WO2003025217A1 - Prognostic method - Google Patents

Abstract

A method of prognosis of a cancer patient comprising subjecting a sample of primary tumour tissue of defined type and grade from the patient to quantitative analysis of KAI1 gene expression and determining whether KAI1 expression level is elevated or reduced relative to a threshold value for the defined tumour tissue type and grade; wherein elevated expression is indicative of restraint of future tumour progression and reduced expression is indicative of future tumour metastasis. A method of detection of tumour presence within tissue of a patient comprising subjecting a sample of tissue of defined type to quantitative analysis of KAI1 gene expression and determining whether KAI1 expression level is elevated or reduced relative to a threshold value for tissue of the defined type; wherein elevated expression is indicative of tumour presence within the tissue.

Description

PROGNOSTIC METHOD

FIELD OF THE INVENTION

The present invention relates to methods of cancer prognosis and detection, and in particular, but not exclusively, to methods of prostate cancer prognosis and detection. The invention also relates to kits for use in such methods.

BACKGROUND OF THE INVENTION

Prostate cancer is a leading cause of male morbidity and mortality. Despite advances in diagnostic techniques and screening allowing for the earlier detection of prostate cancer, current methods for predicting individual patient outcome that utilise standard histological staging are generally unreliable as guides to optimal therapeutic management. Some prognostic indication is possible in men with clinically localised disease by correlation of histological grade and levels of prostate-specific antigen (PSA). In this context the Gleason system of histological grading is generally adopted, particularly for patients with well and poorly differentiated cancers (29-32). Unfortunately, more than 75% of patients with clinically localised prostate cancer have moderately differentiated tumours (33) in relation to which the PSA/grade correlation is not appropriate. Even when combined with other prognostic techniques (diagnostic biopsy, Gleason grade and digital rectal examination) PSA levels cannot be relied upon to predict the extent and outcome of cancer in individual patients.

A leading cause of treatment failure and mortality associated with prostate cancer is metastatic disease. Understanding the molecular basis of prostate cancer progression and metastasis should therefore allow for improved prognostication and the design of more selective therapeutic strategies, which may be adopted at an early stage to more effectively control disease progression. An important advance in our understanding of prostate cancer metastasis has been the cloning of the prostate cancer metastasis suppressor gene KAIl from human chromosome l lpll.2 (1). Sequencing of the KAIl cDNA clone revealed complete identity to the human lymphocyte surface antigen CD82, variously designated as R2, 14A, 4F9 or C33 in several laboratories, using different monoclonal antibodies (2-5). The predicted protein encoded by KAI1/CD82 has been characterised as a member of the transmembrane 4 (TM4) family, a recently described group of structurally related cell surface proteins characterised by four highly conserved membrane spanning domains (6). The biological function of the TM4 family is unclear, although various data suggest a role in cell growth, adhesion and motility (6); consistent with these functions, other TM4 family molecules have been implicated in tumour progression (7) and in the negative regulation of tumour cell metastasis (8).

Current evidence suggests that KAIl gene expression may serve as a marker of metastatic human prostate cancer. Firstly, Dong et al (1) showed by Northern analysis that KAIl RNA expression is high in normal human prostate tissue, but is reduced in human cell lines derived from prostate cancer metastases. Furthermore, the study demonstrated that transfection of highly metastatic rat prostate cancer cells with KAIl suppresses the ability of the cells to form lung metastases in nude mice.

More recently, studies using the C33 monoclonal antibody for the immunohistochemical detection of KAIl protein demonstrated downregulation of KAIl protein in primary and metastatic prostate cancer. Dong et al (9) found reduced expression of KAIl protein in 70% of 49 untreated primary prostate cancers, 90% of 15 metastatic prostate cancers relapsed from androgen ablation therapy and in all of the lymph node metastases studied. This work formed the basis for international patent publication WO 96/34117. Using an identical immunohistochemical protocol, Ueda et al (10) found that KAIl protein expression in primary prostate cancer is inversely correlated with Gleason pattern and clinical stage. Molecular analysis has failed to detect KAIl gene mutations or allelic loss in primary and metastatic prostate cancers suggesting that downregulation of KAIl protein is not commonly due to either of these mechanisms (9). It is generally accepted in most solid cancers that KAIl expression decreases as the cancer progresses and this has been shown in non-small cell lung (34), breast (35), hepatocellular (36) and bladder (37) cancers and invasive squamous cell carcinomas from the lung, head, neck and cervix (38).

Earlier work from the laboratory of the present inventors suggest a biphasic pattern of KAIl mRNA expression in primary human prostate cancers dependent upon histological grade, when compared to non-malignant prostate tissue (39). That is, there was an elevation in low grade primary tumours and reduction in high grade primary tumours.

The present inventors have now shown that not only do levels of KAIl expression vary inversely as a function of histological grade, but that within tumour grades for specific tissue types there is variation of KAIl expression. Furthermore, the inventors have determined that this variation of KAIl expression within a tumour of defined grade and tissue type is predictive of tumour metastasis. This understanding gives rise to far greater prognostic accuracy than was previously available and will allow clinicians to better manage individual patient therapies so that appropriate interventions may be taken when required, and more often avoided when not required. Furthermore, the present inventors have identified elevated KAIl expression in normal tissue surrounding cancerous lesions. Therefore, even if biopsy tissue samples are negative for cancer, the determination of increased KAIl expression in normal tissues surrounding the tumour (which may in fact not be able at that stage to be detected using conventional techniques) will offer a means of verifying negative biopsy results, to thereby eliminate or at least reduce false negative results. The diagnostic and prognostic methods according to the present invention therefore provide powerful tools for clinicians in tumour detection, control of tumour metastasis and maintaining patient quality of life.

SUMMARY OF THE INVENTION

According to one embodiment of the present invention there is provided a method of prognosis of a cancer patient comprising subjecting a sample of primary tumour tissue of defined type and grade from the patient to quantitative analysis of KAIl gene expression and determining whether KAIl expression level is elevated or reduced relative to a threshold value for the defined tumour tissue type and grade; wherein elevated expression is indicative of restraint of future tumour progression and reduced expression is indicative of future tumour metastasis.

The threshold value may be determined by statistical analysis of KAIl gene expression levels in respect of samples of primary tumour tissue of defined type and grade previously taken from patients in relation to whom the status of tumour progression is known.

According to another embodiment of the present invention there is provided a method of prognosis of a cancer patient comprising subjecting a sample of primary tumour tissue of defined type and grade from the patient to quantitative analysis of KAIl gene expression and subjecting a sample of normal tissue of the defined type and from the same patient, that is not derived from adjacent to the tumour, to quantitative analysis of KAIl gene expression, determining the quotient of KAIl gene expression from the tumour tissue and KAI gene expression from the normal tissue and determining whether the quotient is elevated or reduced relative to a quotient threshold value for the defined tumour tissue type and grade; wherein an elevated quotient is indicative of restraint of future tumour progression and a reduced quotient is indicative of future tumour metastasis.

According to another embodiment of the present invention there is provided a method of prognosis of a cancer patient comprising subjecting a sample of normal tissue derived from adjacent to tumour tissue of the same type, to quantitative analysis of KAIl gene expression and determining whether KAIl expression level is elevated or reduced relative to a threshold value for tissue of that type located adjacent to tumour tissue; wherein elevated expression is indicative of restraint of future tumour progression and reduced expression is indicative of future tumour metastasis.

According to a further embodiment of the present invention there is provided a method of detection of tumour presence within tissue of a patient comprising subjecting a sample of tissue of defined type to quantitative analysis of KAIl gene expression and determining whether KAIl expression level is elevated or reduced relative to a threshold value for tissue of the defined type; wherein elevated expression is indicative of tumour presence within the tissue.

The samples of tumour and normal tissue may be taken by biopsy or from biological fluids obtained from the patient. In preferred embodiments of the invention the biological fluids are urine, blood, lymph, saliva, tears or semen.

In a preferred embodiment the tumour tissue is prostate, breast, lung, colon, liver, brain, kidney, trachea, rectum, stomach, pancreas, cervix, uterus, ovary, testicle, thyroid or epidermal tissue. Preferably the tumour tissue is prostate, breast or lung tissue, particularly preferably it is prostate tissue.

In one embodiment of the invention quantitative analysis of KAIl gene expression is conducted by determining whether KAIl protein level is elevated or reduced relative to a threshold value for KAIl protein level. In one embodiment an ELISA assay is conducted to determine whether KAIl protein levels are elevated or reduced. In another embodiment immunohistochemical analysis is conducted to determine whether KAIl protein levels are elevated or reduced.

In another preferred embodiment determination of whether KAIl protein level is elevated or reduced is by conducting an assay wherein labelled KAIl antibody is bound to a support and is contacted with protein from the tumour tissue, wherein following binding of KAIl protein to the labelled KAIl antibody and detection of the label, a quantitative measure of KAIl expression is obtained. Preferably the label is radioactive, fluorescent or colourimetric.

In another embodiment of the invention quantitative analysis of KAIl gene expression is conducted by determining whether KAIl mRNA level is elevated or reduced relative to a threshold value for KAIl mRNA level. RT-PCR analysis may be conducted to determine whether KAIl mRNA levels are elevated or reduced. In one embodiment of the invention RNA from tumour tissue of defined type and grade is subjected to RT-PCR amplification using a primer or primers complementary to at least a fragment of KAIl cDNA; the amplified cDNAs are then separately hybridized to a labelled probe complementary to at least a fragment of KAIl cDNA; the hybridized labels are detected to give a quantitative measure of KAIl mRNA in the tumour tissue which is compared to a threshold value for the defined tumour tissue type and grade. Preferably the label is radioactive, fluorescent or colourimetric.

In another embodiment of the invention there is provided a kit for use in a method of prognosis as outlined above. Such a kit may for example include ELISA, immunohistochemistry or RT-PCR compounds and reagents, such as antibodies, oligonucleotide sequences, solid supports, labels, enzymes, buffer solutions and the like. The kit may additionally include appropriate instructions for conducting the prognostic methods according to the invention.

BRIEF DESCRIPTION OF THE FIGURES

The invention will be further described, by way of example, with reference to the following figures:

Figure 1. Immunohistochemical detection of KAIl in BPH not associated with cancer, BPH associated with prostate cancer and well, moderately and poorly differentiated prostate cancers. Each specimen is shown in pairs: panel a) represents the specimen incubated with primary anti-human KAIl antibody, panel b) represents the same specimen tested for non-specific staining (i.e. control slide). Panels la) and b) well differentiated prostate cancer; Panels 2a) and b) moderately differentiated prostate cancer; Panels 3 a) and b) poorly differentiated prostate cancer; Panels 4a) and b) BPH not associated with cancer and Panels 5 a) and b) BPH associated with prostate cancer. Magnification x400 for all panels. Figure 2. KAIl staining scores for BPH not associated with cancer and well, moderately and poorly differentiated prostate cancers. The score is the product of two parameters: 'Stain density multiplied by scan area' and 'Ratio of target area to scan area', as measured using the MCID system (see Methods section). The solid line represents the mean score within each specimen category. The broken line represents the 95 % upper score limit for BPH not associated with cancer.

DETAILED DESCRIPTION OF THE INVENTION

Throughout this specification and the claims which follow, unless the context requires otherwise, the word "comprise", and variations such as "comprises" and "comprising", will be understood to imply the inclusion of a stated integer or step or group of integers or steps but not the exclusion of any other integer or step or group of integers or steps.

The reference to any prior art in this specification is not, and should not be taken as, an acknowledgement or any form of suggestion that that prior art forms part of the common general knowledge in Australia.

The present invention relates to a method of prognosis of a cancer patient. What is meant by this is that by following the method according to the invention it is possible to determine the likelihood of a primary tumour metastasising to form secondary tumours elsewhere within the patient's body. This is obviously of great importance in formulating a treatment plan that will serve to prolong the patient's length and quality of life.

The invention also provides a method for detection of a tumour in a tissue, particularly in the instance where this tumour is at a size such that it cannot be readily detected using conventional techniques. This detection technique may conveniently be utilised in conjunction with other conventional techniques, such as biopsy. While the inventors have noted that for primary tumour tissue of defined type and grade the level of KAIl gene expression (when compared to a threshold value) is predictive of metastasis, and that the same can be said for normal tissue adjacent to tumour tissue, it is also the case that generally high KAIl expression levels (relative to a threshold value for a particular tissue) are predictive for the presence of tumour. Generally the analytical steps associated with the detection method correspond to those of the prognostic methods that are more specifically discussed herein.

In conducting the prognosis method of the invention it is necessary for a sample of primary tumour tissue from the patient to be obtained. In another embodiment the prognostic method may be conducted by taking a sample of normal tissue of the same type as the tumour tissue that is adjacent to the tumour tissue. Such samples may be obtained by way of regular biopsy or operative techniques that are well known in the surgical field, and will of course vary depending upon the location of the tumour and the organ concerned. It may in fact be the case that the tissue sample is taken from the patient some time before conducting the prognostic method and that it is then stored in an appropriate manner until the prognostic method is conducted. The methods according to the present invention can be applied to a wide range of human tumours, such as for example prostate, breast, lung, colon, liver, brain, kidney, trachea, rectum, stomach, pancreas, cervix, uterus, ovary, testicle, thyroid or epidermal tumours. Accordingly, the primary tissue sample can be drawn from tumours within any of the tissues mentioned. In preferred embodiments of the invention the tissue sample is taken from prostate, breast or lung tissue, most preferably from the prostate. It is also possible for the sample of tumour tissue to be obtained from biological fluids including blood, lymph, saliva, tears and semen, in the case where cells from the defined tissue are likely to be present in a biological fluid. For the purpose of threshold determination normal tissue may be obtained in the same manner as the tumour tissue, and should be obtained from tissue of the same type which is of course not affected by a tumour and which is not directly adjacent to tumour.

As explained above it is possible to predict the likely progression of a primary tumour based upon the level of expression within tumour cells of defined type and grade (or indeed within adjacent normal tissue of the same type) of the KAIl gene (that is, the level of production of KAIl protein) in comparison to a threshold value. The level of expression can be quantitated either directly or alternatively indirectly by, for example, detection of levels of KAIl mRNA, within the tissue concerned. There are a broad range of means by which such quantitative analysis can be conducted, but by way of example it is mentioned that the analysis can be conducted by way of immunohistochemistry analysis, an ELISA assay or other assays involving antibodies or other compounds that will bind the KAIl protein or cDNA or even nucleotide sequences that are complementary to at least a fragment of KAIl cDNA and will therefore bind to this cDNA. The mention of these types of quantitative analysis methods is not intended to be limiting upon the scope of the invention, and in its broadest sense the quantitative analysis methods envisaged are simply limited to techniques that will allow a determination of the level of KAIl gene expression.

The quantitative analysis may also be conducted by contacting an extract obtained from the tumour or normal tissue sample (preferably an extract of isolated protein) against a solid support to which KAIl antibody has been applied. Solid supports such as beads, tubes or polymer particles are well known in the chromatography art. After binding of the protein to the antibody there are then many means of obtaining a signal through which the level of binding of protein to antibody can be measured as a means of assaying the level of protein expressed within the tumour or normal tissue cells. For example there may be a label attached to the antibody such as a radioactive, fluorescent or colourimetric label which can be arranged so that the signal emitted by the label is only detected in the situation where protein and antibody are bound together. It is also possible to adopt the use of indicator molecules that will emit a signal (for example such as colour, fluorescence, etc.) when they come into contact with protein that is bound to antibody. Similar detection means can be adopted in relation to assays that quantify relative amounts of mRNA in tumour and normal tissue of the same type. Such techniques may involve the isolation of RNA from samples of tumour or normal tissue and amplification by RT-PCR to produce a KAIl cDNA sequence or fragment thereof using appropriate oligonucleotide primers. Such techniques may involve the hybridisation of the KAIl cDNA or fragment to a labelled probe nucleotide sequence that is complementary to at least a fragment of the KAIl cDNA. The hybridized labelled probe can then be detected and quantified. Preferably probes utilised in such techniques will be at least five nucleotides in length, particularly preferably at least ten nucleotides in length, still more preferably at leat 20 nucleotides in length, in order to substantially prevent non-specific binding of the probe to the KAIl cDNA sequence.

Threshold value may, for example, be determined if raw KAIl gene expression levels are being monitored, by use of the arithmetic mean and/or standard deviation of KAIl gene expression levels from samples of tumour tissue of the same type and grade. In the case of the more powerful quotient threshold, a quotient will be calculated from KAIl gene expression of tumour against normal tissue of the same type. KAIl expression or quotient can then be correlated against the appropriate threshold value and preferably statistical analysis, such as a Chi² test, will be conducted to determine significant variation above or below threshold value.

The threshold value determined in relation to the tumour tissue type, which is indicative of tumour progression, will be specific for defined tumour grades. By this it is intended to mean that the primary tumour tissue under analysis is to be categorised in a consistent manner to the population samples of tumour tissue from which the threshold value has been determined. For example, tumour tissue may be graded into the categories of well differentiated, moderately differentiated and poorly differentiated, as will be well understood by persons skilled in the art. Preferably, however, a more powerful grading approach will be adopted, such as for example in the case of prostatic tumours, the Gleason classification system (29). Other well recognised grading systems may be adopted for tumours of other tissue types. Numerous possible means of quantifying KAIl gene expression levels are available and would be well known to persons skilled in the art. Examples of such techniques include Northern analysis, Western analysis, immuno precipitation using a labelled probe and immunohistochemical quantification (immuno- image analysis).

In one embodiment of the invention the quantitated KAIl expression level for the tumour tissue of defined type and grade (or from adjacent normal tissue of the same type) will be divided by the quantitated KAIl expression level obtained for normal tissue of the same type that is located remote from the tumour, preferably from the same patient, to produce a quotient. Whether or not a raw KAIl expression level or a quotient is quantified for the primary tumour tissue of defined grade and type, this value will be compared to a threshold value calculated by statistical analysis from samples of the same primary tumour tissue grade and type, which has been demonstrated to be indicative of tumour progression. This demonstration of tumour progression will be achieved by reviewing KAIl levels from samples obtained from patients whose tumour progression status is known. That is, KAIl levels may be plotted in respect of patients suffering from tumours of known type and grade who have been shown over the course of time to have had either metastatic or non- metastatic disease. Similarly, a separate "quotient threshold" can be determined from patient populations by statistically analysing KAIl expression levels from samples obtained from patients known to have had metastatic or non-metastatic disease and dividing this by KAIl expression levels for normal tissue of the same type surrounding (but not directly adjacent to) tumour in the same patient.

Within this specification by the term "restraint of future tumour progression" it is intended to mean that as a result of an elevated quotient or elevated expression of the KAIl gene within the tumour tissue concerned (or in adjacent normal tissue of the same type), relative to the appropriate threshold, it is unlikely for metastasis of the tumour to subsequently occur. In contrast the phrase "indicative of future tumour metastasis" indicates that due to a reduced quotient or reduced expression of the KAIl gene within the tumour tissue concerned (or adjacent normal tissue of the same type), relative to the appropriate threshold, it is likely that the primary tumour will metastasise to form secondary tumours in other parts of the body.

Antibodies useful in methods according to the invention can be produced according to standard techniques known in the art, as for example explained in "Immunochemical methods in cell and molecular biology", R J Mayer and J H Walker, Academic Press, 24- 28 Oval Road, London, UK, 1987, the teaching of which is included herein by way of reference. Antibodies concerned may be monoclonal or polyclonal, preferably monoclonal. The invention will now be further described with reference to the following non-limiting examples.

EXAMPLE 1 Materials and methods for KAIl mRNA determination in prostate cancer

Specimens

Total RNA from the human cancer cell lines LNCaP, DU-145, and PC-3, all derived from human prostate cancer metastases, was kindly provided by Dr W. Tilley (Flinders Medical Centre, South Australia). Archived formalin-fixed, paraffin-embedded specimens of benign prostatic hyperplasia (n=5) and primary prostatic carcinoma (n=35) were retrieved from files at the Department of Anatomical Pathology of the Austin and Repatriation Medical Center. All specimens were obtained surgically from previously untreated patients with localized disease who underwent resection of the prostate at the Austin and Repatriation Medical Centre between 1993 and 1995. Specimens were fixed in 10 per cent buffered formalin and embedded in paraffin wax. All of the original haematoxylin and eosin (H&E)-stained sections made from these blocks were reviewed by a senior pathologist and specimens selected for study. Tumour grade was classified according to the Gleason system (11). The specimens were classified as low grade (Gleason score 2-6) and high grade (Gleason score 7-10). Fresh normal human prostate, confirmed histologically, was also obtained at autopsy within 24 h of death and stored at -70°C until ready for use. Foci of prostatic carcinoma were identified by light microscopy and dissected from archived specimens using a sterile scalpel blade and re-embedded in paraffin wax. A 5 μm section from each dissected specimen contiguous to those used for RNA extraction was stained with H&E and the diagnosis was confirmed histologically. The same dissection procedure was used to isolate comparably sized areas of benign prostatic hyperplasia. Polymerase chain reaction and Southern blotting

Total RNA from fresh normal prostate tissue was extracted using the acid-guanidinium thiocyanate method described by Chomczynski and Sacchi (12). For archived formalin- fixed, paraffin-embedded prostate tissue, 40 x 5 μm serial sections containing the microdissected tissue were cut from each paraffin block with a microtome and collected into 50 ml polypropylene conical tubes (Greiner). Tissue sections were deparaffinized by two 10-min washes in 1.5 ml of xylene at room temperature and then pelleted by centrifugation at 13000g for 5 min. The tissues were rehydrated and xylene was removed by washing twice with 1.5ml of 100 per cent ethanol at room temperature and centrifugation at 13000g for 5 min. Total cellular RNA was isolated by homogenizing the precipitate in 500 μl of 4 M guanidinium thiocyanate, 25 mM sodium citrate (pH 7.0), 0.1 M -mercaptoethanol, and 0.05 per cent N-lauroylsarcosine. 0.1 volume of 2 M sodium acetate (pH 4.0), 1 volume of water-saturated phenol, and 0.4 volume of chloroform- isoamyl alcohol (in a ratio of 24:1) were then added sequentially and the suspension was centrifuged at lOOOOg for 30 min at 4°C. The upper aqueous phase was transferred to a fresh 1.5 ml micro fuge tube, mixed with an equal volume of isopropanol in the presence of 20 μg of glycogen (Boehringer Mannheim, Victoria, Australia) as a carrier and placed at -70°C for at least 30 min to precipitate the RNA. After centrifugation at 13000g for 30 min at 4°C, the pelleted RNA was redissolved in 300 μl of denaturing solution and reprecipitated with an equal volume of isopropanol for 30 min at -70°C. The RNA was centrifuged again, resuspended in 75 per cent ethanol, and incubated at room temperature for 10 min. Following a final centrifugation, the pelleted RNA was vacuum-dried for 10 min, dissolved in 10 μl of DEPC-treated water, and then stored at -70°C until ready for further use. Due to the small amounts of archived tissue used for RNA extraction, direct RNA qualification was not possible, although microdissected tissue section sizes were comparable and treated identically for RNA extraction.

Extracted RNA (100 ng from frozen normal prostate and 3 μl from formalin-fixed, paraffin-embedded tissue) was diluted to a total volume of 11 μl with DEPC-treated water, heated to 70°C for 2 min, and then chilled on ice. The reaction mixture was then brought to a final volume of 20 μl containing 40 U of rRNasin (Promega, Victoria, Australia), 10 μM random hexamers, 1 mM dNTPs (dGTP, dATP, dTTP, and dCTP) (Clontech, NSW, Australia), 200 U of Moloney murine leukaemia virus reverse transcriptase (M- MuLV, Promega, Victoria, Australia), 25 mM Tris-HCl 37.5 mM KC1, 1.5 mM MgC12, and 5 mM dithiothreitol (pH 8.3) and incubated at 42°C for 1 h, followed by 5 min at 70°C to inactivate the reverse transcriptase.

The reverse transcription product (cDNA) was then amplified by PCR using two different oligonucleotide primer pairs designed to detect the KAIl gene and that of the housekeeping gene β-actin. PCR amplification of the cytoplasmic /3-actin gene was used to assess cDNA integrity and as a reference to allow semi-quantitation of KAIl mRNA. The first KAIl primer pair [KAIl-(l)] used forward

(5'GTACTGCATGGAGAAGGTGCAGGG-3' SEQ LD NO:l) and reverse (5ΑAGCAGATGGACAGGACCATCCCC-3' SEQ LD NO:2) primers to amplify a 112 bp fragment (base positions 809-920). The second KAIl primer pair [KAIl-(2)] used forward (5'GTACAATCGCCCTGAGCTCACCTA-3' SEQ ID NO:3) and reverse (5'- GCCTGCACCTTCTCCATGCAGCCC-3' SEQ ID NO:4) primers to amplify a 174 bp fragment (base positions 657-830). For amplification of jS-actin, the forward primer was 5'TGGCATTGCCGACAGGATGCAGAA-3' SEQ ID NO:5 and the reverse was 5'- CTCGTCATACTCCTGCTTGCTGAT-3 ' SEQ LD NO:6, designed to span intron "E" (172 bp for cDNA size). For each PCR, cDNA was amplified in a total volume of 20 μl containing the cDNA was amplified in a total volume of 20 μl containing the cDNA preparation (1 μl for frozen tissue and 3 μl for formalin-fixed, paraffin-embedded tissue); 200 μM of each dGTP, dATP, dTTP, and dCTP (Clontech, NSW, Australia); 0.5 μM of each of the appropriate forward and reverse primers for KAIl or B-actin; 1.5 mM MgC12, and 0.5 unit of "Red Hot" Taq polymerase (Advanced Biotechnologies, Victoria, Australia). For cDNA prepared from fresh tissue, amplification was performed using 25 cycles of PCR, whereas 40 cycles were required to detect PCR products from archived tissue. A modification of a high stringency PCR protocol adapted for the specific detection of low-level mRNA, as previously described in our laboratory (13) was used. Briefly, the reaction mixture was initially heated to 94°C for 3 min, followed by 5 cycles at 94°C for 30 s, 70°C for 30 s, and 74°C for 1 min, followed by 25-40 cycles at 94°C for 30 s, 55°C for 30 s, and 74°C for 1 min, and a final extension time of 10 min at 74°C. Negative control reactions lacking cDNA were included to monitor contamination and preparations of fresh normal prostate tissue were amplified as a positive reaction control. After amplification, 10 μl of each PCR product was combined with 1 μl of loading dye (6 x, being 0.25 per cent bromophenol blue, 0.25 per cent xylene cyanol, and 15 per cent Ficoll 400) and analysed by electrophoresis on 2.5 per cent agarose gels (DNA grade agarose, Progen Industries Limited, Australia) containing 0.5 mg/ml ethidium bromide for 1.5 h at 65 V in 1 x TAE buffer (367 mM Tris, 360 mM boric acid, 8 mM EDTA, pH 8.0). Product size was determined relative to standard DRIgestTM size markers (Pharmacia Biotech, Inc., Victoria, Australia). DNA PCR products were then transferred to Hybond N+ nylon membranes (Amersham Lifesciences Pty Ltd., NSW, Australia). Agarose gels were incubated for 30 min in 0.4 M NaOH and DNA transferred to membranes with 0.4 M NaOH by upward capillary action for at least 5 h. Membranes were air-dried and stored at -20°C until hybridized.

Radioactive oligonucleotide DNA probes for hybridization to KAIl and /3-actin were designed to sequences internal to the PCR primers used. The probes for detecting the PCR amplification products for KAIl-(l), KAIl-(2), and 3-actin corresponded to bases 864-893 (5'CGTGGGCGTGGGTGTGGCCATCATCGAGCT3' SEQ ID NO:7), 751-780 (5'CCCGGCAACAGGACCCAG AGTGGCAACCAC3 ' SEQ ID NO:8), and 2872-2901 (5'GAGCGCAAGTACTCCGTGTGGATCGGCGGC3' SEQ LD NO: 9), respectively. The probes were dissolved in distilled water at 75 ng/μl and 5' end labelled at 37°C for 30 min in a reaction containing 7.5 ng of DNA probe, 10 mM Tris-HCl, 10 mM MgC12, 5 mM DTT, 0.1 mM spermidine (pH 1.6), and 5μl of γ-[32P]ATP (3000-5000 Ci/mmol) (Progen, Australia) and 0.5 unit of T4 polynucleotide kinase (Progen Industries Ltd., Australia).

Membranes were washed once in 2 x SSC and pre-hybridized for 15 min at 42°C in rapid- hyb buffer (Amersham Lifesciences, NSW, Australia). Hybridization was performed at 42°C for 1.5 h in the same buffer with radioactively labelled jS-actin and KAIl probes at final concentrations of not less than 1.87 ng/ml. Following hybridization, the membrane was rinsed at 42°C, once in 2 x SSC, 0.1 per cent SDS for 10 min and once in 1 x SSC , 0.1 per cent SDS for 15 min. Membranes were then imaged with autoradiography by exposure to X-ray film (BioMax, Kodak, Integrated Sciences, Victoria, Australia) for at least 2 h at -70°C. Autoradiographs were quantitated by computer-assisted densitometry using the MCLD software (Imaging Research Inc., Ontario, Canada). The hybridizing intensity of each KAIl band was normalized to its respective /3-actin band.

Statistical analysis

Data were analysed with one-way analysis of variance (ANOVA) followed by Tuckey's multiple comparisons, p values less than 0.05 were considered statistically significant. Where experiments were repeated, the results are expressed as the mean ± the standard error of the mean (SEM).

EXAMPLE 2

Immunohistochemical analysis of KAIl protein expression in prostate cancer

Materials and Methods

Tissue specimens

Formalin-fixed, paraffin-embedded sections from 46 patients who underwent transurethral resection of the prostate (TURP) from 1984 to 1999 were analysed. Specimens included 13 well, 15 moderately and 12 poorly differentiated prostate cancers and 6 cases of benign prostatic hyperplasia (BPH), also called 'BPH not associated with cancer'. Cancer specimens were chosen on the basis that they contained some areas of non-malignant surrounding tissue (called 'BPH associated with prostate cancer'), which were used for intraspecimen comparison of KAIl expression. Cancer tissues were examined microscopically and graded histologically according to the Gleason system (22) (well differentiated: Gleason grade 2-4, moderately differentiated: Gleason grade 5-7, poorly differentiated: Gleason grade 8-10). Immunohistochemistry

Immunohistochemical analysis of KAIl protein expression in prostate cancers of different histological grade was performed on 4μm sections obtained from formalin-fixed, paraffin- embedded TURPs. Slides were deparaffmised in xylene and rehydrated in 100% and 95% ethanol, respectively. Antigen retrieval was carried out by microwaving in lOmM citric acid buffer, pH 6.0, in a microwave (900W Panasonic) for 10 minutes at medium-high setting. Slides were allowed to cool for 20 minutes before the immunohistochemical procedure was continued using the avidin/biotin-based LSAB+/ΗRP kit (DAKO Corporation, Carpinteria/California), according to the manufacturer's protocol with minor alterations. All incubations were carried out at room temperature in a humidified chamber. Slides were washed in 1% BSA/PBS solution, pH 7.4, between all incubations listed below except after the peroxidase block, which involved excess block being tapped off, and distilled water being used after the last alcohol wash, microwave step and for all steps after application of the chromogen.

Endogenous peroxidase activity was quenched by incubation in 3% hydrogen peroxide. 'DAKO Protein Block Serum-Free' was applied for 5 minutes to reduce background staining prior to specimens being incubated with 1:250 preparation (in PBS) of polyclonal rabbit anti-human KAIl (C-16) antibody (Santa Cruz Biotechnology, Santa Cruz/California) for 2 hours. This was followed by sequential incubation of specimens in an anti-goat/rabbit/mouse biotinylated link antibody and peroxidase-labelled streptavidin. Protein expression was visualised after incubation with 3,3XDiaminobenzidine (DAB) substrate-chromogen solution, yielding a brown end-product, and counterstaining with haematoxylin. As a negative control, slides were incubated with 1% BSA/PBS solution in place of KAIl antibody solution.

Immunohistochemical quantification

Specimens were scored for KAIl protein expression using the Microcomputer Imaging

Device (MCID)-M2 program (Imaging Research Incorporated, Ontario). Digitised colour images, obtained using the Sony Progressive 3CCD colour video camera (Model DXC-

9100P, Sony Corporation/ apan), were analysed for levels of DAB staining. DAB staining (target area) as measured as a function of the level of brown staining measured with the following parameter settings:- Scan Area: Hue 0.00-36.56 and 340.31-359.00 (Scale: 0=red, 60=yellow, 120=green, lδO^cyan, 240=blue, 300=magenta), Intensity and Saturation 0.00-1.00, Study selection: Grain count. The following data was measured for each digitised image: Sample selection: 'Density (D) x Area (A)' and 'Proportional Area¹. DxA (units: intensity x pixel) is the density of the measured grains multiplied by the area of the sample window. Proportional area is the ratio of the total target area to the scanned area. Each sample window was the size of the default measurement box used by the program; the size of the box allowed for measurements within a region half the size of an average epithelial cell. Only cytoplasmic regions of the cell were measured. Twenty windows within ten adjacent fields (200x magnification) in each specimen were measured, hi samples with different cancer patterns, such as moderately differentiated specimens which may contain regions of both well and poorly differentiated tissue, equal numbers of fields in each area were measured (to a total of 10 fields). D x A ad proportional area were multiplied to give the KAIl staining score for each specimen, which is representative of KAIl protein expression within the sample. Scores were averaged for each category of cancer and the standard deviation (SD), standard error of the mean (SEM) and 95% confidence interval were also calculated. Statistical analysis of the differences between KAIl scores in BPH not associated with cancer, BPH associated with prostate cancer and all prostate cancers was performed using an unpaired, two sample t-test. A paired t-test was performed for comparisons between all prostate cancer specimens and BPH associated with these cancers.

Results A pattern of diffuse staining of the cytoplasm of prostatic epithelial cells was observed (Figure 1). Examples of staining in BPH not associated with cancer and well, moderately and poorly differentiated prostate cancers and BPH associated with prostate cancer are shown in Figure 1. Quantification of staining for all specimens, using the MCLD program, is shown in Figure 2. Scores are also presented in Table 1. Categorisation of tissue scoring is described below. Table 1

KAIl staining scores for BPH not associated with cancer, BPH associated with prostate cancers and well, moderately and poorly differentiated prostate cancers. P values from two sample t-test.

BPH not associated with cancer versus all prostate cancer specimens.

The combined pool of prostate cancers studied (n=40) exhibited quite a variable range of staining, from very low to very high KAIl protein expression (Figure 1; panels 2-4), compared to that seen in BPH not associated with cancer (Figure 1; panel 1). Immunohistochemical stores of KAIl expression (as described in the Methods section) in the cancer specimens ranged from 0.28 to 49.48 (Figure 2), with an average of 8.67+ 1.65 (mean+SEM) (Table 1), whilst scores in BPH ranged from 1.16 to 9.13 (Figure 2), with an average of 3.59+ 1.21 (Table 1). The difference in KAIl expression between BPH not associated with prostate cancer and prostate cancers was found to be significant (p=0.019), i.e. prostate cancers exhibited significantly higher KAIl protein expression than that seen in BPH, with 27.5% exhibiting KAIl expression beyond the 95% upper limit observed in BPH (Table 1). BPH not associated with cancer versus individual categories of prostate cancer (well, moderately and poorly differentiated specimens).

In prostate cancers, variable levels of staining were seen within and between each category of well, moderately and poorly differentiated cancer specimens (Figure 1; panels 2, 3 and 4, respectively). Increased KAIl protein expression in well (score: 9.25±3.89) and moderately (9.78+2.74) differentiated prostate tumours, compared to BPH not associated with cancer (3.59+ 1.21), was observed (Table 1, Figure 2, Figure 1; panel 1 vs panels 2, 3 and 4). In poorly differentiated cases, a less obvious increase in protein levels (6.64± 1.33) was observed, compared to BPH not associated with cancer (Table 1, Figure 2). Scores for well differentiated cancers ranged from 0.90 to 49.48, in moderately differentiated cancers ranged from 0.28 to 37.84 and in poorly differentiated cancers ranged from 1.61 to 16.01 (Figure 2). Thus, compared to BPH not associated with cancer, increased KAIl levels were seen in all three categories of specimens (well, moderately and poorly differentiated tumours), though in poorly differentiated cancers a reduction towards levels seen in BPH not associated with cancer was noted (Figure 1; panels 1 vs 4, Figure 2, Table 1). Wliile no significant differences were found amongst KAIl expression in well, moderately or poorly differentiated prostate cancers, the difference between KAIl expression in the BPH specimens not associated with cancer and moderately differentiated cancers approached significance (P-0.053) (Table 1). It was observed that 15.4%, 33.3% and 33.3% of well, moderately and poorly differentiated cancers exhibited KAIl expression beyond the 95% upper limit observed in BPH not associated with prostate cancer (Table 1).

BPH not associated with cancer versus BPH associated with prostate cancers Interestingly, BPH associated with prostate cancers (Figure 1; panel 5, KAIl score: 9.19+ 1.50 (Table 1), range: 0.82 to 46.09 (Figure 2)) showed increased KAIl staining compared to BPH not associated with cancer (KAIl score 3.59+ 1.21 (Table 1)), Figure 1; panel 1 vs 5. This difference was found to be highly significant (p=0.008). There appeared to be a trend of increased KAIl expression in BPH associated with prostate cancers as the cancers became more advanced, i.e. slight increases in KAIl expression in the benign tissue as the cancers progressed from well to poor levels of differentiation. There was, however, no significant difference between KAIl levels in the BPH associated with cancers across these cancer grades. Of the BPH associated with prostate cancers, 35% of these tissues exhibited KAIl expression beyond the 95% upper limit observed in BPH not associated with cancer (Table 1).

BPH associated with prostate cancers versus well, moderately and poorly differentiated prostate cancers

No difference in KAIl expression was found between prostate cancers and BPH associated with prostate cancers, whether compared to the combined pool of prostate cancers or individual categories of prostate cancer (well, moderately and poorly differentiated specimens) (Figure 1; panels 2, 3 and 4 vs panel 5, Figure 2). BPH associated with prostate cancer (score: 9.19+ 1.50) was observed to exhibit KAIl levels similar to that seen in well (9.25+3.89) and moderately (9.78+2.74) differentiated prostate cancers (Table 1).

Threshold and Chi Determination

KAIl gene expression for tumours as a function of KAIl expression in surrounding normal tissue, expressed as a ratio, was determined based on arithmetic means of immunohistochemical image analysis results. The quotient threshold of 1.34 was determined.

It was then shown that 18 of 28 samples taken from tumours which had metastasised at the time of, or within two years after sampling, had a quotient of less than the 1.34, the threshold value.

It was also shown that 9 of 11 samples taken from tumour tissue which did not metastasise within the following two years had a quotient greater than the quotient threshold value.

Statistical analysis of the results using the Chi² test (p < 0.01) shows the results are highly significant and that values of KAIl gene expression quotient below the quotient threshold are highly predictive of non-metastasis of tumour tissue. Discussion

Current screening strategies for prostate cancer have not reduced mortality rates. One problem is that prostate cancer is an unpredictable disease and quite heterogeneous behaviour can be seen in the clinical setting (23). Evidence would not suggest that increased detection results in a higher proportion of cancers that will not cause excess morbidity and mortality (24-28). Of further concern, however, is the finding that a significant portion of men (16-33%) have undetected micrometastatic disease at the time of surgery (29,30) and pathologically detectable extracapsular disease after surgery is found in 40 to 60%) of patients (31-34). Such patients require systemic treatment if recurrence is to be avoided, highlighting the need for improved screening modalities. Currently, no method exists to distinguish those cancer cells which possess the metastatic phenotype within the primary tumour population from those that do not. As such, current prognostic indicators in prostate cancer are unable to accurately determine individual outcomes of prostate cancer patients or those at highest risk of metastases. The need for new prognostic markers is thus quite urgent. The use of metastasis suppressor genes, such as KAIl, as biological markers of the acquisition of metastatic ability is attractive since they encode products that function at the transition point from "localised" to "metastatic" cancer.

It is generally accepted that KAIl expression in most solid cancers decreases as the cancer progresses, with metastatic deposits having the lowest KAIl levels. This is true of non- small cell lung (35), breast (36), hepatocellular (37) and bladder (38) cancers and invasive squamous cell carcinomas from the lung, head and neck and cervix (39). Despite this general observation, however, changes in KAIl expression in a number of different primary cancers is not so straightforward. Several studies, reporting the pattern of KAIl expression in oesophageal (40-42), gastric (40,43), pancreatic (44-46), colon (47-50) and prostate (19-21,51) cancers, suggest either no change in KAIl expression, progressive downregulation or a biphasic pattern of expression during cancer progression. In a particular study of oesophageal and gastric cancers, it was found that KAIl mRNA levels were unchanged from normal levels in metastatic and non-metastatic cases (40), contrary to other findings that KAIl is decreased during the progression of both cancers (41-43). In the colon, an immunohistochemical study has suggested a biphasic pattern of KAIl during cancer progression, with ore intense KAIl staining in early stage cancers and reduced levels in later stages of the disease, compared to levels in normal tissue, though KAIl could not be correlated with tumour grade (47). A biphasic pattern is also evident in pancreatic cancers; more than 80% of these cancers expression KAIl mRNA at levels above that seen in noπnal surrounding tissue (45,52). Early tumours (stage I, II) are seen to have significantly higher KAIl levels than more advanced cancers (stages III, IV) in which lymph node or distant metastases were present (45,52). Unusually, however, poorly differentiated pancreatic tumours exhibited significantly higher KAIl mRNA levels than well or moderately differentiated tumours (52).

The pattern of KAIl expression in prostate cancer is also somewhat unclear. Some studies have shown an inverse relationship between KAIl expression and tumour grade (Gleason score) (20), sage (20) and clinical metastatic behaviour (19). In contrast, recent findings from our own laboratory have shown a biphasic pattern of KAIl mRNA expression in primary prostate cancers according to histological grade, compared to non-malignant prostate tissues (BPH not associated with cancer); KAIl was significantly elevated in low grade prostate tumours and reduced in high grade tumours (21). The current immunohistochemical study, comparing BPH not associated with cancer with well, moderately and poorly differentiated cancers, appears to mirror and therefore complements our mRNA findings. The lower grades of cancer (well and moderately differentiated tumours) tended to have higher levels of KAIl expression than that seen in BPH not associated with cancer or in more advanced, high grade (poorly differentiated) cancers. Overall, it was found that prostate cancers expressed significantly higher KAIl protein than BPH not associated with cancer (P=0.019), which approached significance in moderately differentiated prostate cancers (P=0.053). An interesting and new finding from the present study is the observation of a significant difference in KAIl expression between the BPH tissues not associated with cancer and the BPH tissues associated with prostate cancers, the latter expressing KAIl at significantly higher levels (P=0.0008). This might indicate that the BPH glands surrounding prostate cancers, though appearing histologically benign in nature, may already be on the pathway leading towards cancer. Because KAIl exhibits metastasis suppressor ability, the increase in KAIl expression seen in BPH associated with prostate cancers may serve as a protective mechanism to restrain further cancer progression and, potentially, subsequent metastasis development. Alternatively, this effect may be as a result of changes in the microenvironment due to the presence of cancer in the prostate or the by-product of a local field effect, where alterations in KAIl expression in the cancerous region affects surrounding normal tissue. This so-called 'field

^' effect', the influence of changes in one part of an organ producing equivalent effects in another part of the organ, has been observed in head and neck (53,54), lung (54-56), uterine cervix (56-60), breast (56,58,61-63), as well as in prostate cancers (64-68). In prostate cancer, the findings that prostatic intraepithelial neoplasia, a likely prostate cancer precursor lesion (69-76), and prostate cancer can arise independently within the same prostate (65-67) supports the notion that a field effect may underlie prostate carcinogenesis.

In the present study, increased KAIl protein levels were seen in BPH associated with prostate cancer lesions and across all grades of primary prostate cancers, compared to BPH not associated with cancer, though KAIl levels in poorly differentiated cancers appeared to fall towards levels in BPH not associated with cancer. It is conceivable that differential KAIl expression in primary prostate cancers might have important prognostic implications for the development of subsequent metastases. Thus, patients with the highest levels of KAIl expression may be afforded greater protection from development of metastases than those patients with low-level KAIl expression. This might allow clinicians to target therapies more intensively in those patients with primary prostate cancer who are considered to be at greatest risk of subsequent metastases, thus favourably impacting upon morbidity and mortality associated with prostate cancer. Further immunohistochemical work and correlation of the present findings with patient clinical outcomes are needed to confirm such a hypothesis. The finding of conserved KAIl expression correlating with good prognosis in non-small cell lung cancer patients (35), and similar findings in pancreatic (44), oesophageal (41) and breast (36) cancer patients, preempts a possible prognostic role for KAIl in prostate cancer. Though the findings of this study point to a potentially important role for KAIl measurement in prostate cancers in the clinical setting, it is important to note that there are still aspects of KAIl expression and regulation of the signalling pathways associated with this protein which remain unknown. What is known, however, is that chromosome 11, which contains the KAIl gene, plays an important role in metastasis suppressor activity via other genes, such as the cell adhesion molecule CD44, at lip 13 (77), and the new breast caner metastasis suppressor BRMS1, at llql3.1-13.2 (78). Expression of BRMS1 and specific isoforms of CD44 have been found to inhibit metastasis in various human cell lines (77-80). Like KAIl, CD44 is also downregulated during prostate cancer progression in humans (81). The mechanism by which KAIl expressions downregulated during progression is still unknown, though it is known that allelic loss (19, 82) and mutations (19) are not involved.

In conclusion, the current finding of an apparent biphasic pattern of KAIl protein expression during prostate cancer progression, compared to levels in BPH not associated with cancer, is exciting and of great potential relevance in terms of diagnosis and prognosis of prostate cancers. Furthermore, increased KAIl levels in BPH associated with prostate cancer may yield an important means of identifying lesions which are likely to contain coexisting (otherwise undetected) cancers.

REFERENCES

1. Dong JT, Lamb PW, Rinker-Schaeffer CW, et al. KAIl, a metastasis suppressor gene for prostate cancer on human chromosome l lpl l.2. Science (Wash DC)

1995; 268; 884-886.

2. Gavgitsch HW, Hofer E, Huber NE, Schnabal E, Baumruker TA. A new superfamily of lymphoid and melanoma cell protein with extensive homology to Schistasonia niansoni antigen Sm23. Eur J Immunol 1991 ; 21 :377-383.

3. Gil ML, Vita N, Lebel-Binay S, et al. A member of the tetra spans transmembrane protein superfamily is recognised by a monoclonal antibody raised against an HLA class I-deficient, lymphokine activated killer-susceptible, B lymphocyte line. J Immunol 1992; 148: 2826-2833.

4. Nojima Y, Hirosc T, Tachibana K, et al. The 4F9 antigenis a member of the tetra spans transmembrane protein family and functions as an accessory molecule in T- cell activation and adhesion. Cell Immunol 1993; 152: 249-260.

Imai T, Fukudome K, Takagi S, et al. C33 antigen recognised by monoclonal antibody inhibitory to human T cell leukemia virus type I-induced syncytium formation is a member of a new family of transmembrane proteins, including CD9, CD37, CD53 and CD63. J. Immunol 1992; 149: 2879-2886.

Wright MD, Tomlinson MG. The ins and outs of the transmembrane 4 superfamily, Immumol Today 1994; 15: 588-594.

7. Hotta H, Ross AH, Huebner K, et al. Molecular cloning and characterisation of an antigen associated with early stages of melanoma tumour progression, Cancer Res

1988; 48: 2955-2962. 8. Ikeyama S, Koyama M, Yamaoko M, Sasada R, Miyake M. Suppression of cell motility and meastasis by transfection with human motility-related protein (MRP1/CD9) DNA. J Exp Med 1993; 177:1231-1237.

9. Dong JT, Suzuki H, Pin SS, et al. Down-regulation of the KAIl metasatasis suppressor gene during the progression of human prostatic cancer infrequently involves gene mutation or allelic loss. Cancer Res 1996; 56: 4387-4390.

10. Ueda U, Ichikawa T, Tamaru J, et al. Expression of the KAIl protein in benign prostatic hyperplasia and prostate cancer. Am J Pathol 1996; 149: 1435-1440.

11. Gleason DF. Histologic grading of prostate cancer: a perspective. Hum Pathol 1992; 23: 273-279.

12. Chomczynski P. Sacchi N. Single step method of RNA isolation by acid guanidinium thiocyanate-phenol-chloroform extraction. Anal Biochem 1987; 162: 156-159.

13. Major B, Cures A, Frauman AG. The full length and splice variant thyrotropin receptor is expressed exclusively in skeletal muscle of extraocular origin: a link to the pathogenesis of Graves' ophthalmopathy. Biochem Biphys Res Commun 1997; 230: 493-496.

14. Scardino, P.T., Weaver, R. and Hudson, M.A., Early detection of prostate cancer, Hum Pathol, 23(3), 211, 1992.

15. Partin, A.W., Pound, C.R., Clemens, J.Q. et al., Serum PSA after anatomic radical prostatectomy. The Johns Hopkins experience after 10 years, Urol Clin North Am, 20(4), 713, 1993. 16. Lu-Yao, G.L., Potoskyl, A.L., Albertsen, P.C. et al, Follow-up prostate cancer treatments after radical prostatectomy: a population-based study, J Natl Cancer Inst, 88(3-4), 166, 1996.

17. Wasson, J.H., Cushman, C.C., Bruskewitz, R.C. et al., A structured literature review of treatment for localized prostate cancer. Prostate Disease Patient Outcome Research Team, Arch Font Med, 2(5), 487, 1993.

18. Paulson, D.F., Moul, J.W., Robertson, J.E. et al., Postoperative Radiotherapy of the prostate for patients undergoing radical prostatectomy with positive margins, seminal vesical involvement and/or penetration through the capsule, J Urol, 143(6), 1178, 1990.

19. Hanks, G.E. and Dawson, A.K., The role of external beam radiation therapy after prostatectomy for prostate cancer, Cancer, 58(11), 2406, 1986.

20. Higashiyama, M., Kodama, K., Yokouchi, H. et al., KAI1/CD82 expression in non- small cell lung carcinoma is a novel, favourable prognostic factor - an immunohistochemical analysis, Cancer, 83(3), 466, 1998.

21. Huang, C.I., Kohno, N., Ogawa, E. et al., Correlation of reduction in MRP-1/CD9 and KAI1/CD82 expression with recurrences in breast cancer patients, Am J Pathol, 153(3), 973, 1998.

22. Guo, X.Z., Friess, H., Di Mola, F.F. et al., KAIl, a new metastasis suppressor gene, is reduced in metastatic hepatocellular carcinoma, Hepatology, 28(6), 1481, 1998.

23. Ow, K., Delprado, W., Fisher, R. et al, Relationship between expression of the KAIl metastasis suppressor and other markes of advanced bladder cancer, J Pathol, 191(1), 39, 2000. 24. Geradts, J., Maynard, R., Birrer, M.J. et al., Frequent loss of KAIl expression in squamous and lymphoid neoplasms - an immunohistochemical study of archival tissues, Am J Pathol, 154(6), 1665, 1999.

25. Guo, X.Z., Friess, H., Maurer, C. et al., KAIl is unchanged in metastatic and nometastatic esophageal and gastric cancers, Cancer Res, 58(4), 753, 1998.

26. Dong, J.T., Suzuki, H., Pin, S.S. et al., Down-regulation of the KAIl metastasis suppressor gene during the progression of human prostatic cancer infrequently involves gene mutation or allelic loss, Cancer Res, 56(19), 4387, 1996.

27. Ueda, T., Ichikawa, T., Tamaru, J. et al., Expression of the KAIl protein in benign prostatic hyperplasia and prostate cancer, Am J Pathol, 149(5), 1435, 1996.

28. Bouras, T. and Frauman, A.G., Expression of the prostate cancer metastasis suppressor gene KAIl in primary prostate cancers: A biphasic relationship with tumour grade, J Pathol, 188(4), 382, 1999.

29. Gleason, D.F., Classification of prostatic carcinomas, Cancer Chemother Rep, 50(3), 125, 1966.

30. Boring, C.C., Squires, T.S. and Tong, T., Cancer statistics, 1993, CA Cancer J Clin, 43(1), 7, 1993.

31. Albertsen, P.C., Hanley, J.A., Gleason, D.F. et al., Competing risk analysis of men aged 55 to 74 years at diagnosis managed conservatively for clinically localized prostate cancer, JAMA, 280(11), 975, 1998. 32. Hankey, B.F., Feuer, E.J., Clegg, L.X. et al, Cancer surveillance series: interpreting trends in prostate cancer—part I: Evidence of the effects of screening in recent prostate cancer incidence, mortality, and survival rates, J Natl Cancer Inst, 91(12), 1017, 1999.

33. Feuer, E.J., Merrill, R.M. and Hankey, B.F., Cancer surveillance series: interpreting trends in prostate cancer—part II: Cause of death misclassificationa nd the recent rise and fall in prostate cancer mortality, J Natl Cancer Inst, 91(12), 1025, 1999.

34. Etzioni, R., Legler, J.M., Feuer, E.J. et al., Cancer surveillance series: interpreting trends in prostate cancer— part III: Quantifying the link between population prostate-specific antigen testing and recent declines in prostate cancer mortality, J Natl Cancer Inst, 91(12), 1033, 1999.

35. Johansson, J.E., Holmberg, L., Johansson, S. et al., Fifteen-year survival in prostate cancer. A prospective, population-based study in Sweden, JAMA, 277(6), 467, 1997.

36. Miyazaki, T., Kato, H., Shitara, Y. et al., Mutation and expression of the metastasis suppressor gene KAIl in esophageal squamous cell cacinoma, Cancer, 89(5), 955, 2000.

37. Uchida, S., Shimada, Y., Watanabe, G. et al, Motility-related protein (MRP- 1/CD9) and KAI1/CD82 expression inversely correlate with lymph node metastasis in oesophageal squamous cell carcinoma, Br J Cancer, 79(7-8), 1168, 1999.

38. Hinoda, Y., Adachi, Y., Takaoka, A. et al., Decreased expression of the metastasis suppressor gene KAIl in gastric cancer, Cancer Lett, 129(2), 229, 1998. 39. Sho, M., Adachi, M., Taki, T. et al., Transmembrane 4 superfamily as a prognostic factor in pancreatic cancer, bit J Cancer, 19(5), 509, 1998.

40. Friess, H., Guo, X.Z., Berberat, P. et al., Reduced KAIl expression in pancreatic cancer is associated with lymph node and distant metastases, Int J Cancer, 79(4),

349, 1998.

41. Guo, X., Friess, H., Graber, H.U. et al., KAIl expressionin is up-regulated in early pancreatic cancer and decreased in the presence of metastases, Cancer Res, 56(21), 4876, 1996.

42. Maurer, C.A., Graber, H.U., Friess, H. et al., Reduced expression of the metastasis suppressor gene KAIl in advanced colon cancer and its metastases, Surgery, 126(5), 869, 1999.

43. Takaoka, A., Hinoda, Y., Satoh, S. et al., Suppression of Invasive Properties of Colon Cancer Cells By a Metastasis Suppressor KAIl Gene, Oncogene, 16(11), 1443, 1998.

44. Lombardi, D.P., Geradts, J., Foley, J.F., et al., Loss of KAIl expression in the progression of colorectal cancer, Cancer Res, 59(22), 5724, 1999.

45. Huang, C.L., Taki, T., Adachi, M. et al., MRP-1/CD9 and KAI1/CD82 expressionin normal and varius cancer tissues, Int J Oncol, 11(5), 1045, 1997.

46. Ward, J.M., Konishi, N., Ohshima, M. et al., Expression of KAIl in paraffin- embedded normal, hyperplastic and neoplastic prostate and prostate carcinoma cell lines, Pathol Int, 48(2), 87, 1998. 47. Guo, X.Z., Friess, H., Graber, H.U., et al., KAIl expression is up-regulated in early pancreatic cancer and decreased in the presence of metastases, Cancer Res, 56(21), 4876, 1996.

48. Giles, G., Staples, M., McCredie, M. et al., Multiple primary melanomas: an analysis of cancer registry data from Victoria and New South Wales, Melanoma Res, 5(6), 433, 1995.

49. Neugut, A.I., Sherr, D., Robinson, E. et al., Differences in histology between first and second primary lung cancer, Cancer Epidemiol Biomarkers Prev, 1(2), 109,

1992.

50. Franklin, W.A., Gazdar, A.F., Haney, J. et al., Widely dispersed p53 mutation in respiratory epithelium. A novel mechanism for field carcinogenesis, J Clin Invest, 100(8), 2133, 1997.

51. Shamsuddin, A.M., Tyner, G.T. and Yang, G. Y., Common expression of the tumor marker D-galactose-beta-[l~>3]-N-acetyl-D-galactosamine by different adenocarcinoas: evidence of field effect phenomenon, Cancer Res, 55(1), 149, 1995.

52. Chu, T.Y., Shen, C.Y., Lee, H.S. et al., Monoclonality and surface lesion-specific microsatellite alterations in premalignant and malignant neoplasia of uterine cervix: a local field effect of genomic instability and clonal evolution, Genes Chromosomes Cancer, 24(2), 127, 1999.

53. Nohammer, G., Bajardi, F., Benedetto, C. et al., Histophotometry of protein thiols and disulphides in tissue samples from the human uterine cervix and the skin reveals a "field effect" as well as an "extended field effect" of malignant tumours, Ada Histochem Suppl, 38, 247, 1990. 54. Taskin, M., Lallas, T.A., Shevchuk, M. et al., p53 expression in adenomyosis in endometrical carcinoma patients, Gynecol Oncol, 62(2), 241, 1996.

55. Benedetto, C, Bajardi, F., Ghiringhello, B. et al., Quantitative measurements of the changes in protein thiols in cervical intraepithelial neoplasia and in carcinoma of the human uterine cervix provide evidence for the existence of a biochemical field effect, Cancer Res, 50(20), 6663, 1990.

56. Cummings, M.C., Aubele, M., Mattis, A. et al., Increasing chromosome 1 copy number parallels histological progression in breast carcinogenesis, Br J Cancer,

82(6), 1204, 2000.

57. Mai, K.T., Morphological evidence for field effect as a mechanism for tumour spread in mammary Paget's disease, Histopathology, 35(6), 567, 1999.

58. Walker, A.N. and Fechner, R.E., Papillary carcinoma arising from ectopic breast tissue in an axillary lymph node, Diagn Gynecol Obstet, 4(2), 141, 1982.

59. Olapade-Olaopa, E.O., MacKay, E.H., Taub, N.A. et al., Malignant transformation of human prostatic epithelium is associated with the loss of androgen receptor immunoreactivity in the surrounding stroma, Clin Cancer Res, 5(3), 569, 1999.

60. Bostwick, D.G., Shan, A., Qian, J. et al., Independent origin of multiple foci of prostatic intraepithelial neoplasia: comparison with matched foci of prostate carcinoma, Cancer, 83(9), 1995, 1998.

61. Foster, C.S., Bostwick, D.G., Bonkhoff, H. et al., Cellular and molecular pathology of prostate cancer precursors, ScandJ Urol Nephrol Suppl, 205, 19, 2000. 62. Crissman, J.D., Sakr, W.A. Hussein, M.E. et al., DNA quantitation of intraepithelial neoplasia and invasive carcinoma of the prostate, Prostate, 22(2), 155, 1993.

63. Bastacky, S.I., Wojno, K.J., Walsh, P.C. et al., Pathological features of hereditary prostate cancer, J Urol, 153(3 Pt 2), 987, 1995.

64. Bostwick D.G. and Brawer, M.K., Prostatic intra-epithelial neoplasia and early invasion in prostate cancer, Cancer, 59(4), 788, 1987.

65. Macoska, J.A., Micale, M.A., Sakr, W.A. et al., Extensive genetic alterations in prostate cancer revealed by dual PCR and FISH analysis, Genes Chromosomes Cancer, 8(2), 88, 1993.

66. Sakr, W.A., Macoska, J.A., Benson, P. et al., Allelic loss in locally metastatic, multisampled prostate cancer, Cancer Res, 54(12), 3273, 1994.

67. Bostwick, D.G., Amin, M.B., Dundore, P. et al., Architectural patterns of high- grade prostatic intraepithelial neoplasia, Hum Pathol, 24(3), 298, 1993.

68. McNeal, J.E. and Bostwick, D.G., Intraductal dysplasia: a premalignant lesion of the prostate, Hum Pathol, 17(1), 64, 1986.

69. Qian, J.Q., Bostwick, D.G., Takahashi, S. et al., Chromosomal anomalies in prostatic intraepithelial neoplasia and carcinoma detected by fluorescence in situ hybridization, Cancer Res, 55(22), 5408, 1995.

70. Alers, J.C, Krijtenburg, P.J., Vissers, K.J. et al., Interphase cytogenetics of prostatic adenocarcinoma and precursor lesions - analysis of 25 radical prostatectomies and 17 adjacent prostatic intrapeithelial neoplasias, Genes

Chromosomes Cancer, 12(4), 241, 1995. 71. Davidson, D., Bostwick, D.G., Qian, J. et al., Prostatic intraepithelial neoplasia is a risk factor for adenocarcinoma: predictive accuracy in needle biopsies, J Urol, 154(4), 1295, 1995.

72. Gao, A.C., Lou, W., Dong, J.T. et al., CD44 is a metastasis suppressor gene for prostatic cancer located on human chromosome l lpl3, Cancer Res, 57(5), 846, 1997.

73. Seraj, M.J., Samant, R.S., Verderame, M.F. et al., Functional evidence for a novel human breast carcinoma metastasis supressor, BRMS1, encoded at chromosome llql3, Cancer Res, 60(11), 2764, 2000.

74. Gunthert, U., Stauder, R., Mayer, B. et al., Are CD44 variant isoforms involved in human tumour progression?, Cancer Surv, 24, 19, 195.

75. Rudy, W., Hofmann, M., Schwartz-Albiez, R. et al., The two major CD44 proteins expressed on a metastatic rat tumor cell line are derived from different splice variants: each one individually suffices to confer metastatic behaviour, Cancer Res, 53(6), 1262, 1993.

76. Kallakury, B.N., Yang, F., Figge, J. et al., Decreased levels of CD44 protein and mRΝA in prostate carcinoma. Correlation with tumor grade and ploidy, Cancer, 78(7), 1461, 1996.

77. Tagawa, K, Arihiro, K, Takeshima, Y. et al., Down-regulation of KAIl messenger RΝA expression is not associated with loss of heterozygosity of the KAIl gene region in lung adenocarcinoma, Jpn J Cancer Res, 90(9), 970, 1999.

Claims

THE CLAIMS DEFINING THE INVENTION ARE AS FOLLOWS:

1. A method of prognosis of a cancer patient comprising subjecting a sample of primary tumour tissue of defined type and grade from the patient to quantitative analysis of KAIl gene expression and determining whether KAIl expression level is elevated or reduced relative to a tlireshold value for the defined tumour tissue type and grade; wherein elevated expression is indicative of restraint of future tumour progression and reduced expression is indicative of future tumour metastasis.

2. The method according to claim 1 wherein the primary tumour tissue is prostate, breast, lung, colon, liver, brain, kidney, trachea, rectum, stomach, pancreas, cervix, uterus, ovary, testicle, thyroid or epidermal tissue.

3. The method according to claim 1 wherein the primary tumour tissue is prostate, breast or lung tissue.

4. The method according to claim 1 wherein the primary tumour tissue is prostate tissue.

5. The method according to any one of claims 1 to 4 wherein quantitative analysis of

KAIl gene expression is conducted by determining whether KAIl protein level is elevated or reduced relative to a threshold value for KAIl protein level.

6. The method according to claim 5 wherein an ELISA assay is conducted to determine whether KAIl protein level is elevated or reduced.

7. The method according to claim 5 wherein immunohistochemistry analysis is conducted to determine whether KAIl protein level is elevated or reduced.

8. The method according to claim 5 wherein determination of whether KAIl protein level is elevated or reduced is by conducting an assay wherein labelled KAIl antibody is bound to a support and is contacted with protein from the tumour tissue, wherein following binding of KAIl protein to the labelled KAIl antibody and detection of the label, a quantitative measure of KAIl expression is obtained.

9. The method according to claim 8 wherein the label is radioactive, fluorescent or colourimetric.

10. The method according to any one of claims 1 to 4 wherein quantitative analysis of KAIl gene expression is conducted by determining whether KAIl mRNA level is elevated or reduced relative to a threshold value for KAIl mRNA level.

11. The method according to claim 10 wherein RT-PCR analysis is conducted to determine whether KAIl mRNA level is elevated or reduced.

12. The method according to claim 11 wherein RNA from tumour tissue of defined type and grade is subjected to RT-PCR amplification using a primer or primers complementary to at least a fragment of KAIl cDNA; the amplified cDNAs are then separately hybridized to a labelled probe complementary to at least a fragment of KAIl cDNA; the hybridized labels are detected to give a quantitative measure of KAIl mRNA in the tumour tissue which is compared to a threshold value for the defined tumour tissue type and grade.

13. The method according to claim 12 wherein the label is radioactive, fluorescent or colourimetric.

14. A kit for use in a method according to any one of claims 1 to 13.

15. A method of prognosis of a cancer patient comprising subjecting a sample of primary tumour tissue of defined type and grade from the patient to quantitative analysis of KAIl gene expression and subjecting a sample of normal tissue of the defined type and from the same patient, that is not derived from adjacent to the tumour, to quantitative analysis of KAIl gene expression, determining the quotient of KAIl gene expression from the tumour tissue and KAI gene expression from the normal tissue and determining whether the quotient is elevated or reduced relative to a quotient threshold value for the defined tumour tissue type and grade; wherein an elevated quotient is indicative of restraint of future tumour progression and a reduced quotient is indicative of future tumour metastasis.

16. The method according to claim 15 wherein the tumour tissue is prostate, breast, lung, colon, liver, brain, kidney, trachea, rectum, stomach, pancreas, cervix, uterus, ovary, testicle, thyroid or epidermal tissue.

17. The method according to claim 15 wherein the primary tumour tissue is prostate, breast or lung tissue.

18. The method according to claim 15 wherein the primary tumour tissue is prostate tissue.

19. The method according to any one of claims 15 to 18 wherein quantitative analysis of KAIl gene expression is conducted by determining KAIl protein levels.

20. The method according to claim 19 wherein an ELISA assay is conducted to determine KAIl protein levels.

21. The method according to claim 19 wherein immunohistochemistry analysis is conducted to determine KAIl protein levels.

22. The method according to claim 19 wherein determination of KAIl protein levels is by conducting an assay wherein labelled KAIl antibody is bound to a support and is contacted with protein from the tumour tissue, wherein following binding of KAIl protein to the labelled KAIl antibody and detection of the label, a quantitative measure of KAIl expression is obtained.

23. The method according to claim 22 wherein the label is radioactive, fluorescent or colourimetric.

24. The method according to any one of claims 15 to 18 wherein quantitative analysis of KAIl gene expression is conducted by determining KAIl mRNA levels.

25. The method according to claim 24 wherein RT-PCR analysis is conducted to determine whether KAIl mRNA levels.

26. The method according to claim 25 wherein RNA from tumour tissue of defined type and grade is subjected to RT-PCR amplification using a primer or primers complementary to at least a fragment of KAIl cDNA; the amplified cDNAs are then separately hybridized to a labelled probe complementary to at least a fragment of KAIl cDNA; and the hybridized labels are detected to give a quantitative measure of KAIl mRNA in the tumour tissue.

27. The method according to claim 26 wherein the label is radioactive, fluorescent or colourimetric.

28. A method of prognosis of a cancer patient comprising subjecting a sample of normal tissue derived from adjacent to primary tumour tissue of the same type, to quantitative analysis of KAIl gene expression and determining whether KAIl expression level is elevated or reduced relative to a threshold value for tissue of that type located adjacent to tumour tissue; wherein elevated expression is indicative of restraint of future tumour progression and reduced expression is indicative of future tumour metastasis.

29. The method according to claim 28 wherein the primary tumour tissue is prostate, breast, lung, colon, liver, brain, kidney, trachea, rectum, stomach, pancreas, cervix, uterus, ovary, testicle, thyroid or epidermal tissue.

30. The method according to claim 28 wherein the primary tumour tissue is prostate, breast or lung tissue.

31. The method according to claim 28 wherein the primary tumour tissue is prostate tissue.

32. The method according to any one of claims 28 to 31 wherein quantitative analysis of KAIl gene expression is conducted by determining whether KAIl protein level is elevated or reduced relative to a threshold value for KAIl protein level.

33. The method according to claim 32 wherein an ELISA assay is conducted to determine whether KAIl protein level is elevated or reduced.

34. The method according to claim 32 wherein immunohistochemistry analysis is conducted to determine whether KAIl protein level is elevated or reduced.

35. The method according to claim 32 wherein determination of whether KAIl protein level is elevated or reduced is by conducting an assay wherein labelled KAIl antibody is bound to a support and is contacted with protein from the tissue adjacent tumour tissue, wherein following binding of KAIl protein to the labelled KAIl antibody and detection of the label, a quantitative measure of KAIl expression is obtained.

36. The method according to claim 35 wherein the label is radioactive, fluorescent or colourimetric.

37. The method according to any one of claims 28 to 31 wherein quantitative analysis of KAIl gene expression is conducted by determining whether KAIl mRNA level is elevated or reduced relative to a threshold value for KAIl mRNA level.

38. The method according to claim 37 wherein RT-PCR analysis is conducted to determine whether KAIl mRNA level is elevated or reduced.

39. The method according to claim 38 wherein RNA from tissue adjacent tumour tissue is subjected to RT-PCR amplification using a primer or primers complementary to at least a fragment of KAIl cDNA; the amplified cDNAs are then separately hybridized to a labelled probe complementary to at least a fragment of KAIl cDNA; the hybridized labels are detected to give a quantitative measure of KAIl mRNA in the tissue adjacent tumour tissue which is compared to a threshold value for the tissue type.

40. The method according to claim 39 wherein the label is radioactive, fluorescent or colourimetric.

41. A method of detection of tumour presence within tissue of a patient comprising subjecting a sample of tissue of defined type to quantitative analysis of KAIl gene expression and determining whether KAIl expression level is elevated or reduced relative to a tlireshold value for tissue of the defined type; wherein elevated expression is indicative of tumour presence within the tissue.

42. The method according to claim 41 wherein the tissue is prostate, breast, lung, colon, liver, brain, kidney, trachea, rectum, stomach, pancreas, cervix, uterus, ovary, testicle, thyroid or epidermal tissue.

43. The method according to claim 41 wherein the tissue is prostate, breast or lung tissue.

44. The method according to claim 41 wherein the tissue is prostate tissue.

45. The method according to any one of claims 41 to 44 wherein quantitative analysis of KAIl gene expression is conducted by determining whether KAIl protein level is elevated or reduced relative to a threshold value for KAIl protein level.

46. The method according to claim 45 wherein an ELISA assay is conducted to determine whether KAIl protein level is elevated or reduced.

47. The method according to claim 45 wherein immunohistochemistry analysis is conducted to determine whether KAIl protein level is elevated or reduced.

48. The method according to claim 45 wherein determination of whether KAIl protein level is elevated or reduced is by conducting an assay wherein labelled KAIl antibody is bound to a support and is contacted with protein from the tmnour tissue, wherein following binding of KAIl protein to the labelled KAIl antibody and detection of the label, a quantitative measure of KAIl expression is obtained.

49. The method according to claim 48 wherein the label is radioactive, fluorescent or colourimetric.

50. The method according to any one of claims 41 to 44 wherein quantitative analysis of KAIl gene expression is conducted by determining whether KAIl mRNA level is elevated or reduced relative to a threshold value for KAIl mRNA level.

51. The method according to claim 50 wherein RT-PCR analysis is conducted to determine whether KAIl mRNA level is elevated or reduced.

52. The method according to claim 51 wherein RNA from tissue of defined type is subjected to RT-PCR amplification using a primer or primers complementary to at least a fragment of KAIl cDNA; the amplified cDNAs are then separately hybridized to a labelled probe complementary to at least a fragment of KAIl cDNA; the hybridized labels are detected to give a quantitative measure of KAIl mRNA in the tissue which is compared to a threshold value for the defined tissue type.

53. The method according to claim 52 wherein the label is radioactive, fluorescent or colourimetric.

4. A kit for use in a method according to any one of claims 41 to 53.