the polypeptide under investigation.

In one embodiment the polypeptide is expressed by a transgenic organism.

In one embodiment the agent is an antagonist of a biological activity of a polypeptide encoded by a gene selected from the group consisting of IQGAPl, Homer 1, IGFBP2 and ClQLl.

In one embodiment the agent is a monoclonal antibody which binds to the polypeptide.

In a preferred embodiment the agent is an antagonist of a biological activity of a polypeptide encoded by a gene selected from the group consisting of IQGAPl, Homer 1, IGFBP2 and ClQLl. Examples of potential antagonists include an antibody, or in some cases, an oligonucleotide, which binds to the polypeptide.

Potential antagonists include small molecules, such as those that bind to and occupy the binding site of a target receptor thereby making the receptor inaccessible, such that normal biological activity is prevented or reduced. Examples of small molecules include but are not limited to small peptides or peptide-like molecules.

Kits

The identification by the present inventors of genes that are abnormally expressed in tumours of the CNS and subsequent methods for diagnosis of such tumours also leads to the provision of components for use in such methods. As such, invention also provides kits for use in diagnosing a tumour of the CNS, the kit comprising at least one probe specific for a gene selected from the group consisting of IQGAPl, Homer 1, IGFBP2 and ClQLl. In a preferred embodiment the kits comprises a plurality of antibodies specific for polypeptides present at elevated expression levels in tumours of the CNS compared to normal tissue, such as IQGAPl, Homerl, IGFBP2 and ClQLl.

Any suitable probe or probes may be included in the kit, such as one or more probes selected from the group consisting of a nucleic acid sequence and an antibody or both. The kit may comprise any number of additional components. By way of non- limiting example, the additional components may include (i) one or more reference probe(s); (ii) one or more detection reagent(s); (iii) one or more agent(s) for immobilising a polypeptide on a solid support; (iv) a solid support material; (v) instructions for use of the kit or a component(s) thereof in a method for diagnosing a tumour of the CNS. In one embodiment the kit comprises one or more probe(s) immobilised on a solid support, such as a biochip.

As used herein, the term "kit" refers to any delivery system for delivering materials. In the context of detection assays, such delivery systems include systems that allow for the storage, transport, or delivery of reaction reagents (for example, oligonucleotides, enzymes, antibodies, attachment materials, labels, reference samples, supporting material, etc. in the appropriate containers) and/or supporting materials (for example, buffers, written instructions for performing the assay etc.) from one location to another. For example, kits include one or more enclosures, such as boxes, containing the relevant reaction reagents and/or supporting materials. As used herein, the term "fragmented kit" refers to a delivery system comprising two or more separate containers that each contain a subportion of the total kit components. The containers may be delivered to the intended recipient together or separately. For example, a first container may contain an enzyme for use in an assay, while a second container contains oligonucleotides. Indeed, any delivery system comprising two or more separate containers that each contains a subportion of the total kit components are included in the term "fragmented kit." In contrast, a "combined kit" refers to a delivery system containing all of the components of a reaction assay in a single container (e.g., in a single box housing each of the desired components). The term "kit" includes both fragmented and combined kits.

Pharmaceutical compositions

Compounds of the invention, such as polypeptides, nucleic acid sequences, antibodies and agents capable of modulating the expression of a gene, may be prepared as pharmaceutical compositions. It will be understood that compounds of the invention need not always be in the form of a pharmaceutically acceptable composition. For example, compounds and compositions for use in diagnostic applications may or may not be in a pharmaceutically acceptable form. In general, suitable compositions may be prepared according to methods which are known to those of ordinary skill in the art and accordingly may include a pharmaceutically acceptable carrier, diluent, excipient and/or adjuvant.

These compositions can be administered by standard routes. In general, the compositions may be administered by the parenteral (e.g., intravenous, intraspinal, subcutaneous or intramuscular), oral or topical route. Preferably administration is by the parenteral or oral route. More preferably administration is by the oral route.

The carriers, diluents, excipients and adjuvants must be "acceptable" in terms of being compatible with the other ingredients of the composition, and not deleterious to the recipient thereof. Examples of pharmaceutically acceptable carriers or diluents are demineralised or distilled water; saline solution; vegetable based oils such as peanut oil, safflower oil, olive oil, cottonseed oil, maize oil, sesame oils, arachis oil or coconut oil; silicone oils, including polysiloxanes, such as methyl polysiloxane, phenyl polysiloxane and methylphenyl polysolpoxane; volatile silicones; mineral oils such as liquid paraffin, soft paraffin or squalane; cellulose derivatives such as methyl cellulose, ethyl cellulose, carboxymethylcellulose, sodium carboxymethylcellulose or hydroxypropylmethylcellulose; lower alkanols, for example ethanol or iso-propanol; lower aralkanols; lower polyalkylene glycols or lower alkylene glycols, for example polyethylene glycol, polypropylene glycol, ethylene glycol, propylene glycol, 1,3- butylene glycol or glycerin; fatty acid esters such as isopropyl palmitate, isopropyl myristate or ethyl oleate; polyvinylpyrridone; agar; carrageenan; gum tragacanth or gum acacia, and petroleum jelly. Typically, the carrier or carriers will form from 10% to 99.9% by weight of the compositions.

The composition may include agents which increase the bioavailability or therapeutic duration of the active compound or compounds.

The compositions of the invention may be in a form suitable for administration by injection, in the form of a formulation suitable for oral ingestion (such as capsules, tablets, caplets, elixirs, for example), in the form of an ointment, cream or lotion suitable for topical administration, in a form suitable for delivery as an eye drop, in an aerosol form suitable for administration by inhalation, such as by intranasal inhalation or oral inhalation, in a form suitable for parenteral administration, that is, subcutaneous, intramuscular or intravenous injection.

For administration as an injectable solution or suspension, non-toxic parenterally acceptable diluents or carriers can include, Ringer's solution, isotonic saline, phosphate buffered saline, ethanol and 1,2 propylene glycol.

Some examples of suitable carriers, diluents, excipients and adjuvants for oral use include peanut oil, liquid paraffin, sodium carboxymethylcellulose, methylcellulose, sodium alginate, gum acacia, gum tragacanth, dextrose, sucrose, sorbitol, mannitol, gelatine and lecithin. In addition these oral formulations may contain suitable flavouring and colourings agents. When used in capsule form the capsules may be coated with compounds such as glyceryl monostearate or glyceryl distearate which delay disintegration.

Adjuvants typically include emollients, emulsifiers, thickening agents, preservatives, bactericides and buffering agents.

Solid forms for oral administration may contain binders acceptable in human and veterinary pharmaceutical practice, sweeteners, disintegrating agents, diluents, flavourings, coating agents, preservatives, lubricants and/or time delay agents. Suitable binders include gum acacia, gelatine, corn starch, gum tragacanth, sodium alginate, carboxymethylcellulose or polyethylene glycol. Suitable sweeteners include sucrose, lactose, glucose, aspartame or saccharine. Suitable disintegrating agents include corn starch, methylcellulose, polyvinylpyrrolidone, guar gum, xanthan gum, bentonite, alginic acid or agar. Suitable diluents include lactose, sorbitol, mannitol, dextrose, kaolin, cellulose, calcium carbonate, calcium silicate or dicalcium phosphate. Suitable flavouring agents include peppermint oil, oil of wintergreen, cherry, orange or raspberry flavouring. Suitable coating agents include polymers or copolymers of acrylic acid and/or methacrylic acid and/or their esters, waxes, fatty alcohols, zein, shellac or gluten. Suitable preservatives include sodium benzoate, vitamin E, alpha-tocopherol, ascorbic acid, methyl paraben, propyl paraben or sodium bisulphite. Suitable lubricants include magnesium stearate, stearic acid, sodium oleate, sodium chloride or talc. Suitable time delay agents include glyceryl monostearate or glyceryl distearate.

Liquid forms for oral administration may contain, in addition to the above agents, a liquid carrier. Suitable liquid carriers include water, oils such as olive oil, peanut oil, sesame oil, sunflower oil, safflower oil, arachis oil, coconut oil, liquid paraffin, ethylene glycol, propylene glycol, polyethylene glycol, ethanol, propanol, isopropanol, glycerol, fatty alcohols, triglycerides or mixtures thereof.

Suspensions for oral administration may further comprise dispersing agents and/or suspending agents. Suitable suspending agents include sodium carboxymethylcellulose, methylcellulose, hydroxypropylmethyl-cellulose, poly-vinyl-pyrrolidone, sodium alginate or acetyl alcohol. Suitable dispersing agents include lecithin, polyoxyethylene esters of fatty acids such as stearic acid, polyoxyethylene sorbitol mono- or di-oleate, -stearate or - laurate, polyoxyethylene sorbitan mono- or di-oleate, -stearate or -laurate and the like.

The emulsions for oral administration may further comprise one or more emulsifying agents. Suitable emulsifying agents include dispersing agents as exemplified above or natural gums such as guar gum, gum acacia or gum tragacanth.

Methods for preparing parenterally administrable compositions are apparent to those skilled in the art, and are described in more detail in, for example, Remington's Pharmaceutical Science, 15th ed., Mack Publishing Company, Easton, Pa., hereby incorporated by reference herein.

The topical formulations of the present invention, comprise an active ingredient together with one or more acceptable carriers, and optionally any other therapeutic ingredients. Formulations suitable for topical administration include liquid or semi-liquid preparations suitable for penetration through the skin to the site of where treatment is required, such as liniments, lotions, creams, ointments or pastes, and drops suitable for administration to the eye, ear or nose. Formulations suitable for topical administration may be provided as a transdermal patch.

Drops according to the present invention may comprise sterile aqueous or oily solutions or suspensions. These may be prepared by dissolving the active ingredient in an aqueous solution of a bactericidal and/or fungicidal agent and/or any other suitable preservative, and optionally including a surface active agent. The resulting solution may then be clarified by filtration, transferred to a suitable container and sterilised. Sterilisation may be achieved by: autoclaving or maintaining at 90⁰C-IOO⁰C for half an hour, or by filtration, followed by transfer to a container by an aseptic technique. Examples of bactericidal and fungicidal agents suitable for inclusion in the drops are phenylmercuric nitrate or acetate (0.002%), benzalkonium chloride (0.01%) and chlorhexidine acetate (0.01%). Suitable solvents for the preparation of an oily solution include glycerol, diluted alcohol and propylene glycol. Lotions according to the present invention include those suitable for application to the skin or eye. An eye lotion may comprise a sterile aqueous solution optionally containing a bactericide and may be prepared by methods similar to those described above in relation to the preparation of drops. Lotions or liniments for application to the skin may also include an agent to hasten drying and to cool the skin, such as an alcohol or acetone, and/or a moisturiser such as glycerol, or oil such as castor oil or arachis oil.

Creams, ointments or pastes according to the present invention are semi-solid formulations of the active ingredient for external application. They may be made by mixing the active ingredient in finely-divided or powdered form, alone or in solution or suspension in an aqueous or non-aqueous fluid, with a greasy or non-greasy basis. The basis may comprise hydrocarbons such as hard, soft or liquid paraffin, glycerol, beeswax, a metallic soap; a mucilage; an oil of natural origin such as almond, corn, arachis, castor or olive oil; wool fat or its derivatives, or a fatty acid such as stearic or oleic acid together with an alcohol such as propylene glycol or macrogols. The composition may incorporate any suitable surfactant such as an anionic, cationic or non-ionic surfactant such as sorbitan esters or polyoxyethylene derivatives thereof. Suspending agents such as natural gums, cellulose derivatives or inorganic materials such as silicaceous silicas, and other ingredients such as lanolin, may also be included. The compositions may also be administered in the form of liposomes. Liposomes are generally derived from phospholipids or other lipid substances, and are formed by mono- or multi-lamellar hydrated liquid crystals that are dispersed in an aqueous medium. Any non-toxic, physiologically acceptable and metabolisable lipid capable of forming liposomes can be used. The compositions in liposome form may contain stabilisers, preservatives, excipients and the like. The preferred lipids are the phospholipids and the phosphatidyl cholines (lecithins), both natural and synthetic. Methods to form liposomes are known in the art, and in relation to this specific reference is made to: Prescott, Ed., Methods in Cell Biology, Volume XIV, Academic Press, New York, N.Y. (1976), p. 33 et seq., the contents of which is incorporated herein by reference.

Examples

The examples are intended to serve to illustrate this invention and should not be construed as limiting the general nature of the disclosure of the description throughout this specification. Methods Tumour Sampling and Data Collection

Glioma samples were collected from patients who underwent surgery at Royal North Shore Hospital, North Shore Private Hospital, and Prince of Wales Private Hospital, NSW, Australia. Approval for this study was obtained from the Human Research Ethics Committees of the participating institutions. No patients received chemotherapy or radiotherapy prior to surgery. Surgically removed gliomas were snap- frozen in liquid nitrogen immediately and stored at -80⁰C until RNA extraction. Frozen tissue was available for 71 gliomas (37 samples included in the microarray analysis and an additional 34 samples included in the qPCR). Formalin-fixed, paraffin-embedded glioma blocks were provided by the Department of Anatomical Pathology, Royal North Shore Hospital. Paraffin-embedded tissue was available for the 71 frozen tumour samples and an additional 72 glioma samples (total of 143 gliomas). All gliomas were obtained from the initial surgery and graded according to the

WHO 2000 criteria by experienced neuropathologists. The following subtypes were included in this study: astrocytoma grade II (All), astrocytoma grade III (AIII), glioblastoma multiforme (GBM), oligodendroglioma grade II (Odgll), oligodendroglioma grade III (OdglH) and oligoastrocytoma grade III (OAIII). Clinical information collected included patient age, gender, resection type (debulking or biopsy), tumour location and percentage Ki67 staining. Data on the loss of heterozygosity (LOH) of chromosome Ip and 19q, routinely tested in oligodendroglioma patients, was also available. Survival time (days) was counted from the patient admission date. The census date for survival was 31^st July 2006. Only gliomas that were surgically debulked were included in the survival analysis. Pathology reports for all patients included in this study were reviewed for the presence of necrosis and microvascular proliferation (MVP). In our data set, 96% of gliomas with described necrosis also had MVP present. Gliomas with both Necrosis and MVP were grouped together and referred to as high grade gliomas while, gliomas in which necrosis or MVP were absent were grouped together and referred to as low grade gliomas. The characteristics of the 143 study patients and the regrouping of the WHO classified gliomas into high grade and low grade groups are listed in Table 1.

Microarray analysis using a two color platform involves comparative hybridization of the sample of interest against a reference, for example tumour versus normal. It was logistically and ethically not possible to obtain sufficient matched normal brain tissue. Therefore a commercial total brain RNA mix was used (Ambion, Inc. USA). The commercial RNA was isolated from a donor who had no history of cancer, or brain disorders. All rumours were hybridized against this normal total brain reference.

Preparation of RNA samples

Total RNA was extracted from fresh frozen brain tumour tissue using QIAZOL reagent (Qiagen GmbH, Strasse, Germany) and purified by precipitation with 2.5 mol/L Lithium chloride according to the manufacturer's protocols (Ambion, Inc., Austin, TX, USA). Purified RNA was quantitated by UV absorbance at 260 and 280nm and assessed qualitatively by formaldehyde agarose gel electrophoresis. Adequate amounts of RNA for the microarray experiments were available for 17 high grade gliomas and 20 low grade gliomas.

Microarray Experimental Procedure Microarray slides printed with the Compugen 19,000 human oligonucleotide library were obtained from the Adelaide Microarray Facility, University of Adelaide, Australia. An array list, H19K array list v3.01.gal, was provided with the chips. Slides from one print batch only were used in these experiments.

For each glioma, 20μg of purified RNA from both the tumour and the normal total brain control were reversed transcribed using Superscript III (Invitrogen® San Diego, CA, USA) and anchored oligo-(dT20VN) in the presence of Cy5-dCTP and Cy3-dCTP (Amersham Inc., Piscataway NJ, USA) as described previously (15). After direct labelling, the two probes were hybridized for 20 hours at 5O⁰C to a microarray slide. The slides were then washed, immediately dried, and scanned with a 10-μm resolution on a GenePix 4000A scanner (Molecular Devices, Sunnyvale, CA, USA) at wavelengths 635 and 532 nm for Cy5- and Cy3- labelled probes, respectively. The resulting TIFF images were analyzed by GenePix Pro 4.0 software (Molecular Devices, Sunnyvale, CA, USA).

Microarray Data Analysis Both univariate and multivariate approaches were used to detect differential expression between tumour groups. Prior to statistical analysis, pre-processing and normalisation of the raw GenePix data was required to remove artifacts and systematic effects due to processing of the arrays. Pre-processing of the data was carried out utilizing R statistical software Version 2.0.1 libraries contained in Bioconductor, the open source software providing tools for the analysis of genomic data (http://www.bioconductor.orgΛ. No filtering or background correction was performed on the data and normalization was carried out by printTipLoess.

The univariate analysis was carried out using the package Limma which has been specifically designed for detecting differential expression in microarray data. The software application was developed by Smyth (2004) and involves the use of a moderated t statistic. The top 100 ranked differentially expressed genes were identified for each tumour group comparison.

Multivariate statistical methods are more appropriate for high dimensional data such as microarray intensities and use of these methods avoids the need to adjust for multiple testing by controlling for the family-wise error rate or False Discovery Rate.

Two multivariate methods developed by CSIRO, GeneRave (Kiiveri, 2003) and SDDA were utilized to select genes that discriminate between high grade and low grade gliomas. GeneRave has a generalised linear model (GLM) framework and is coupled with a Bayesian approach to variable selection. The GeneRave method involves fitting logistic regression models to build a model that gives the best separation between groups.

GeneRave utilises specialised model fitting EM algorithms and achieves almost unbiased estimation of error rates and model significance through cross-validation and permutation.

The GeneRave algorithms perform well where there is little prior knowledge and have been shown to be computationally fast and scale up well to handle large numbers of variables.

SDDA is a technique based on linear discriminant analysis which is used to build multiclass classifiers by assuming the gene intensity measures come from a multivariate normal distribution with means differing by a class fixed (unknown) covariance matrix.

Validation of genes using quantitative PCR

Target genes selected from the microarray analysis were validated using quantitative real time PCR (qPCR) in 25 high grade gliomas and 23 low grade gliomas (where adequate amounts of RNA were available). cDNA was synthesised from 5μg of RNA using random hexamers as described previously (Haven et al., 2004). All qPCR were performed using a 5' nuclease technique with specific TaqMan® Gene Expression Assays (Applied Biosystems (ABI), Foster City, CA, USA) and TaqMan® Universal PCR Master Mix, NO AmpErase UNG (Applied Biosystems, Foster City, CA, USA). All qPCR were performed on a Rotorgene 3000 (Corbett Research, Mortlake, NSW, Australia). Ribosomal 18s RNA was chosen as the endogenous control for normalization. Differences between classes were assessed statistically using REST-XL©- version 2 (Relative Expression Software Tool) (Pflaff et al., 2002) where relative expression ratios are computed based on the PCR efficiency and crossing point differences. The Student t- test analysis was used to evaluate the statistical significance of the mRNA expression levels of the target genes between high grade and low grade gliomas (Stata statistical software version 8.2 (StataCorp Texas, USA)).

Antibodies A mouse monoclonal antibody (MAb) against IQGAPl (BD Transduction Laboratories, Macquarie University Research Park, NSW, Australia, 2113) was used at a concentration of 1 :300. The negative control, Mouse IgGl was purchased from Dako (Dako Inc., CA, USA). A goat polyclonal antibody against IGFBP2 (C-18, Santa Cruz Biotechnology, Inc., Santa Cruz, USA) was used at a concentration of 1 :150. Normal goat IgG was purchased from Santa Cruz (Santa Cruz Biotechnology, Inc., Santa Cruz, USA).

Immunohistochemistry

Immunohistochemical (IHC) studies were performed on serial 4μm sections from 143 paraffin-embedded glioma blocks (73 high grade gliomas and 70 low grade gliomas) and normal brain samples. One section from each sample was stained with haematoxylin and eosin (H&E) to facilitate histological assessment. Sections were deparaffinized, rehydrated, treated with EDTA retrieval solution (pH 9.0) or citrate based retrieval solution (pH 6.0) for 20 minutes at 95⁰C and blocked with 0.3% hydrogen peroxide prior to the application of primary antibodies (Dako Aust. Pty Ltd, Botany, Australia). The IQGAPl antibody was detected using the Envision™+ Dual Link Peroxidase Detection System (Dako Aust. Pty Ltd, Botany, Australia) and IGFBP2 was detected using the LSAB™+ Streptavidin Peroxidase Detection System (Dako Aust. Pty Ltd, Botany, Australia). DAB+ liquid stable substrate system was used for visualization. All IHC staining was performed on an autostainer (Autostainer Plus, Dako Inc., CA, USA). Sections were counterstained with haematoxylin. Two types of negative controls, substituting the matched mouse IgG isotype and goat non immune IgG in the staining protocol, were used.

Immunostaining results were evaluated semi-quantitatively by three independent observers and a neuro-pathologist. IQGAPl immunostaining was scored according to: 0- negative staining; 1- weak cytoplasmic staining (<5% of examined tumour cells); 2- moderate cytoplasmic staining (<20% of examined tumour cells); 3- moderate to strong cytoplasmic staining (<25% of examined tumour cells) and 4- strong cytoplasmic staining (>25% of examined tumour cells). Gliomas that scored >3 were regarded as positive. Gliomas that scored <2 were regarded as negative for IQGAPl protein expression. IGFBP2 immunostaining was scored according to: 0- negative staining; 1- weak cytoplasmic staining (<5% of examined tumour cells); 2- moderate cytoplasmic staining (<25% of examined tumour cells) and 3- strong membranous and cytoplasmic staining (>25% of examined tumour cells). Gliomas that scored >2 for IGFBP2 protein expression were regarded to be positive, and gliomas scoring <1 were regarded as negative. Chi squared (χ²) analysis was used to evaluate the statistical significance of the immunostaining results (Stata statistical software version 8.2). The agreement (average kappa and range) among the observers was assessed using Stata (Stata statistical software version 8.2). The kappa value (K) of >0.80 indicates excellent interobserver agreement in excess of chance, while a K- value of <0.20 indicates a poor agreement.. Logistic regression analysis was used to assess the association between IGFBP2 and IQGAPl protein expression levels and high or low grade status (Stata statistical software version 8.2).

To determine if IQGAPl and IGFBP2 protein expression levels could be used as predictive markers of long term survival in patients diagnosed with GBM, 13 patients were recruited with a confirmed diagnosis of a GBM and survival greater than 3 years after initial diagnosis. The average age of these patients was 45.8 years and the average survival was 3.7 years. As a comparative group, 37 GBM patients were recruited with an average age of 49.5 years and an average survival of 1.1 years. Paraffin sections (4μm) were obtained for all LTS and STS patients and immunostained for IQGAPl and IGFBP2 according to the protocol outlined above.

Survival Analysis Cox proportional hazards regression analysis was performed to assess IGFBP2 and IQGAPl protein expression levels as prognostic indicators for survival adjusted for age (R statistical software Version 2.0.1). The proportional hazards assumption was tested and found to be acceptable. Kaplan-Meier survival analysis was used to generate survival curves and estimates of median survival times. The log rank test was used to compare survival curves for samples split by age, grade status, tumour type, IQGAPl protein expression and IGFBP2 protein expression (Stata statistical software version 8.2).

RESULTS Survival Analysis of the Glioma Dataset

To determine the correlation between WHO tumour grade and survival, we plotted Kaplan-Meier survival curves of the 143 patients entered in this study (Fig. IA). These results confirm that histologic (WHO) grade is a good predictor of survival, with significantly shorter survival observed in patients with GBM. There is a subgroup of patients within the OdgIII and OAIII grades who have similar survival to GBM in the first year however other patients within these same grades show much longer survival (Fig. IA). To identify genes associated with this more aggressive biological behaviour, we assigned the gliomas («=143) into two broad groups; high grade (n=73) containing 52 GBMs, 8 OAIIIs and 13 OdgIIIs or low grade («=70) containing 4 OAIIIs, 13 Odgllls, 20 Odglls, 20 AIIIs and 13 Alls. Kaplan-Meier plots of these two major groups show that high grade gliomas had a median survival of 337 days (Fig. IB). Only 10% of high grade gliomas were alive at 5 years compared to 80% of low grade gliomas. Separating the gliomas according to necrosis and MVP has highlighted the separation of OAIII and OdgIIIs into 2 distinct groups. The survival curves also confirm that necrosis and microvascular proliferation are important independent predictors of survival.

Microarray Data Analysis

Gene expression profiles of the 17 high grade gliomas (12 GBMs, 3 OAIIIs and 2 OdgIIIs) when compared to 20 low grade gliomas (5 Alls, 5 AIIIs, 2 OAIIIs, 3 OdgIIIs and 5 Odglls) showed significant upregulation of 185 genes and the downregulation of 42 genes. The two algorithms, SDDA and GeneRave were applied to the full gene data set to identify smaller sets of expressed genes uniquely associated with the high grade gliomas. In this study we utilised SDDA to identify 7 gene combinations with known biological function [Homerl and IQGAPl, LGALSl and LRRC20, IGFBP2 on its own, CARHSPland COPZ2] that could discriminate between the two broad groups of gliomas (Table 2). GeneRave was utilised to identify 7 gene combinations with known biological function [IGFBP2 and ClQLl, CHBLl (commonly referred to by its protein product, YKL40), SERPINA3, SPPl and RBPl, LGALSl on its own]. We chose the gene sets; Homer 1 and IQGAPl and IGFBP2 and ClQLl for further validation and analysis based on their rankings as genes with the highest predictive accuracy of identifying gliomas with necrosis and MVP by GeneRave and SDDA (79% and 84%, respectively; Table 2). IGFBP2 was of particular interest because both algorithms nominated this gene as the top discriminator of high grade and low grade gliomas (Table 2). By plotting the gene expression values of the 37 gliomas, the scatter plots demonstrate excellent separation of high grade gliomas from the low grade gliomas using IQGAPl and Homer 1 (Fig. 2A) and IGFBP2 and ClQLl (Fig. 2B).

qPCR validation of gene sets qPCR was used to quantify mRNA levels of Homer 1, IQGAPl, IGFBP 2 and

ClQLl in 48 gliomas and normal brain. The validation dataset comprised 14 tumours used in the microarray analysis, complemented with an additional 34 untested tumours («=48). The total number used in the validation included 25 high grade gliomas (13 GBMs, 7 OAIIIs and 5 Odgllls) and 23 low grade gliomas (1 OAIII, 3 Odgllls, 6 Odglls, 7 AIII and 9 Alls). Upregulation of IQGAPl and IGFBP2 was observed in the high grade gliomas (PO.001 for both target genes) (Fig. 2C). In this high grade group the mRNA expression of IQGAPl was 4-fold higher compared to low grade gliomas while IGFBP2 showed 12-fold higher mRNA expression in the high grade group compared to low grade gliomas. No significant difference in mRNA expression was observed between the two glioma groups with Homerl and ClQLl gene sets (Fig. 2C; P=0.465, P=0.721 respectively).

Immunohistochemical expression of IQGAPl and IGFBP2

The interobserver agreement for IQGAPl and IGFBP2 expression scores in the 143 gliomas tested among the three observers was K- values of 0.91 and 0.97, for each respective protein scored. Logistic regression analysis confirmed that protein expression scores of IQGAP and IGFBP2 were strongly associated with high grade gliomas (PO.001). For each unit increase in IQGAPl or IGFBP2 score, the odds that the tumour will have necrosis and MVP increases 2.4 and 4.7 fold respectively (Fig. 3).

Expression and localisation of IQGAPl in normal brain and gliomas

No cytoplasmic or membranous immunostaining of IQGAPl was observed in the normal glial tissue or in 6 out of 70 (9%) of the low grade gliomas (Fig. 3, Panel A). Some uptake was noted in red blood cells within vascular spaces and in the endothelial cells. Widespread weak cytoplasmic IQGAPl (Scores 1-2) was observed in 45 out of 70 (64%) low grade gliomas tested (Fig. 3, Panel A, Table 3). Interestingly, there was strong positive immunostaining of IQGAPl also observed in the endothelial cells in these tumours. The immunostaining of IQGAPl in the high grade glioma group was intense throughout the sections examined (Fig. 3, Panel B; Table 3). Sixty-three out of 73 (86%) high grade gliomas showed strong IQGAPl protein expression, with accentuation beneath the cell membrane (Scores >3). In contrast to the low grade gliomas there was no staining in the endothelial cells of the high grade gliomas (Fig. 3, Panel B, GBM 1). Of note was the IQGAPl immunostaining of the pseudopalisading cells surrounding the central necrotic focus (Fig. 3, Panel B, GBM 2).

The immunostaining of IQGAPl in the 143 gliomas grouped according to their WHO classification is summarised in Table 5. As predicted from the high grade grouping scheme, 79% of GBMs («=52) were positive for IQGAPl (score >3). Only 15% of the low grade gliomas, All (n=\ 3) and OdgIIs (n=20) showed IQGAPl protein expression. Large variability in IQGAPl immunostaining was evident within each of the grade III groupings, AIII, OdgIII and OAIII glioma types. IQGAPl immunostaining was observed in 20% of AIII («=20), 32% of OdgIII (#i=26) and 50% of OAIII («=12). Representative images of the negative and positive IQGAPl immunostaining of pure oligodendrogliomas (OdgIII) and mixed gliomas (OAIII) are shown in Figure 3 (Panel A, B).

Expression and localisation of IGFBP2 in normal brain and gliomas

No immunostaining of IGFBP2 was observed in normal brain (glia, endothelium or red blood cells) or in 43 out of 70 (61%) low grade gliomas (Fig. 3 Panel C; Table 4). In 16 (23%) of low grade gliomas, minor immunostaining represented by faint haloes of accentuated cytoplasmic reactivity surrounding the nuclei, was observed in less than 5% of the tumour sections studied. Strong cytoplasmic immunostaining of IGFBP2 was observed in 64 out of the 73 (88%) high grade gliomas tested however the distribution was patchy throughout the tumour sample (Fig. 3, Panel D; Table 4). In some samples, IGFBP2 immunostaining was visible in over 50% of the glioma section examined. Strongly positive areas were often located close to the necrotic regions and staining was especially prominent in pseudopalisading cells (Fig. 3, Panel D, GBM 2).

Immunostaining of IGFBP2 in the 143 gliomas (WHO classification) is summarised in Table 5. Positive IGFBP2 protein expression was observed in 92% of GBM (score >2). Only 3 out of a total of 13 All and 3 out of 20 OdgII low grade gliomas showed positive IGFBP2 immunostaining. Large variability in the IGFBP2 immunostaining was again evident within each of the grade III groupings, AIII, OdgIII and OAIII. IGFBP2 immunostaining was noted in 15 of AIII («=20), 32% of OdgIII 5 (n=26) and 58% of OAIII («=12). Representative images of the negative and positive IGFBP2 staining observed in the pure oligodendrogliomas (OdgIII) and mixed gliomas (OAIII) are shown in Figure 3, Panel C and D.

Survival Analysis io Univariate analysis (log-rank test) of the prognostic significance of the following variables; age, pure oligodendroglioma (OdgII and OdgIII) or astrocytic component (All, AIII, OAIII and GBM), IQGAPl and IGFBP2 protein expression scores showed that all variables were highly significantly related to survival time. The Kaplan-Meier curves illustrating the effect of these features on survival are shown in Figure 4. A patient's age

I₅ over 60 years and gliomas with an astrocytic element were strongly associated with poorer survival (Fig. 4A, B). Patients with high scores of IQGAPl (Score >3) and IGFBP2 (Score>2) experienced shorter median survival times (median 351 and 303 days, respectively) than patients who scored negative for IQGAPl (Scores <2) and IGFBP2 (Scores <1) (Fig. 4C, D).

₂o A Cox proportional hazards regression model incorporating age, IQGAPl protein expression (>3) and IGFBP2 protein expression (>2) showed that the best predictors of poor survival were increased age and IGFBP2 protein expression (Table 6). There was evidence that IQGAPl protein expression was also predictive of poorer survival but this did not reach significance (P=O.095). Compared to the reference age group of 20-39

25 years, patients aged 60+ had a 3.5 times greater risk of death. Patients with high protein expression of IGFBP2 had a 1.7 times greater risk of death.

The survival curves shown in Figure 5 illustrate the impact of the IQGAPl and IGFBP2 protein expression markers on patient outcome when the 143 gliomas were grouped according to their WHO classification. Because of low numbers of All and

30 OdgII (refer to Table 5), comparisons of these low grade gliomas were not performed. The median survival for patients diagnosed with a GBM did not differ significantly in regard to whether the sample stained positive (IQGAP 1+ve) or negative with IQGAPl (IQGAP 1-ve) (Fig. 5A). However, at 3 years, 22% of GBM/IQGAPl-ve patients in our study were alive compared to less than 10% of GBM/IQGAPl+ve patients. There were no AIII/IQGAP+ve patients alive after 6 years compared to 90% survival in AIII/IQGAPl-ve patients, after 8 years (Fig. 5A). Significantly shorter survival was observed in OdgIII/IQGAPl+ve patients. Only 20% OdgIII/IQGAPl+ve patients were still alive at 5 years compared to 70% of OdgIII/IQGAPl-ve patients (Fig. 5B). In patients who were OAIII/IQGAPl+ve, 25% were still alive at 2 years compared to 70% survival in patients with OAIII/IQGAPl-ve (Fig. 5B).

The median survival of patients with AIII/IQGAP+ve gliomas (1919 days) (Log Rank PO.001) and OdgIII/IQGAP+ve gliomas (427 days) (Log Rank P=0.081) were better than GBM (315 days). The median survival of patients with OAIH/IQGAP+ve gliomas was 242 days (Log Rank P=0.890) (Fig. 5).

The median survival for GBM patients who were positive or negative for IGFBP2 protein expression (IGFBP2+ve/-ve) did not significantly differ (182 days with IGFBP2 expression compared to 186 with no IGFBP2 expression; Fig. 5C). In a small number of GBM/IGFBP2-ve patients («=4), longer survival of over 3 years was observed. In patients who were AIII/IGFBP2+ve, 70% were alive after 5 years, compared to 95% of patients who were AIII/IGFBP2-ve (Fig. 5C). Significantly poorer survival was observed in the OdgIII/IGFBP2+ve gliomas. Approximately 30% of OdgIII/IGFBP2+ve patients were alive after 3 years compared to 75% survival in patients with OdgIII/IGFBP2-ve gliomas (Fig. 5D). No patients with OAIII/IGFBP2+ve gliomas were alive compared to 80% survival in patients with OAIII/IGFBP2-ve gliomas after 2 years (Fig. 5D).

The median survival of patients with AIII/IGFBP2+ve gliomas (1919 days) (Log Rank P=0.018), OdgIII/IGFBP2+ve gliomas (427 days) (Log Rank P=0.218) and OAIII/IGFBP2+ve gliomas (333 days) (Log Rank P=0.890) were better than GBM (315 days) (Fig. 5). An independent study was conducted on a larger sample group of GBM patients that were matched for age but differed significantly in their survival outcome. Of the 13 patients with a recorded survival of greater than 3 years (defined as long term survivors, LTS), 11 (85%) were immunohistochemically negative for IGFBP2. In contrast, 76% (28 out of 37) patients who were matched for age but displayed shorter overall survival (average 1.1 years; defined as short term survivors, STS) showed high expression of IGFBP2 (scores 2+) (Fig. 6). Similar trends were observed when patients were tested for IQGAPl protein expression. Thirty-one percent (4 out of 13) LTS patients were immunohistochemically negative for IQGAPl while 84% (31 out of 37) patients who showed shorter survival were immunohistochemically positive for IQGAPl (scores 3+) (Fig. 6). A significant relationship between absent IGFBP2 and absent IQGAPl in predicting long term survival was observed (R²: 0.4563; p=0.0009).

DISCUSSION Developing reliable prognostic markers for patients diagnosed with high grade gliomas has been challenging, despite overwhelming evidence that wide variability exists in the clinical behaviour, and thus in survival times, of morphologically identical tumours. We searched for biomarkers that could be used as an adjunct to the WHO 2000 grading system to improve prognostic accuracy in glioma. The overexpression of IQGAPl and IGFBP2, at both the gene and protein levels, was highly correlated with gliomas exhibiting necrosis and MVP. These markers were generally associated with significantly shorter adjusted median survival. When either of one of these markers was used in conjunction with the WHO system, both IQGAPl and IGFBP2 readily identified a subgroup of AIII, Odglll, and OAIII whose prognosis was relatively poor. These results suggest that IQGAPl and IGFBP2 are clinically useful prognostic markers and provide additional information to the WHO system.

The multivariate algorithms, GeneRave and SDDA, are designed to find small sets of expressed genes that are biologically meaningful and can act as strong discriminators between two sample groups. This is the first study to apply these algorithms to a brain tumour sample set. By broadly separating the gliomas into two groups using necrosis and MVP, this allowed us sufficient power to perform multivariate analysis. Thirteen genes with known biological function were identified by GeneRave and SDDA. The two most significantly overexpressed genes associated with necrosis and MVP were IQGAPl and IGFBP2. We chose to validate these two genes in a large, semi-independent glioma sample set.

Protein expression of IQGAPl and IGFBP2 was observed in over 80 percent of high grade gliomas and was associated with shorter survival. The application of these markers to the grade III tumours (AIII, Odglll and OAIII) clearly identified a subset of IQGAP 1/IGFBP2 positive patients who had significantly poorer survival outcome. Although AIII tumours separated into distinct survival groups, overall median survival for AIII with IQGAPl and IGFBP2 protein expression was 5 years. There was no significant difference between the median survival of the OAIII gliomas with IQGAPl and IGFBP2 expression from GBM. The presence of an astrocytic element in glioma was associated with significantly shorter survival (Refer to Fig. 4B). However, this finding most likely implies that the astrocytic component of the oligoastrocytomas in our data set could be GBM. IQGAPl and IGFBP2 protein expression could be used to better identify oligoastrocytoma that behave similarly to GBM.

Although not statistically significant, the shorter survival observed in OdgIII was still better than for GBM (Refer to Fig. 5). Loss of heterozygosity (LOH) of chromosomes Ip and 19q, most commonly detected in pure and mixed oligodendroglial tumours, has been associated with better survival (Cairncross et al., 1998). The loss of Ip and 19q was not a frequent occurrence in our sample set. Of the 26 patients with OdgIII in our study, 8 gliomas were identified with lp/19q loss. These gliomas stained both positive («=3) and negative («=5) for IQGAPl and IGFBP2, respectively. The 3 gliomas identified with lp/19q loss, coupled with IQGAP 1/IGFBP2 protein expression, did not show any survival advantage over the other OdgIII gliomas. For the OAIII dataset («=12), LOH data was available for 50% of samples and none of these showed IpI 9q loss. Due to the small number of samples identified with lp/19q loss, the power to detect any survival benefit in the oligodendrogliomas and mixed gliomas from this deletion is too low and its relative impact on the prognostic markers, IQGAPl and IGFBP2 can not be concluded.

As expected, almost 80% of GBMs were IQGAPl positive and over 90% of GBMs were IGFBP2 positive. However, patients diagnosed with a GBM but who were negative for IQGAPl and IGFBP2 showed an improved survival outcome. This is clearly demonstrated in Figures 5A and 5C. Despite the extensive documentation of long term survival (LTS) in small populations of GBM patients, no current histological parameters have been elucidated that could identify such a group. From our data set of 52 GBMs, 3 GBM patients were identified with a recorded survival of greater than 3 years. In two of these patients, no IQGAPl or IGFBP2 protein expression was detected. One of these patients is still alive after 5 years. The prognostic utility of absent IQGAPl and IGFB P2 protein expression in GBM patients was further demonstrated in a larger, independent study of 13 LTS patients and an aged-matched group of 37 STS patients. Long term survival (LTS) was significantly correlated with GBM samples devoid of IGFBP2 and IQGAPl protein expression. This study clearly demonstrates utility for IQGAPl and IGFBP2 to supplement the more traditional diagnostic markers to offer additional prognostic and predictive information.

The association of IQGAPl with poor survival in gliomas has not previously been reported. Our immunohistochemical analysis revealed strong positive cytoplasmic staining for IQGAPl in high grade gliomas, however, in contrast, IQGAPl expression was specific only to the endothelial cell structures in the low grade gliomas (astrocytic and oligodendroglial). This observation is consistent with a recent report describing IQGAPl expression in rat brain and human glioma samples (Balenci et al., 2006). The authors suggested that IQGAPl protein expression was restricted to gliomas of astrocytic origin. However, our study demonstrates that there is a subgroup of pure oligodendrogliomas that express IQGAPl and behave poorly.

IQGAPl is a scaffolding protein that has a multifunctional role in normal tissue. It has been shown to be a target molecule of Cdc42 and Racl small GTPases and negatively regulates the E-cadherin-based cell-cell adhesion. Abrogation of cell-cell adhesion is a key event in the invasive phenotype of many cancers and the cadherin superfamily of adhesion molecules (E-, P- and N-cadherin) have been associated with glioma invasion (Demuth et al., 2004). In addition, IQGAPl plays a role in cellular motility and morphogenesis by interacting directly with cytoskeletal, cell adhesion and signal transduction proteins. Overexpression of IQGAPl in the breast cancer cell line MCF-7 results in significant increases in cell invasive capacity while down regulation of IQGAPl in ovarian cancer cells by IQGAPl -specific small interfering RNAs leads to a loss of migratory ability in these cells (Bourguignon et al., 2005). It seems likely that IQGAPl may also play a significant role in glioma migration and could be involved in the rapid dissemination of glioma cells throughout the brain. There is an increasing body of evidence from mouse models supporting glioma initiation as a result of neural progenitor cell transformation. In a recent study, neoplastic IQGAPl positive cells were isolated from rat glioblastoma and subsequently expanded in culture. These IQGAPl positive cells possessed cancer stem-like progenitor cell characteristics and were highly aggressive.

IGFBP2 protein expression has been shown to be a key signature marker for GBM and there have been numerous studies linking IGFBP2 with poor prognosis (Wang et al., 2003). However, there have been no reports demonstrating the use of IGFBP2 protein expression as a marker of aggressive biological behaviour in WHO grade III gliomas. In addition, this is the first report suggesting a better survival outcome for GBM patients who do not express IGFBP2.

In our study we found IGFBP2 immunostaining to be very patchy in its distribution and associated with the pseudopalisading cells surrounding the necrotic foci and in areas where cellular degeneration was evident in both astrocytic and oligodendroglial tumours. The localisation of IGFBP2 immunostaining to the pseudopalisading cells has previously been reported in whole tissue sections (Godard et al., 2003). Laser capture microscopy has been used to isolate pseudopalisading cells from GBMs and demonstrated upregulation of gene transcripts involved in glycolysis and cell- cycle control in the pseudopalisading cells.. A role in angiogenesis has also been suggested for pseudopalisading cells due to the high expression levels of VEGF as a result of increased transcriptional levels of HIFl -α. In addition, the chemokine receptor, CXCR4, has been found to consistently co-localize with HIF- 1-α expression in pseudopalisading glioma cells surrounding areas of necrosis. The overexpression of IGFBP2 is typically observed in the advanced stages of cancer and it seems plausible that its heightened expression is related to the increasing abundance of necrosis. An immunoprecipitation study showed that IGFBP2 binds to integrin α5 suggesting that IGFBP2 functions to enhance elevated migration rates via an integrin-mediated pathway (Wang et al., 2006). Tissue RNA levels of integrin α5 have been reported to be significantly higher in hypoxic conditions than under normoxic.

In conclusion, we have shown that the protein expression of IQGAPl and IGFBP2 is strongly associated with poor survival in astrocytoma and oligodendroglioma. Although the association of IGFBP2 has previously been described with poor prognosis, the association of IQGAPl is a novel finding. The findings of our study are significant as IQGAPl and IGFBP2 protein expression markers could complement the WHO classification system to permit more precise delineation of the existing tumour grades and identify potentially more aggressive glioma subtypes. Importantly, these markers were able to identify a subset of GBM patients who had shown long term survival of over 3 years after initial diagnosis. At present, there are no markers with the capacity to predict long term survival in such patients. Another clinically useful role of these markers would be to improve accuracy in the prediction of biological grade and aggressive potential, particularly in small biopsies of gliomas that are surgically inoperable. Functional studies of IQGAPl and IGFBP2 and their relationship with each other and relative roles played in glioma biology may lead to a better understanding and may provide potential targets for anti-tumour therapy.

Summary

Clinical treatment decisions and the survival outcomes of patients with gliomas are directly impacted by accurate tumour classification. New and more reliable prognostic markers are needed to better identify the variable duration of survival among histologically defined glioma grades. Microarray expression analysis and immunohistochemistry were used to identify biomarkers associated with gliomas with more aggressive biological behaviours. The protein expression of IQGAPl and IGFBP2, when used in conjunction with the WHO grading system, readily identified and defined a subgroup of grade III glioma patients whose prognosis was poor. In addition, in patients with GBM, where IQGAPl and IGFBP2 were absent, long term survival of over 3 years was observed. The use of these markers confirmed a non-uniform distribution of survival in WHO grade III and FV tumours. Thus IQGAPl and IGFBP2 immunostaining supplements current histological grading by offering additional prognostic and predictive information.

Any description of prior art documents herein, or statements herein derived from or based on those documents, is not an admission that the documents or derived statements are part of the common general knowledge of the relevant art in Australia or elsewhere.

While the invention has been described in the manner and detail as above, it will be appreciated by persons skilled in the art that numerous variations and/or modifications including various omissions, substitutions, and/or changes in form or detail may be made to the invention as shown in the specific embodiments without departing from the spirit or scope of the invention as broadly described. The present embodiments are, therefore, to be considered in all respects as illustrative and not restrictive.

List of References Kleihues, P., Cavenee, WK Pathology and Genetics of tumours of the nervous system. World Health Organization classification of tumours. Lyon: IARC, 2000.

Haven, C, Howell VM, Eilers PH, Dunne R, Takahashi M, van Puijenbroek M,

Furge K, Kievit J, Tan MH, Fleuren GJ, Robinson BG, Delbridge LW, Philips J, Nelson

AE, Krause U, Dralle H, Hoang-Vu C, Gimm O, Morreau H, Marsh DJ, Teh BT. Gene expression of parathyroid tumours: molecular subclassifϊcation and identification of the potential malignant phenotype. Cancer Research, 64: 7405-7411, 2004.

Velazquez-Fernandez, D., Laurell, C, GeIi, J., Hoog, A., Odeberg, J., Kjellman, M., Lundeberg, J., Hamberger, B., Nilsson, P., and Backdahl, M. Expression profiling of adrenocortical neoplasms suggests a molecular signature of malignancy. Surgery, 138: 1087-1094, 2005.

Kiiveri A Bayesian approach to variable selection when the number of variables is very large. Hayward, California, 2003. s Pfaffi, M. W., Horgan, G. W., and Dempfle, L. Relative expression software tool

(REST) for group- wise comparison and statistical analysis of relative expression results in real-time PCR. Nucleic Acids Res, 30: e36, 2002.

Brown, M. D. and Sacks, D. B. IQGAPl in cellular signaling: bridging the GAP. Trends Cell Biol, 16: 242-249, 2006. o Sun, W., Zhang, K., Zhang, X., Lei, W., Xiao, T., Ma, J., Guo, S., Shao, S., Zhang,

H., Liu, Y., Yuan, J., Hu, Z., Ma, Y., Feng, X., Hu, S., Zhou, J., Cheng, S., and Gao, Y. Identification of differentially expressed genes in human lung squamous cell carcinoma using suppression subtractive hybridization. Cancer Lett, 212: 83-93, 2004.

Morris, L. E., Bloom, G. S., Frierson, H. F., Jr., and Powell, S. M. Nucleotides variants within the IQGAPl gene in diffuse-type gastric cancers. Genes Chromosomes Cancer, 42: 280-286, 2005.

Dong, P., Nabeshima, K., Nishimura, N., Kawakami, T., Hachisuga, T., Kawarabayashi, T., and Iwasaki, H. Overexpression and diffuse expression pattern of IQGAPl at invasion fronts are independent prognostic parameters in ovarian carcinomas.0 Cancer Lett, 2005

Cairncross JG, Ueki K, Zlatescu MC, et al. Specific genetic predictors of chemotherapeutic response and survival in patients with anaplastic oligodendrogliomas. J Natl Cancer Inst 1998;90: 1473-79.

Godard S, Getz G, Delorenzi M, et al. Classification of human astrocytic gliomas₅ on the basis of gene expression: a correlated group of genes with angiogenic activity emerges as a strong predictor of subtypes. Cancer Res 2003 ;63:6613-25.

Smyth GK. Linear models and empirical Bayes methods for assessing differential expression in microarray experiments. Stat App Gen MoI Biol 2004;3:3.

Balenci L, Clarke ID, Dirks PB, et al. IQGAPl protein specifies amplifying cancer0 cells in glioblastoma multiforme. Cancer Research 2006;66:9074-83

.Demuth T, Berens, M. Molecular mechanisms of glioma cell migration and invasion. J Neurooncol 2004;70:217-28.

Bourguignon LY, Gilad E, Rothman K, et al. Hyaluronan-CD44 interaction with IQGAPl promotes Cdc42 and ERK signaling, leading to actin binding, EIk-I /estrogen receptor transcriptional activation, and ovarian cancer progression. J Biol Chem 2005;280: 11961-72.

Wang H, Wang H, Shen W, et al. Insulin-like growth factor binding protein 2 enhances glioblastoma invasion by activating invasion-enhancing genes. Cancer Res 2003 ;63:4315-21.

Wang G, Hu L, Fuller GN, et al. An interaction between insulin-like growth factor binding protein 2 and integrin alpha 5 is essential for IGFBP2-induced cell mobility. J. Biol Chem 2006;281 :14085-91.

Stupp R, Mason WP, van den Bent MJ, Weller M, Fisher B, Taphoorn MJ, et al. Radiotherapy plus concomitant and adjuvant temozolomide for glioblastoma. N Engl J Med 2005;352:987-996

Claims

1. A method of diagnosing a tumour of the central nervous system (CNS) in an individual comprising: a) determining the expression of at least one gene selected from the group consisting of IQGAPl, Homer 1, and ClQLl or determining the expression of

5 at least two genes selected from the group consisting of IQGAPl, Homer 1, IGFBP2, and ClQLl in a biological sample from said individual; and b) comparing said expression with that of at least one reference sample; wherein a difference in said expression is diagnostic of a tumour of the CNS.

2. The method according to claim 1 , wherein the tumour is a brain tumour.

I₀ 3. The method according to claim 1 wherein the tumour is selected from the group consisting of glioblastoma multiforme, astrocytomas, oligodendrogliomas, ependymomas and mixed gliomas.

4. The method according to claim 1, wherein the method comprises one or more of (i) discriminating high grade tumours and low grade tumours; (ii) predicting is responsiveness of the individual to therapy; (iii) predicting survival or prognosis of the individual; and (iv) grading a tumour.

5. The method according to claim 1, wherein the method comprises determining the level of expression of one or more gene sets selected from the group consisting of (i) Homerl and IQGAPl; (ii) IGFBP2 and ClQLl; (iii) IQGAPl and IGFBP2; (iv) IGFBP2

20 and Homerl; and (v) IQGAPl and ClQLl.

6. The method according to claim 1, further comprising determining the expression of at least one gene selected from the group consisting of LGALSl, LRRC20, KIAA0599, KPNA5, NFYB, CARHSPl, COPZ2, ARS, HS7SLP, CH13L1, HSxS138, SERPINA3, SPPl, and RBPl.

₂₅ 7. The method according to claim 1, wherein absence of expression of IQGAPl and IGFBP2 is diagnostic of long term survival in an individual having glioblastoma multiforme.

8. The method according to claim 1, wherein the biological sample is brain tissue.

30 9. The method according to claim 1, wherein the reference sample is selected from the group consisting of (i) brain tissue obtained from normal tissue of the subject individual; (ii) normal brain tissue obtained from a similar or identical region of the brain of a second individual; (iii) whole brain nucleic acid preparation; (iv) brain tissue type specific nucleic acid preparation.

10. The method according to claim 9, wherein the nucleic acid preparation is selected from the group consisting of total RNA, cDNA, and polyA⁺ RNA.

11. The method according to claim 1 , wherein determining the expression of a gene comprises determining the level of one or more of a nucleic acid sequence or fragment thereof corresponding to said gene or of a polypeptide or fragment thereof encoded by said gene.

12. The method according to claim 1 wherein determining expression comprises contacting said sample with at least one antibody specific to a polypeptide encoded by said gene or a fragment thereof.

13. A method of treating a tumour of the CNS in an individual comprising a) diagnosing a tumour of the central nervous system (CNS) in said individual by a method comprising determining the expression of at least one gene selected from the group consisting of IQGAPl, Homer 1, and ClQLl or determining the expression of at least two genes selected from the group consisting of IQGAPl, Homerl, IGFBP2, and ClQLl in a biological sample from said individual, and comparing said expression with that of a reference sample; wherein a difference in said expression is diagnostic of a tumour of the CNS; and b) formulating a therapeutic regime suitable for the treatment of an individual having said diagnosed tumour; and c) administering said therapeutic regime to said individual.

14. A kit for use in diagnosing a tumour of the CNS, the kit comprising at least one probe specific for a gene or gene product selected from the group consisting of IQGAPl, Homerl, and ClQLl.

15. The kit according to claim 14 further comprising at least one probe specific for IGFBP2 or a gene product thereof.

16. The kit according to claim 14 further comprising at least one probe specific for a gene or gene product selected from the group consisting of LGALS 1 , LRRC20, KIAA0599, KPNA5, NFYB, CARHSPl, COPZ2, ARS, HS7SLP, CHl 3Ll, HSxS 138, SERPIN A3, SPPl, and RBPl.

17. The kit according to claim 14, wherein the probe is selected from the group consisting of a nucleic acid and an antibody.

18. The kit according to claim 14, further comprising one or more additional components selected from the group consisting of (i) one or more reference probe(s); (ii) one or more detection reagent(s); (iii) one or more agent(s) for immobilising a polypeptide on a solid support; (iv) a solid support material; (v) instructions for use of the kit or a component(s) thereof in a method for diagnosing a tumour of the CNS.

19. The kit according to claim 14, comprising one or more probe(s) immobilised on a solid support, such as a biochip.

20. A biochip comprising at least one probe specific for a gene or gene product selected from the group consisting of IQGAPl, Homer 1, and ClQLl.

21. The biochip according to claim 20 further comprising at least one probe specific for IGFBP2 or a gene product thereof.

22. The biochip according to claim 20 further comprising at least one probe specific for a gene or gene product selected from the group consisting of LGALSl, LRRC20, KIAA0599, KPNA5, NFYB, CARHSPl, COPZ2, ARS, HS7SLP, CHl 3Ll, HSxS138, SERPINA3, SPPl, and RBPl.

23. A method for the treatment of a tumour of the CNS in a subject wherein said tumour is characterised by elevated expression of IQGAPl, the method comprising administering a therapeutically effective amount of an agent selected from the group consisting of (i) an inhibitor of expression of IQGAPl; and (ii) an antagonist of a biological activity of a polypeptide encoded by IQGAP 1.

24. A method for the treatment of a tumour of the CNS in a subject wherein said tumour is characterised by decreased expression of Homerl, the method comprising administering a therapeutically effective amount of an agent selected from the group consisting of (i) a stimulant of expression of Homerl ; and (ii) a stimulant of a biological activity of a polypeptide encoded by Homerl .

25. A method for the treatment of a tumour of the CNS in a subject wherein said tumour is characterised by elevated expression of ClQLl, the method comprising administering a therapeutically effective amount of an agent selected from the group consisting of (i) an inhibitor of expression of ClQLl; and (ii) an antagonist of a biological activity of a polypeptide encoded by C 1 QL 1.

26. A method for screening for an agent capable of modulating the expression of one or more genes selected from the group consisting of IQGAPl, Homerl, IGFBP2 and ClQLl, the method comprising exposing a cell to a candidate agent and comparing the expression of the gene in the presence of the agent to the expression of the gene in the absence of the agent, whereby a difference in expression of the gene in the presence compared to the absence of the agent is indicative of an agent capable of modulating the expression of the gene.

27. The method according to claim 26, wherein the agent is an antisense or siRNA.

28. A method for screening for an agent capable of modulating a biological activity of a polypeptide, or fragment thereof, encoded by a gene selected from the group consisting of IQGAPl, Homer 1, IGFBP2 and ClQLl, the method comprising exposing said polypeptide or said fragment to a candidate agent under conditions suitable for the expression of the biological activity and comparing the biological activity of the polypeptide or fragment in the presence of the agent to the biological activity in the absence of the agent, whereby a difference in biological activity of the polypeptide or fragment in the presence compared to the absence of the agent is indicative of an agent capable of modulating a biological activity of the polypeptide or fragment.

29. The method according to claim 28, wherein the agent is an antagonist of a biological activity.

30. The method according to claim 28, wherein the agent is a monoclonal antibody.

TABLE 1. Patient Characteristics and Histopathologic Features

multiforme, Odgll, grade III, MVP,

TABLE 2. Genes identified by GeneRave and SDDA programs to be significant discriminators between high grade and low grade gliomas.

TABLE 3. Protein expression scores of IQGAPl in gliomas classified as low grade or high grade.

IQGAP1 staining *« Low grade gliomas ^<" ^m ' ■>" "iHigh grade gliomas?^*1- *^*

, _m score_ft i • (ή≡'73) '<^■ ft

% %

0 9 2

1 21 4

2 43 8

13 11

14 75 4

X ² test: 64.31 p<0.000

TABLE 4. Protein expression scores of IGFBP2 in gliomas classified as low grade or high grade.

IGFBP2 staining low grade gliomas ,Hϊfv f score " -% (n = 7θfζ^f ^^{" }}

% %

0 61 9

1 23 3

2 16 14

3 0 74

X ² test: 38.98 p<0.000

TABLE 5. Protein expression of IQGAPl and IGFBP2 in 143 gliomas according to their WHO classification

%, ,

Abbreviations All, astrocytoma grade II, AMI, astrocytoma grade III, GBM, glioblastoma multiforme, Odgll, oligodendroglioma grade II, Odglll, oligodendroglioma grade III, OAIII oligoastrocytoma grade III

TABLE 6. Multivariate Cox regression analysis of IQGAPl and IGFBP2 cytoplasmic expression (based on maximum scores of 4 and 3 only), and age.

Variables Relative 95% Cl P-value Risk

Age (40-49 years) 0 958 0 33-2 81 0 940 Age (50-59 years) 2 288 0 88-5 92 0 088

Age (60+ years) 3 537 » , 1 50-8 36 0 004 IQGAP1 (score≥3) 1 311 0 95-1 80 0 095 IGFBP2 (score≥2)~ ^" WiT^" _ 1 28-2 29 0^"0OO

Abbreviations Cl, confidence interval Table 7