Animal models of human prostate cancer progression

ES 2 290 964 T3

DESCRIPTION

Animal models of the evolution of human prostate cancer.

Throughout the present application, various publications are referenced in parentheses. Full citations to these publications can be found at the end of the specification immediately prior to the claims.

Background of the invention

Prostate cancer is the most common cause of cancer in men. In 1996, 317,000 new cases of prostate adenocarcinoma were diagnosed and more than 41,400 men died of the disease (Karp et al., 1996). Only lung cancer has a higher mortality. A man's chance of developing invasive prostate cancer in his lifetime is 1 in 6 or 15.4%. By the age of 50, a man has a 42% chance of developing prostate cancer and a 2.9% chance of dying from the disease. Although advances have been made in the early diagnosis and treatment of locally confined tumors, prostate cancer is incurable once it has metastasized. Patients with metastatic prostate cancer on hormone therapy will eventually develop an androgen-resistant (androgen-independent) state that will lead to disease progression and death.

The leading cause of prostate cancer mortality and mortality is the result of androgen-independent metastatic tumor growth. As a result, there is great interest in defining the molecular basis for advanced-stage disease in the hope that these ideas may improve the therapeutic options for these patients. However, evolution in this area has been difficult for numerous reasons. For example, the availability of prostate tissue for molecular studies is limited because most prostate tumors are small. Furthermore, there is tremendous heterogeneity in surgical prostatectomy tumor samples, prostate cancer explants are difficult to grow in vitro in a reducible manner, and there are a limited number of immortalized prostate cancer cell lines.

There is, therefore, an interest in finding alternative procedures that allow the stable growth of prostate cancer tissue, which in turn would allow investigation of the evolution of prostate cancer in vivo, provide a stable supply of prostate cancer tissue, prostate and would provide a model of metastatic expansion of prostate cancer that would exactly simulate or mimic the biology of the disease.

There is also a need for more reliable procedures and prognostic and informative staging in the treatment of advanced prostate cancer. Clinically staging prostate tumors are based on rectal examination to determine whether the tumor remains within the confines of the prostatic capsule (locally confined) or extends beyond (locally advanced) in combination with serum PSA determinations. and ultrasound-guided transrectal biopsies. However, none of these techniques has proven to be reliable in predicting the course of the disease.

The main sites of prostate cancer metastasis are the regional lymph nodes and bones. Bone metastasis occurs at hematopoietically active red bone marrow sites, including the lumbar spine, ribs, pelvis, proximal long bones, sternum, and skull. The bone metastasis of prostate cancer differs from that of other tumors that frequently colonize the bone and are characterized by a net gain in bone formation (osteoblastic) rather than the predominant resorption in bone metastasis of the breast cancer and melanoma.

Until recently, bone metastasis was believed to be a late stage in the course of the disease. However, the recent development of highly sensitive techniques (such as RT-PCR for prostate-specific genes) to detect prostate cancer cells has revised this idea. Prostate cancer cells have been detected in the peripheral blood and bone marrow of patients with advanced stage disease using RT-PCR analysis for PSA mRNA (Ghossein et al., 1955; Seiden et al., 1994; Wood et al., 1994; Katz et al., 1994) or immunomagnetic bead selection for PSA protein (Brandt et al., 1996). When positive, these tests show that prostate cancer cells account for approximately 0.1-1.0% of circulating blood cells. Furthermore, it is now evident that a small number of prostate cancer cells circulate in the peripheral blood and lodge in the bone marrow even in patients with early stage, low-risk disease (Olsson et al., 1997; Deguchi et al. , 1997; Katz et al., 1996). Interestingly, these cells tend to disappear in most patients after radical prostatectomy (Melchior et al., 1997). These results suggest that the main tumor site is a constant source for seeding the marrow, and that only a small subset of these cells has the capacity to grow in a metastatic lesion. This concept is consistent with estimates in animal models for other tumor types that only about 1 in 10,000 circulating cancer cells are capable of accommodating and productively colonizing other organs (Fidler et al., 1990).

The factors involved in the evolution of advanced prostate cancer to bone metastasis are poorly defined. All anatomical, local bone marrow, and tumor cell factors are believed to play a role. Baston described the general vertebral venous system consisting of a network of valveless longitudinal veins running parallel to the spine and the extensive, direct anastomosis with rib veins, pelvis

ES 2 290 964 T3 and brain (Bastón, 1942). Prostate cancer cells that enter the prostate veins can be transported through this plexus directly to these organs without entering the inferior vena cava on their way to the lungs. This putative mechanism of metastasis both by clinical documentation of prostate cancer metastasis models compared to other tumors and by animal models in which occlusion of the inferior vena cava during injection into the tail vein of tumor cells increased the frequency of vertebral metastasis (Nishijama et al., 1992; Coman and DeLong, 1951).

Although vascular anatomy is an essential component of the spread of prostate cancer to bone, the selective pattern of all skeletal metastasis cannot be fully explained. Bone, which represents between 5 and 10% of cardiac output, is a frequent metastatic point that would be expected based on blood-circulation criteria (Berettoni and Carter, 1986). The bone marrow consists of two clearly identifiable components: the hematopoietic cells that comprise most of the cellular elements, and the stromal component that is made up of highly vascular connective tissue. Hematopoietic cells are transient in the bone marrow; during maturation they move in the bloodstream. The stroma, however, remains and serves as the structure on which hematopoietic cells can differentiate and mature. One of the important factors in prostate cancer cells that stop at these points is probably their adherence to the stroma of the bone marrow. It has been shown both in vitro and in vivo that tumor cells will preferentially adhere to stromal cells of the organs to which they metastasize (Haq et al., 1992; Netland and Zetter, 1985; Zetter et al., 1992) . When rat prostate cancer cells (MatLyLu) are injected into the left ventricle of clonal rats, metastases develop in the vertebral body; these metastases are then harvested, disintegrated, and reinjected. Cell lines created after 6 similar passages through animals strongly and preferentially adhered to bone marrow stroma and endothelial cells (Haq et al., 1992). A similar strategy has increased the frequency of metastasis of the prostate cancer cell line LNCaP in immunocompromised mice (Thalmann et al., 1994).

It is critical that appropriate in vivo models for prostate cancer bone metastasis are developed to more fully explore the mechanistic aspects of this process. To date, most work in this area has focused on three human prostate cancer cell lines, PC-3, DU-145, and LNCaP (Lee et al., 1993). All three develop subcutaneous nodules in immunocompromised mice, and sublines with variable metastatic properties have been derived (Shervin et al., 1988,1989; Wang and Sterarns, 1991; Kozlowski et al., 1988). However, none of these sublines have been shown to reproduce reproducibly in osteoblastic lesions typical of prostate cancer. A major limitation of the DU-145 and PC-3 cell lines is the lack of prostate specific antigen (PSA) and androgen receptor (AR) expression (Kaighn et al., 1979; Gleave et al., 1992 ), which increases with respect to relevance to clinical prostate cancer. The LNCaP cell line is androgen sensitive and expresses PSA, but contains a mutation in the androgen receptor that alters the specificity of the ligand.

The introduction of a human prostate cancer xenograft into an immunocompromised mouse is described elsewhere (Pretlow, et al., 1993; Van Weerden et al., 1996; Lubaroff et al., 1995). Sato et al. (1996) describe subcutaneous human prostate cancer xenografts in SCID mice using a prostate cancer cell line. Ellis et al. (1996) describe BALB / c-nu / nu (athymic) mice having human prostate cancer xenografts from a patient diagnosed with stage D1 adenocarcinoma of the prostate. Nagabhushan et al. (1996) describe the CWR22 prostate cancer model in nude mice. Hsieh et al. (1993) describe the development of a model for the evolution of human prostate cancer. Wu et al. (1994) demonstrate that co-injection of LNCaP cells with bone fibroblasts leads to the evolution of prostate cancer in vivo. Kjonniksen et al. (1994) describe the validity and utility of human tumor models created by intratibial inoculation of cells in hairless rats.

Summary of the invention

In a first aspect, the invention provides a method for simulating the evolution of human prostate cancer from primary tumor formation to micrometastasis in an animal model comprising (a) generating a human prostate cancer xenotransplantation in an immunodeficient SCID mouse. locally implanting advanced or metastatic prostate cancer tissue, or a cell suspension thereof from man, into an immunodeficient SCID mouse; and (b) allowing the xenotransplantation to develop for a time sufficient to allow detection of prostate cancer cells within and external to the implant site in the immunodeficient SCID mouse thereby simulating the evolution of human prostate cancer from the formation of the primary tumor to micrometastasis in the animal model.

In a second aspect, The invention provides a method for simulating the evolution of osteoblastic bone metastasis in human prostate cancer in a mouse model comprising (a) injecting into the tibial bone marrow cavity of an immunodeficient SCID mouse cancer cell of prostate cells prepared from a prostate cancer xenotransplantation generated by implanting into an immunodeficient mouse SCID locally advanced or metastatic human prostate cancer tissue or a cell suspension thereof; and (b) allowing the injected cells to develop and form an osteoblastic bone lesion that stimulates the evolution of osteoblastic bone metastasis in human prostate cancer in the mouse model.

In a third aspect, the invention provides a method for identifying a gene that minimizes the evolution of human prostate cancer, comprising the steps of (a) introducing the gene into an immunodeficient SCID mouse.

ES 2 290 964 T3 bearing a human prostate cancer xenotransplant generated by implanting locally advanced or metastatic prostate cancer tissue or a cell suspension thereof in the present immunodeficient SCID mouse, in which the xenotransplantation simulates the clinical transition from sensitive disease androgen even independent of androgen; (b) transducing the xenograft cells with the gene in vivo; (c) evaluating the presence of micrometastases in the present immunodeficient SCID mouse by detecting prostate cancer cells in peripheral blood, bone marrow, lymph nodes or other sites distant from the xenotransplantation site; in which the effect of the gene on the evolution of micrometastatic prostate cancer is determined is determined in relation to a reference immunodeficient SCID mouse that carries the human prostate xenotransplantation and in which the gene has not been introduced, so the Reference mouse xenotransplantation is not transduced; and (d) selecting a gene that facilitates the evolution of prostate cancer in the present mouse compared to the evolution of the cancer in the control mouse.

In a fourth aspect, the invention provides a method for identifying a gene that facilitates the evolution of human prostate cancer, comprising the steps of (a) introducing the gene to a present immunodeficient SCID mouse bearing a human prostate cancer xenotransplantation. generated by implanting locally advanced or metastatic prostate cancer tissue or a cell suspension thereof in a present immunodeficient SCID mouse, wherein xenotransplantation simulates the clinical transition from androgen-sensitive to androgen-independent disease; (b) transducing the xenograft cells with the gene in vivo; (c) evaluating the presence of micrometastases in the present immunodeficient SCID mouse by detecting prostate cancer cells in peripheral blood, bone marrow, lymph nodes, or other sites distant from the xenotransplantation site; in which the effect of the gene on the evolution of micrometastatic prostate cancer is determined in relation to a reference immunodeficient SCID mouse that carries the human prostate xenotransplantation and in which the gene has not been introduced so that the mouse xenotransplantation reference is not transduced; and d) selecting a gene that facilitates the evolution of prostate cancer in the present mouse relative to the evolution of prostate cancer in the reference mouse.

In a fifth aspect, the invention provides a method for evaluating the effect of a composition or treatment on prostate cancer, comprising (a) (a) providing an immunodeficient SCID mouse comprising a human prostate cancer xenotransplantation from tissue of locally advanced or metastatic prostate cancer or a cell suspension thereof; (b) subjecting the mouse to the composition or treatment; and (c) determining the effect of the composition or treatment on the growth of the xenograft in said mouse.

In a sixth aspect, the invention provides a method of producing a hybrid mouse model of human prostate cancer comprising (a) implanting locally advanced or metastatic prostate cancer tissue or a cell suspension thereof from a human into a first immunocompromised SCID mouse at a site that vascularizes implanted human prostate cancer tissue; (b) determining whether human prostate cancer tissue or cell suspension thereof results in the mouse model that simulates the clinical transition from androgen-dependent to androgen-independent disease; and (c) selecting the hybrid mouse model that simulates the clinical transition from androgen-dependent to androgen-independent disease.

In a seventh aspect, the invention provides an assay for evaluating the effect of a human prostate cancer treatment comprising (a) applying the treatment to an immunodeficient SCID mouse bearing a subcutaneous human prostate cancer xenograft generated by implanting cancer tissue. locally advanced or metastatic prostate or a cell suspension thereof from a human in the immunodeficient SCID mouse; (b) determine the effect of the treatment on the growth of the xenotransplantation in said mouse.

In an eighth aspect, the invention provides an assay for evaluating the effect of a gene of interest for human prostate cancer comprising (a) introducing the gene into an immunodeficient SCID mouse bearing a human prostate cancer xenograft generated by the locally advanced or metastatic prostate cancer tissue by implanting or a cell suspension thereof in the immunodeficient SCID mouse; (b) transducing the xenograft cells with the gene in vivo; (c) evaluating the presence of metastases in the immunodeficient SCID mouse by detecting prostate cancer cells in peripheral blood, bone marrow, lymph nodes, or other sites distant from the xenotransplantation site; in which the effect of the gene on the evolution of micrometastatic prostate cancer in relation to a reference immunodeficient SCID mouse carrying a human prostate xenotransplant generated with a non-transduced subset of the xenotransplantation cells is determined.

Preferred embodiments of the first to eighth aspects are published in claims 2 to 7, 9, 13 to 18, 20 to 23 and 26 to 29.

The model procedures described herein provide a system to study the molecular biology of prostate cancer, evaluating the influence that various genes and therapeutic compounds present in different stages of the evolution of the disease, evaluating the metastatic potential of cancer cells. prostate, and designing specific therapeutic regimens for the patient.

Brief description of the figures

Fig. 1. Molecular analysis of prostate cancer xenografts for human DNA content and expression of prostate specific antigen (PSA). Fig. 1A: DNA-PCR analysis of genomic DNA isolated from xenografts using specific primers for the human ^ -globin gene. Each sample presented was obtained from

ES 2 290 964 T3 late pass xenotransplantation. The LAPC-5 sample was obtained in which 4 when the human tumor was overgrown by a tumor of murine origin. The two LAPC-4 samples were obtained from the androgen-dependent ("ad") and androgen-independent ("ai") sublines. Fig. 1B: RT-PCR analysis of whole RNA using specific primers for human PSA and human or murine jd-actin ("% human cells") refers to the percentage of LNCaP cells that are diluted 10⁵ mouse cells). A serial dilution of human prostate cancer LNCaP cells in murine NIH 3T3 cells is presented in the left part of the figure, with the percentage of LNCaP cells varying from 100% to 0.0%. The results of the three LAPC xenotransplants are presented on the right.

Fig. 2. Photographs of immunohistochemical analysis of the LAPC-4 xenograft, showing the expression of PSA. Paraffin sections of formalin-fixed tissue from the original tumor sample obtained at the time of surgery (top row) and LAPC-4 xenotransplantation (bottom row) were stained with hematoxylin and eosin (left), an antibody to reference (middle) and an antibody specific for human PSA (right).

Fig. 3. Bar graphs showing the androgen sensitivity of LAPC-3 and LAPC-4 xenografts in vivo. Equally sized implants from LAPC-3 and LAPC-4 xenografts were simultaneously passaged into male or female mice and tumor formation was examined weekly.

Fig. 4. Regression and regrowth of LAPC-4 tumors after castration. Fig. 4A: Line graph showing typical results of two animals in the cohort whose tumor sizes were equivalent at 4 weeks. The course of tumor development in a female mouse is presented for comparison. Fig. 4B: Bar graph showing the mean tumor size (± standard error) of the entire cohort of intact and castrated male mice. The data for each animal are expressed in tumor size compared to the time point of the 4^to week.

Fig. 5. Line graph showing limiting dilution analysis of LAPC-4 graft in male mice.

Fig. 6. Photographs showing the detection of metastatic disease in mice with LAPC4 xenografts. Whole RNA was isolated from the indicated murine tissues and analyzed for expression of PSA (a) or jd-actin (b) using RT-PCR. Tumor and various tissue results from the three representative mice are presented. Tissues from a fourth mouse (control SCID) were analyzed as a negative control. The signal can be quantized by comparison with LNCaP (lane 1). No RNA was added to the negative reference sample (lane 2).

Fig. 7. Photographs showing the detection of LAPC-4 cells in bone by immunohistochemistry after 2 weeks. Frozen sections of the tibia of mice injected with lApC-4 cells were stained with an antibody against cytokeratin-18 (lower panel) or an isotope reference antibody (upper panel). The four cells that stain red are LAPC-4 cells.

Fig. 8. Photographs showing that LAPC-4 produces bone lesions. Hematoxylin and eosin sections of the tibia are presented at 4, 6, and 8 weeks after intratibial injection of LAPC-4 cells. Panel A shows a small focus of tumor formation adjacent to normal bone and hematopoiesis. Panels B and C present the evolutionary increase in new bone formation in response to surrounding tumor cells.

Fig. 9. Radiographic evidence of osteoblastic bone lesions produced by LAPC-4. X-rays of mice were taken at 8 weeks after injection of LAPC-4 cells into the tibia (right panel). The bone demonstrates evidence of erosion of the cerebral cortex with increased bone density in the marrow cavity due to osteoblastic activation.

Fig. 10. Flow cytometric analysis of LAPC-4 cells stained with anti-galectin-6 antibody, showing the level of expression compared to a reference antibody of the isotope.

Detailed description of the invention

Immunocompromised animal hosts

Severe combined immunodeficient (SCID) mice are the animal hosts used in the practice of the invention. Female, male, castrated or uncastrated mice can be used, depending on whether one is interested in studying the effect of androgen availability on the course of tumor growth. In the specific and preferred embodiments described herein, CB 17 scid / scid mice are used.

Models that simulate advanced prostate cancer

In one aspect, the murine xenotransplantation models described herein mimic or resemble primary tumor-forming human prostate cancer. Procedures for preparing advanced stage propagating human prostate tumor tissue as subcutaneous xenografts in immunodeficient SCID mice are also described. In the practice of the invention, prostate cancer xenografts can be created in immunocompromised SCID mice by subcutaneous implantation of fresh human prostate cancer explants surgically removed from patients with locally advanced or metastatic prostate cancer. The point of implantation can be at any subcutaneous point that allows the blood supply to reach the implant, such as the flanks of the host animal. Primary prostate tumor tissue as well as ganglion points

ES 2 290 964 T3 lymphatic, lung, bone and other organ metastases can be used to create the prostate cancer xenografts used in the invention. Prostate tumor explants can be introduced along with a base membrane composition, such as Matrigel (US Pat. No. 5,508,188), an extracellular matrix preparation that has been shown to increase the growth of epithelial tumors in vivo (including prostate cancer cells) (Lim et al., 1993; Noel et al., 1992; Pretlow et al., 1991), as well as other similar types of compositions. Once created, xenotransplantation tumors grow to considerable size, providing substantial volumes of tissue for further utilization. The xenografts used in the invention preserve the human phenotype determined by the expression of human ^ -globin, express the human prostate specific antigen (PSA), and retain androgen sensitivity and metastatic growth characteristics reflecting the clinical situation.

As used herein, the terms "locally advanced prostate cancer" and "locally advanced disease" mean prostate cancers that have spread through the prostatic capsule and mean that they include stage C disease in the prostate system. the American Urological Association (AUA), stage C1-C2 disease in the Whitmore-Jewett system and stage T3-T4 and N + disease in the TNM system (tumor, node, metastasis). In general, surgery is not recommended for patients with locally advanced disease, and these patients have substantially fewer favorable responses compared to patients with clinically localized (organ-confined) prostate cancer. Locally advanced disease is clinically identified by hardening beyond the lateral limit of the prostate, or asymmetry or hardening above the base of the prostate. Locally advanced prostate cancer is pathologically diagnosed after radical prostatectomy if the tumor invades or penetrates the prostate capsule, extends into the surgical margin, or invades the seminal vesicles.

As used herein, the terms "metastatic prostate cancer" and "metastatic disease" mean prostate cancers that have spread to lymph nodes in the area or to distant sites, and means that they include disease in stage D in an AUA system and in the TxNxM + stage in the TNM system. As in the case of locally advanced prostate cancer, surgery is not generally indicated for patients with metastatic disease, and hormonal therapy (androgen removal) is the preferred treatment modality. Patients with metastatic prostate cancer optionally develop an androgen-resistant state within 12 to 18 months of initiation of treatment, and approximately half of these patients die within 6 months. The most common site for prostate cancer metastasis is the bone. Bone metastases from prostate cancer are, on balance, characteristically osteoblastic rather than osteolytic (ie, resulting in pure bone formation). Bone metastases are most frequently found in the spine, followed by the femur, pelvis, rib cage, skull, and humerus. Other common sites of metastasis include the lymph nodes, lungs, liver, and brain. Metastatic prostate cancer is typically diagnosed by open or laparoscopic pelvic lymphadenectomy, whole-body radionuclide scans, skeletal radiography, and / or biopsy of the bone lesion.

This and other aspects of the invention described herein provide tools for studying the pathogenesis and treatment of advanced prostate cancer. For example, SClD mice immunodeficient with subcutaneous xenografts (and others) can be used to evaluate the effect of various prostate cancer treatments (eg, therapeutic compositions, gene therapies, immunotherapies, etc.) on tumor growth and evolution. of the illness. Xenotransplantation cells can be used to identify new genes and genes that are differently expressed in prostate cancer cells, or to analyze the effect these genes have on prostate cancer progression. For example, the genetic compositions of prostate cancer cells from xenotransplantation with different androgen sensitivities (eg. androgen-dependent versus androgen-independent) can be compared to each other as well as to the genetic compositions of normal prostate cells. Also, the genetic compositions of micrometastatic prostate cancer cells can be compared to metastatic prostate cancer cells. Various sampling and nucleic acid subtraction techniques can be used for this purpose, including, for example, representative differential analysis (RDA). Furthermore, prostate cancer xenograft cells can be used for the introduction of various genetic capabilities, including the introduction of various genes, complementary sequences, ribozymes, regulatory sequences that increase or suppress the expression of endogenous genes, and so on.

Furthermore, this aspect provides methods for purifying prostate cancer cells from the typical cell heterogeneous mixture of human prostate cancer biopsy material, further providing methods for generating larger amounts of tumor cells for further use and analysis. In one embodiment, the prostate cancer cell purification method comprises implanting the human prostate cancer biopsy material subcutaneously into a SCID mouse and allowing the implanted material to grow as a xenograft in the mouse. Purified human prostate cancer cells are obtained by harvesting the xenograft. Xenografts can be expanded and further purified by serial propagation in additional immunodeficient SCID mice or by short-term propagation in cell culture. Individual cell suspensions of xenograft tumor tissue or cultured cells can be used to orthotopically seeding intraprostatic tumors, bone tumors, or other organic tumors. Xenograft tumor tissue and cell preparations can be frozen and feasibly recovered for later use.

The invention also utilizes subcutaneous prostate cancer xenografts that preserve stable prostate cancer cell phenotypes by multiple passages in SCID mice. Various ways to

ES 2 290 964 T3 implementation, including androgen-independent and androgen-dependent xenografts, xenografts that express prostate-specific antigen (PSA) at clinically reflex levels, xenotransplants that express the natural androgen receptor (AR), and xenotransplants that present chromosomal abnormalities. Still other embodiments include xenotransplantation that retain all of the above characteristics as well as xenotransplantation that model androgen-independent disease progression. These and other embodiments of the invention are described in greater detail by the following examples. As described in Example 1, numerous subcutaneous xenotransplants were created successfully from tumor tissue explants extracted from the prostate gland and bone, lymphatic and lung metastases from patients with stage C or D prostate cancer. they grow and are passaged in SCID mice very frequently and retain definitive features of human prostate cancer, even in late passage. A xenotransplantation, called LAPC-4, has been adapted to tissue culture as a stable cell line and has been in continuous culture for 18 months.

Xenografts such as the LAPC-4 xenograft described in Example 1 are of particular interest. Similar to prostate tumors isolated directly from patients, LAPC-4 cells retain expression of prostate specific antigen (PSA), androgen receptor (AR), and prostate acid phosphatase by more than 20 passages. Furthermore, LAPC-4 xenotransplantation is unique among prostate cancer model systems as its AR does not contain any mutation in DNA or ligand-binding domains and AR expression is conserved in independent LAPC-4 sublines. of androgen. In addition, LAPC-4 xenotransplantation models the transition from androgen-dependent to androgen-independent disease as well as the development of micrometastatic disease. For example, LAPC-4 tumors that have been passaged in male mice retain androgen-dependent growth characteristics, whereas tumors that have been passaged in castrated male or female mice acquire a stable androgen-independent phenotype. . These sublines can be easily expanded using the procedures described herein to provide tissue broad for biochemical molecular analysis of cases associated with androgen-independent growth. There are a few other experimental models for the growth of androgen-dependent prostate cancer. Published articles include the widely used LNCaP cell line (Lim et al., 1993; Gleave et al., 1992) and two recently described xenografts, CWR22 (Weinstein et al., 1994) and LuCaP23 (Lin et al., 1996) . LAPC-4 xenotransplantation is unique in that tumors placed under the selective pressure of androgen-loss reproducibility induce an androgen-independent state, providing an opportunity to assess molecular changes associated with androgen independence over time and directly prove its functional importance.

This aspect also provides assays to determine the function or effect of various genes on prostate cancer cells. In one embodiment, the assay comprises isolating prostate cancer cells from a prostate cancer xenograft (e.g. g., subcutaneous, intraprostatic), transducing the cells with the gene of interest so that the transduced cells express or overexpress the gene, creating a subcutaneous or intraprostatic xenograft tumor in a SCID mouse with the transduced cells and evaluating the growth of the xenograft resulting. The effect of expressing the gene on xenograft growth can be determined by reference to a reference xenograft created with untranslated prostate cancer cells, preferably isolated from the same paternal xenograft. In another embodiment, the assay comprises generating a prostate cancer xenograft (e.g. g., subcutaneous, intraprostatic), transducing cells with the gene of interest in vivo, and evaluating xenotransplantation growth, wherein the effect of the gene on xenotransplantation growth can be determined by reference to a reference xenotransplant.

Also, this aspect provides analysis to determine the effect of experimental therapeutic compositions or treatments on the growth of prostate cancer cells. In one embodiment, the assay comprises applying the composition or treatment to a SCID mouse bearing a human subcutaneous prostate cancer xenograft and determining the effect of the composition or treatment on the growth of the xenograft. In another embodiment, a SCID mouse carrying an intraprostatic xenograft is used to determine the effect of the composition or treatment.

This aspect may also have various clinical applications, including the use of the model in a procedure to assess the prognosis of a patient with locally advanced or metastatic prostate cancer. For example, in one embodiment, the method comprises implanting a prostate tumor sample from the patient into an immunocompromised SCID mouse subcutaneously, and allowing the implanted sample to grow into a xenograft in the mouse. The growth rates of xenotransplantation can be used as a prognostic indicator. The results of such a test can assist the treating oncologist in determining how to intensively treat a patient.

Models that simulate prostate cancer micrometastasis

In another aspect, the invention uses models and methods to simulate and study the micrometastasis process in human prostate cancer. SCID mice with subcutaneous prostate cancer xenografts show evidence of circulating prostate cancer cells. Therefore, this model duplicates the process of cell migration from the main tumor to the bone marrow and other distant sites of micrometastasis. As detailed in Example 3, 100% of male mice inoculated subcutaneously with xenograft LAPC-4 cells developed localized subcutaneous tumors at 4 to 6 weeks without evidence of bone metastasis. However, when these animals were examined for the presence of micrometastatic disease, up to 50% of the

ES 2 290 964 T3 mice had detectable prostate cancer cells in bone marrow and blood. Using the same semi-quantitative RT-PCR assay that has been applied to large scans of prostate cancer patients, micrometastatic prostate cancer cells were observed at levels comparable to approximately 0.1 to 1.0% of total mouse bone marrow. . Similar results were obtained by immunohistochemical analysis for PSA expression. Therefore, subcutaneous growth of prostate cancer xenografts simulates the clinical observation that prostate cancer cells circulate in the blood and lodge in the bone marrow, even in early stage disease.

In one embodiment, simulating or mimicking prostate cancer micrometastasis comprises creating a subcutaneous prostate cancer xenograft in a SCID mouse and allowing the tumor to grow for a time sufficient to allow detection of cancer cells. of prostate in mouse peripheral blood. The presence of micrometastases is monitored by detecting prostate tumor cells that have migrated to the lymphatic and / or vascular system, bone, lung, liver, and other sites distant from the primary xenograft site. Detection of such cells can be performed, for example, by analyzing the presence of human PSA mRNA in peripheral blood using an RT-PCR analysis for PSA mRNA (such as the analysis described in Example 3).

In another embodiment, the simulation of prostate cancer micrometastasis comprises preparing a single cell suspension of prostate cancer cells from a subcutaneous xenograft tumor grown in a SCID mouse, followed by intraprostatic (orthotopic) injection of the suspension. cell isolated in another SCID mouse. The intraprostatic tumor is allowed to develop long enough to allow detection of prostate cancer cells in the peripheral blood of the mouse or at other sites distant from the orthotopic tumor. Isolated cell suspensions prepared from cultured xenotransplantation cells can also be used for intraprostatic (orthotopic) implantation.

This aspect also provides a framework to determine the effect of certain variables on the development of micrometastasis. Such variables can include the presence or absence of hormones or other growth modulating factors in the tumor environment, the state of expression of various genes in tumor cells, etc. For example, the rate of micrometastasis of androgen-dependent and androgen-independent xenograft variants can be assessed. Said evaluation is described in Example 3, using the androgen-dependent and independent sublines of the LAPC-4 xenograft, which demonstrates a significantly higher rate of micrometastasis in mice with androgen-independent LAPC-4 xenografts.

This aspect provides analysis to determine the function or effect of various genes on the evolution of prostate cancer micrometastasis. In one embodiment, the assay comprises isolating prostate cancer cells from a prostate cancer xenograft (e.g. g., subcutaneous, intraprostatic), transduce cells with the gene of interest so that the transduced cells express or overexpress the gene, use the transduced cells to create a subcutaneous or intraprostatic xenograft tumor in a SCID mouse, and assess the presence and micrometastatic disease levels by detecting prostate cancer cells in blood, bone marrow, lymph nodes, and other sites distant from the primary xenotransplant tumor site. The effect of expressing the gene on the rate of micrometastasis can be determined relative to a reference xenograft created with non-transduced prostate cancer cells, preferably isolated from the same parent xenograft. In another embodiment, the assay comprises generating a prostate cancer xenograft (e.g. g., subcutaneous, intraprostatic), transducing the xenograft cells with the gene of interest in vivo, and evaluating the presence of concentrations of micrometastatic disease, in which the effect of expressing the gene on the rate of micrometastasis can be determined in relation to the Reference xenotransplantation.

Also, this aspect provides analysis to determine the effect or experimental therapeutic compositions or treatments on the evolution of micrometastatic disease. In one embodiment, the assay comprises applying the composition or treatment to a SCID mouse bearing a subcutaneous human prostate cancer xenograft and determining the effect of the treatment on micrometastasis by monitoring the presence and levels of the prostate cancer cells. in the peripheral blood, lymph nodes, bone marrow and / or other distant sites of the xenotransplantation. In another embodiment, a SCID carrying an intraprostatic xenotransplantation is used to determine the effect of treatment on micrometastasis.

This aspect may also have various clinical applications, including the use of the model in a procedure to assess the prognosis of a patient with locally advanced or metastatic prostate cancer. For example, in one embodiment, the method comprises implanting a prostate tumor sample from the patient into an immunodeficient SCID mouse subcutaneously and leaving the implanted sample to develop as a xenograft in the mouse. The growth rates of xenotransplantation and the development of micrometastasis can be used as prognostic indicators. The results of such a test can assist the treating oncologist in determining how to intensively treat a patient.

Models that simulate metastatic prostate cancer

In another aspect, the invention uses models and methods to mimic and study the development of macrometastatic osteoblastic bone lesions (bone metastases) in prostate cancer. Subcutaneous growth of xenograft tumors results in detectable micrometastases, indicating that cells from the tumors of the

ES 2 290 964 T3 xenotransplantation in SCID mice have the ability to exit the growth point of the main tumor, circulate in the blood and lodge in the bone marrow, reflecting the human clinical situation.

In one embodiment, simulating the development of prostate cancer bone metastasis comprises injecting a single cell suspension of prostate cancer cells prepared from a subcutaneous prostate cancer xenotransplantation that develops in SCID mice into the prostate. from another SCID host mouse, and allow the resulting orthotopic tumor to grow long enough to allow detection of bone metastases in the mouse. Alternatively, subcutaneous prostate cancer xenografts can be created with such single cell preparations and allowed to develop. Detection of bone metastasis can be accomplished by various methods, including histologically, immunohistochemically, and radiographically.

Subcutaneous and orthotopic tumors typically develop rapidly, reaching a size that requires the host animal to be euthanized in about 4 to 6 weeks. Accordingly, alternative methods are also provided that increase the number of prostate cancer cells in the bone marrow, thereby obviating this limitation. In one embodiment, an isolated cell suspension prepared from xenograft tumor cells, or from xenograft cells in tissue culture, is injected directly into the bone marrow cavity (eg, tibia). ) from a SCID mouse. The development of micrometastases, bone tumor growth, and osteoblastic activity can be monitored in a number of ways, including by immunohistochemistry and in situ hybridization of bone sections or by radiographic imaging.

As described in Example 6, an isolated cell suspension of 10,000 xenograft tumor cells prepared from a subcutaneous tumor was injected into the tibia of a SCID mouse. A small subset of the injected cells was detectable at 2 weeks, followed by small foci of bone tumor growth in a few isolated areas at 4 weeks, followed by extensive gross bone tumor growth, destruction of the bone cortex, and bone formation. pure new bone at 6 to 8 weeks. Accordingly, cells isolated from the human prostate cancer xenografts of the invention are capable of proliferation in the microenvironment of the SCID mouse bone marrow cavity.

The above procedure provides an excellent model to simulate the formation of osteoblastic bone lesions and the progression to this stage of the disease. The model can be used not only to study the molecular and cellular cases involved in the evolution of this stage of prostate cancer, but also to test the effect of various experimental therapeutic genes, proteins, and other compounds. Furthermore, the model can be used as an analysis to assess the metastatic and osteoblastic potential of prostate cancer cells obtained from human patients.

Accordingly, this aspect also provides analysis to determine the role or effect of various genes in the progression of prostate cancer bone metastasis. In one embodiment, the assay comprises isolating prostate cancer cells from a prostate cancer xenograft (e.g. g., subcutaneous, intraprostatic, bone), transducing cells with the gene of interest so that the transduced cells express or overexpress the gene, introducing the transduced cells into the bone marrow cavity of a SCID mouse and monitoring in the marrow the presence and levels of osteoblastic macrometastatic lesions. The effect of expressing the gene on the development and growth of bone metastasis can be determined by referring to a control animal receiving non-transduced prostate cancer cells, preferably isolated from the same parent xenograft. In another embodiment, the assay comprises generating a bone marrow xenograft in a SCID mouse by injecting a single cell suspension of prostate cancer cells prepared from a subcutaneous or intraprostatic xenograft created in another SCID mouse, transducing the xenotransplant cells. of bone marrow with the gene of interest in vivo, and evaluating the effect of the gene on the presence and levels of osteoblastic macrometastatic lesions.

Furthermore, this aspect provides analysis to determine the effect of experimental therapeutic compositions or treatments on the progression of bone metastasis from prostate cancer. In one embodiment, the analysis comprises applying the composition or treatment to a SCID mouse receiving an intratibial injection of prostate cancer xenograft cells and determining the effect of the composition or treatment on the evolution of bone metastasis by monitoring the bone marrow of the tibia the presence and levels of prostate cancer cells and / or osteoblastic macrometastatic lesions. The presence of prostate cancer cells in the bone marrow can be detected by various means, including histology, immunochemistry, or by analyzing the presence of PSA mRNA or protein. The presence of osteoblastic macrometastatic lesions can be detected using histological, radiographic, or other imaging techniques.

This aspect may also have several clinical applications, including using the model in a procedure to assess the prognosis of a patient with locally advanced prostate cancer, and in particular, to predict the probability that a patient will progress to metastatic disease. For example, in one embodiment, the method comprises injecting a single cell suspension prepared from the patient's prostate biopsy material directly into the bone marrow of an immunocompromised mouse and then monitoring the bone marrow for the development of lesions. bone. The growth rate of the bone lesion and osteoblastic activity can be used as prognostic indicators. The results of such a test can help the treating oncologist determine how to intensively treat a patient with locally advanced disease.

ES 2 290 964 T3

Likewise, the effect of various therapeutic strategies to locally treat advanced or metastatic disease in a given patient can be predicted. For example, the effect of a treatment strategy can be predicted by applying the treatment to an immunocompromised mouse that receives a bone marrow injection of the patient's prostate cancer cells. The effect of treatment can be monitored by comparing the rate and extent of bone lesion growth and osteoblastic activity in the test mouse to the corresponding rates in an untreated control mouse receiving a corresponding bone marrow injection. Furthermore, this procedure can be used to determine the efficacy of a treatment strategy on androgen independent prostate cancer cells using a castrated male or female immunodeficient mouse in order to select androgen independent clones in the patient's tumor material. The results of these tests can help a treating oncologist determine which of several alternative therapies should be used to treat a certain patient's condition.

Short-term culture of xenograft tumor cells

Xenotransplantation tumor cells can be propagated using short duration in vitro tissue culture techniques well known in the art. Furthermore, different clonal populations of a xenograft tumor can be isolated by tissue culture techniques. In this regard, methods for preparing isolated cell suspensions from xenograft tumor tissue samples are described. In one embodiment, the xenograft tumor tissue is surgically removed from a subcutaneous xenograft tumor, separated, and proteolytically digested, using the procedure described in Example 2 or similar procedures. The cells can then be suspended in a solution of Matrigel, other base membrane compositions, saline, or other buffers. Such preparations are useful for creating new tumors in SCID recipient mice, for example, by subcutaneous inoculation, intraprostatic injection, or by injection directly into the bone marrow metastasis. Cell suspensions can be prepared from subcutaneous, intraprostatic, bone, or other orthotopic tumors that develop in SCID mice.

Procedures for Expansion and Purification of Prostate Cancer Cell Populations

Also described are procedures for expanding advanced-stage prostate cancer cells, procedures for preparing relatively pure populations of prostate cancer cells from heterogeneous populations of cells, and procedures for preparing specific prostate cancer cells from the in vivo or in vitro stage. The main tumor samples are heterogeneous in their cellular compositions, and are usually contaminated with normal and stromal cells. Furthermore, it is difficult to obtain substantial populations of prostate cancer cells from human tissue biopsy material. In contrast, cells collected from subcutaneous prostate tumors that develop in SCID mice predominantly comprise prostate cancer cells. Thus, the models used in the invention provide a vehicle for purifying advanced stage human prostate cancer cells from heterogeneous biopsy material.

Serial passage of xenograft tumors in additional mice can be used to further enhance the prostate cancer specificity of the xenograft cellular composition. Also, the ability to serially propagate, such as by serial propagation, relatively pure human prostate cancer cells in immunodeficient mice provides a method of obtaining large numbers of defined prostate cancer cells.

In one embodiment, tissue harvested from xenograft tumors is enriched for prostate cancer cells by subsequent passage into additional SCID mice. In another embodiment, xenograft tumor cells can be cultured in vitro. In another embodiment, isolated cell suspensions of prostate cancer cells can be prepared from such cultured cells or directly from xenograft tumor tissue. Isolated cell suspensions prepared from subcutaneous xenograft tumors digested with proteases retain the biological properties of the precursor tumors. Isolated cell suspensions can be used to demonstrate, for example, new subcutaneous tumors, intraprostatic tumors, or bone tumors. As shown in the experiments published in Example 2, only 10 xenotransplantation cells can be seeded into a new subcutaneous tumor.

Selective factors can be added to the medium in which tumor cell enrichment is taking place in order to expand cells with a specific phenotype. For example, the presence of androgen in the medium in vivo can be monitored by chemical or surgical castration procedures well known in the art in order to select for androgen-dependent or independent prostate cancer cells. Alternatively, female mice can be used to expand androgen independent cells. Also, in an in vitro medium, the absence of androgen in the growth medium can be used to select for androgen-independent prostate cancer cells.

Furthermore, the presence of cell surface proteins in subcutaneous xenograft tumor cells can be used to distinguish and isolate human prostate cancer cells from other cells. In particular, antibodies against cell surface proteins expressed differently in prostate cancer cells (as compared to their expression on murine marrow cells) can be used to isolate prostate cancer cells from tumor tissue of the xenotransplantation, from cells in culture, etc., using antibody-based cell sorting or affinity purification techniques. Most preferred for antibody-based cell sorting are antibodies against cell surface proteins that are specific for cancer of the

ES 2 290 964 T3 human prostate. However, antibodies to other human proteins can be employed effectively provided they do not show significant cross-reactivity with the murine homologue of the protein. An example of such a protein is human galectin-6.

The ability to generate large numbers of relatively pure, advanced-stage human prostate cancer cells that can grow in cell culture or as xenograft tumors in SCID mice provides many advantages, including, for example, allowing the evaluation of various transgenes or compounds. experimental therapeutics on the growth or other phenotypic characteristics of a relatively homogeneous population of prostate cancer cells. Furthermore, this also enables the isolation of highly enriched preparations of human prostate cancer-specific nucleic acids in amounts sufficient for various molecular manipulations. For example, large amounts of such nucleic acid preparations will aid in the identification of rare genes with biological relevance for the course of prostate cancer disease.

Another valuable application of this aspect of the invention is the ability to analyze and experiment with relatively pure preparations of viable prostate tumor cells cloned in individual patients with locally advanced or metastatic disease. In this way, for example, individual patient's prostate cancer cells can be expanded in a limited biopsy sample and then determine the presence of diagnostic and prognostic genes, proteins, chromosomal aberrations, gene expression profiles, or other. relevant genotypic and phenotypic characteristics, without the potentially confusing variable of contaminating cells. Furthermore, neoplastic aggressiveness and metastatic potential can be evaluated in said cells in the subcutaneous, orthotopic and bone tumor models of the invention. This aspect of the invention provides a method for experimentation with alternative treatment modalities with a view toward customary optimal patient-specific treatment regimens. Also, patient-specific prostate cancer vaccines and cellular immunotherapeutics can be created from such cell preparations.

The prostate cancer models used in the invention further provide methods for isolating specific prostate cancer cells for the isolation stage, including micrometastatic and osteoblastic prostate cancer cells. In one embodiment, macrometastatic cells are isolated from hematopoietic tissues such as bone marrow or blood using affinity purification or antibody-based cell sorting techniques. In another embodiment, osteoblastic prostate cancer cells are isolated from the bone marrow of SCID mice with osteoblastic bone lesions. The presence of these bone lesions can be detected histologically, immunohistochemically, or radiographically. Stage-specific prostate cancer cells can be further expanded and purified by subsequent reimplantation in SCID mice. For example, osteoblastic prostate cancer cells can be subpassed in vivo by reinjection into the bone marrow or in vitro using the defined bone stroma as a growth substrate.

As shown by the experimental work presented in Example 3, a small number of macrometastatic prostate cancer cells can be detected and isolated from the bone marrow of SCID mice with subcutaneous prostate cancer xenografts. Although these prostate cancer cells represent less than about 1% of the cells in the host mouse bone marrow, they can be isolated and expanded using cell purification procedures, such as those discussed above. In one embodiment, bone marrow harvested from subcutaneous xenograft mice is incubated with a human specific monoclonal antibody to galectin-6 and a secondary antibody conjugated to magnetic lentils. The prostate cancer cells are then isolated using Miltenyi Magnetic Minimacs columns (Sunnyvale, CA) to magnetically retain the antibody positive cells on the column while allowing the antibody negative cells to flow through. A small number of macrometastatic prostate cancer cells isolated in this way can be propagated in vivo by subcutaneous inoculation of Matrigel suspension cells into SCID mice. Osteoblastic prostate cancer cells can also be isolated directly from bone marrow lesions.

The ability to purify and expand stage-specific prostate cancer cells can have several clinical applications. For example, stage-specific prostate cancer cells can be isolated within clinical material using cell sorting or purification techniques and then expanded as subcutaneous, intraprostatic, or bone tumors in SCID mice, depending on the specific target. In one embodiment, micrometastases are isolated from the patient's serum, formulated into individual cell suspensions, and injected subcutaneously with the goal of expanding these cells generally. In an alternative embodiment, macrometastatic cell preparations are injected into the bone marrow of a SCID mouse in order to selectively expand these cells with osteoblastic characteristics. Prostate cancer cells that have been passaged in this way can become conditioned by various factors in the bone marrow microenvironment and can form osteoblastic lesions that can then be harvested for further use or analysis.

Continuous cell lines

Continuous human prostate cancer cell lines can be created by culturing xenograft cell preparations. In one embodiment, the cell line comprises human prostate cancer cells grown in a subcutaneous xenograft. In a specific embodiment described in the title of Example 9, the LAPC-4 cell line was created by cultivating an isolated cell suspension prepared from the LAPC-4 xenograft. The LAPC-4 cell line expresses PSA, androgen receptor (AR), and is androgen dependent. The

ES 2 290 964 T3 LAPC-4 cell line has been growing in continuous culture for 1.5 years, and retains phenotypic characteristics that correlate with human clinical situation more closely than any other available human prostate cancer cell line.

Cell lines can be used for many purposes. By way of example and not by way of limitation, cell lines can be used as a source of large amounts of nucleic acids and proteins, as a tool to identify and evaluate experimental therapeutic transgenes, proteins, and other compounds, and as a tool to identify and isolate genes. specific to the prostate or expressed differently. Genes that can regulate prostate cancer growth can be assessed by growth overexpression of transduced cells in vitro or in vivo. The effects of genes on micrometastasis and the development of osteoblastic bone lesions can be evaluated in vivo by subcutaneous, intraprostatic, or intratibial inoculation of the transduced cells.

Examples

The invention is described in greater detail and illustrated by the following examples and the experimental details therein. This section is published as an aid to understanding the invention, but is not intended to limit the claimed invention, nor should it be construed as limiting the claimed invention.

Example 1

Generation of subcutaneous human prostate cancer xenografts that simulate the evolution of prostate cancer

Materials and procedures

Patients: All clinical material was extracted from patients with locally advanced or metastatic disease (stage C or D) after obtaining written permission according to an IRB approved protocol. Most patients have undergone some form of androgen removal therapy (medical or surgical and have progressive disease at the time the tissue samples were obtained.

Animals: CB17 scid / scid (SCID) mice were raised at UCLA under sterile conditions as previously described (Aldrovandi et al., Nature 363: 732-736 (1993)). The biopsy material obtained at the time of surgery was placed on ice and immediately transferred to the SCID mouse facility for implantation. A scalpel was used to mince the tissue into 2 to 3 mm sections³ which were then implanted subcutaneously into the flanks of SCID mice. Mice were anesthetized with methoxyflurane prior to implantation, initial implants were made with 100 to 200 µl of Matrigel (Collaborative Research, Bedford, MA) injected around the implant. Matrigel is an extracellular matrix preparation useful for enhancing the growth of epithelial tumors in vivo (Pretlow (1993), supra; Noel et al., Biochemical Pharmacology 43: 1263-1267 (1992) and Lim et al., Prostate 22: 109 -118 (1993)). Once the xenotransplantation was attenuated, it was passaged 2 to 3 times, Matrigel was no longer used for serial propagation. Androgen removal was performed by surgical castration under anesthesia. Tumor sizes were determined by weekly caliper height, width, and thickness measurements. Slow-release testosterone granules (Innovative Research of America, Sarasota, FL) were implanted subcutaneously, as recommended by the manufacturer, in some experiments. Xenotransplants were stored viably in liquid nitrogen by freezing tissue sections cut into 1 to 2 mm³ in medium containing DMSO.

PCR analysis, histology and immunochemistry: DNA from tumor tissue was prepared using SDS detergent extraction and proteinase K digestion as described by Sambrook et al., Molecular Cloning: A Laboratory Manual, Cold Spring Harbor Laboratory Press, edition 2 (1989) . RNA was prepared using a commercially available kit containing guanidine thiocyanate and jd-mercaptoethanol (RNAgents Total RNA Isolation System, Promega). To avoid contamination of the bulk tissue preparations, the tissue homogenizer and all surgical instruments used at the time of necropsy were cleaned by repeated rinses in HCl, DEPC-treated water, and ethanol. DNA PCR analyzes for human ^ -globin (Aldrovandi et al., Supra and Saiki et al., Science 230: 1350-1354 (1985)) and RT-PCR analyzes for PSA (Pang et al., Hum. Gene Ther. 6: 1417-1426 (1995)) were performed as described above. In summary, PCR analysis using specific primers for the human ^ -globin gene was performed for 30 cycles with 100 ng of genomic DNA isolated from LAPC xenografts. One tenth of each reaction mixture was analyzed by agarose gel electrophoresis and observed by staining with ethidium bromide. Murine 3T3 cells were used as a negative control. The quality of all RNA samples was confirmed by ethidium bromide staining for ribosomal RNA and by RT-PCR using parab'-acline primers (Pang et al., Supra) as a reference. Details of the primer sequences can be found in the original references. RT-PCR analysis for PSA expression was performed on 100 ng of complete RNA using primers specific for human PSA. The same RNA samples were analyzed using primers that recognize human or murine '-actin to confirm equivalent loadings on gels. Immunochemical staining for PSA was performed using polyclonal antisera against PSA (Dako) as described (Hsu et al., Am. J. Clin. Path. 75: 734-738 (1981)).

Androgen receptor DNA sequencing: Exons 2 to 8 of the androgen receptor were sequenced from genomic DNA using intron-specific PCR primers (Marcelli et al., Mol. Endocrinol. 90: 1105-1115

ES 2 290 964 T3 (1990)). PCR products were initially identified by SSCP using appropriate positive and negative controls as described (Sutherland et al., J. Urol. 156: 828-831 (1996)). This technique has been shown to detect mutations in clinical samples of prostate cancer even when tumor cells represent only 20% of the population used to construct genomic DNA. All SSCP abnormalities were analyzed by sequencing. Two independent DNA samples were analyzed in two independent laboratories for the presence of any mutation.

Cytogenetics: Tumor tissue was aseptically transported in DMEM culture medium enriched with 10% fetal calf serum by overnight carrier at the University of Utah for cytogenetic preparation and analysis. Briefly, the tissue was ground and washed in Hanks balanced salt solution (free of Ca⁺⁺ and Mg ++), was resuspended in RPMI medium enriched with fetal bovine serum and cells were stopped at metaphase with 0.001 jug / ml of colcemide for 16 hours. Cytogenetic harvests were performed using standard procedures and, after treatment with hypotonic KC (0.075 M) and fixation with 3: 1 methanol / acetic acid, sections and chromosomes were prepared with trypsin G-banding / Wrights staining.

Results

Advanced-stage prostate cancer explants can be serially propagated in SCID mice

Locally advanced or metastatic tumor tissue biopsies were obtained from a total of 15 patients with locally advanced or metastatic prostate cancer (stage C, D1, or D2) who underwent palliative surgical procedures due to disease complications. The biopsy material was immediately transferred from the surgical room to the SCID mouse facility, cut into 2 to 3 mm sections³ and was implanted subcutaneously in SCID mice in the presence of Matrigel. Tumor growth was scored positive only if the explant exhibited a sustained two to three-fold increase in size. In addition to the histological studies, two molecular tests were performed on each xenotransplantation to verify the human origin of the tumors. These include a PCR analysis on genomic DNA using primers specific for the human / 1-globin gene and a quantitative RT-PCR analysis on tumor RNA using primers specific for the human PSA gene. Expression analysis by PSA was also used to verify the prostate origin of xenotransplantation.

The results obtained from the subcutaneous implantation of tumor tissue samples from two independent series of these 15 patients are individually described below (i.e. the LAPC-1 series to LAPC-8 and the LAPC-9 series to LAPC-15). .

LAPC-1 to LAPC-8 Series

Explants from six to eight patients (designated LAPC 1-8 for Los Angeles prostate cancer) formed tumors after a latent period that ranged from 2 to 10 months (Table 1). The six explants that grew were passaged into secondary recipients in an attempt to create permanent xenotransplants. Two of these (LAPC-1 and LAPC-5) were terminated after 3 to 4 passages because they were unable to detect human DNA or PSA expression in tumors. These explants were probably overdeveloped by cells of murine origin because they contained human DNA content and expressed PSA during the initial passages of LAPC-5 (Table 1, columns 6 and 7).

The remaining four explants (LAPC-3, 4, 7, and 8) were successfully propagated as subcutaneous xenografts in secondary recipients for between 4 and 20 (or more) passages. RT-PCR was used to measure PSA mRNA expression levels compared to LNCaP, a prostate cancer cell line known to express PSA mRNA and protein. This analysis is semi-quantitative and is capable of detecting the expression of PSA mRNA from 100 LNCaP cells diluted 10⁵ mouse cells (1 in 1,000 or 0.1%) (Fig. 1B, top panel). Four of the six xenografts (LAPC-3, 4, 5 and 8) expressed human PSA at levels between 1% and 100% of the level found in LNCaP cells (Fig. 1B, top panel). Simultaneous RT-PCR analysis using paraβactin primers confirmed that equivalent levels of RNA were present in each reaction (Fig. 1B, upper panel). Fig. 2 presents a histological comparison of the original LAPC-4 tumor sample obtained at the time of surgery for the same tumor after passage as xenograft in male mice. Hematoxylin and eosin stained sections (Fig. 2, left panels) present a monotonous population of anaplastic cells that stain positive for PSA using immunohistochemical analysis (Fig. 2, right panels). These findings demonstrate that advanced-stage prostate cancer explants can be serially propagated in SCID mice and retain definitive tissue-specific gene expression.

ES 2 290 964 T3

TABLE 1: SUMMARY OF IMPLANTS IN SCID / LAPC-1 MICE UP TO LAPC-8 SERIES¹

<td colspan="2">Patient implant (stage of disease)</td><td>Biopsy point</td><td>Growth</td><td>Time interval for tumor growth</td><td>Passes</td><td>State of DNA human²</td><td>PSA status³</td><td>Grades</td>

<td>LAPC-1 (stage D)</td><td></td><td>liver found</td><td>Yes</td><td>2 months</td><td> 5</td><td>negative pass</td><td>negative</td><td>Murine tumor overgrowth after serial passage</td>

<td>LAPC-2 (stage D)</td><td></td><td>ganglion lymphatic found</td><td>do not</td><td>without growth at 2 years</td><td> 1</td><td></td><td></td><td>no cells</td>

<td>LAPC-3 (stage D)</td><td></td><td>channel prostatic TURP</td><td>Yes</td><td>10 months</td><td> 3</td><td>positive</td><td>positive</td><td>positive for PSA outside the implantation site (n = 2) PSA-positive cells in the marrow</td>

<td>LAPC-4 (stage D)</td><td></td><td>ganglion lymphatic found</td><td>Yes</td><td>3 months</td><td> &gt;8</td><td>positive</td><td>positive</td><td>bone, spleen, blood in 50% of the mice (n = 12) overgrowth</td>

<td>LAPC-5 (stage D)</td><td></td><td>ganglion lymphatic found</td><td>Yes</td><td>9 months</td><td> 5</td><td>positive, then negative in pass 4</td><td>positive, then negative in pass 4</td><td>by tumor of murine origin in passage 4</td>

<td>LAPC-6 (stage C)</td><td></td><td>prostate</td><td>do not</td><td>without growth at 9 months</td><td> 1</td><td> -</td><td> -</td><td> -</td>

<td>LAPC-7 (stage C)</td><td></td><td>prostate</td><td>Yes</td><td>3 months</td><td> 2</td><td>positive</td><td>positive</td><td> -</td>

<td>LAPC-8 (stage D)</td><td></td><td>ganglion lymphatic found</td><td>Yes</td><td>10 months</td><td> 2</td><td>positive</td><td>positive</td><td>no cells positive to PSA outside the point of</td>

_implantation (n = 1) ¹ locally advanced (stage C) or metastatic (stage D) disease;

² determined by PCR of genomic DNA for human β-globin;

³ determined by RT-PCR and / or immunohistochemistry

Two of the xenotransplants in this series, LAPC-3 and LAPC-4, show constant histological and molecular characteristics of prostate cancer for more than 6 and 8 passages, respectively. Both xenotransplants can be viable frozen as tumor explants and recovered from freezing with almost 100% efficiency. An LAPC-4 xenograft cell line was created by serial passage of crushed xenograft tissue, treated with trypsin in Iscove culture medium enriched with 20% fetal calf serum. The LAPC-4 cell line has continued to be confirmed for more than 20 passages and has been in continuous culture for more than 18 months. These cells continue to express PSA, form tumors in SCID mice, and retain androgen sensitivity.

LAPC-9 to LAPC-15 Series

A second series of xenotransplantation experiments was performed by implanting tissue samples from an additional seven prostate cancer patients with advanced stage disease (C or D) (Table 2). Four of these seven implants have produced the generation of androgen-sensitive xenografts that express PSA and that are capable of serial propagation in more mice (LAPC-9, 12, 14 and 15). LAPC-9 xenotransplantation, generated from a bone tumor biopsy from a patient with hormone-resistant metastatic disease, demonstrates a highly androgen-sensitive phenotype (PSA levels decrease to 0 after castration) and has been subjected to passes and maintained in a viable way for approximately 1 year. LAPC-14 xenotransplantation, generated from a biopsy of a prostate tumor from a patient with metastatic disease, demonstrates abrasive growth characteristics and exhibits a high degree of androgen sensitivity (growth was substantially enhanced by the addition of testosterone) .

ES 2 290 964 T3

TABLE 2: SUMMARY OF IMPLANTS IN SCID / LAPC-9 MICE UP TO LAPC-15 SERIES¹

<td>Implantation of patient (stage of disease)</td><td>Biopsy site</td><td>the</td><td>Growth</td><td>Interval of time for tumor growth</td><td>Passes</td><td>State of human DNA²</td><td>Condition PSA³</td><td>from Notes</td>

<td>LAPC-9 (stage D)</td><td colspan="2">femoral tumor</td><td>Yes</td><td>5 weeks</td><td> 5</td><td>positive</td><td>positive</td><td>patient dependent of the hormone resistant androgen</td>

<td>LAPC-10 (stage C / D)</td><td colspan="2">prostate (protatectomy)</td><td>do not</td><td> -</td><td> -</td><td> -</td><td> -</td><td>Gleason 9</td>

<td>LAPC-11 (stage D)</td><td colspan="2">femoral tumor</td><td>do not</td><td></td><td></td><td></td><td></td><td>1 week after lupron 2 weeks after flutamide</td>

<td>LAPC-12 (stage D)</td><td>TURP prostate</td><td>of</td><td>Yes</td><td>13 weeks</td><td> 1</td><td>positive</td><td>positive</td><td>resistant to hormone metastatic</td>

<td>LAPC-13 (stage C / D)</td><td>trans rectal biopsy</td><td></td><td>do not</td><td> -</td><td> -</td><td> -</td><td> -</td><td>Gleason 7</td>

<td>LAPC-14 (stage D)</td><td>attenuation negative TURP prostate</td><td>of of</td><td>Yes</td><td>4 weeks</td><td> 2</td><td>positive</td><td>positive</td><td>PSA metastatic &lt;0.2 androgen treated with lupron</td>

<td>LAPC-15 (stage D)</td><td>TURP prostate</td><td>of</td><td>Yes</td><td>12 weeks</td><td> 1</td><td>positive</td><td>positive</td><td>treated with casedex treated with lupron for bone</td>

¹ locally advanced (stage C) or metastatic (stage D) disease;

² determined by PCR of genomic DNA for human β-globin;

³ determined by RT-PCR and / or immunohistochemistry

LAPC-3 and LAPC-4 xenografts contain chromosomal abnormalities

Extensive histogenetic studies of human prostate cancer have been difficult due to the heterogeneity of clinical material obtained in surgery and the limited growth of prostate tumor cells in vitro. To determine whether passage of prostate tumor tissue in SCID mice should facilitate karyotypic analysis, early tumors were analyzed by passage of LAPC-3 and LAPC-4 xenografts using standard cytogenetic techniques. A high mitotic yield was obtained from the tumor samples from both xenografts and all metaphase cells contained human chromosomes. The karyotypes of the detailed compound are indicated in Table 3. LAPC-4 modal chromosome number was 89, suggesting a hypotetraploid line, since LAPC-3 modal chromosome number was 69, suggesting that this line is almost triploid, even the presence of four copies of many chromosomes increases the possibility of reduction of the tetraploid. Both xenotransplantations present described numerical and structural chromosomal abnormalities such as loss of Y and 16. In addition, both xenografts contain a deletion on chromosome 12p12, a karyotypic abnormality that has not been previously described in prostate cancer.

ES 2 290 964 T3

TABLE 3: CYTOGENETIC ANALYSIS OF LAPC-3 AND LAPC-4 XENOTRASPLANTS

<td rowspan="2">Xenotransplantation</td><td colspan="3">Number of Passes in the Metaphases Karyotype Number</td>

<td>time of analysis (number of independent tumors)</td><td>analyzed</td><td>chromosomes manners</td>

<td>LAPC-3</td><td>pass 2 (1 tumor); pass 3 (2 tumors)</td><td> 80</td><td>69 68-81, XXY + add (l) (p22), -2, +3, +4, +5, del (6Xq21) x2, +7. +9, +9. -11, del (12) (p12), -13. -13, +14, t (14: 14Xq10: q10), -16, +18, +19, +20 [cp80]</td>

<td>LAPC-4</td><td>pass 3 (2 tumors)</td><td> 44</td><td>89 76-92, XX, -Y, + Y, add (8Xp23), +9, del (12Xp12), -14, -16, -18, -21, + mar1,</td>

+ mar2 [cp44]

Evolution of LAPC-4 xenotransplantation for androgen independence

Prostate cancer cells are exquisitely sensitive to the growth-stimulating effects of androgen, but androgen-independent disease optionally develops in patients under selective pressure for androgen deprivation. The mechanism for this transition to androgen-independent growth is unknown. The question whether this phase of the disease could be modeled in SCID mice was determined using the LAPC-4 xenograft, which reproducibly forms tumors in serial passage in male mice with the frequency of 100%.

The androgen dependence of xenotransplantation was measured in vivo by comparing growth rates after implantation in intact male mice with those from castrated male mice or female mice. For LAPC-4, the mean time to tumor formation in castrated male mice or female mice (n = 10) was 13.4 weeks versus 4.3 weeks in intact males (n = 14) (Fig. 3) . Growth retardation in female mice was reversed by implantation of a slow release testosterone pellet at 90 days (Fig. 3). The androgen independence of tumors growing in castrated female or male mice was confirmed by secondary transfer experiments. Once created, these tumors develop in 4 to 5 weeks in castrated male, female, and male mice.

The LAPC-3 xenotransplantation exhibited growth characteristics similar to the androgen-independent sublines of LAPC-4. After an initial latent period for passage 1, LAPC-3 tumors grew in 7 to 8 weeks independently of the receptor's hormonal background (Fig. 3), clearly creating this xenotransplantation as androgen-independent.

Clinically, antiandrogen therapy produces temporary amelioration of the disease in most patients with advanced prostate cancer. To determine whether a similar phenomenon is observed in the mouse model, the effect of acute androgen deprivation on the growth of tumors created in male mice was examined. The equivalent size LAPC-4 xenograft implants were passaged into a cohort of 14 male mice, all of which easily developed measurable tumors after four weeks. Half of these mice underwent castration, then the tumor sizes in each group were determined weekly by caliper measurement of the tumor diameters in three dimensions. Tumors in uncastrated mice doubled in size within 2 to 3 weeks (Fig. 4). In contrast, the castrated mice exhibited a decrease in tumor size within one week of approximately 50 percent that persisted for 2 to 3 weeks. These tumors resumed growth after a variable latent period (3 to 8 weeks) and eventually developed the same size seen in non-castrated mice. These results demonstrate that LAPC-4 xenotransplantation exhibits androgen-dependent growth, that androgen-independent sublines can develop, and that this xenotransplantation simulates the clinical transition from androgen-sensitive to androgen-independent disease.

LAPC-3 and LAPC-4 express natural androgen receptors

To determine if similar mutations are present in LAPC-3 and LAPC-4, exons 2 to 8 of the androgen receptor gene, spanning the DNA-binding and ligand-binding domains of the receptor, were sequenced. Single chain configuration polymorphism (SSCP) analysis was also performed. Each exon was amplified by PCR from genomic DNA from early and late passage tumors and analyzed using previously characterized mutant and wild-type androgen receptor DNA as positive controls (Sutherland et al., 1996). The results demonstrate that both LAPC-3 and LAPC-4 contain wild-type sequences in exons 2 to 8. Furthermore, these sequences remain wild-type in the androgen-independent LAPC-4 sublines. Immunoblot analysis confirmed the expression of an androgen receptor protein of the appropriate size. These results provide definitive evidence that androgen-independent progression of prostate cancer can occur in the absence of androgen receptor mutations in DNA or in the ligand-binding domains.

ES 2 290 964 T3

LAPC-4 cells can be efficiently transduced with retroviruses: LAPC-4 xenograft cells can be successfully transduced by retroviruses temporarily packaged in 293T cells with an enveloped amphotropic protein. LAPC-4 cells were infected with retrovirus stocks expressing the cell surface protein Thy-1 and the expression detected by flow cytometry using an antibody against Thy-1. The results presented Thy-1 expression in up to 50% of cells 48 hours after infection, indicating successful retroviral mediated transduction of the Thy-1 gene in LAPC-4 cells.

Example 2

Preparation of cell suspensions isolated from xenotransplantation cells

Materials and procedures

Cell suspensions isolated from subcutaneous LAPC-4 tumors were prepared as follows. After removing tissue from the xenotransplantation of SCID mice, the tissue was cut into 1 to 2 mm sections.³ although the tissue was soaked in 1x Iscoves medium, the tissue was then sliced ​​and centrifuged at 1.3 K rpm for 4 minutes, the supernatant was resuspended in 10 ml of ice cold 1x Iscoves medium and centrifuged at 1.3 K rpm for 4 minutes. The pellet was then resuspended in 1X Iscoves with 0.1% pronase E and incubated for 18 minutes at room temperature with gentle shaking followed by incubation on ice for 2 to 4 minutes. The mixture was then filtered using a 200 µm nylon mesh filter. The filtrate was centrifuged at 1.3 K for 4 minutes, and pronase was removed from the aspirated pellet by resuspending in 10 ml of Iscoves and centrifuging again. The resulting pellets were resuspended in PrEGM pre-incubated at 37 degrees C. Cell counts were determined, and limiting dilutions were formulated as indicated in Fig. 5.

Results

The results of the limiting dilution analysis of the tumor graft using isolated cell suspension of LAPC-4 xenotransplantation cells are presented in Fig. 5. The results demonstrate that isolated cell suspensions of xenotransplantation cells can form subcutaneous tumors in male mice after injection of only 10 LAPC-4 cells and that these cells retain the androgen sensitivity of precursor tumors.

Example 3

Simulation of the course for micrometastasis in SCID mice with subcutaneous tumors

Materials and procedures

LAPC-4 xenotransplantation was used in this study. This xenograft originated from a lymph node containing metastatic prostate cancer cells, and 100% of male mice inoculated subcutaneously with LAPC-4 cells developed localized tumors after 4 to 6 weeks without evidence of bone metastasis. The presence of micrometastases in SCID mice implanted with LAPC-4 tumors was determined by analyzing prostate cancer cells in peripheral blood using RT-PCR analysis for PSA mRNA. Simultaneous RNA PCR studies using / i-aclin primers demonstrated equivalent RNA loading. To confirm that positive PSA mRNA signals were not due to contamination with tumor cells during the necropsy procedure or during RNA preparation, samples were simultaneously prepared from a control mouse that was not implanted with a xenograft. . No expression of PSA was detected in the control mice, even after prolonged autoradiography exposure times (Fig. 6). Bone marrow, spleen, liver, lung and kidney tissue from mice implanted with subcutaneous LAPC4 tumors was also analyzed for the presence of prostate cancer cells using RT-PCR to detect PSA mRNA.

Results

The examples of the analyzes of two mice (Fig. 6A, mice n ° 213 and n ° 241) demonstrate the detection of PSA mRNA in the blood at a level between 0.1 and 1.0%, which is comparable to the levels described in clinical studies. Other organs were positive in various mice, including bone marrow (mice 213 and 241), lung (mouse 214), and spleen (data not shown). Results from 12 LAPC-4 xenograft animals (Table 4) demonstrate that 50 percent of mice have PSA mRNA positive cells (PSA expression level by RT-PCR 0.1 percent or greater) detected in peripheral blood, bone marrow or spleen. The level of expression was approximately quantified by comparison to a series of LNCaP cells in murine fibroblasts and ranged from 0.1% to 1.0%. Interestingly, the frequency of detecting micrometastatic disease was higher (80%) in female mice or in male mice castrated before implantation compared to intact males (27%). These results suggest that the transition to androgen-independent disease is associated with a higher metastatic frequency, a hypothesis that is also supported by clinical experience.

ES 2 290 964 T3

<td colspan="2">TABLE 4: FREQUENCY OF DETECTION OF PSA POSITIVE CELLS IN HEMATOPOIETIC TISSUES OF SCID MICE WITH LAPC-4</td>

<td>Group</td><td>Number of mice with PSA-positive cells in Hematopoietic Organs by Total Number Analyzed</td>

<td>Males of integrity</td><td> 2/7 (29%</td>

<td>Castrated males</td><td> 4/5 (80%)</td>

<td>(or females)</td><td></td>

<td>Total</td><td> 6/12(50%)</td>

Example 4

Generation of intraprostatic tumors with xenotransplantation cells

Materials and procedures

Isolated subcutaneous xenograft cell suspensions were prepared as described in Example 2. SCID mice were anesthetized with ketamine / xylazine prior to implantation. Transverse incisions were made in the lower abdomen of mice, incisions were made in the muscles of the abdominal wall and the bladder, and the seminal vesicles were incised to expose the dorsal prostate. Approximately 10,000 LAPC-4 cells in suspension in 10 µl PrEGM were injected slowly into the dorsal prostate under the capsule using a 30 gauge needle, and the incisions were closed using a running suture.

Results

Intraprostatic injection of isolated cell suspensions prepared from LAPC-4 and LAPC-9 xenografts and LAPC-4 cell line resulted in orthotopic tumors in recipient SCID mice with 100% efficacy.

Example 5

Simulation of the progression to the metastatic stage of prostate cancer in SCID mice with intraprostatic tumors

Materials and procedures

Cell suspensions isolated from LAPC-4 xenograft cells were prepared and used to create orthotopic tumors in the prostate of SCID mice as described in the previous example. The presence of metastases was determined by histological examination and by RT-PCR to detect PSA mRNA between 8 and 12 weeks after injection.

Results

The results, presented in Table 5 below, indicate high frequencies of lymph and lung metastases as well as a significant frequency of bone marrow metastasis formation. An increased frequency of bone metastasis was observed in a subset of the mice pretreated with a combination of radiation and NK cell depletion. Similar results were obtained using LAPC-9 xenotransplantation.

<td colspan="2">TABLE 5: METASTASIS MODEL AFTER ORTOTOPIC INJECTION OF LAPC-4</td>

<td>TUMOR POINT</td><td>FREQUENCY</td>

<td>Local tumor</td><td> 100%</td>

<td>Pelvic lymph nodes</td><td> 90%</td>

<td>Lung</td><td> 90%</td>

<td>Bone marrow</td><td> 30%</td>

ES 2 290 964 T3

Example 6

Simulation of the evolution to osteoblastic bone metastasis in SCID mice inoculated intratibially with single cell suspensions of xenograft cells

Materials and procedures

Tibia injection assay: Prostate cancer cells were isolated from a subcutaneous xenograft LAPC-4 tumor and prepared as isolated cell suspensions as described in Example 2. Ten thousand LAPC-4 cells were surgically injected into suspension in 1 pl of Matrigel in each proximal tibial metastasis of a cohort of SCID mice using a 27 gauge needle. Three mice were sacrificed at each of weeks 2, 4, 6, 8, and 12 after injection. Serum PSA levels were periodically analyzed by ELISA. At 2 weeks, frozen bone sections were immunohistochemically analyzed for cytokeratin-18 staining with an antibody specific for human cytokeratin-18 or an isotope reference antibody. Longitudinal sections of the tibiae of mice sacrificed at weeks 4, 6, and 8 were analyzed for tumor growth by hematoxylin and eosin (H + E) staining of the decalcified sections in paraffin. Radiographs of the mice were taken at necropsy to monitor for evidence of osteoblastic bone lesions.

Results

At 2 weeks, small numbers of human prostate cancer cells were observed by immunohistochemical staining with anti-cytokeratin-18 antibody (Fig. 7). 18 cytokeratin positive cells were observed scattered throughout the medullary canal. These data indicate that the majority of LAPC-4 cells injected into the mouse tibia are either killed or migrated to other positions since only a small subset of injected cells can be detected at this time point.

At 4 weeks, small foci of tumor growth were observed in a few isolated areas, usually adjacent to normal bone spicules, by H + E histology (Fig. 8A) and PSA could not be detected in serum. At the 6 and 8 week time points, more extensive tumor growth was observed throughout the medullary cavity along with a progressive increase in new bone formation indicative of osteoblastic activity within the medullary cavity in response. to surrounding tumor cells (Fig. 8B and C). Serum PSA concentrations rose markedly at this time point.

For 8 weeks, the bone lesions were visible by radiography as a mixture of osteoblastic and osteolytic lesions with dominant bone formation similar to the clinical observation in human prostate cancer. Referring to Fig. 9, the left panel presents a radiograph of a normal mouse tibia with the sharp, well-defined cortex and relatively radiopaque medullary cavity. The right panel is an X-ray of the tibial medullary cavity injected with LAPC-4 xenotransplantation cells, exhibiting a heterogeneous increase in bone density due to osteoblastic activity and destruction of an area of ​​the cortex. These results indicate that LAPC-4 xenograft cells can proliferate in murine bone, suggesting that the crossover between bone stroma and prostate cancer cells may occur across species.

Example 7

Isolation of prostate cancer cells from the bone marrow of SCID mice with subcutaneous xenografts

The presence of the cell surface protein galectin-6 in LAPC-4 cells was demonstrated by incubating intact LAPC-4 cells with a specific human monoclonal antibody against galectin-6 or reference isotope. The antibody was observed by flow cytometry after incubation with a FITC-conjugated secondary antibody. The flow cytometry results present galectin-6 expression at a level that is at least an order of magnitude higher than background (Fig. 10). Similar experiments performed in mouse bone marrow did not show any galectin staining.

As described in Example 3, small amounts of prostate cancer cells can be detected in the bone marrow of SCID mice with subcutaneous xenografts at 4 to 6 weeks after incubation, representing slightly less than 1% of the cells. in the marrow. This population of prostate cancer cells can be isolated from bone marrow using the Miltenyi Magnetic Minimacs (Sunnyvale, CA) antibody-based affinity purification system and anti-galectin-6 antibody as follows. Twenty mice with subcutaneous LAPC-4 tumors were euthanized after xenograft implantation. Bone marrow was removed from the tibiae and femurs by spraying the bone marrow cavities with saline. The marrow was mixed and incubated with a human specific monoclonal antibody against galectin-6 and a secondary antibody conjugated to magnetic lentils and fed through the Minimacs column as recommended by the manufacturer. LPAC-4 cells will remain in the column, while mouse bone marrow cells will pass through it. The purified LAPC-4 cells can then be removed from the column and expanded by seeding subcutaneous tumors in SCID mice.

ES 2 290 964 T3

Example 8

Isolation of prostate cancer cells from the bone marrow of SCID mice injected into the tibia with xenograft cells

Intratibial tumors were created in SCID mice using LAPC-4 cells as described in Example 6. LAPC-4 cells growing in bone marrow are recovered from mice after necropsy at 12 weeks by spraying the bone marrow cavity of the warm with saline solution and collecting the cells. At 12 weeks post injection, approximately 90% of the cells recovered are prostate tumor cells with some residual murine bone marrow cells. This population of cells can be further purified for prostate cancer cells using the galectin-6 / magnetic antibody affinity purification method as described in Example 7. Example 9

LAPC-4 cell line preserves expression of PSA, androgen receptor, and prostatic acid phosphatase through multiple passages

Materials and procedures

A continuous cell line was created from LAPC-4 xenograft by serial passage of chopped xenograft tissue, treated with trypsin in Iscove culture medium enriched with 20% fetal calf serum.

Results

LAPC-4 cells grown in continuous culture in vitro have maintained the expression of PSA, the androgen receptor and prostatic acid phosphatase for more than 20 passages. Furthermore, LAPC-4 cells do not contain mutations in DNA or in the ligand-binding domains of the androgen receptor, which is a new feature among known prostate cancer models. The only other PSA-expressing cell line, LNCaP, expresses an androgen receptor with a point mutation in the ligand-binding domain. Furthermore, LAPC-4 cells continue to express the androgen receptor in androgen-independent sublines, analogous to the results obtained from the analysis of clinical material. The LAPC-4 cell line is androgen dependent as tumors grow rapidly in male mice but not in castrated female or male mice. The LAPC-4 cell line has been created for more than 20 passages and has been in continuous culture for more than 18 months. These cells continue to express PSA, form tumors in SCID mice, and retain androgen sensitivity.

Example 10

Experimentation of the biological effects of experimental genes in androgen-independent in vivo culture

Some genes raised in hormone-resistant prostate cancer may contribute to androgen-independent pathogenesis. BcI-2, for example increased in many advanced prostate cancers, has been shown to provide androgen independence to the androgen-dependent LNCaP prostate cancer cell line (Raffo et al., 1995). According to this example, the contribution of experimental genes for the androgen-independent phenotype can be accessed in vivo.

LAPC-4 androgen-dependent tumor explants are grown in tissue culture and form androgen-dependent tumors upon reinjection in SCID mice.

Current bioassays for androgen-dependent and independent growth rely almost exclusively on the LNCaP prostate cancer cell line, because it is the only cell line available that exhibits androgen-dependent characteristics. In order to solve the problem of cell lines that have undergone long-term passages with the potential for multiple mutations in vitro, the LAPC-4 xenotransplantation was grown in short-term culture and then reinjected into mice to form tumors. . The explanted tumors were then genetically engineered and the effects of these manipulations were measured in vivo.

LAPC-4 tumors were cut into small pieces and cultured in medium with 15% fetal calf serum. The growth of both epithelial cells and fibroblasts was noted after 2 to 3 days. The cells were then grown to confluence and could be successfully passaged to remove the original tumor pieces. RT-PCR confirmed the continuous expression of PSA. They were then reinjected 1 χ 10⁷ cells in intact male or castrated SCID mice. Similar to the initial experiments, the injected cells formed tumors in an androgen-dependent fashion, requiring long periods of time to form tumors in the castrated mice.

LAPC-4 cultures can be transduced with retrovirus

In order to test the infectivity of retrovirus-explanted LAPC-4 cells, these cells were transduced with a retroviral vector containing a truncated nerve growth factor receptor (NGFR) gene. A PG13 packaging cell line, containing the gibbon monkey leukemia virus envelope

ES 2 290 964 T3 (GALV), was used to generate a high titer virus. The retrovirus virions produced in this way have the unique property of infecting man, but not murine cells, thus preventing introduction of the transgene into mouse stromal cells (Bauer et al., 1995). After infection, cells were stained with an antibody directed against NGFR and analyzed by FACS analysis. Five to 10% of the cells were transduced. Negative controls from murine fibroblasts showed no infection, while human 293T cells were efficiently transduced.

Biological analysis for up-regulated cDNA in androgen-independent prostate cancer

Experimental cDNAs can be cloned at the 5 'position of the retroviral vector pSRalfa used extensively in our laboratory (Afar et al., 1994). A reporter gene, either NGFR, LacZ, or human codon optimized green fluorescent protein (GFP), would be inserted downstream. The plasmid can be transfected into the PG13 packaging cell line, virus harvested, and titers measured. LAPC-4 cells can be infected after the first passage and then propagated without selection until sufficient numbers are available for injection. Expression of the transgene can be confirmed either by FACS analysis or by Northern blot analysis using the RDA cDNA clone as a probe.

Two different types of experiments can be performed. In the first, infected cells are injected into the flanks of intact male SCID mice. Once tumors form on both sides of an individual mouse, a tumor is removed and the mouse is then castrated. The explanted tumor is analyzed to quantify the percentage of infected cells. This can be done either by LacZ staining or by FACS analysis for GFP or NGFR. The authors predict that 5-10% of the cells will carry the transgene. The remaining tumor can be analyzed in a similar way once it retrogrades and regrows (ie, approximately 4 to 8 weeks after castration). If the transgene provides a survival advantage or androgen independence for infected cells, one would expect to see that the percentage of cells with transgene increases after hormonal removal. Many mice can be injected in each mount and positive results confirmed by repeat.

In a second series of experiments, infected cells can be implanted into intact and castrated male mice in parallel after quantifying the frequency of infection. The resulting tumors (at 4 and 12 weeks, respectively) are analyzed for insertion frequency as described above. Again, the authors hope that "androgen-independent" genes provide an androgen-independent growth advantage and dominate the resulting tumor. Furthermore, it is possible that a given experimental gene is time-staged for tumor formation in neutered males. This can also be measured. Finally, it is possible that a given gene can produce intensive androgen-dependent growth. This can also be quantified in this analysis by comparing time to tumor formation and insertion frequency before and after injection in intact male mice.

These analyzes can be validated with positive references. In particular, bci-2, c-myc, and c-met can be used, as these have been consistently associated with androgen independence.

References

Aldrovani, GM et al. The SCID-hu mouse as a model for HIV-1 infection. Nature 363: 732-736 (1993).

Bastón, OV The role of the vertebral veins in metastatic processes. Ann int Med. 16: 38-45 (1942).

Berettoni, BA and Carter, JR Mechanisms of cancer metastasis to bone. J. Bone Joint Surg. 68A: 308-312 (1986).

Brandt, B. et al. Isolation of prostate-derived single cells and cell clusters from human peripheral blood. Cancer Res. 56: 4556-4561 (1996).

Cornan, DR and DeLong, RP The role of the vertebral venous system in the metastasis of cancer to the spinal column. Cancer 4: 610-618 (1951).

Deguchi, T. et al. Detection of micrometastatic prostate cancer cells in the bone marrow of patients with prostate cancer. Brit. J. Cancer 75: 634-638 (1997).

ELLIS WJ ET AL: "Characterization of a novel androgen-sensitive, prostate-specific antigen-producing prostatic carcinoma xenograft: LuCaP 23". CLINICAL CANCER RESEARCH: AN OFFICIAL JOURNAL OF THE AMERICAN ASSOCIATION FO CANCER RESEARCH. UNITED STATES JUN 1996, vol. 2, no.6, June 1996.

Fidler, IJ Critical factors in the biology of human cancer metastasis. Cancer Res. 50: 6130-6138 (1990).

Ghossein, RA et al. Detection of circulating tumor cells in patients with localized and metastatic prostate carcinoma: Clinical implications. J. Clin. Oncol. 13: 1195-2000 (1995).

ES 2 290 964 T3

Gleave, ME et al. Serum prostate specific antigen levels in mice bearing human prostate LNCaP tumors are determined tumor volume and endocrine and growth factors. Cancer Res. 52: 1598-1605 (1992).

Hac, M et al. Rat prostate adenocarcinoma cells disseminate to bone and adhere preferentially to bone marrowderived endothelial cells. Cancer Res. 52: 4613-4619 (1992).

HSIEH JT ET AL: “An androgen-independent model of human prostate cancer progression: Molecular and cellular characterization”. PROCEEDINGS OF THE AMERICAN ASSOCIATION FOR CANCER RESEARCH ANNUAL, vol. 34, 1993.

Hsu, SM et al. A comparative study of the peroxidase-antiperoxidase method and an avidin-biotin complex method for studying polypeptide hormones with radioimmunoassay antibodies. Am J. Clin. Pathol. 75: 734-738 (1981).

Kaighn, ME et al. Establishment and characterization of a prostatic carcinoma cell line. Invest Urol. 17: 16-23 (1979).

Karp, JE, et al. Prostate cancer Prevention: Investigational approaches and opportunities. Cancer Res. 56: 55475556 (1996).

Katz, AE et al. Molecular staging of prostate cancer with the use of an enhanced reverse transcriptase-PCR assay. Urology 43: 765-775 (1994).

Katz, AE et al. The role of reverse-transcriptase polymerase chain reaction assay for prostate-specific antigen in the selection of patients for radical prostatectomy. Urology Clinics of North America 23: 541-549 (1996).

JONNIKSEN INGE ET AL: “Validity and usefulness of human tumor models established by intratibial cell inoculation in nude rats”. CANCER RESEARCH, vol. 54. n ° 7. 1994.

Kozlowski, JM et al. Prostate cancer and the invasive phenotype: Application of new in vivo and in vitro approaches. In: Tumor Progression and Metastasis, edited by Fidler, IJ and Nicholson, G. Alan R. Liss, Inc., New York, 1988, p. 189-231.

Lee, C. et al. In vitro and In vivo approaches to study metastasis in human prostatic cancer. Canc. Met Rev. 12: 2128 (1993).

Lim, DJ et al. Growth of an androgen-sensitive human prostate cancer cell line. LNCaP, in nude mice. Prostate 22: 109-118 (1993).

Liu, AY et al. Prostatic cell lineage markers: Emergence of BCL2 + cells of human prostate cancer xenograft LuCaP 23 following castration. Intl. J. Cancer 65: 85-89 (1996).

LUBAROFF ET AL: "Survival of human prostate carcinoma, benign hyperplastic prostate tissues, and IL-2-activated lymphocytes in SCID mice" THE PROSTATE vol. 27, 1995.

Maricelli, M. et al. Definition of the human androgen receptor gene structure permits the identification of mutations that cause androgen resistance: Premature termination of the receptor protein at amino acid residue 588 cause complete androgen resistance. Mol. Endocrinol. 909: 1105-1116 (1990).

Melchior, SW et al. Clinical relevance of prostate cells in the bone marrow of patients with clinically localized carcinoma of the prostate (CAP). J. Urology 157: 1718 (1997 (Abstract).

NAGABHUSHAN M ET AL: "CWR22: the first human prostate cancer xenograft with strongly androgen-dependent and relapsed strains both in vivo and in soft agar". CANCER RESEARCH. UNITED STATES 1 J UL 1996, vol. 56, n ° 13, July 1.

Netland, PA and Zetter, BR Metastatic potential of B16 melanoma cells after in vitro selection for organspecific adherence. J. Cell Biol. 10: 720-724 (1985).

Nishijima, Y. et al. Clinical significance of the vertebral vein in prostate cancer metastasis. Adv. Exp. Med. Biol. 324: 93-100 (1992).

Noel, A. et al. Basement membrance components (Matrigel) promote the tumorigenicity of human breast adenocarcinoma MCF7 cells and provide an in vivo model to assess the responsiveness of cells to estrogen. Biochemical Pharmacology 43: 1263-1267 (1992).

Olsson, CA et al. Reverse transcriptase-polymerase chain reaction assays for prostate cancer. Urology Clinics of North America 24: 367-378 (1997).

ES 2 290 964 T3

Pang, S. et al. Prostate tissue specificity of the prostate-specific antigen Promoter isolated from a patient with prostate cancer. Hum. Gene Ther. 6: 1417-1426 (1995).

Pretlow, TG et al. Transplantation of human prostatic adenocarcinoma into nude mice in Matrigel. Cancer Res. 51: 3814-3817 (1991).

PRETLOW et al., JOURNAL OF THE NATIONAL CARCINOMA INSTITUTE, March 3, 1993, Vol. 85, No. 5, "Xenografts of Primary Human Prostatic Cancer."

Saiki, RK, et al. Enzymatic amplification of b-globin genomic sequences and restriction site analysis for diagnosis of sickle cell anemia. Science 230: 1350-1354 (1895).

Sambrook, J. et al. Molecular Cloning: A Laboratory Manual, Ed.2. Cold Spring Harbor Press, Clod Spring Harbor, New York, 1989.

SATO et al., PROCEEDINGS OF THE AMERICAN UROLOGICAL ASSOCIATION, May 1996, Vol. 155, Suppl. 338A, abstract no. 109, "A Metastatic and Androgen-Sensitive Human Prostate Cancer Animal Model Using Orthotopic Inoculation of LNCaP Cells in SCID Mice".

Seiden, MV et al. Detection of circulating tumor cells in men with localized prostate cancer. J. Clin. Oncol. 12: 2364-2639 (1994).

Shevrin, DH et al. Development of skeletal metastasis by human prostate cancer in athymic nude mice. Clin. Exp. Metastasis 6: 401-409 (1988).

Shevrin, DH et al. Patterns of metastasis by the human prostate cancer cell line PC-3 in athymic nude mice. The Prostate 15: 187-194 (1989).

Sutherland, RW et al. Androgen receptor gene mutations are rarely associated with isolated penile hypospadias. J. Urol. 156: 828-831 (1996).

Thalmann, GN et al. Androgen-independent cancer progression and bone metastasis in the LNCaP model of human prostate cancer. Cancer Res. 54: 2577-2581 (1994).

VAN WEERDEN ET AL: “Development of seven new human prostate tumor xenograft models and their histopathological characterization” AMERICAN JORUNAL OF PATHOLOGY vol. 149, n ° 3, September 1996.

Wang, M. and Stearns, ME Isolation and characterization of 3 PC-3 prostatic tumor sublines which preferentially metastasize to select organs in SCID mice. Differentiation 48: 115-125 (1991).

Wainstein, MA et al. CWR22: Androgen-dependent xenograft model derived from a primary human prostatic carcinoma. Cancer Res. 54: 6049-6052 (1994).

Wood, DP et al. identification of bone marrow micrometastases in patients with prostate cancer. Cancer 74: 25332540 (1994).

WU HC ET AL: “Derivation of androgen-independent human LNCaP prostatic cancer cell sublines: role of bone stromal cells”. INTERNATIONAL JOURNAL OF CANCER. JOURNAL INTERNATIONAL DU CANCER. UNITED STATES May 1, 1994, vol. 57, n ° 3. May 1, 1994.

Zetter, BR et al. The cellular basis for prostate cancer metastasis. Adv. Exp. Med. Biol. 324: 39-43 (1992).