WO2012010904A1

WO2012010904A1 - Mammalian model for amplification of cancer stem cells

Info

Publication number: WO2012010904A1
Application number: PCT/GB2011/051390
Authority: WO
Inventors: Anne Collins
Original assignee: Procure Therapeutics Limited
Priority date: 2010-07-23
Filing date: 2011-07-21
Publication date: 2012-01-26
Also published as: GB201301321D0; GB2495059A

Abstract

The disclosure relates to a non-human mammalian model and a process to produce a non-human mammalian model for the analysis of cancer, in particular for the analysis of cancer stem cells, and the use of the model in the identification and validation of therapeutic agents useful in the treatment of cancer, in particular but not limited to, prostate cancer.

Description

MAMMALIAN MODEL FOR AMPLIFICATION OF CANCER STEM CELLS

Introduction The disclosure relates to a non-human mammalian model and a process to produce an non-human mammalian model for the analysis of cancer, in particular for the analysis of cancer stem cells, and the use of the model in the identification and validation of therapeutic agents useful in the treatment of cancer, in particular but not limited to, prostate cancer; and also the use of the non-human mammalian model as a host for the amplification followed by isolation of cancer stem cells, particularly prostate cancer stem cells such as scarce CD24⁺ prostate cancer stem cells .

Background The concept of a cancer stem cell within a more differentiated tumour mass, as an aberrant form of normal differentiation, is now gaining acceptance over the current stochastic model of cancer in which all tumour cells are equivalent both in growth and tumour-initiating capacity [Hamburger AW, Salmon SE: Primary bioassay of human tumour stem cells. Science 1977, 197: 461463; Pardal R, Clarke MF, Morrison SJ: Applying the principles of stem cell biology to cancer; Nat. Rev. Cancer 2003, 3: 895902.] For example, in leukaemia, the ability to initiate new tumour growth resides in a rare phenotypically distinct subset of tumour cells [Bonnet D, Dick J.E. Human acute myeloid leukemia is organized as a hierarchy that originates from a primitive hematopoietic cell. Nat. Med. 1997, 3: 730737] which is defined by the expression of CD34+CD38 surface antigens and have been termed leukemic stem cells (LSC).

Similar tumour-initiating cells have also been found in 'solid' cancers such as breast [AIHajj M, Wicha MS, BenitoHernandez A, Morrison SJ, Clarke MF: Prospective identification of tumorigenic breast cancer cells. Proc Natl Acad Sci U S A 2003, 100: 39833988], brain [Singh SK, Hawkins C, Clarke ID, Squire JA, Bayani J, Hide T, Henkelman RM, Cusimano MD, Dirks PB: Identification of human brain tumour initiating cells. Nature 2004, 432: 396401 ], lung [Kim CF, Jackson EL, Woolfenden AE, Lawrence S, Babar I., Vogel S, Crowley D, Bronson RT, Jacks T: Identification of bronchioalveolar stem cells in normal lung and lung cancer. Ce// 2005, 121 : 823-835] colon [O'Brien CA, Pollett A, Gallinger S, Dick JE: A human colon cancer cell capable of initiating tumour growth in immunodeficient mice. Nature 2007, 445: 1061 10; RicciVitiani L, Lombardi DG, Pilozzi E, Biffoni M, Todaro M, Peschle C, De Maria R: Identification and expansion of human colon cancer initiating cells. Nature 2007, 445: 1 1 1 1 15]; and gastric cancers [Houghton J, Stoicov C, Nomura S, Rogers AB, Carlson J, Li H, Cai X, Fox JG, Goldenring JR, Wang TC: Gastric cancer originating from bone marrow derived cells. Science 2004, 306: 15681571 ].

In WO2005/089043 we describe the isolation of prostate cancer stem cells which have been directly isolated from lymph node and prostate glands from a series of patient samples. These stem cells express markers that characterise the cells with stem cell properties. The following markers are typically expressed as prostate stem cell markers; human epithelial antigen (HEA), CD44, α₂β ^ήι and CD133 and represents around 0.1 % of the total cell tumour mass.

As is apparent the cancer stem cell comprises a small percentage of the total tumour cell mass. The analysis of the biology of the cancer stem cell is fraught with problems regarding isolation and characterization of cancer stem cells from tumour tissue. In addition the behaviour of the isolated cancer stem cell is difficult to assess when separated from bulk tumour cells. There is therefore a need to provide a validated in vivo non-human mammalian model for the analysis of cancer stem cell biology and also to test therapeutic agents specific for the cancer stem cell and/or bulk tumour.

Non-human mammalian models for the study of tumour initiation and growth are known in the art. For example, WO2008/061674 discloses a transgenic animal model that is modified in the Aph-1b gene, a cell membrane receptor that interacts with presenilin and nicastrin as a functional component of the λ secretase complex and is shown to be associated with initiation of tumour formation. WO2008/074880 discloses a transgenic animal model for the study of lung cancer initiation and progression. The model allows the site specific modification of the mouse genome to ablate genes considered involved in tumour formation. WO2008/153743 discloses a further mouse transgenic model for the study of lymphoma and the identification of genes that predispose the mouse to develop lymphoma. WO2008/021393 discloses a yet further mouse model for the study of hepatocarcinoma that uses RNA interference to identify genes involved in tumour formation. A problem associated with these models is that they do not study human tumour initiation and growth but rather the formation of murine tumours. This is unsatisfactory. Current models used for studying prostate cancer biology and drug evaluation generally consist of xenografts in immune-deficient mice of well established human prostate cancer cell lines that have been adapted to in vitro growth, for example LNCaP and PC- 3. Such models have been useful for testing new therapeutics but they have severe shortcomings: they are highly anaplastic representing the extreme end of advanced cancers and importantly they do not reflect the hierarchies observed in solid tumours. These limitations make it impossible to predict patient's response to anticancer drugs in the clinic. In view of this we have developed more relevant models illustrated by xenografting primary prostate cancer tissue into more permissive immune-deficient mice (Rag2^"/_Y C transgenics).

The primary tissues are engrafted into Rag2^{" _}y C ^{_ ~} mice with an initial incidence of 19%, which is improved with increased tumour grade. Furthermore, once a xenograft has been established tumour incidence is increased to 90% from each passage of tumour tissue and an incidence rate of 70% is achieved when mouse cells are depleted, and single tumour cells are grafted, giving a robust xenograft model. These xenografts have been shown to have a consistent genotype, in relation to its origin, through routine screening, with further characterisation revealing the majority of cells have a trans- amplifying phenotype (CD44⁺CK18⁺a₂3i ) and contain a minor population that express the luminal markers CD24 and androgen receptor. Importantly the stem cell markers CD133 and CD1 17 are also expressed. These xenograft models have been shown to express known stem cell targets and allow the measurement of tumour incidence, growth, recurrence and metastasis, making them invaluable tools in cancer research, in particular prostate cancer research.

Statements of Invention

According to an aspect of the invention there is provided a method for the amplification of cancer stem cells in a non-human mammalian model comprising the steps:

i) obtaining a tissue sample comprising cancerous cells from a human subject;

ii) forming a preparation comprising the transplanted tissue and a cell support substrate and transplanting the preparation into an immune deficient non-human mammal;

iii) providing conditions that allow the growth of one or more tumours in said non-human mammal; dissecting the tumour and depleting non-human mammalian cells from said tumour to provide an enriched human tumour cell sample wherein said enriched tumour cell sample comprises a population of cancer stem cells that express one or more cancer stem cell markers; optionally determining the expression of at least one or more cancer stem cell markers;

vi) forming a preparation comprising a cell support substrate and the enriched human tumour cell sample and transplanting the preparation into a second immune deficient non-human mammal to allow the growth of one or more tumours to further amplify the cancer stem cell population.

In a preferred method of the invention said tissue sample comprises cancerous prostate cells. In a preferred method of the invention said cancerous prostate cells are derived from a primary prostate tumour.

In an alternative preferred method of the invention said cancerous prostate cells are derived from a secondary prostate tumour.

In a preferred method of the invention said population of prostate cancer stem cells express at least one cancer stem cell marker selected from the group consisting of: CD133, CD1 17, CD44 and CD24. In a preferred method of the invention said prostate tumour is a Gleason score of 6 or higher; preferably a Gleason score of 7, and above (8 or 9, 10).

The prognosis of prostate cancer in a subject that has been diagnosed with prostate cancer is via an established system called the Gleason Grading System. The system attempts, with other variables, to determine the prognosis for prostate cancer suffers and to determine a suitable treatment regime for the subject. Gleason grading [1 -5] ranks the morphology of prostate cells in a biopsied sample obtained from a subject. The grading determines the most common cellular morphology followed by the second most common. For example, a Gleason grade of 1 would refer to cells of uniform size shape and would describe cells that are well differentiated. A Gleason grade of 4 would indicate irregular cell masses and structure, the cells of which are poorly differentiated. The pathologist would add these two grades together to obtain a Gleason score. The higher the Gleason score the more aggressive the cancer and therefore the poorer the prognosis. In a preferred method of the invention said immune deficient mammal is a rodent; preferably a mouse.

In a preferred method of the invention said mouse is a transgenic mouse deficient in the production of Natural Killer [NK] cells.

In a preferred method of the invention said transgenic deficient mouse is a Rag2^_/" yC ^~'^~ transgenic mouse.

Commonly the NOD/SCID mouse is used to determine human stem and tumour cell frequencies. However, this transplantation assay significantly underestimates the frequency of normal stem and human cancer cells with tumourigenic potential, due to NK cell activity. Mice lacking NK cells, such as the Rag2^v" yC transgenics, has resulted in the generation of xenografts (from primary prostate tumours) that have histopathological features similar to those normally encountered in the clinic. Importantly, the xenografts are hierarchically organised.

In a preferred method of the invention said cell support substrate comprises collagen.

Matrigel is an example of a collagen based cell support substrate and is a preferred substrate for production of the non-human mammalian model.

In a preferred method of the invention said transplantation is sub-cutaneous, subperitoneal, or kidney capsule located.

In an alternative preferred method of the invention said transplantation is orthotopic. In a preferred method of the invention said non-human mammal is administered simultaneously or sequentially an agent that promotes or enhances the initiation and/or growth of said cancerous cells from said human subject. The agent, for example a hormone, can be administered prior to transplantation, simultaneously with transplantation or sequentially post-transplantation.

In a preferred method of the invention said agent is a hormone.

In a preferred method of the invention said hormone is dihydrotestosterone.

In a preferred method of the invention said cancerous human cells are transfected with a vector that encodes a reporter molecule operably linked to a transcription promoter.

In a preferred method of the invention said reporter is a chemiluminescent reporter, for example a fluorescent or luciferase reporter.

The analysis of promoter activity in a tissue can be conveniently monitored by fusing a promoter to a nucleic acid that encodes a "reporter" protein or polypeptide. Examples are well known in the art and include enzymes such as β glucuronidase. Reporters that are proteinaceous fluorophores are also known in the art. Green fluorescent protein, GFP, is a fluorescent protein isolated from coelenterates, such as the Pacific jellyfish, Aequoria victoria. Its role is to transduce, by energy transfer, the blue chemiluminescence of another protein, aequorin, into green fluorescent light. GFP can function as a protein tag, as it tolerates N- and C-terminal fusions to a broad variety of proteins many of which have been shown to retain native function. Most often it is used in the form of enhanced GFP in which codon usage is adapted to the human code. Other proteinaceous fluorophores include yellow, red and blue fluorescent proteins. These are commercially available from, for example Clontech (www.clontech.com). A yet further example is firefly luciferase. The use of fluorescent reporters allows the in vivo analysis of tumour cell initiation, growth and metastasis.

According to a further aspect of the invention there is provided a non-human mammalian model obtained or obtainable by the method according to the invention.

According to a further aspect of the invention there is provided a method for testing the efficacy of one or more anti-cancer agents comprising:

i) providing a non-human mammal according to the invention;

ii) administering an anti-cancer agent[s] to be tested; iii) determining the effect of the agent[s] on tumour initiation and/or growth of cancer cells and/or cancer stem cells and optionally comparing the effect with a non-human mammal according to the invention that has not been administered the agent[s].

i) obtaining an enriched human tumour cell sample obtained according to the invention;

ii) contacting the sample with an agent[s] to be tested ex vivo;

ii) transplanting the treated sample into a second non-human mammal; and v) determining the effect of the agent[s] on tumour initiation and/or growth of cancer cells and/or cancer stem cells and optionally comparing the effect with an enriched human tumour cell sample that has not been contacted with said agent[s].

In a preferred method of the invention said agent[s] is a small interfering RNA.

A technique to specifically ablate gene function is through the introduction of double stranded RNA, also referred to as small inhibitory or interfering RNA (siRNA), into a cell which results in the destruction of mRNA complementary to the sequence included in the siRNA molecule. The siRNA molecule comprises two complementary strands of RNA (a sense strand and an antisense strand) annealed to each other to form a double stranded RNA molecule. The siRNA molecule is typically derived from exons of the gene which is to be ablated. The mechanism of RNA interference is being elucidated. Many organisms respond to the presence of double stranded RNA by activating a cascade that leads to the formation of siRNA. The presence of double stranded RNA activates a protein complex comprising RNase III which processes the double stranded RNA into smaller fragments (siRNAs, approximately 21 -29 nucleotides in length) which become part of a ribonucleoprotein complex. The siRNA acts as a guide for the RNase complex to cleave mRNA complementary to the antisense strand of the siRNA thereby resulting in destruction of the mRNA.

In an alternative preferred method of the invention said agent[s] is an antibody. "Antibody" includes monoclonal and polyclonal antibodies or immunoglobulins and also antibody fragments that include antibody binding fragments [e.g. Fab, Fab₂, F(ab')₂, Fv, Fc, Fd, scFvs]. "Antibody" also includes chimeric, humanized and human antibodies In a further preferred method of the invention said agent is a small organic molecule.

In a preferred embodiment of the invention said small organic molecule is a peptide, preferably a modified peptide [e.g. a cyclic peptide]. According to an aspect of the invention there is provided an immune deficient non- human mammal comprising an enriched human tumour cell sample according to the invention.

In a preferred embodiment of the invention said immune deficient mammal is a rodent; preferably a mouse.

In a preferred embodiment of the invention said mouse is a transgenic mouse deficient in the production of Natural Killer [NK] cells. In a preferred embodiment of the invention said transgenic deficient mouse is a Rag2^_/" yC ^_/~ transgenic mouse.

In a preferred embodiment of the invention said enriched human tumour cell sample is derived from a prostate tumour.

In a preferred embodiment of the invention said prostate tumour is a metastatic prostate tumour.

In a preferred embodiment of the invention said prostate tumour is a Gleason score of 6 or higher; preferably a Gleason score of 7, and above (8 or 9, 1 0).

Throughout the description and claims of this specification, the words "comprise" and "contain" and variations of the words, for example "comprising" and "comprises", means "including but not limited to", and is not intended to (and does not) exclude other moieties, additives, components, integers or steps. Throughout the description and claims of this specification, the singular encompasses the plural unless the context otherwise requires. In particular, where the indefinite article is used, the specification is to be understood as contemplating plurality as well as singularity, unless the context requires otherwise.

Features, integers, characteristics, compounds, chemical moieties or groups described in conjunction with a particular aspect, embodiment or example of the invention are to be understood to be applicable to any other aspect, embodiment or example described herein unless incompatible therewith.

An embodiment of the invention will now be described by example only and with reference to the following figures:

Figure 1 illustrates a genetic fingerprint of xenograft H016 at passage 4. STRs of nine loci are compared with patient lymphocytes;

Figure 2 illustrates the histology of xenografts derived from human prostate cancer tissue biopsies implanted subcutaneously: H&E of a xenograft derived from patient Y042 (A). Tissue section of Y042 stained with pan cytokeratin. Note tumour cells invading through the mouse liver stroma (B). Macroscopic image of liver metastases, from a subcutaneous graft of Y042 (C);

Figure 3 illustrates the characterisation of xenografts from patient Y042 by flow cytometry. Expression of CD44⁺ (A),AR7CD24⁺ (yellow population), AR⁺ CD24^"(red population), CD247AR ⁺⁺ (blue population) (B), CK18⁺ (C) and CD133⁺ (D);

Figure 4 illustrates the histology of xenografts derived from 10⁵ Lin^" cells implanted subcutaneously: H&E of a xenograft derived from patient H016 (A). Small glandular structures are apparent and there is invasion into skeletal muscle (arrow). H & E of a xenograft derived from patient Y019 (B). Note invasion of tumour cells into blood vessel (arrow);

Figure 5 illustrates differential expression by qRT-PCR. Validation of differential gene expression in stem cells versus committed basal cells (A) and in cancer versus benign cells (B) as detected by Affymetrix microarray or qRT-PCR; and Figure 6 illustrates characterisation of pappalysin expression on xenografts by flow cytometry: Isotype control (A), H016 (B), Y019 (C) and PEY042 (D).

Materials and Methods Genetic Finger Printing of Xenograft.

Xenografts are routinely genotyped using the STR (Short Tandem Repeats) of nine loci (8 STR) and Amelogenin (Powerplex 1 .2 System; Promega), and compared with patient blood lymphocytes and xenograft passages. Briefly, tumour tissue is digested overnight in collagenase (200IU/ml) at 37 ^<Ό, followed by trypsin digest for 30 mins. The cells are resuspended in MACS buffer and mouse, blood lineage cells are depleted using mouse lineage specific microbeads (Milteny Biotec). DNA is then extracted, using the Qiagen DNeasy blood and tissue kit, and microsatellites are analysed according to the manufacture's instructions (Promega Powerplex 1.2 System) .

Xenograft Histology

Tumour tissue was fixed in formalin and subsequently embedded in paraffin blocks. Tissue sections were subsequently dewaxed followed by a haematoxylin eosin stain or antigen unmasked, using citrate buffer, for anti pan-cytokeratin staining. Briefly, sections were rehydrated, non specific sites bloked and then incubated with anti-pan cytokeratin nonoclonal antibody. Binding was visualised by incubating with a secondary antibody conjuaged to horse radish peroxidase and developed with 3,3^'- diaminodbenzidine (DAB).

Xenograft Flow Cytometry

Briefly, tumour tissue is digested overnight in collagenase (200IU/ml) at 37 ^<Ό, followed by a trypsin digest for 30 mins. The cells are resuspended in MACS buffer and mouse, blood lineage cells are depleted using mouse lineage specific microbeads (Milteny Biotec). Cells are then incubated with monoclonal antibodies to CD44, CD24, CK5/14, AR, CD133 or CD1 17 either directly conjugated to fluorescent molecules or via a direclty conjugated secondary antibody. At least 250,000 event are counted on a flow cytometer. Histology of Xenoqrafted Enriched Human Cells

Tumour tissue was fixed in formalin and subsequently embedded in paraffin blocks. Tissue sections were subsequently dewaxed followed by a haematoxylin eosin stain

Microarray Gene Expression Profiling

Expression profiling studies that lead to the identification of pappalysin as a differentially expressed gene in prostate cancer stem cells have been described previously (Birnie et al, 2008). Briefly, RNA was extracted from selected populations of a2 i high/CD133+ and a231 low/CD133- primary prostate epithelial cells from malignant and non-malignant cultures. Gene expression profiling was carried out using Affymetrix HGU133plus2 microarrays. RNA samples were labelled, hybridised to the array and scanned according to the manufacturers standard protocols. The gene expression profiles of a231 high/CD133+ and a231 low/CD133- prostate cancer cells were compared with profiles from a231 high/CD133+ and a2p1 low/CD133- benign prostate epithelial cells to identify differentially expressed genes. gRT-PCR

Reverse transcription was carried out on 50ng of fractionated cell RNA to generate cDNA. Real Time PCR was carried out using the Taqman gene expression system (Applied Biosystems, Warrington, UK) according to the manufacturer's protocol with the exception that a reduced total reaction volume was used. For the validation of microarray differential expression results a final volume of 25μΙ was used. All reactions were carried out in triplicate in 96-well PCR plates on an ABI Prism 7300 sequence detection system (Applied Biosystems). Standard thermal cycling conditions included a hot start of 5 minutes at 50 ^<Ό, 10 minutes at 95°C, followed by up to 50 cycles of: 95 °C 15 s, 60°C for 1 minute. Data analysis was carried out using ABI SDS software and Microsoft Excel. For the initial validation of microarray differential expression results mRNA expression levels were normalised relative to the geometric mean of three endogenous control genes (GAPDH, ITGB1 and PPT1A). Differential expression was calculated by the AACt method. For subsequent expression measurement in cell lines GAPDH and RPLP0 were used as endogenous control genes and all expression values normalised to the mean of these two genes. For basic expression measurement values were expressed as the ratio of test gene Ct values: mean control Ct value, results are the mean of three independent experiments. Relative expression was calculated with respect to PNT2 cells. Pappalvsin Expression

Pappalysin protein expression was measured in xenografts by staining with an antibody against PAPP-A and analysing by flow cytometry. In xenografts pappalysin expression was measured simultaneously in CD133+ stem cells by co-staining with anti-PAPP-A and an antibody against CD133. Cells were trypsinised for 30 minutes at 37°C and washed in MACS buffer (2mM EDTA, 0.5% FCS, PBS). Cells were then incubated in MACS blocking buffer (PBS, 0.3% BSA heat treated at 80°C for 5 minutes) for 10 minutes, then incubated with CD133-APC antibody (clone 293C3, Miltenyi Biotec) according to the manufacturer's instructions plus anti-PAPP-A (ab59088, Abeam, Cambridge, UK) for 60 min. Following incubation, cells were washed and secondary antibody (anti-Rabbit-Alexa488, Invitrogen, Paisley, UK) applied for 20 min. Finally, cells were rinsed in MACS buffer then resuspended in PBS. Samples were analysed on a DakoCytomation CyAn ADP instrument (Dako UK Ltd, Cambridgeshire, UK). FACS results were analysed with Summit Software, v4.3 (Dako UK Ltd, Cambridgeshire, UK). Gates were set to remove debris and doublets based on forward/side scatter and pulse width, respectively.

Example 1

Prostate tissue biopsies are obtained from men undergoing radical prostatectomy and transurethral resection (TUR) for prostate cancer, with informed consent. Some patients undergo neoadjuvant therapy prior to TUR. Approximately 25 tumour samples are obtained annually. The specimens are examined, sectioned and selected by pathologists for histological analysis prior to xenografting. Tissue pieces are grafted subcutaneously into recipient Rag2^{" "}y C ^{_ "} mice, together with 90 day hormone release pellets (DHT), optional. Xenografts are continually being generated from tissue biopsies, resulting in several lines that are now available for preclinical testing (Table 1 ). Tumourigenic incidence is approximately 19% (Table 2) (at least 3x higher than that achieved with NK⁺ mice), and is significantly higher (64%) from high grade tumours. Example 2

Xenografts are routinely genotyped using the STR (Short Tandem Repeats) of nine loci (8 STR) and Amelogenin (Powerplex 1 .2 System; Promega), and compared with patient blood lymphocytes (Figure 1 A) and xenograft passages (Figure 1 B). We have shown that xenografts maintain the genotpye of origin after successive passages in vivo. Xenografts are maintained continuously in the Rag2^"/"y C ^"y" mouse, but at each passage tumour cells are isolated, mouse cells are depleted (Lin ) and single cell suspensions are frozen in liquid nitrogen. Tumour incidence is approximately 70% from frozen stocks and possible genetic changes are routinely monitored using the Promega kit described above.

Example 3

Xenografts are graded by a consultant pathologist and Figure 2 is an example of a xenograft derived from patient Y042. In general, the resultant xenografts are poorly differentiated, invasive tumours (metastases are often observed in the liver of NK null mice). Characteristics of prostate cancer histology, such as small glandular structures, are observed.

Lineage depletion of mouse cells from tumours yields high purity populations of approximately 1 0000 CD44⁺ and 10000 CD24⁺AR⁺ human cells. Characterisation of these xenografts, by flow cytometry, has revealed that the majority of cells express CD44 (a basal cell marker; Figure 3A). Interestingly, CD24 and androgen receptor (luminal cell markers) are also expressed (Figure 3B), but unlike prostate cancers (in situ) they are a minority population, despite supplementing mice with DHT pellets. Although CD44 is expressed throughout, the high molecular weight keratins (CK5 and CK14) are only expressed in a rare population (<1 %; Figure 3C), which suggests that the majority of tumour cells are amplifying cells (CD447CK187a₂3i^h'^9h) rather than committed basal cells given their high mitotic index. The stem cell markers CD1 33 and CD1 1 7 (c-kit) are also expressed in all xenografts tested. Both are rare populations (0.1 % for CD1 33 and 0.02% for CD1 17). The results for CD1 33 are depicted in Figure 3D. Example 4

The results of our characterisation suggest that amplification of intermediate (basal- luminal phenotype) cells occurs in vivo which is similar to in vitro culture of normal and tumour prostate cells. After initial grafting, tumour incidence is high (>90%) from each tumour tissue passage, and approximately 70% incidence is achieved following injection of a Lin^" population (single cell suspension of tumour cells following mouse cell depletion). The tumours are identical to the parental line as all phenotypes are observed in similar proportions. Figure 4 is an example of tumours generated from 10⁵ Lin^" cells. Example 5

The generation of a cancer stem cell gene expression signature by microarray analysis has been reported previously (Birnie et al., 2008). When gene expression profiles from stem cells and committed basal cells isolated from primary cultures of benign and malignant prostate epithelial cells were compared, pappalysin (PAPP-A) showed the largest increase in gene expression in stem cells relative to committed basal cells (3.83 fold). PAPP-A was also found to be upregulated in cancer relative to benign cells; 3.26 fold if the comparison was made across all samples and 2.48 fold if the comparison was made in stem cells only. These relative overexpression values were highly statistically significant, even in a heterogeneous patient population. These results were confirmed by qRT-PCR, where comparison of PAPP-A expression in stem cells relative to committed basal cells showed an increase in PAPP-A expression comparable to that observed by microarray. Similarly, comparison of expression in cancer cells relative to benign controls showed consistent results between microarray and qRT-PCR experiments (Figure 5). These results identified pappalysin as a putative target in cancer stem cells.

As described above, the transplantation of human tissue xenografts subcutaneously into recipient Rag2^"/_7 C ^{_ ~} mice, is a useful tool for pre-clinical testing of prostate cancer therapeutics. The availability of these models enables us to validate potential novel therapeutics for this and other identified targets. Analysis of several xenografts by flow cytometry has shown significant levels of surface PAPP-A expression. Figure 6 shows PAPP-A surface expression for isotype control (A), xenograft H016 (B), xenograft Y019 (C) and xenograft PE042 (D). These findings show that we have developed a mouse model which expresses a known target that has been shown to be up-regulated in cancer stem cells in previous experiments, thus our xenograft models are consistent with primary samples from patients.

Claims

1 A method for the amplification of cancer stem cells in a non-human mammalian model comprising the steps:

i) obtaining a tissue sample comprising cancerous cells from a human subject;

ii) forming a preparation comprising the transplanted tissue and a cell support substrate and transplanting the preparation into an immune deficient non-human mammal; iii) providing conditions that allow the growth of one or more tumours in said non-human mammal;

iv) dissecting the tumour and depleting non-human mammalian cells from said tumour to provide an enriched human tumour cell sample wherein said enriched tumour cell sample comprises a population of cancer stem cells that express one or more cancer stem cell markers; optionally

v) forming a preparation comprising a cell support substrate and the enriched human tumour cell sample and transplanting the preparation into a second immune deficient non-human mammal to allow the growth of one or more tumours to further amplify the cancer stem cell population.

2. The method according to claim 1 wherein said tissue sample comprises cancerous prostate cells.

3. The method according to claim 2 wherein said cancerous prostate cells are derived from a primary prostate tumour.

4. The method according to claim 2 wherein In an alternative said cancerous prostate cells are derived from a secondary prostate tumour.

5. The method according to any one of claims 2-4 whererin said population of prostate cancer stem cells express at least one cancer stem cell marker selected from the group consisting of: CD133, CD1 17, CD44 and CD24.

6. The method according to claim 5 wherein said prostate cancer stem cells express CD133 and CD44.

7. The method according to any one of claims 2-6 wherein said prostate tumour is a Gleason score of 6 or higher.

8. The method according to any one of claims 1 -8 wherein said non-human immune deficient mammal is a rodent.

9. The method according to claim 8 wherein said non-human immune deficient mammal is a mouse.

10. The method according to claim 9 wherein said mouse is a transgenic mouse deficient in the production of Natural Killer [NK] cells.

1 1 . The method according to claim 10 wherein said transgenic deficient mouse is a Rag2^{_ "} yC ^"/_ transgenic mouse.

12. The method according to any one of claims 1 -1 1 wherein said cell support substrate comprises collagen.

13. The method according to any one of claims 1 -12 wherein said transplantation is sub-cutaneous, sub-peritoneal, or kidney capsule located.

14. The method according to any one of claims 1 -12 wherein said transplantation is orthotopic.

15. The method according to any one of claims 1 -14 wherein said non-human mammal is administered simultaneously or sequentially an agent that promotes or enhances the initiation and/or growth of said cancerous cells from said human subject.

16. The method according to claim 15 wherein said agent is a hormone.

The method according to claim 16 wherein said hormone is dihydrotestosterone.

18. The method according to any one of claims 1 -17 wherein said cancerous human cells are transfected with a vector that encodes a reporter molecule operably linked to a transcription promoter.

19. A non-human mammalian model obtained or obtainable by the method according to any one of claims 1 -18.

A method for testing the efficacy of one or more anti-cancer agents comprising:

i) providing a non-human mammal according to claim 19; ii) administering an anti-cancer agent[s] to be tested;

iii) determining the effect of the agent[s] on tumour initiation and/or growth of cancer cells and/or cancer stem cells and optionally comparing the effect with a non-human mammal according to the invention that has not been administered the agent[s].

21 . A method for testing the efficacy of one or more anti-cancer agents comprising:

i) obtaining an enriched human tumour cell sample obtained according to claim 1 [iv]; ii) contacting the sample with an agent[s] to be tested ex vivo;

iii) transplanting the treated sample into a second non-human mammal; and iv) determining the effect of the agent[s] on tumour initiation and/or growth of cancer cells and/or cancer stem cells and optionally comparing the effect with an enriched human tumour cell sample that has not been contacted with said agent[s].

The method according to claim 21 wherein said agent[s] is a small interfering

23. The method according to claim 22 wherein In an alternative preferred method of the invention said agent[s] is an antibody or active binding fragment of an antibody.

24. The method according to claim 22 wherein said agent is a small organic molecule.

25. The method according to claim 24 wherein said small organic molecule is a peptide.

26. An immune deficient non-human mammal comprising an enriched human tumour cell sample according to claim 1 [iv].

27. The immune deficient non-human mammal according to claim 26 wherein said immune deficient mammal is a rodent; preferably a mouse.

28. The immune deficient non-human mammal according to claim 27 wherein said mouse is a transgenic mouse deficient in the production of Natural Killer [NK] cells.

29. The immune deficient non-human mammal according to claim 28 wherein said transgenic deficient mouse is a Rag2^_/" yC ^_/~ transgenic mouse.

30. The immune deficient non-human mammal according to any one of claims 26-29 wherein said enriched human tumour cell sample is derived from a prostate tumour.

31 . The immune deficient non-human mammal according to claim 30 wherein said prostate tumour is a metastatic prostate tumour.