WO1999020308A1

WO1999020308A1 - A method for inhibiting metastasis and tumor cell growth by clinically modulating tumor expression of fas ligand

Info

Publication number: WO1999020308A1
Application number: PCT/US1998/022390
Authority: WO
Inventors: Emilio Barbera-Guillem
Original assignee: Biocrystal Ltd.
Priority date: 1997-10-23
Filing date: 1998-10-22
Publication date: 1999-04-29
Also published as: AU1273699A

Abstract

A method for impairing metastasis and tumor growth in an individual having solid nonlymphoid tumor that comprises FasL(-) tumor cells. The method comprises administering to the individual a therapeutical1y effective amount of a pharmaceutical composition that contacts and causes FasL expression by the tumor cells. Also disclosed is a method for treating FasL(-) tumor cells to become FasL comprising contacting the FasL(-) tumor cells with a therapeutically effective amount of a pharmaceutical composition that acts to cause FasL expression. The FasL+ tumor cells, in nonadherent conditions, may impair metastasis and tumor growth by contacting and inducing Fas mediated cytotoxicity of Fas expressing tumor cells, and/or Fas expressing B cells involved in a pro-tumor immune response. The composition for causing FasL expression in FasL(-) tumor cells comprises a biological modifier or an expression vector encoding FasL.

Description

A METHOD FOR INHIBITING METASTASIS AND TUMOR CELL GROWTH BY CLINICALLY MODULATING TUMOR EXPRESSION OF FAS LIGAND

FIELD OF THE INVENTION

The present invention is related to novel methods for anticancer therapy of certain tumor types in humans. More particularly, the present invention is related to the methods for impairing or inhibiting metastasis and tumor cell growth by inducing the expression of Fas ligand (FasL) in FasL-negative tumor cells.

BACKGROUND OF THE INVENTION

Regulation of cell number is determined by a balance between cell proliferation and cell death. One general mechanism of activation induced cell death seen in the process of development, differentiation, and cell turnover is apoptosis. Apoptosis is a characteristic form of cell death involving activation of one or more internally controlled pathways leading to autodigestion. Characteristic changes typical of a cell undergoing apoptosis include cell shrinkage and loss of contact with neighboring cells, cyto- skeletal disruption, membrane blebbing and alterations in the plasma membrane, and endonuclease activity-associated degradation of DNA.

Apoptosis can be induced by the binding and cross- linking of a cell surface receptor known as Fas. Human Fas (also known as APO-1 and CD95) is a cell surface protein consisting of 325 amino acids with a signal sequence at the NH₂-terminus and a membrane spanning region in the middle of the molecule. Fas appears to be constitutively expressed on cells of a varied, but limited, number of normal tissues, including skeletal muscle, liver, skin, heart, lung, kidney, and reproductive tissues (Leithauser et al., 1993, Labora tory Invest . 69:415; atanabe-Fukunaga et al., 1992, J. Immunol . , 148:1274). Malignant cells of hematologic or nonhematologic origin have also been demonstrated to express Fas (See, e.g., Leithauser et al., 1993, supra ) .

Fas-mediated apoptosis (also known as Fas-mediated cytotoxicity) requires cross-linking of Fas with either agonistic anti-Fas antibody, with cell bound FasL (Fas- ligand), or with soluble FasL (see, e.g. Alderson et al.,

1995, J. Exp . Med. 181:71-77; Yonehara et al., 1989, J. Exp . Med. 169:1747-1756; Suda et al., 1994, J. Exp . Med. 179: 873) . FasL is a type II transmembrane protein of the tumor necrosis factor family. Depending on the tumor type, FasL cell surface expression is variable; e.g., detectable in some tumors and absent in others. For those tumors expressing FasL, it has been suggested that such expression provides a mechanism of immune privilege of the tumors; i.e. a means by which the tumor evades immune-induced tumor cell depletion (Walker et al., 1997 J. Immunol . 158:4521-4). For example, FasL+ hepatocellular carcinomas were shown to kill Fas+ T lymphocytic cells in coculture (Strand et al., 1996, Na t . Med. 2:1361-6); FasL+ human colonic adenocarcinoma cell lines induced apoptosis of Fas+ T lymphocytic cells in coculture (SW480, Shiraki et al., 1997, Proc . Na tl . Acad. Sci . USA 94:6420-5; SW620, O'Connell et al., 1996, J. Exp . Med. 184:1075-82); FasL+ human lung carcinoma cell lines killed Fas+ T lymphocytic cells in coculture (Niehans et al., 1997, Cancer Res . 57:1361-6); and FasL+ melanoma cells induced apoptosis of Fas+ target cells in coculture (Hahne et al., 1996, Science 274:1363-6). These data suggest that FasL expression by tumor cells enhances tumorigenesis by killing Fas expressing immune effector cells (e.g., activated or tumor-reactive T cells) and surrounding Fas expressing tissue cells (e.g., hepatocytes; see Shiraki et al., 1997, supra ) . Further, cancer cells found to be FasL+ and Fas+ fail to undergo Fas-mediated apoptosis after treatment with agonistic anti-Fas antibody (O'Connell et al., 1996, supra) suggesting that tumor-expressed Fas did not transmit an apoptotic signal. Resistance to Fas- mediated apoptosis after anti-Fas antibody treatment has also been observed in nonhematopoietic tumors (Owen-Schaub et al., 1994, Cancer .Res. 54:1580-1586), human hepato a cells (Ni et al., 1994, Exp . Cell Res . 215:332-7), breast carcinoma (Keane et al., 1996, Cancer Res . 56:4791-8). Thus, that tumor cells (e.g., from breast, colon, testis, and liver) expressing Fas appear to have lost their sensitivity to anti-Fas mediated cytotoxicity, suggests that tumor cells escape the normal induction of apoptosis that occurs in these tissues (Micheau et al., 1997, J. Natl . Cancer Inst . 89:783-789).

Hence, a need still exists for a method to impair or inhibit metastasis and tumor cell growth, and with less systemic toxicity than the current standard treatments comprising chemotherapy and/or radiation therapy.

SUMMARY OF THE INVENTION

Accordingly, it is a primary object of the present invention to provide methods for treating individuals having FasL(-) tumors by inducing the expression of Fas ligand (FasL) in FasL(-) tumor cells, thereby impairing or inhibiting metastasis and tumor cell growth.

It is another object of the present invention to provide a method for treating individuals having FasL(-) tumors, wherein the method comprises inducing FasL expression in nonadherent FasL(-) tumor.

It is another object of the present invention to provide a method for impairing metastasis in individuals having FasL(-) tumors by inducing FasL expression in nonadherent tumor cells, and wherein the method is facilitated, at least in part, by Fas-mediated cytotoxicity of the tumor and by Fas-mediated cytotoxicity of tumor promoting B cells local or regional to the tumor.

The foregoing objects are based on the novel discoveries that metastatic cells and nonadherent tumor cells (under non-anchorage conditions) which are Fas+ can be induced to apoptosis by contact with FasL+ tumor cells; and that inducing FasL expression on FasL(-) tumor cells does not confer immune privilege to these cells m vivo, but instead targets the tumor cells for destruction. Additionally, as disclosed herein, and also m disclosed in detail n co-pending U.S. applications Serial Nos.

60/073882, 60/077970, and 60/083155, B lymphocytes exposed to shed tumor antigen can promote tumor progression. FasL+ tumors can contact and induce cytotoxicity of these Fas+ tumor promoting B cells. In one embodiment of the present invention, the method comprises inducing FasL expression in FasL(-) tumor cells (including nonadherent tumor cells) in an individual by administering to that individual a therapeutically effective amount of a pharmaceutical composition comprising a FasL inducing composition that comprises (a) a biological modifier for inducing or upregulatmg the cell-surface expression of FasL by the treated tumor cells; or (b) or an expression vector for inducing or upregulatmg the cell-surface expression of FasL by the treated tumor cells; or (c) a FasL inducing conjugate targeted to such tumor cells, wherein the conjugate comprises an affinity ligand coupled to a biological modifier; or (d) a FasL inducing conjugate targeted to such tumor cells, wherein the conjugate comprises an expression vector encoding FasL for inducing or upregulatmg the cell- surface expression of FasL by the treated tumor cells.

These and further features and advantages of the mvention will be better understood from the description of the preferred embodiments when considered m relation to the figures m which:

BRIEF DESCRIPTION OF THE DRAWINGS

FIG. 1A is a bar graph illustrating an apoptosis index of B16F10 melanoma cells grown m monolayer plotted versus the concentration of agonistic anti-Fas antibody. FIG. IB is a graph illustrating the number of colonies of B16F10 melanoma cells cultured in suspension, as plotted against various concentrations of agonistic anti-Fas antibody.

FIG. 1C is a graph illustrating the percentage of colony inhibition of B16F10 melanoma cells cultured in suspension, as plotted against various concentrations of agonistic anti- Fas antibody.

FIG. 2A is a illustration of a gel showing the absence or presence of amplified product representing FasL sequences expressed by B16F10; B16F10 cells transfected with pCDNA3 only; and B16F10 cells transfected with pCDNA3 containing the FasL cDNA.

FIG. 2B is an illustration of flow cytometric analysis for FasL expression was detected on B16F10 cells transfected with pCDNA3 only; and B16F10 cells transfected with pCDNA3 containing the FasL cDNA insert.

FIG. 3 is a bar graph illustrating the ability to form colonies in non-anchorage conditions for B16F10 cells; B16F10 cells transfected with pCDNA3 only ("Fas+/FasL- B16"); and B16F10 cells transfected with pCDNA3 with FasL cDNA insert ("Fas+ FasL+ B16" clones #3, #4, and #7) .

FIG. 4 is a bar graph illustrating tumor growth (as measured by spleen weight) in mice injected with B16F10 cells ("B16F10"); or Fas+/FasL- B16 cells; or PBS injection; or Fas+/FasL+ B16 clones #3, #4, and #7. FIG. 5 is a bar graph illustrating the number of lung metastases in mice receiving Fas+, FasL+ B16 cells as compared to the number of lung metastases in mice receiving Fas+, FasL- cells. FIG. 6 is a bar graph illustrating tumor growth (as measured by spleen weight) in mice injected with 3LL cells ("3LL"); or 3LL cells transfected with pCDNA3 only ("Fas-/FasL- 3LL"); or PBS; or 3LL cells transfected with pCDNA3 containing FasL cDNA ("Fas-/FasL+ 3LL") . FIG. 7 is a bar graph illustrating the number of lung metastases in mice receiving 3LL cells or Fas-/FasL+ 3LL cells . FIG. 8 is a bar graph illustrating relative populations of T lymphocytes, CD4 cells, CD8 cells, B lymphocytes, in mice injected with either B16F10 cells or Fas+/FasL+ B16 cells. FIG. 9 is a bar graph illustrating metastasis inhibition by a control group, C57BL/6 mice, muMT/muMT C57BL/6 mice, and nu/nu mice.

FIG. 10 is a bar graph illustrating the number of lung metastases, and splenic tumor burden in mice receiving B16F10 cells, irradiated B16F10 cells, Fas+/FasL+ B16, or irradiated Fas+/FasL+ B16; followed by a subsequent challenge with B16F10 cells.

FIG. 11 is a graph illustrating a comparison between spleen tumor size in B cell deficient (C57B/μMT/μMT) mice as compared to B cell competent (C57BL/6) mice. FIG. 12 is a bar graph illustrating a comparison between sub-cutaneous tumor growth in B cell deficient (C57B/μMT/μMT) mice as compared to B cell competent (C57BL/6) mice. FIG. 13 is a bar graph illustrating the average number of lung metastases in B cell deficient (C57B/μMT/μMT) mice as compared to B cell competent (C57BL/6) mice. FIG. 14 is a bar graph illustrating in vivo spleen tumor cell growth and liver metastasis (combined score) in the presence of splenic B lymphocytes from tumor bearing mice (T-SpL) , B lymphocytes from tumor (B-TIL) , and splenic B lymphocytes from normal mice (N-Spl) .

FIG. 15 is a bar graph illustrating in vi tro tumor growth of Met 129 tumor cells alone; or co-incubated with either CD8+ cells, CD4+ cells, CD8+ cells and CD4+ cells, B-TIL, CD8+ cells and B-TIL, or B-TIL and CD8+ cells and CD4+ cells. FIG. 16 is a bar graph illustrating the primary tumor (spleen) scores in mice treated with either saline, irrelevant goat IgG, or goat anti-mouse IgG and anti-mouse IgM. FIG. 17 is a bar graph illustrating the extra-regional

(lymph node) metastasis scores in mice treated with either saline, irrelevant goat IgG, or goat anti-mouse IgG and anti-mouse IgM.

FIG. 18 is a bar graph illustrating the liver metastasis scores in mice treated with either saline, irrelevant goat IgG, or goat anti-mouse IgG and anti-mouse IgM.

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS Definitions

The term "affinity ligand" is used herein, for purposes of the specification and claims, to mean a molecule which has binding specificity and avidity for a determinant associated with solid, nonlymphoid tumors. In general, affinity ligands are known to those skilled in the art to include, but are not limited to, lectins, monoclonal antibodies ("mAb", including chimeric or genetically modified monoclonal antibodies which may be preferable for administration to humans), peptides, and aptamers. The term "monoclonal antibody" is also used herein, for purposes of the specification and claims, to include immunoreactive fragments derived from a mAb molecule, which fragments retain all or a portion of the binding function of the whole mAb molecule. Such immunoreactive fragments are known to those skilled in the art to include F(ab')₂, Fab', Fab, Fv, scFV, Fd¹ and Fd fragments. Methods for producing the various fragments from mAbs are well known in the art (see, e.g., Plϋckthum, 1992, Immunol . Rev. 130:152-188). For example, F(ab')₂ can be produced by pepsin digestion of the monoclonal antibody, and Fab' may be produced by reducing the disulfide bridges of F(ab')₂ fragments. Fab fragments can be produced by papain digestion of the monoclonal antibody, whereas Fv can be prepared according to methods described in U.S. Patent No. 4,642,334. Single chain antibodies can be produced as described in U.S. Patent No. 4,946,778. The construction of chimeric antibodies is now a straightforward procedure (Adair, 1992, Immunological

Reviews 130: 5-40,) in which the chimeric antibody is made by joining the murine variable region to a human constant region. Additionally, "humanized" antibodies may be made by joining the hypervariable regions of the murine monoclonal antibody to a constant region and portions of variable region (light chain and heavy chain) sequences of human immunoglobulins using one of several techniques known in the art (Adair, 1992, supra ; Singer et al., 1993, J. Immunol . 150:2844-2857). Methods for making a chimeric non- human/human mAb are described in general and in detail in U.S. Patent No. 5,736,137. Aptamers can be made against B cell determinants using methods described in U.S. Patent No. 5,789,157. See, for example, Table 2 herein for molecules illustrative of affinity ligands that may be useful in the compositions and methods of the present invention. The term "lymphoid tissue" is used herein, for purposes of the specification and claims, to mean a tissue which contains localized areas of antigen presenting cells (e.g., follicular or germinal center dendritic cells) and B lymphocytes, and in which can be induced an immune response involving B cells. An example of such localized areas comprises germinal centers. Such lymphoid tissues comprise tissues including, but not limited to, lymph nodes; milky patches in the mesenterium of the intestine; omentum; appendix; Peyer' s patches; loose connective tissue (e.g., associated with vessels in the walls of the aorta) ; lymphatic vessels; submucosal spaces; subserosa spaces; peritoneal cavity; ligaments (e.g., gastrohepatic ligament); artherosclerotic plaques containing trapped B cells; and epineura . The term "B cells" is used herein, for purposes of the specification and claims, to mean nonmalignant (nonlymphomic) B lymphocytes which express Fas on their surface (Fas+) such as may be displayed by germinal center B cells or B cells infiltrating a tumor. In a preferred embodiment, such B cells express both Fas and cell surface- bound immunoglobulin comprising antibody against shed tumor antigen, such as may be displayed by germinal center B cells in lymphoid tissues regional or distal to a solid, non- lymphoid tumor, or B cells infiltrating a solid, nonlymphoid tumor. The term "B cells" also includes mature B cells, and shed tumor antigen-specific memory B cells primed to be activated by shed tumor antigen to proliferate and secrete tumor promoting factor.

The term "metastases" or "metastatic tumor cell" is used herein, for purposes of the specification and claims, to mean a metastasis from a primary tumor wherein the primary tumor is a solid, non-lymphoid tumor, as will be more apparent from the following embodiments.

The terms "solid, non-lymphoid tumor" or "tumor" are used hereinafter, for purposes of the specification and claims, to mean any primary tumor of nonhematopoietic and ductal epithelial cell origin, including, but not limited to, tumors originating in the liver, lung, brain, lymph node, bone marrow, adrenal gland, breast, colon, pancreas, stomach, or reproductive tract (cervix, ovaries, endometrium etc.); and which produces shed tumor antigen (e.g., serous, or endometroid, or mucinous tumors) .

The term "nonadherent tumor cells" is used herein, for purposes of the specification and claims, to mean a metastatic tumor cell such as may be found moving through tissues of a body; a tumor cell circulating in blood, lymph or other body fluids; a tumor cell having a high potential to metastasize (e.g., express-ing IL-2Rα as described in more detail in U.S. Patent No. 5,536,642); or a solid, non- lymphoid tumor cell which is non-adherent as existing in non-anchorage conditions in a tissue environment. This includes tumor cells growing in clusters without visible intercellular connective matrix or desmoplastic or angio- genic processes, such as exhibited by metastatic growth in the lymphatic sinuses. Non-anchorage conditions, for example, exist during lung metastases formation during cell arrest, and at some points in colony formation. Such tumor ^"cells are non-adherent at points when they circulate freely in the blood or lymph systems.

The term "shed tumor antigen" is used herein, for purposes of the specification and claims, to mean a glycoprotein which:

(a) is released from a primary tumor or its metastases, thereby becoming soluble and allowing movement into lymphoid tissues regional or distal to the primary tumor or its metastases; (b) comprises either a cryptic antigen, or an antigen comprising a point mutation;

(c) interacts with a B cell surface receptor (by itself or as presented by follicular or germinal center dendritic cells), in activating (by itself or in the presence of another B cell stimulatory factor, such as may be displayed by presenting follicular or germinal center dendritic cells) the B cells to proliferate and produce/secrete tumor promoting factor; and

(d) may induce a dominant T cell dependent immune response resulting in the production and secretion of tumor promoting factor by such B cells or their progeny (e.g., plasma cells) .

With regard to the tumor antigen being soluble, the tumor antigen is noncellular ("shed") tumor. Non-cellular tumor antigen comprises soluble tumor antigen that is not an integral part of a living tumor cell. Such shed tumor antigen exists in a form selected from the group consisting of free form (shed tumor antigen alone), in an immune complex form (shed tumor antigen bound to anti-shed tumor antigen antibody) , in a form as presented on the surface of a follicular or germinal center dendritic cells (antigen presenting cell) , in a form as bound to the cell surface of B cells, and as a form in tumor cell membranes existing apart from living tumor cells (i.e., soluble membrane complexes representing portions of dead tumor cells) .

With regard to the shed tumor antigen being a cryptic antigen, and glycoprotem in composition; and for purposes of illustration, and not limitation; exemplifying such shed tumor are mucms and mucm-like molecules. For a review of the structure of the family of mucm molecules, see Finn et al. (1995, Immunol . Rev. , 145:62-89). Briefly, mucms are high molecular weight glycoprotems (e.g., greater than about 100 kiloDaltons (kD) m molecular mass) of which a significant portion of the polypeptide backbone comprises a domain composed of a tandomly repeating peptide subunits (e.g. about 20 to about 125 repeats). Mucms are found on normal ductal epithelial cells in sequestered locations that are not normally exposed to the immune system (e.g., restricted to the lumen of duct); and hence, tolerance has not been established. Exposing the mucm to the immune system, such as by compromising the integrity of the normal ductal epithelium, could result in immune recognition of mucm as a cryptic antigen. Thus, a cryptic antigen is used herein to refer to a high molecular weight glycoprotem which occurs in normal tissue, but m a sequestered location which does not normally expose the glycoprotem to the immune system. However, m processes such as transformation or tumor development, the glycoprotem is presented to the immune system and thereby becomes an antigen. The mucm expressed by tumor cells generally has the same polypeptide backbone as mucm expressed by normal epithelial cells. Exposing the mucm to the immune system, such as by compromising the integrity of the normal ductal epithelium, could result in immune recognition of mucm as a cryptic antigen. Thus, a cryptic antigen is used herein to refer to a high molecular weight glycoprotem which occurs m normal tissue, but in a sequestered location which does not normally expose the glycoprotem to the immune system. However, in processes such as transformation or tumor development, the glycoprotem is presented to the immune system and thereby becomes an antigen. Further, due to various factors (e.g., the increased production of mucm, lack of availability of glycosyltransferases) , tumor cells produce mucin in an underglycosylated (incompletely glycosylated) form and/or in a form of altered glycosylation (e.g., with a terminal sialic acid group) . An immune response against mucin produced by tumor cells is thought to be primarily directed against one or more epitopes on the mucin glycoprotein which is exposed to the immune system as a result of under-glycosylation or altered glycosylation. Thus, because of the underglycosylation or altered glycosylation in growing tumors, the shed tumor mucin has epitopes not normally found on mucin or not normally exposed to the immune system. Such epitopes may include, but are not limited to, carbohydrate epitopes comprising the sialyl Tn (sTn) antigen (comprising the NeuAc portion of NeuAcα— 6GalNAc l—> O-Ser- or Thr), the Tn antigen (comprising the GalNAc portion of NeuAcα→ δGalNAcαl→ O-Ser- or Thr), the T antigen, and other sialic acid containing epitopes (e.g., NeuAc α2 on the carbohydrate chains NeuAcα2- 6Gal—>0-Ser- or Thr; NeuAcα2→ 3Gal→0-Ser- or Thr; or NeuAcα2→ 3GalNAc-»0- Ser- or Thr) . An example of a mucin-like glycoprotein which is differentially glycosylated by tumor cells and is shed by tumor cells is SSEA-1 antigen.

Tumor-associated glycoproteins, and characterizations such as nature of carbohydrate chain structure and/or monoclonal antibody binding, are known to those skilled in the art (see, e.g., Table V of Hakomori, 1989, Adv. Cancer Res . 52:257-331). Tumor-associated glycoproteins which are known to those skilled in the art as being found in a soluble form include, but are not limited, to the human equivalents of those presented in Table 1. Table 1

1- Lee et al., 1992 J. Formos . Med. Assoc . 91:760-3.

2- Arai et al., 1990, Jpn . J. Clm . Oncol . 20:145-53.

3- Takai et al., 1991, Nippon Shokakibyo Gakka i Zasshi , 88:170-4.

4-Meden and Kuhn, 1997, Eur . J. Obstet . Gynecol . Reprod .

Biol . 71:173-9.

5-Kohno et al., 1989, Cancer Res . 49:3412-9.

For purposes of illustration, and not limitation, in a preferred embodiment of the present invention, the shed tumor antigen comprises one or more antigens on the gene product of the MUC-1 gene (also known as polymorphic epithelial mucm) . In this preferred embodiment, the epitopes of the shed tumor antigen to which anti-shed tumor antigen-antibody are directed, include the sTn antigen, Tn antigen, and other sialic acid containing epitopes (e.g., NeuAc α.2 on the carbohydrate chains NeuAcα2- 6Gal—>0-Ser- or Thr; NeuAcα2→ 3Gal-0-Ser- or Thr).

With regard to a dominant T cell dependent response that may be induced by the shed tumor antigen, it is known that the immune response induced by tumor cell-associated mucm is predominantly cellular (CD8+) , with little or no antibody produced. In contrast to mucm bound to the surface of whole tumor cells, shed mucm induces an antibody and T helper cell response (TH₂) , but not cytotoxic T cell responses (Apostolopoulos et al., 1994, Cancer Res . 54: 5186) . However, it was not known that such an immune response may promote tumor cell growth and metastasis. In that regard, it appears that shed tumor antigen may be an lmmunodommant tumor antigen as compared with other cell- associated antigens of the tumor presented to the immune system in the process of tumoπgenesis . The carbohydrate portion of the shed tumor antigen may play a role m potentiating (e.g., adjuvant-like effect) this dominant T cell dependent response against an epitope of the shed tumor antigen. More particularly, shed tumor antigen may induce a T cell dependent response, the eventual result of which may be that a significant amount of the antibody response is produced against the shed tumor antigen, relative to that induced against any other single tumor cell-associated antigen ("dominant" response). The result of this specific type of immune response is a form of tolerization of the immune system to some tumor antigens other than shed tumor antigen; and hence, the inhibition of the development of an effective antitumor humoral immune response.

The term "hyperplastic" with reference to germinal centers or lymphoid tissues, is used herein, for purposes of the specification and claims, to mean a reactive process which includes an expansion in the size of germinal centers or germinal center equivalents (see, e.g., Weidner et al., 1982, Arch. Derma tol . Res . 272:155-161), and an infiltration and/or proliferation of B cells, in lymphoid tissue regional (draining) or distal to a primary tumor or its metastases, wherein the lymphoid tissue contains shed tumor antigen.

The term "tumor promoting factor" is used herein, for purposes of the specification and claims, to mean one or more soluble molecules released/secreted from B cells or their progeny (plasma cells) , wherein

(a) production of tumor promoting factor by B cells is induced by an interaction between the B cells and the shed tumor antigen in the presence of another B cell stimulatory factor (e.g., T helper cell antigen presentation molecule or as presented by follicular or germinal center dendritic cells in lymphoid tissue) ;

(b) tumor promoting factor comprises one or more molecules which mediate inflammation, and consists primarily of an anti-shed tumor antigen antibody, but may also consist of one or more cytokines, or a combination thereof; and

(c) tumor promoting factor, as an IgG antibody, acts indirectly by interacting with, and binding to shed tumor antigen in forming immune complexes, wherein the immune complexes act on host cells which are mediators of inflammation (e.g., granulocytes) ; and/or directly (acting on the tumor cell itself) to mediate tumor progression including, but not limited to, promoting tumor growth and/or metastasis, and/or advancing stage of malignancy. In either interaction, the tumor promoting factor may mediate tumor progression by one or more mechanisms which may include, but are not limited to, the formation of immune complexes which induce a cascade of inflammatory processes which promote tumor development; down-regulate T helper cells which normally may mount an immune response against tumor cell- associated antigens, thereby inhibiting development of an antitumor immune response; inhibiting tumor cell-associated antigen presentation to human tumor-specific cytotoxic lymphocytes; increasing expression on primary tumor cells of cell-surface molecules which promote metastasis; by cross- linking Fc gamma receptors on tumor cells, activates tyrosine kinase production which induces tumor proliferation and/or an increase in tumor production and secretion of shed tumor antigen; and facilitating a local environment which mediates spread and/or development of metastases beyond the primary tumor and to lymphoid tissues regional or distal to the primary tumor.

The terms "preferentially expressed" is used herein, in conjunction with a determinant of a tumor cell of a solid, nonlymphoid tumor, and particularly of a nonadherent tumor cell, for purposes of the specification and claims, to mean a tumor-associated molecule expressed on the surface of the tumor cell, wherein the level of expression (measured directly or indirectly, and including by presence or by activity) of such determinant is at least 3 to 4 times that expressed by normal cells of that tissue type or by normal cells found in the blood and lymph circulatory system. Thus, preferential expression may include detection of expression of the determinant on the surface of a nonadherent tumor cell, and absence of detection of the same molecule on normal cells of that tissue type or by normal cells found in the blood and lymph circulatory system; or a log greater expression of the same molecule on a nonadherent tumor cell as compared to expression on normal cells of that tissue type or by normal cells found in the blood and lymph circulatory system. Determinants known to those skilled in the art which are preferentially expressed by tumor cells may include, but are not limited to, the determinants recognized by the monoclonal antibodies (mAb) and lectins listed in Table 2.

TABLE 2

specific tumor type mAb

HEA125 or HT29-15 colorectal carcinoma BCD-B4 or NCRC-11 breast carcinoma anti-PSA antibody or PD41 prostate carcinoma

ALT-04 lung carcinoma

HMD4 or COC183B2 ovarian carcinoma

HEA-125, MM46, or 9.2.27 melanomas MGb 2, or ZCE 025 gastric carcinoma

BW494 pancreatic carcinoma

CEB9 uterine cervical carcinoma

HSAN 1.2 neuroblastoma

HISL-19 neuroendocrinomas TRA-1-60 seminoma

B72.3, CC49, hCTMOl, HbSTn, TKH2, mucinous adeno- JTlOe, BCP7, and others (see, e.g., carcinomas including U.S. Patent No. 5,512,443, and Tumor pancreas, ovary, Biology, 19:1-152, 1998) stomach, colorectum, breast, uterus, gallbladder, and prostate.

Lectins

Sambucus nigra agglutinin (SNA) , mucinous adenocarcinomas Carcinoscorpius rotunda cauda sialic acid binding lectin, Peanut agglutinin (PNA), Maackia amurensis leukoagglutinin (MAL) , soybean agglutinin,

Rana catesbeiana lectin, Amaranthin (ACA) , Bauhinia purpurea (BPA) , Vicia villosa (VVA)

Helix pomatia agglutinin (HPA) gastric cancer

The term "individual" is used herein, for purposes of the specification and claims, to mean a mammal; and preferably a human, including an individual having a primary tumor comprising a solid, non-lymphoid tumor and/or its metastases, or an individual who has been treated for a solid, nonlymphoid tumor and thereby inherently carries a risk of recurrence because of circulating tumor cells. In either case, the individual is at risk for developing, or has developed, a pro-tumor immune response. In one embodiment of the method and compositions according to the present invention, inducing FasL expression in FasL(-) solid, nonlymphoid tumors may target, contact, and induce Fas-mediated cytotoxicity in, B cells localized in hyperplastic lymphoid tissues and/or infiltrating a solid, nonlymphoid in such an individual. The term "FasL inducing conjugate" is used herein, for purposes of the specification and claims, to mean a composition comprised of (a) at least one affinity ligand according to the present invention; and (b) at least one biological modifier, or an expression vector, for inducing or upregulating FasL express-ion, or a combination of such a vector and biological modifier. The methods for coupling an affinity ligand to various compositions, such as may be used to couple a biological modifier or a expression vector, are well known to those skilled in the art (See, for example, antibody conjugates as reviewed by Ghetie et al., 1994, Pharmacol . Ther. 63:209-34). Often such methods utilize one of several available hetero-bifunctional reagents used for coupling or linking molecules. Such bifunctional linkers include, but are not limited to, N-hydroxy succmimide-based linkers cystamme, glutaraldehyde, and diamino hexane. The term "biological modifier" is used herein for purposes of the specification and claims, to mean a composition that induces or upregulates FasL expression in solid, nonlymohoid tumor cells (including nonadherent tumor cells thereof) that are, before treatment, FasL(-); and whereupon such tumors express FasL and when they are nonadherent, they may induce Fas-mediated apoptosis upon contact with Fas+ tumor cells and/or Fas+ B cells. Such compositions may include, but are not limited to, tumor necrosis factor alpha (TNF-α) , staphylococcal enterotoxm B, dexamethasone, mterleukm-1 (IL-1), HTLV-I Tax (trans- activator protein) , halopeπdol, gonadotropm releasing hormone (GnRH) for GnRH receptor-bearing tumors, and other compositions which can be identified using the methods for detection of FasL expression described herein. These methods include reverse transcπptase-polymerase chain reaction (RT-PCR) , lmmunohistochemical staining, lmmuno- fluorescence, flow cytometry, and functional bioassays comprising treated cells coculture with Fas+ target cells in observing for Fas-mediated apoptosis. For the purposes of the specification and claims, "biological modifier" does not refer to such antitumor drugs such as fludarabme, doxo- rubicm, cisplatm, etoposide, methotrexate, staurosporme, or topotecan because although these drugs may induce FasL expression, these antitumor drugs mediate apoptosis in a Fas-independent pathway (Eischen et al., 1997, Blood 90:935- 943; Villunger et al . , 1997, Cancer Research 57:3331-3334). The term "FasL negative" or "FasL(-)" is used herein for purposes of the specification and claims, to mean solid, non-lymphoid tumor cells, and particularly nonadherent tumor cells thereof, which lack detectable expression of FasL either on the surface of the cell or at the mRNA level, as determined within the limits of detection by methods conventionally used by those skilled in the art to detect FasL expression including, but not limited to, RT-PCR, lmmunohistochemical staining, immunofluorescence flow cytometry, and functional bioassays, as will be more apparent from the following embodiments.

The term "FasL positive" or "FasL+" is used herein for purposes of the specification and claims, to mean solid, non-lymphoid tumor cells, particularly nonadherent tumor cells thereof, having detectable expression of FasL on the surface of the cell, as determined by methods conventionally used by those skilled in the art to detect FasL expression including, but not limited to, RT-PCR, lmmunohistochemical staining, immunofluorescence flow cytometry, and functional bioassays, as will be more apparent from the following embodiments.

The term "vector" or "expression vector" is used herein for purposes of the specification and claims, to mean vectors used in accordance with the present invention as a vehicle for introducing into and expressing in a mammalian cell a gene encoding FasL. As known to those skilled in the art, such vectors can be selected from plasmids, viruses, and retroviruses . For a recent review of vectors useful in gene therapy of cancer, see Weichselbaum and Kufe (1997, Lancet , 349 : S10-S12) . The features of a vector which make it useful m the methods of the present invention include that it have a selection marker for identifying vector which has inserted therein a gene encoding FasL; restriction sites to facilitate cloning of a gene encoding FasL; and the ability to enter and/or replicate in mammalian cells. In a one embodiment, the vector further comprises an activation- ducible cis-acting regulatory element for upregulatmg FasL expression (Egr-3; SEQ ID NO: 16 see, e.g., Mittelstadt and Ashwell, 1998, Mol . Cell . Biol . 18:3744-3751), wherein the regulatory element is operatively linked to the gene encoding FasL in a manner permitting upregulation (for example, Egr-3 can be located approximately 200 bp upstream of initiation codon) . Examples of a preferred vector for the in vivo introduction of a recombinant vector into mammalian cells include, but are not limited to viral vectors. Virus- based vectors are one preferred vehicle as they infect cells in vivo, wherein during the infection process the viral genetic material is transferred into the cells. A retro- viral vector, such as a plasmid containing AAV (Adeno- associated virus) sequences, has been described previously (see for example Chatterjee et al., 1992, Science, 258:1485- 1488; U.S. Patent No. 5,252,479, herein incorporated by reference) . In one embodiment, the AAV vector contains inverted terminal repeats (ITR) with a selection marker such as the gene encoding neomycin resistance, an SV40 promoter, a polylinker, and other plasmid sequences. A promoter in the ITR drives the expression of the neomycin phospho- transferase gene, whereas the SV40 promoter drives expression of the operably linked FasL gene to be expressed. The inverted terminal repeats of the AAV vector provide a means for integrating the vector, and sequences inserted therein, into the chromosome as the repeats serve as a sequence which has been shown to insert site-specifically, rather than randomly, into chromosomes. Examples of other vectors for the in vi tro or in vivo introduction into mammalian cells include retroviral vectors (Miller et al., 1989, BioTechniques 7:980-990; Korman et al., 1987, Proc .

Na tl . Acad. Sci . USA 84:2150-54), papovavirus episomes (U.S. Patent No. 5,624,820, herein incorporated by reference), and adenovirus vectors (U.S. Patent No. 5,585,362, herein incorporated by reference) . Such vectors can utilize tissue-specific promoters in targeting expression to tumor cells of particular tissue types. For example, the alpha-1- ^"antitrypsin promoter and the albumin promoter are promoters activated primarily in liver tissue; and thus, may be used to target expression of FasL in tumor cells of hepatic origin. Similarly, the α-fetoprotein promoter may be used to target expression of FasL in hepatomas. The DF3/MUC-1 promoter may be used to target expression of FasL in breast cancer cells.

A drawback to systemic therapies is the lack of selectively delivering the therapy to its intended target, diseased tissue, rather than to normal tissue. In that regard, activation-induced cell death of tumors has been complicated by the apparent resistance of Fas+ tumor cells to Fas-mediated cytotoxicity (see, e.g., O'Connell et al., 1996, supra ) . Additionally, current theory is that FasL expression by tumor cells enhances tumorigenesis (immune privilege or tumor evasion) by killing Fas expressing immune effector cells and surrounding Fas expressing tissue cells (Strand et al., 1996, supra ; Shiraki et al., 1997, supra ; O'Connell et al., 1996, supra Niehans et al., 1997, supra ; Hahne et al., 1996, supra ) .

The present invention relates to the discoveries that Fas expressed by nonadherent tumor cells (including circulating tumor cells and metastatic cells) can, unexpectedly, transduce an apoptotic signal when cross- linked by FasL; that nonadherent tumor cells develop fratricidal Fas/FasL mediated apoptosis; that the metastatic capacity of a malignant tumor can be abrogated if Fas and FasL are coexpressed in tumor cells that are nonadherent; and that FasL expression in nonadherent tumor cells may contact B cells involved in a pro-tumor immune response, and may induce Fas-mediated cytotoxicity of such B cells, thereby inhibiting or impairing the B cell involvement in progression of solid, nonlymphoid tumor and metastasis. In one embodiment of the present invention, the method comprises inducing FasL expression in FasL(-) solid, nonlymphoid tumor cells in an individual by administering m a site-directed delivery to such tumor a therapeutically effective amount of a pharmaceutical composition comprising a biological modifier. The biological modifier, upon contact with such Fas+ tumor cells, upregulates or induces the cell-surface expression of FasL, thereby making the treated Fas+ tumor cells also FasL+. Thus, those Fas+ and FasL+ tumor cells in non-anchorage conditions in the treated site, or that metastasize from the treated site, may participate in the fratricidal Fas/FasL mediated apoptosis, and may also contact and induce Fas-mediated cytotoxicity of Fas+ B cells involved in a pro-tumor immune response. In another illustration of this embodiment, the composition further comprises an affinity ligand linked to the biological modifier, in forming a FasL inducing conjugate. The affinity ligand of the FasL inducing conjugate has binding specificity for a tumor-associated molecule (determinant) preferentially expressed by solid, nonlymphoid tumor cells, and particularly by nonadherent tumor cells. Thus, the affinity ligand facilitates selective delivery of the biological modifier to its intended target. Thus, those Fas+ and FasL+ tumor cells in non-anchorage conditions in the treated site, or that metastasize from the treated site, may participate in the fratricidal Fas/FasL mediated apoptosis, and may also contact and induce Fas-mediated cytotoxicity of Fas+ B cells involved in a pro-tumor immune response. In these embodiments, the composition may further comprise a pharmaceutically acceptable carrier. Such pharmaceutically acceptable carriers are known in the art to include, but are not limited to, physiological solutions, liposomes, and other delivery vehicles.

In another embodiment of the present invention, the method comprises inducing FasL expression in FasL(-) Fas+ solid, nonlymphoid tumor cells in an individual by administering to that individual a therapeutically effective amount of a pharmaceutical composition comprising a vector. ^"The vector, in accordance with the present invention, is a vehicle for introducing into and expressing in the targeted tumor cells a gene encoding FasL. The vector, upon entry into such tumor cells, upregulates or induces the cell- surface expression of FasL, thereby making the treated tumor cells FasL+. Thus, those Fas+ and FasL+ tumor cells in non- anchorage conditions in the treated site, or that metastasize from the treated site, may participate in the fratricidal Fas/FasL mediated apoptosis, and may also contact and induce Fas-mediated cytotoxicity of Fas+ B cells involved in a pro-tumor immune response.

In another illustration of this embodiment, the composition further comprises a affinity ligand linked to the vector, in forming a FasL inducing conjugate. The affinity ligand of the FasL inducing conjugate has binding specificity for a tumor-associated molecule preferentially expressed by a tumor cell, particularly by a nonadherent tumor cell. Thus, the affinity ligand facilitates selectively delivery of the vector to its intended target: Fas+ solid, nonlymphoid tumor cells, including nonadherent tumor cells (e.g., circulating tumor cells and metastatic tumor cells) . In this embodiment, the composition may further comprise a pharmaceutically acceptable carrier. Such pharmaceutically acceptable carriers are known in the art to include, but are not limited to, physiological solutions, liposomes, other delivery vehicles, and compositions which facilitate infection or transfection of the tumor cell by the vector (e.g., microparticles which permit or enhance uptake or introduction of vector into the target cells) .

For purposes of the description, the methods of the present invention will be illustrated in the following examples. In the following examples used to illustrate the invention, it is important to note that mice have been validated as a model for the evaluation of antitumor agents because the model has been shown to reflect the clinical effectiveness of antitumor agents in original patients treated with these agents; and reflects antitumor effects from the agents, such as tumor regression or inhibition of tumor growth, as consistent with the activity against the corresponding types of clinical cancer (See for example, Neuwalt et al., 1985, Cancer Res . 45:2827-2833; Ovejera et al., 1978, Annals of Clin . and Lab . Science 8:50).

EXAMPLE 1 This Example illustrates the difference in the ability of tumor-expressed Fas to transmit an apoptotic signal, depending on cell adhesion status. The B16F10 melanoma cell line, well-characterized and of high metastatic potential, was selected as the FasL(-), Fas+ tumor cells in this illustration. Lack of FasL expression by B16F10 melanoma cells was confirmed by Western blot, Northern blot, and RT-PCR analyses. Fas expression by B16F10 melanoma cells was detected by flow cytometry, and RT-PCR analyses. The ability of Fas+, expressed by these melanoma cells, to transmit an apoptotic signal was tested by cross-linking of Fas with agonistic anti-Fas antibody (IgG antibody- clone Jo2, commercially available from Pharmingen) . B16F10 cells were cultured in a confluent monolayer on 24 well plates, and then incubated for 48 hours with increasing concentrations of the anti-Fas antibody Jo2. The concentration of antibody added was either 0 ng/ml, 100 ng/ml, 1000 ng/ml, or 10,000 ng/ml. After antibody treatment, 1 μg/ml of ethidium bromide was added to the tissue culture medium, and the number of apoptotic cells (com- prising red fluorescent cells with pyknotic nucleus) were counted using flow cytometry (see, e.g., Gorczyca et al., 1993, Cancer Res . 53:3186-92). An apoptosis index was calculated by dividing the number of apoptotic cells by the number of living cells. The average results for each concentration added are shown in FIG. 1A. As shown in FIG. 1A, Fas cross-linkage with an agonistic anti-Fas antibody did not induce evident apoptosis. Further, Fas cross- linkage with an agonistic anti-Fas antibody did not inhibit growth of B16F10 melanoma cells in an in vi tro culture. These results suggest that the tumor-expressed Fas does not efficiently transmit an apoptotic signal.

Fas+ B16F10 cells were also tested for their ability to transmit an apoptotic signal by culturing the cells in suspension (non-anchorage conditions) and then cross-linking Fas with agonistic anti-Fas antibody. The melanoma cells (10⁴ cells per well in a 24 well plate) were cultured in a suspension in a fibrin clot (1.5 ml), and then incubated for 48 hours with increasing concentrations of the anti-Fas antibody Jo2 (at 0 ng/ml, 100 ng/ml, 1000 ng/ml, or 10,000 ng/ml). After antibody treatment, the number of cell colonies per well were counted under phase contrast microscopy for determining incidence of apoptosis. The number of colonies were compared to the number of colonies in the control (no antibody added) in determining the % number of colonies (see, FIG. IB) . The percentage of colony inhibition was calculated using the formula: 100 x [Nc - Ne] /Nc wherein Nc is the number of colonies in the control, and Ne is the number of colonies in each concentration of antibody (>0 ng/ml) (see FIG. 1C) . As shown by FIGs. IB and 1C, Fas cross-linkage with an agonistic anti-Fas antibody induced dose-related reductions in the number of B16F10 cell colonies (FIG. IB), and a corresponding increase in the % colony inhibition (FIG. 1C) . That colony formation in non- anchorage conditions was significantly reduced in the presence of agonistic anti-Fas antibody indicates that tumor-expressed Fas can transmit apoptotic signals in tumor cells grown in non-anchorage conditions.

EXAMPLE 2 This Example illustrates the differences in tumor growth and metastatic behavior in Fas/FasL normal mammals of tumor cells either FasL+ or FasL(-). In this illustration, B16F10 melanoma cells were transfected with a mammalian expression vector containing FasL cDNA. The nucleotide sequence of the human FasL gene is provided herein as SEQ ID NO: 1 (as described previously, e.g., Takahashi et al., 1994, Int . Immunol . 6:1567-74). Other mammalian FasL nucleotide sequences are also known to those skilled in the art (Peitsch and Tschopp, 1995, Mol . Immunol . 32:761-72; Suda et al., 1993, Cell 75:1169-78). As an example, murine FasL cDNA was subcloned into pCDNA3 (commercially available from Invitrogen) downstream and operably linked to the cytomegalovirus (CMV) promoter using a restriction enzyme ( Jbal). Restriction enzyme digestion of plasmid DNA from individual clones with PstI distinguished clones in the positive orientation (bands of 4.08 kb, 1.91 kb and 0.357 kb; "positive orientation" indicates the formation of an expression vector capable of expressing in mammalian cells FasL message under the control of the CMV promoter) , versus clones in the reverse orientation (bands of 4.08 kb, 1.68 kb, and 0.613 kb) . B16F10 cells were transfected with plasmid DNA containing the FasL gene in a positive orientation, using a transfection reagent (lipo-fectin) . Selection for transfected cells was performed by the addition of G418 (neomycin) to the culture. Following cell selection, the transfected cells were cloned by limiting dilution in the presence of neomycin. Expression of FasL by the transfected B16F10 clones was then confirmed by RT-PCR (mRNA level) and by flow cytometry (protein level) . As a control, the same process of transfection and cloning was performed using pCDNA3 alone (e.g., without the FasL cDNA insert) .

FasL mRNA was detected from the transfected B16F10 cells by isolating RNA from the cells by lysis in guanidine thiocyanate followed by phenol chloroform extraction and ethanol precipitation. cDNA was synthesized using AMV reverse transcriptase and primers according to the manufacturers directions. FasL cDNA was amplified by using polymerase chain reaction and SEQ ID NO: 2 as the sense primer and SEQ ID NO: 3 as the antisense primer. The reactions were carried out in a 50μl volume with 0.1 μM of each primer, 50 μM dNTP, and 1.5 mM MgCl₂. Denaturing was done at 96°C for 15 seconds, annealing at 55°C for 30 seconds and polymerization for 72°C for 3 minutes, for 40 cycles. The result of the polymerase chain reaction using these two primers was an amplified product of 538 base pairs (bp) . The 538 bp amplified product was purified by agarose gel electrophoresis and visualized with ethidium bromide staining. As controls, the RT-PCR technique was performed using RNA from B16F10 cells, and from B16F10 cells transfected with pCDNA3 only. As shown in FIG. 2A, B16F10 (untransfected) cells do not show any FasL mRNA expression (lanes 1 and 2); B16F10 cells transfected with pCDNA3 only do not show any FasL mRNA expression (lanes 5 and 6) ; whereas B16F10 cells transfected with pCDNA3 containing the FasL cDNA insert show significant FasL mRNA expression (lanes 3 and 4) .

FasL expression was detected on transfected B16F10 cells by harvesting the cells in culture by gentle scraping. Cells were washed in phosphate buffered saline (PBS) , fixed with 3% paraformaldehyde, and then incubated with bio- tinylated anti-FasL (Jo2) for 30 minutes at 4°C, and washed in PBS (containing 2% FCS) . Conjugate (phycoerythrin- conjugated streptavidin) was incubated with the cells for 30 minutes at 4°C. Cells were washed again in PBS, and flow cytometric analysis was performed at 610 nm. Cells in each sample were simultaneously measured for forward light scatter, side scatter, and green fluorescence and red fluorescence emissions. The data was stored and analyzed using standard methods of analysis. Quantitation of FasL expression was based on an examination of 10,000 cells for each determination as measured by semiautomatic evaluation. As shown in FIG. 2B, B16F10 cells transfected with pCDNA3 only do not show any FasL expression (lined peak) ; whereas B16F10 cells transfected with pCDNA3 containing the FasL cDNA insert show FasL expression (black peak) . To determine whether Fas/FasL coexpression in tumor cells could influence growth of the tumors in vi tro, compared was the ability to grow these cells in adherent (monolayer) conditions. One thousand transfected B16F10 cells (pCDNA3 with FasL cDNA insert; Fas+/FasL+) were cultured in 1.5 ml of tissue culture medium supplemented with 10% FBS per well in 24 well plates. As controls, either one thousand B16F10 cells (untransfected) or one thousand B16F10 cells transfected with pCDNA3 only (Fas+/FasL (-) ) were cultured in 1.5 ml of tissue culture medium supplemented with 10% fetal bovine serum (FBS) per well in 24 well plates. In comparing the growth in mono- layers, clones of the Fas+/FasL+ B16F10 cells were able to grow in monolayer with the same efficiency in culture as the Fas+/FasL(-) control B16F10 cells (untransfected or trans- fected with pCDNA3 only) .

To determine whether Fas/FasL coexpression in tumor cells could influence growth of the tumors in vi tro, compared was the ability to grow these cells in nonadherent (non-anchorage) conditions. One thousand transfected B16F10 cells (pCDNA3 with FasL cDNA insert; Fas+/FasL+) were cultured in 1.5% agarose per well in 24 well plates. As controls, either one thousand B16F10 cells (untransfected) or one thousand B16F10 cells transfected with pCDNA3 only (Fas+/FasL (-) ) were cultured in 1.5% agarose per well in 24 well plates. As shown in FIG. 3, B16F10 cells (untransfected; "B16F10"), and B16F10 cells transfected with pCDNA3 only ("Fas+/FasL- B16") showed the same efficiency in forming colonies in non-anchorage conditions; e.g., after 5 days, cells develop an average of about 150 colonies. Each colony is a compact cluster of several hundreds of cells. However, three separate clones of B16F10 cells transfected with pCDNA3 with FasL cDNA insert ("Fas+ FasL+ B16" clones #3, #4, and #7) failed to develop a significant number of colonies. When a colony did develop from these transfected cells, the colony comprised a small cell cluster of about 5 to about 20 loosely associated cells. A conclusion from these results is that in nonadherent (e.g., non-anchorage) conditions, Fas/FasL coexpressing tumor cells develop fratricidal Fas/FasL mediated apoptosis. Thus, one in vivo mechanism of inducing activation induced cell death in nonadherent FasL(-) Fas+ tumors is to induce FasL expression. The Fas+/FasL+ tumor cells may then be susceptible to Fas dependent apoptosis by interaction with Fas+/FasL+ cells from the same tumor ("fratricidal Fas/FasL mediated apoptosis"). To further evaluate the effect that FasL expression has on non-adherent tumor cells in vivo, a model for metastatic growth was used. It is known by those skilled in the art that direct intrasplenic implantation of melanoma cells (e.g. B16F10) yield large splenic tumors. For example, injection of 10⁵ B16F10 melanoma cells intra- splenically yields large splenic tumors in 14 days (Barbera- Guillem et al . , 1993, Int . J. Cancer 53:298). For 5 days after the injection, the B16F10 cells grow in the spleen forming clusters without visible intercellular connective matrix, thereby mimicking metastatic growth in the lymphatic sinuses (i.e., non-anchorage conditions). After this early period, desmoplastic and angiogenic reactions complete the structure of the tumor tissue (Alonso-Varona et al., 1996, Bull . Cancer Paris 83:27). Assuming that impairment of in vi tro growth of Fas/FasL expressing tumor cells correlates with growth in vivo, intrasplenic tumor formation should also be impaired irrespective of the influence of lymphocyte deletion.

In this in vivo standard experimental model, five groups of C57BL/6 mice were injected intrasplenically with 10⁴ tumor cells; and a sixth group did not receive tumor cells (control) . One group received B16F10 cells; a second ^"group received B16F10 cells transfected with pCDNA3 only (Fas+, FasL(-)); and groups three, four, and five received B16F10 cells transfected with pCDNA3 containing FasL cDNA (Fas+, FasL+; either one of clones 3, 4, and 7) . Fourteen days postinjection, spleens from the three groups of mice were evaluated for tumor growth by measuring spleen weight, by visual observation, and by histological evaluation. For spleen weight determinations, the spleens were removed; dried by immersion in 100% ethanol for seven days during which period the ethanol evaporated; and the dried spleens were weighed and an average for the group reported. As shown in FIG. 4, spleen weight was significantly increased, and macroscopic tumor growth was observed, in mice injected with B16F10 cells ("B16F10") or B16F10 cells transfected with pCDNA3 only ("Fas+/FasL- B16"), as compared to the spleen of a control group of mice receiving only a PBS injection. However, the spleens of mice receiving B16F10 cells transfected with pCDNA3 containing FasL cDNA ("Fas+/FasL+ B16" clones #3, #4, and #7) did not show a significant increase in weight. Visually and histologically, the spleens of the Fas+/FasL+ B16 injected mice showed a normal structure with numerous isolated tumor cells with apoptotic nuclei. These analyses are indicative of a failure of tumor progression. This is an unexpected result because it has been reported that FasL+ expression by tumor cells represents a significant advantage for tumor survival and tumor growth (e.g, via immune privilege). In contrast, the results of this standard animal model indicate that FasL+ expression by tumor cells, at least in nonadherent condi- tions, represents a mechanism by which tumor cell growth is inhibited or impaired. These results in vivo confirm the results obtained in vi tro, and further support a method for impairing metastasis in individuals having FasL(-) Fas+ tumors by inducing FasL expression in such nonadherent tumors. As demonstrated herein, the method is facilitated, at least in part, by Fas-mediated cytotoxicity of the tumor (fratricidal Fas/FasL mediated apoptosis).

EXAMPLE 3

This Example further illustrates the differences in tumor growth and metastatic behavior in Fas/FasL normal mammals of Fas+ tumor cells that are either FasL+ or FasL(-). Methods and compositions for producing Fas+FasL(-) B16 transfected cells (containing pCDNA3) and Fas+FasL+ B16 transfected cells (containing pCDNA3 with FasL cDNA insert) have been described herein in Example 2. It is known by those skilled in the art that B16F10 melanoma cells have a characteristic ability, common among melanomas, to develop lung metastases (Fidler and Nicoloson, 1976, J. Na tl . Cancer Inst . 57:1199; L. Biancone et al., 1996, J. Exp . Med. 183:581). Lung metastases formation involves cell arrest (non-anchorage conditions), extravasation (anchorage condition) and colony formation (anchorage/non-anchorage conditions). To evaluate the effect of FasL expression by tumors on lung metastasis, a model for metastatic growth was used.

It is known by those skilled in the art that injection of B16F10 melanoma cells via the tail vein of mice typically results in the formation of lung metastases. Thus, this experimental animal model for in vivo metastatic growth was used to compare the effect of FasL expression by tumors on lung metastasis. One group of C57BL/6 mice was injected via the tail vein with 10⁵ B16F10 cells transfected with pCDNA3 only ("Fas+/ FasL- B16") . A second group of C57BL/6 mice was injected via the tail vein with 10⁵ B16F10 cells transfected with pCDNA3 containing FasL cDNA

("Fas+/FasL+ B16"). Lungs from the two groups of mice were evaluated for tumor growth by visual observation. Fourteen days postinjection, the lungs from mice receiving B16F10 cells developed numerous macroscopic lung metastases, whereas macroscopic lung metastases were few or absent in mice receiving Fas+/FasL+ B16 cells. The experiment was repeated using an inoculum of either 10⁴ or 10⁵ B6F10 or Fas+, FasL+ B16 clones. Three weeks postinjection, the lungs were removed, fixed in ethanol, and sliced (approximately 0.5 mm thick) for histological examination. The number of metastases per mouse was counted on lung slices using a lOx microscope, and the average count of metastases per mouse for each test group was calculated. As shown in FIG. 5, there is a significant reduction in the number of lung metastases in mice receiving Fas+, FasL+ B16 cells as compared to the number of lung metastases in mice receiving B6F10 cells (Fas+, FasL-). Using the same experimental animal model for in vivo metastatic growth, two groups of mice were monitored for an extended period of time. One group was injected via the tail vein with B16F10 cells, whereas the other group was injected with Fas+, FasL+ B16 cells. After 30 days postinjection, all mice in the group injected with B16F10 cells died. Postmortem analysis of the lungs disclosed sufficient metastatic growth consistent with being the cause of death of this group of mice. However, all mice in the group injected with Fas+, FasL+ B16 cells survived the 30 day period.

It can be concluded from this experimental animal model of metastatic growth in vivo that FasL+ expression by Fas+ tumor cells, at least in nonadherent conditions, represents a mechanism by which metastases are inhibited or impaired, rather than being a significant advantage for tumor survival and tumor growth. These results confirm the results demonstrated in Example 2 herein, and further support a method for impairing metastasis in individuals having FasL(-) Fas+ tumors by inducing FasL expression in such nonadherent tumor cells in vivo .

EXAMPLE 4

This Example further illustrates that the metastasis inhibitory effect of FasL expressing Fas+ tumor cells in non-anchorage conditions is, at least in substantial part, Fas/FasL mediated. That is, that fratricidal apoptosis comprises a substantial portion of such observed metastasis inhibitory effect, and that Fas+ coexpression is necessary for the fratricidal activity. In this illustration, used were the in vi tro culture methods, experimental animal model, and compositions described in Examples 2 and 3 herein. However, in this example, the tumor cells used were Lewis lung carcinoma cells (3LL). The 3LL cells used were shown to be Fas- by both flow cytometry and by RT-PCR. Further, agonist anti-Fas antibody, Jo2, did not induce apoptosis of these cells when the cells were cultured in monolayers (anchorage conditions) nor inhibit colony formation when cultured in fibrin clots (non- anchorage conditions) . The 3LL cells were transfected to express FasL by using pCDNA3 containing a FasL cDNA insert. 3LL clones expressing FasL were identified by RT-PCR. When Fas-/FasL+ 3LL clones were cultured in vi tro in non- anchorage conditions, they were able to produce colonies comparable to those produced by 3LL cells and by 3LL cells transfected with pCDNA3 only (Fas-/FasL (-) ) . A conclusion from these results is that in the absence of Fas expression, the expression of FasL alone does not induce the fratricidal effect .

In the experimental animal model for metastatic growth, 3LL cells or transfected 3LL cells were directly implanted in the spleen. After fourteen days, the spleens were harvested, dried, and weighed as described above. As shown in FIG. 6, spleen weight was significantly increased, and macroscopic tumor growth, was observed in mice injected with 3LL cells ("3LL") or 3LL cells transfected with pCDNA3 only ("Fas-/FasL- 3LL"), as compared to the spleen of a control group of mice receiving only a PBS injection. Spleens of mice receiving 3LL cells transfected with pCDNA3 containing FasL cDNA ("Fas-/FasL+ 3LL") showed a significant increase in weight as compared to the controls, and macroscopic tumor growth was evident. While the Fas-/FasL+ ^"3LL cells produced tumor of smaller size than 3LL, Fas- /FasL+ 3LL cells produced significantly more tumor growth in the spleen than that observed for Fas+/FasL+ B16 cell implantation (see FIG. 4). In the experimental animal model for lung metastases, 3LL cells or transfected 3LL cells were injected into mice via the tail vein. After fourteen days, the lungs were analyzed for the presence of metastases by drying, and weighing the lungs as described above for the spleen. As shown in FIG. 7, lung weight and the number of metastases was significantly increased in mice injected with 3LL cells as compared to the lungs of a control group of mice receiving only a PBS injection. Lungs of mice receiving 3LL cells transfected with pCDNA3 containing FasL cDNA ("Fas-/FasL+ 3LL") produced a significantly lower number of lung metastases as compared to mice receiving 3LL cells .

Taken together, the studies using 3LL cells and 3LL transfected cells indicate that there is another mechanism, in addition to the Fas-mediated cytotoxicity of the tumor (fratricidal Fas/FasL mediated apoptosis), involved in impairment of metastasis. One possibility is that expression of FasL by circulating tumor cells induces changes in the cellular microenvironment which counter the progression of metastases.

EXAMPLE 5

This Example illustrates that expression of FasL by nonadherent tumor cells can interact with Fast B cells to induce Fas-mediated cytotoxicity, thereby inhibiting the B cell involvement in promoting tumor progression and metastasis. Further, expression of FasL by nonadherent tumor cells also induces a change in the CD4/CD8 ratio that may play a role in the metastasis inhibitory effect of FasL expressing nonadherent tumor cells (such as in non-anchorage conditions). In this illustration, used were the experimental animal models, and compositions described in Examples 2 and 3 herein, to assess for alterations in lymphocyte populations. Three groups of C57BL/6 mice were injected subcutaneously. One group received 10⁵ B16F10 cells; and another group received 10⁵ Fas+/FasL+ B16 transfected cells. A third group received saline only, as a control. Three weeks postinjection, the spleens of each group of mice were removed, dispersed, and mononuclear cells selected by density gradient. The mononuclear cells were stained with fluorescent labeled monoclonal antibodies to detect CD3 (pan T lymphocyte), CD4 (T helper cells), CD8 (T suppressor cells), and CD19 (pan B lymphocytes) surface markers as detected and quantitated by flow cytometry. The relative frequency of each cell type is expressed as a percentage of total positively stained cells +; standard deviation. As shown in FIG. 8, the spleens of mice receiving Fas+/FasL+ B16 cells had a significant increase of the T cell to B cell ratio, and an increased CD4 to CD8 ratio, as compared to the spleens of mice receiving either B16F10 cells, or saline control. These results indicate that expression of FasL on tumor cells induces certain systemic changes in lymphocyte populations, with a relative increase in T cell numbers (primarily CD4+) , and a relative reduction in B lymphocyte populations. This is an unexpected result because it has been reported that FasL+ expression by tumor cells confers immune privilege to the tumor cells by mediating apoptosis of activated T cells (see, e.g., Strand et al., 1996, supra ; Niehans et al., 1997, supra ; O'Connell et al., 1996, supra ; and Shiraki et al., 1997, supra ) . In contrast, the results of this standard animal model indicate that FasL+ expression by tumor cells, at least in non-anchorage conditions, represents (a) a mechanism by which systemically T cells are either directly or indirectly activated to mediate inhibition or impairment of metastasis (b) a mechanism by which B cells are reduced thereby mediating inhibition or impairment of metastasis. To assess which of these altered lymphocyte populations (T or B cells) are effector cells of, at least part of, the metastasis inhibitory effect observed with FasL expressing tumor cells, specific immunodeficient mice were used. One group of athymic (T cell deficient) nude nu/nu mice was injected intrasplenically with 5 x 10⁵ Fas+/FasL+ B16 cells. One group of muMT/muMT ("B cell deficient"; i.e., do not develop competent B cell system) mice was injected intrasplenically with 5 x 10⁵ Fas+/FasL+ B16 cells. One group of C57BL/6 (immunocompetent ) mice was injected intrasplenically with 5 x 10⁵ Fas+/FasL+ B16 cells. One group (control) received PBS only. One week postinjection, all groups were injected via the tail vein with 10⁵ B16F10 cells suspended in PBS. Two weeks after injection of B16F10 cells, lung metastases were counted under phase contrast microscopy. The metastasis inhibitory effect of the tumor cells was calculated using the formula: [100 x (number of metastases in control - number of metastases in the test) /number of metastases in the control] .

As illustrated in FIG. 9, the control group, by definition, showed no metastasis inhibitory effect; and the immunocompetent mice ("C57BL/6") and B cell deficient mice ("muMT/muMT C57BL/6") showed very high inhibitory effects on development of metastases related to the subsequent B16F10 cell injections. In contrast, T cell deficient mice ("nu/nu") did not show significant metastasis inhibitory effects. These results indicate that competent T cells are involved in, or a mediator of, the metastasis inhibitory effect observed in FasL÷ expressing circulating tumor cells. Further, these results show that in this model, Fas+/FasL+ B16 cells produce statistically and significantly less metastases in muMT/ muMT (B cell deficient) mice than in C57BL/6 (B cell competent) mice. This is evidence that (a) B cells can promote tumor progression and metastasis, and (b) a reduction of B cells, such as by Fas-mediated cyto- toxicity induced by contact with FasL(+) tumors, is a mechanism for increasing the metastasis inhibitory effect. It is noted that such interactions between Fas+ B cells and FasL(+) tumor would primarily take place in tumor tissue (infiltrating B cells), and in hyperplastic lymphoid tissues either regional or distal to the site of primary tumor. The latter, is a subpopulation of B cells which are concentrated in germinal centers and which may promote tumor progression and metastasis. For further evidence of B cell involvement in tumor progression and metastasis, see Examples 9-13 herein.

EXAMPLE 6

This Example illustrates that the metastasis inhibitory effect observed of FasL expressing circulating tumor cells is systemic, rather than local. In this illustration, used were the experimental animal models, and compositions described in Examples 2 and 3 herein. Five groups of C57BL/6 were injected intrasplenically. One group received normal saline (control); one group received 10⁵

B16F10 cells; one group received 10⁵ irradiated B16F10 cells; one group received 10⁵ Fas+/FasL+ B16 cells; and one group received 10⁵ irradiated Fas+/ FasL+ B16 cells. Seven days postinjection, all groups were injected, via the tail vein, with 10⁵ B16F10 cells. Two weeks after the challenge with 10^D B16F10 cells, the spleens and lungs of the mice were analyzed for tumor burden by weight, and macroscopically . As illustrated in FIG. 10, the control group had developed extensive lung metastases, while lacking development of splenic tumor. In contrast, the group of mice receiving the splenic injection of B16F10 cells had well developed splenic tumor (averaging between 1500 to 2000 mg/ spleen) , but less extensive lung metastases when compared to the control group. These results suggest that the well-developed splenic tumor (Fas+/FasL (-) ) inhibited the growth of lung metastases through a systemic mechanism. Also illustrated in FIG. 10, the group of mice receiving irradiated B16F10 cells ("RxB16F10") had levels of splenic tumor (very low) and lung metastases (extensive) comparable to the control group. The group of mice injected with Fas+/FasL+ B16 cells did not develop splenic tumors; however, the number of lung metastases was significantly reduced when compared to the controls. When comparing the number of lung metastases, mice injected with Fas+/FasL+ B16 cells (not developing splenic tumors) and mice injected with B16F10 cells (Fas+/ FasL(-); having well developed splenic tumor) were similar in efficacy in inhibiting the growth of lung metastases through a systemic mechanism. Thus, while Fas+/FasL+ B16 cells tumor cells did not develop splenic tumors, these tumor cells had similar efficacy in exerting or inducing a distant inhibitory effect, as compared to well developed splenic tumor (Fas+, FasL(-) cells), on lung metastases .

It is important to note that (a) Fas+, FasL+ B16 cells injected intrasplenically are effective in significantly reducing the number of lung metastases formed from Fas+, FasL+ B16 cells (as shown in FIG. 5); and that (b) Fas+, FasL+ B16 cells injected intrasplenically are effective in significantly reducing the number of lung metastases formed from Fas+, FasL(-) B16F10 cells (as shown in FIG. 10) . These results indicate that through a process reaching systemically, FasL+ Fas+ tumors can directly and/or indirectly mediate a metastasis inhibitory effect (e.g., such as by fratricidal apoptosis) of either FasL+ or FasL(-) tumor cells that express Fas, and/or of Fas expressing B cells involved in a pro-tumor immune response (e.g., such as by Fas mediated cytotoxicity) . As can be further concluded from the results illustrated in FIG. 10, this FasL+ mediated metastasis inhibitory effect requires tumor cell proliferation/turnover, as irradiated Fas+, FasL+ B16 cells ( "RxFas+/FasL+B16" ) appeared ineffective in inhibiting metastasis . EXAMPLE 7

This Example illustrates one embodiment of the method of the present invention. Provided is a method for treating FasL(-) Fas+ tumor cells to become FasL+, wherein the FasL+ Fas+ tumor cells can be used to contact and induce Fas mediated cytotoxicity of Fas expressing cells selected from the group consisting of tumor cells, B cells involved in a pro-tumor immune response, and a combination thereof. The method comprises causing FasL expression in FasL(-) Fas+ nonadherent tumor cells by contacting the tumor cells with a therapeutically effective amount of a pharmaceutical composition comprising a biological modifier. The biological modifier, upon contact with such tumor cells, causes (upregulates or induces) cell-surface expression of FasL, thereby making the treated tumor cells FasL+. Biological modifiers have an endpoint of increasing, in FasL(-) tumor cells (particularly nonadherent tumor cells) the surface expression of FasL thereby making the treated Fas+ tumor cells also FasL+. Methods of determining suitable biological modifiers are described below.

Biological modifiers include, but are not limited to, staphylococcal enterotoxin B ("SEB"), tumor necrosis factor alpha (TNF-α) dexamethasone, interleukin-1 (IL-1), HTLV-I Tax (trans-activator protein) , haloperidol, and gonadotropin releasing hormone (GnRH) for GnRH receptor- bearing tumors. It has been demonstrated that SEB induces the expression of FasL on T cells (Boshell et al., 1996, Immunology, 87:586-592). Such T cells were activated T cells which express T cell receptor (TCR) , particularly of the Vβ subset. It is important to note that metastatic tumor cells in various ways resemble activated T cells, such as by the expression of T cell receptors like TCR and IL-2Rα (U.S. Patent 5,536,642 to the present inventor). Thus, a therapeutically effective amount of SEB may be used as a bio-logical modifier to induce FasL(-) nonadherent tumor cells (e.g., those in non-anchorage conditions, or solid nonlymphoid tumor cells that have a high metastatic potential, or that are metastatic or are circulating) to express FasL. Such a therapeutic-ally effective amount of SEB can be determined by one or more of the methods described below. TNF-α has been shown to induce FasL message in cultured renal cells after an 8 hour exposure to TNF-α (Ortiz-Arduan et al., 1996, Am . J. Physiol . 271 : F1193-201) . A therapeutically effective amount of TNF-α or of any biological modifier for inducing FasL(-) Fas+ tumor cells to express FasL can be determined by one or more of the methods described below.

There are several means by which to administer the biological modifier to the individual to be treated. In the situation in which a primary solid, nonlymphoid Fas+ tumor in a specific organ is detected, a catheter may be inserted into one or more of the major vessels (blood or lymphatic) that enter or exit from that organ using standard methods for inserting the catheter into such vessels, as known to those skilled in the art of chemotherapy (e.g., hepatic arterial infusion; J. Na tl . Cancer Inst . , 1996, 88:252-8). Where the tumor and/or lymphoid tisssues are directly accessible, the pharmaceutical composition may be directly injected into the site of the lymphoid tissue or the site of the tumor. This "site-directed" treatment may comprise either a single infusion, or multiple infusions over time, of a therapeutically effective amount of the biological modifier, as monitored by treatment response and by indicia of any possible local toxicity in the treated organ. The catheter may be operatively connected to a portable pump such that the biological modifier may be administered intermittently or continuously.

In another embodiment, the biological modifier may be administered systemically by intravenous injection, either as a single infusion, or by multiple infusions. The intravenous injection procedure may also be facilitated by the use of a catheter. The embodiment of intravenous injection may be particularly preferred for circulating Fas+ tumor cells present throughout the bloodstream. The response to treatment may be monitored for indicia of any possible local toxicity to the treated individual. It is appreciated by those skilled in the art that a composition, such as biological modifier, when introduced into a blood vessel may come contact with circulating tumor cells disseminating through the blood circulatory system. There are means known to those skilled in the art to facilitate delivery of a therapeutic to its intended target. In that regard, the composition comprising the biological modifier may further comprise a affinity ligand linked to the biological modifier in forming a FasL inducing conjugate. The affinity ligand is at least one composition selected from the group consisting of a lectin, antibody or antibody fragment; and has binding specificity for a tumor associated molecule preferentially expressed by a Fas+ tumor cell, and particularly a Fas+ circulating tumor cell. Thus, the affinity ligand is intended to facilitate the delivery of the biological modifier to its intended target: circulating tumor cells (including metastatic tumor cells). Additionally, the composition comprising either the biological modifier or a FasL inducing conjugate, may further comprise a pharmaceutically acceptable carrier. Such pharmaceutically acceptable carriers are known m the art to include, but are not limited to, physiological solutions, liposomes, and other delivery vehicles.

For biological modifiers which require cell entry, rather than activating a cellular process by interaction with a cell surface molecule to induce FasL expression, it is appreciated by those skilled the art that some affinity ligands, upon binding to the circulating tumor cells, may be endocytosed by the tumor cell. Other methods of promoting the introduction of therapeutic molecules, such as biological modifiers or FasL inducing conjugates, into cells the bloodstream are known in the art. In that regard, methods and devices for ex vivo or in vivo electroporating therapeutic molecules into cells in the blood have been described previously (see, e.g., U.S. Patent Nos. 5,507,724 and 5,545,130). A method according to the present invention for inducing a metastasis inhibitory effect in a patient having FasL(-) Fas+ tumor cells, particularly including FasL(-) Fas+ nonadherent tumor cells, comprises administering to the individual a therapeutically effective amount of a composition comprising a biological modifier such that administered biological modifier eventually comes in contact with such tumor cells and thereby induces the tumor cells to become FasL+. The composition may further comprise a pharmaceutically acceptable carrier, and/or a affinity ligand linked to the biological modifier in forming a FasL inducing conjugate. It will be appreciated by those skilled in the art that therapeutically effective amounts of the composition will depend on the mode of administration, tumor burden of the patient, dose schedule, patient's age, size, and other background factors.

Methods are known to those skilled in the art for identifying compositions, and determining a therapeutically effective amount thereof, which comprise biological modifiers that upregulate or induce the cell-surface expression of FasL in FasL(-) Fas+ nonadherent tumor cells so as to effect Fas-mediated cytotoxicity of Fas+ tumor cells and/or Fas+ B cells that come in contact with the FasL+ tumor cells. For example, and as illustrated with the techniques and compositions described in Example 2 herein, FasL mRNA may be detected from treated FasL(-) nonadherent tumor cells by isolating RNA from the treated cells and then performing RT-PCR. As shown in Table 3, various primer pairs have been used to amplify FasL mRNA to detect FasL expression in tumor cells. TABLE 3

Biological modifiers that upregulate or induce the cell-surface expression of FasL in FasL(-) Fas+ tumor cells may also be detected using flow cytometric methods, as illustrated with the techniques and compositions in Example 2 herein. FasL cell surface expression may be detected on such treated tumors by flow cytometry using a biotinylated anti-human FasL clone NOK-1 (commercially available from

Pharmingen Co.), with secondary labeling using streptavidin- phycoerythrin complex (Micheau et al., 1997, J. Na tl . Cancer Inst . 89:783-789). Biological modifiers that upregulate or induce the cell-surface expression of FasL in FasL(-) Fas+ tumor cells may be detected using immunofluorescence or immunohistochemical staining of such treated tumors. For example, sections of the treated tumor or cultured treated tumor cells may be fixed with either 4% formaldehyde at room temperature or 100% acetone at -20°C. The sections or cells may be incubated with anti-FasL antibody. Detection may be accomplished using a FITC-labeled secondary antibody (Niehans et al., 1997, supra ; Ortiz-Arduan et al., 1996, supra ) or avidin-biotin complex with chromogenic substrate and counterstained (Shiraki et al., 1997, supra ; O'Connell et al., 1996, supra ) .

Biological modifiers that upregulate or induce the cell-surface expression of FasL in FasL(-) Fas+ tumor cells may be detected using Western blot analysis of such treated tumors. For example, treated tumor or cultured treated tumor cells may be solubilized with detergents, subjected to polyacrylamide gel electrophoresis, immunoblotted using an anti-FasL antibody, and detection using a secondary antibody conjugate with subsequent chromogenic substrate development. Biological modifiers that upregulate or induce the cell-surface expression of FasL in FasL(-) Fas+ tumor cells may be detected using a FasL bioassay for detecting Fas/FasL mediated apoptosis. For example, Fas-sensitive Jurkat cells may be labeled with 1 mCi Na ⁵¹Cr at 37°C in tissue culture medium. After washing, the cells may be added at 5 x 10⁴ cells/well in a 96 well plate. FasL(-) circulating tumor cells (e.g., tumor cells in non-anchorage conditions or metastatic tumor cells) may be incubated with the Jurkat cells in the presence or absence of the composition being tested as a biological modifier. The % specific killing may be calculated by measuring release of radioactivity by such treated tumors (Boshell et al., 1996, supra ) . Alternatively, radioactivity need not be used. Rather, the nonadherent phase of the mixed culture may be assayed for cell viability using trypan blue exclusion (Niehans et al., 1997, supra ) , or treatment with propidium iodide and flow cytometric detection (Shiraki et al., 1997, supra ) .

EXAMPLE 8 This Example illustrates another embodiment of the methods of the present invention. Provided is a method for causing FasL expression in FasL(-) Fas+ tumor cells, wherein the FasL(-) Fas+ tumor cells become FasL+ and can be used to contact and induce Fas mediated cytotoxicity of Fas expressing cells selected from the group consisting of tumor cells, B cells involved in a pro-tumor immune response, and a combination thereof. The method comprises contacting the tumor cells with a therapeutically effective amount of a pharmaceutical composition comprising an expression vector; wherein the vector is a vehicle for introducing into, and expressing in, the tumor cells, particularly nonadherent ^"tumor cells, a gene encoding FasL. Methods for making such a vector are described in detail in Example 2 herein. The vector, upon entry into such tumor cells, upregulates or induces the cell-surface expression of FasL, thereby making the treated Fas+ tumor cells FasL+. In another illustration of this embodiment, the composition further comprises an affinity ligand linked to the expression vector, in forming a FasL inducing conjugate. The affinity ligand of the FasL inducing conjugate has binding specificity for a tumor- associated molecule preferentially expressed by a tumor cell, and particularly by a nonadherent tumor cell. Thus, the affinity ligand facilitates selectively delivery of the expression vector to its intended target tumor cells. In this embodiment, the composition further comprises a pharmaceutically acceptable carrier. Such pharmaceutically acceptable carriers are known in the art to include, but are not limited to, physiological solutions, liposomes, other delivery vehicles, and compositions which facilitate infection or transfection of the tumor cell by the vector (e.g., microparticles which permit or enhance uptake or introduction of vector into the target cells).

Using methods known in the art of molecular biology, including methods described above, various promoters and enhancers (see, e.g., SEQ ID NO: 16) can be incorporated into the vector or the DNA sequence encoding FasL to increase the expression of FasL. The selection of the promoter will depend on the vector used, and if tissue specific expression is desired. The promoter is operatively-linked to the DNA sequence encoding FasL, and may be part of the vector sequence or introduced as part of the DNA insert containing the FasL encoding sequences. The vector may include other control elements for efficient gene transcription or message translation, including enhancers, and regulatory signals. Accordingly, FasL encoding sequences can be ligated into an expression vector at a specific site in relation to the vector's promoter, control, and regulatory elements so that when the recombinant vector is introduced into the target tumor cell, the FasL DNA sequences can be expressed.

There are several means by which to administer the expression vector to the individual to be treated. In the situation in which a primary solid, nonlymphoid Fas+ tumor in a specific organ is detected, a catheter may be inserted into one or more of the major vessels (blood or lymphatic) that enter or exit from that organ using standard methods for inserting the catheter into such vessels, as known to those skilled in the art. This "site-directed" treatment may comprise either a single infusion, or multiple infusions over time, of a therapeutically effective amount of the vector, as monitored by treatment response and by indicia of any possible local toxicity in the treated organ.

In another embodiment, the expression vector may be administered systemically by intravenous injection. The intravenous injection procedure may also be facilitated by the use of a catheter. The embodiment of intravenous injection may be particularly preferred for circulating tumors present throughout the bloodstream. The vector may be introduced directly ("direct gene transfer") resulting in expression of the genetic material into the target tumor cells has been demonstrated by techniques in the art such as by injecting intravenously an expression plasmid: cationic liposome complex (Zhu et al., 1993, Science 261:209-211). It is appreciated by those skilled in the art that the vector may be linked to a affinity ligand, and the resultant conjugate may bind to the nonadherent tumor cells, and may be endocytosed by the tumor cell. Other methods of promoting the introduction of therapeutic molecules, such as vectors or vectors linked to affinity ligands (conjugate, into cells in the bloodstream are known in the art. In that regard, methods and devices for ex vivo or in vivo electroporating therapeutic molecules into cells in the blood have been described previously (see, e.g., U.S. Patent Nos. 5,507,724 and 5,545,130).

In another example, the circulating Fas+ tumor cells removed from an individual can be transfected or electroporated by standard procedures known in the art, resulting in the introduction of the expression vector DNA into the cells. The cells containing the recombinant vector DNA may then be selected for using methods known in the art such as via a selection marker expressed in the vector, and the selected cells may then be re-introduced into the individual to express FasL.

EXAMPLE 9

This, and the following examples, provide further evidence of the involvement of B cells in a pro-tumor immune response that promotes tumor progression and metastasis. This Example illustrates that removal of hyperplastic lymphoid tissues, that are a foci of the pro-tumor immune response in humans, results in the clinical benefit of an increase in the 5 year survival rate. A detailed description of the immune corrective surgical procedure to remove such hyperplastic lymphoid tissues is described elsewhere (Barbera-Guillem et al., 1998, Am . J. Surgery, 176: ) .

Briefly, procedures for this clinical trial are as follows. Tumor bearing individuals having primary or recurrent colorectal carcinoma were selected for this study. mAb CC49 was labelled using sodium iodide ¹²⁵I using the 1,3,4, 6-tetrachloro-3-alpha-diphenylglycouril method. Radiolabelled mAb is then purified by chromatography, and sterilized by filtration. ¹²⁵I mAb CC49 is then prepared in phosphate buffered saline in forming the detector antibody. Prior to administration of the detector antibody, multiple doses of a saturated solution of potassium iodide (e.g., 1 ml of a 500 mg/ml solution of KI) were given to the tumor bearing individual to block thyroid uptake of the detector antibody. Detector antibody (e.g., ranging from 0.2 to 10 mg) was administered intravenously. In a time period sufficient to establish a background level of radioactivity in the individual (e.g., 21 to 28 days post-injection), the individual underwent exploratory surgery in which a hand- held gamma detecting probe was used to detect radioactive tissue. In this clinical trial, each of the two groups of tumor-bearing individuals, having colorectal carcinomas homogenous in tumor staging, were surgically treated under a different protocol. A first treatment group comprised 34 patients who were subjected to radioimmunoguided surgery alone. This first treatment group included 7 stage I (AJCC staging) patients, 15 stage II patients, and 12 stage III patients. A second treatment group comprised 24 patients were treated with immune corrective surgery according to the present invention, and subjected to radioimmunoguided surgery. This second treatment group included 6 stage I patients, 9 stage II patients, and 9 stage III patients. For each group, radioimmunoguided surgery was performed to remove neoplastic tissue using a protocol as essentially described previously (see, e.g., U.S. Patent No. 4,782,840). In the immune corrective surgery, lymphoid tissues containing shed tumor antigen (as detected by the detector antibody) were removed from one or more of the areas selected from the group consisting of the gastrohepatic ligament, celiac axis, iliac vessels, retroperitoneum, or a combination thereof; areas or sites from which a surgeon would not traditionally remove tissue. After resection of the hyperplastic lymphoid tissues containing shed tumor antigen, the areas around the excised lymphoid tissues were rescanned for residual radio-activity. The five year survival rate (free of clinically evident neoplastic disease) was calculated for each treatment group, and expressed as a percentage (Table 4, "5 yr."). Table 4 shows a comparison of the 5 year survival rate between tumor bearing individuals undergoing radioimmunoguided surgery only ("Group 1"), undergoing immune corrective surgery accordmg to the present invention, combination with radioimmunoguided surgery ("Group 2" ) ; and as compared to outcomes of traditional surgery as reported by the National Cancer Database ("T.Surg."). As shown in Table 4, tumor bearing individuals who received immune corrective surgery according to the present invention, in combination with radioimmunoguided surgery, showed a statistic-ally significant increase in 5 year survival as compared to tumor bearing individuals receiving either radioimmunoguided surgery alone, or traditional surgery alone. Of particular note is the significant improvement m 5 year survival for patients who had stage III (regionally disseminated) disease at the time of immune corrective surgery accordmg to the present invention. In summary, the results in Table 4 clearly show that by not removing hyperplastic lymphoid tissues, the patients' 5 year survival rate is statistically reduced (as compared to that of patients from whom such hyperplastic lymphoid tissues containing B cells had been removed) . Table 4

Patients # of patients Stage 5 yr .

Group 1 7 I 66%

Group 2 6 I 100%

T.Surg. - I 68%

Group 1 15 ϊϊ 42%

Group 2 9 II 100%

T.Surg. - II 64%

Group 1 12 ϊϊϊ 3Ϊ%

Group 2 9 III 100%*

T.Surg. - III 44%

*- two patients subsequently died of cardiovascular disease.

EXAMPLE 10 This, and following embodiments, provide evidence of the B cell involvement, and the specific type of immune response related thereto, which promotes tumor progression and metastasis. In some of these embodiments, it is important to consider the following concept. Various strains of mice were used as a standard animal model for evaluating whether a germinal center B cell response may be involved in tumor progression, including promoting metastasis. In the tumor bearing mice of B cell competent strains, a similar germinal center B cell response was observed in lymph nodes regional to a primary tumor as observed in tumor bearing humans. In that regard, and to assess whether B cells are effector cells of, at least in part, a tumor promoting immune response, an m vivo standard experimental model was used. One group of C57μMT/μMT ("B cell deficient"; i.e., do not develop competent B cell system) mice was injected intrasplenically with 10 B16F10 melanoma tumor cells. One group of C57BL/6 (B cell competent) mice was injected intrasplenically with 10 B16F10 tumor cells. One group of mice (control) received PBS only.

Fourteen days postinjection, spleens from the three groups of mice were evaluated for tumor growth (progression) by measuring spleen weight and size. For spleen weight determinations, the spleens were removed; dried by immersion 100% ethanol for seven days during which period the ethanol evaporated; and the dried spleens were weighed, and an average for the group reported. As shown in FIGs. 11 and 12 respectively, the spleen size and weight was significantly decreased in B cell deficient (C57μMT/μMT) mice as indicative of a decreased ability of B cell deficient mice to mediate tumor progression. Lung metastases formation involves cell arrest

(non-anchorage conditions) , extravasation (anchorage condition) and colony formation (anchorage/non-anchorage conditions). To assess whether B cells are effector cells (at least in part) of a meta-static effect, a model for metastatic growth was used. It is known by those skilled in the art that injection of B16F10 cells via the tail vein of mice typically results in the formation of lung metastases. One group of C57BL/6 mice was injected via the tail vein with 10⁶ B16F10 cells. One group of C57μMT/μMT mice was injected via the tail vein with 10⁶ B16F10 cells. Fourteen days postinjection, the lungs from the two groups of mice were evaluated for tumor growth macroscopically, and the number of metastases counted. As shown in FIG. 13 (El and E2 are duplicate experiments), the average number of lung metastases was significantly decreased in B cell deficient (C57μMT/μMT) mice as compared to B cell competent (C57BL/6) mice. The significant decrease also represents the lack of several of the B cell deficient mice to develop any detectable metastases in the lung.

These results confirm the results illustrated in FIGs. 11 and 12 herein, that there is a decreased ability of B cell deficient mice to mediate tumor progression (including metastasis) as compared to normal mice. This is an unexpected result because it believed that B lymphocytes appear able to elicit anti-tumor immunity which can be mediated by T cells, and produce circulating antitumor antibodies. A possible explanation of the results illustrated in FIGs. 11-13 is that B cells mediate tumor progression (including metastasis). If this were the case, then it would be expected that the B cell deficient mice which exhibited no metastasis of tumor would still contain some primary tumor. Further experiments showed that alive and competent tumor cells are present in these B cell deficient mice which exhibited no metastasis of tumor. In summary, these results are evidence of a subpopulation of B cells which can mediate tumor progression.

EXAMPLE 11

To assess whether different populations of B lymphocytes could promote growth of the tumors in vivo, tumor growth in CH3 mammary gland tumor bearing mice was compared when the mice were injected every 2 days for a 14 day period with either B lymphocytes (50,000 cells) isolated from normal mouse spleen; B lymphocytes isolated from lymphoid tissues (e.g., spleens) of tumor bearing mice (50,000 cells), or tumor infiltrating B lymphocytes (B-TIL; 50,000 cells) isolated from tumors of tumor bearing mice. Isolations of B lymphocytes were performed by magnetic separation methods known in the art. After the 14 day period, liver metastasis and spleen tumor growth (tumor + metastasis score) were evaluated and scored. As shown in FIG. 14, B-TIL and B lymphocytes from spleens of tumor bearing mice ("T-Spl") each promoted statistically significant tumor growth and metastasis in vivo, whereas B lymphocytes from normal spleen ("N-Spl") did not enhance either tumor growth or metastasis. An important conclusion that can be drawn from these results is that to gain the ability to promote tumor growth, B lymphocytes must first be exposed to tumor antigens (e.g., prior contact with shed tumor antigen) .

EXAMPLE 12

This Example illustrates that B-TIL promotion of tumor growth in vi tro can be mediated via a direct action by B-TIL that does not require a T cell intermediary response; but may also act synergistically with CD4+ T cells, when present. In this illustration, several populations of lymphocytes were isolated using magnetic bead separation techniques from Met 129 tumors removed from C3H mice. Met 129 tumors are high mucin secretors. The lymphocyte populations included tumor infiltrating B lymphocytes (B- TIL) , tumor infiltrating CD4+ lymphocytes (CD4+) , and tumor infiltrating CD8+ lymphocytes (CD8+) . Ten thousand Met 129 cells were cultured in 1.5 ml of tissue culture medium supplemented with 10% fetal bovine serum (FBS) per well in 24 well plates alone, or in the presence of either 10,000 CD8+ cells, in the presence of 10,000 CD4+ cells, in the presence of 10,000 CD8+ cells and 10,000 CD4+ cells, in the presence of 10,000 B-TIL, in the presence of 10,000 B-TIL and 10,000 CD8+ cells, or in the presence of 10,000 CD8+ cells, 10,000 CD4+ cells and 10,000 B-TIL. After 72 hours of co-incubation in monolayer culture, Met 129 tumor cell growth was quantitated using Alcian blue staining; e.g., adherent mucin-producing cells (Met 129 tumor cells) were counted. As shown in FIG. 15, CD8+ cells co-incubated with Met 129 (Met 129 +, CD8+ +, CD4+ -, B -) resulted in a statistically significant reduction in tumor cell growth, and thus appeared to effect Met 129 tumor cell death when compared to the control of Met 129 alone (Met 129 +, CD8+ -, CD4+ -, B -) . Likewise, a slight reduction in tumor growth or no increase in tumor growth, as compared to the control, was observed when Met 129 tumor cells were co-incubated with either CD4+ cells (Met 129 +, CD8+ -, CD4+ +, B -) , in the presence of CD8+ cells and CD4+ cells (Met 129 +, CD8+ +, CD4+ +, B -) , or in the presence of B-TIL and CD8+ cells (Met 129 +, CD8+ +, CD4+ -, B +) . In contrast, statistically significant increased tumor cell growth was observed when B-TIL were co-incubated with Met 129 tumor cells (FIG. 15: Met 129 +, CD8+ -, CD4+ -, B +) , and when CD8+ cells, CD4+ cells and B-TIL were co-incubated with Met 129 tumor cells (FIG. 15: Met 129 +, CD8+ +, CD4+ +, B +) , as compared to growth of the control of Met 129 tumor cells alone (Met 129 +, CD8+ -, CD4+ -, B -) .

In summary, these results indicate that B-TIL alone can promote tumor growth in vi tro via a direct action by B-TIL that does not require a T cell intermediary response (FIG. 15: Met 129 +, CD8+ -, CD4+ -, B +) . However, it appears that B-TIL may also act synergistically with CD4+ T cells in promoting tumor growth (FIG. 15: Met 129 +, CD8+ +, CD4+ +, B +) . In that regard, it is interesting to note that CD4+ cells alone could not exert a significant tumor promoting effect, but CD4+ cells in combination with B-TIL and CD8+ cells mediated greater tumor progression than did B-TIL alone. Further experiments have identified Vα3+, CD4+ cells as CD4+ effectors which can promote tumor growth. These results support the findings of others (Wang and Taniguchi, 1995, J. Immunol . 154:1797-1803) and suggest that Vα3+, CD4+ cells negatively regulate T cell responses controlling tumor growth and metastasis.

EXAMPLE 13

In this illustration, an immunotherapeutic agent was administered to a tumor bearing animal, wherein the immunotherapeutic agent is one which would target mature B cells and/or memory B cells to interrupt the host B cell intermediary (pro-tumor) response in a tumor bearing animal, thereby affecting tumor progression. Twenty C3H mice were injected intrasplenically with 10⁶ Met 129 tumor cells. The injected mice were then divided into three treatment groups. One group of 6 mice was injected with phosphate buffered saline (PBS) at days 5, 7, and 9 following tumor challenge.

A second group consisted of 8 mice injected with an irrelevant (not directed against any specific mouse antigen) goat IgG antibody (170 μg per injection) at days 5, 7, and 9 following tumor challenge. A third group consisted of 6 mice injected with goat anti-mouse IgG and goat anti-mouse IgM (170 μg per injection) at days 5, 7, and 9 following tumor challenge. The goat anti-mouse IgG and IgM was used to deplete the C3H mice of their B cells, thereby interrupting the host B cell-mediated pro-tumor immune response. At 22 days following tumor challenge, the three groups of mice were analyzed for primary tumor growth in the spleen, metastasis to the liver, and extra-regional metastasis (abdominal lymph nodes). Table 5 shows a comparison of primary tumor growth, and the incidence of liver metastasis ("Liver Met.") and extra-regional metastasis ("Extra-R Met.") in the mice treated with PBS ("Control"), mice treated with irrelevant goat IgG ("Goat- IgG") , and mice treated with goat anti-mouse IgG and goat anti-mouse IgM ("Anti-IgG Anti-IgM") . Table 5 shows that there is a statistically significant reduction in the incidence of metastasis in the immunotherapeutically treated (B cell-depleted) mice ("Anti-IgG Anti-IgM") as compared to the control group or group receiving irrelevant IgG.

Table 5

Spleen tumor was scored and compared among the three groups of mice. As shown in FIG. 16, treatment of the tumor bearing mice with either goat IgG, or goat anti-mouse IgG and goat anti-mouse IgM, had little effect on the growth of the primary tumor in the spleens of the treated animals. In contrast, shown in FIGs. 17 and 18 respectively, mice receiving immunotherapeutic treatment with goat anti-mouse IgG and goat anti-mouse IgM showed a statistically significant reduction in the incidence of extra-regional metastasis and liver metastasis as compared to the extra- regional metastasis and liver metastasis exhibited by either the control group of mice or the group of mice treated with irrelevant goat IgG.

In summary, the results illustrated in Table 5, and FIGs. 17 and 18 further support the finding that a host B cell response (pro-tumor immune response) is required for statistically significant promotion of metastasis. Such host B cells may include those present in hyperplastic lymphoid tissues, wherein such B cells have been activated by shed tumor antigen. Additionally, the results illustrated in Table 5, and FIGs. 17 and 18 further support methods and compositions for the reduction of B cells (e.g., by Fas- mediated cytotoxicity induced by contact with FasL+ tumor cells) to inhibit the B cell promotion of tumor progression and metastasis in an individual.

The foregoing description of the specific embodiments of the present invention have been described in detail for purposes of illustration. In view of the descriptions and illustrations, others skilled in the art can, by applying, current knowledge, readily modify and/or adapt the present invention for various applications without departing from the basic concept, and therefore such modifications and/or adaptations are intended to be within the meaning and scope of the appended claims.

What is claimed is:

Claims

1. A method for treating FasL(-) tumor cells to become FasL+, wherein the FasL+ Fas+ tumor cells can be used in nonadherent conditions to contact and induce Fas mediated cytotoxicity of Fas expressing cells selected from the group consisting of tumor cells, B cells involved m a pro-tumor immune response, and a combination thereof, wherein the method comprises contacting the FasL(-) tumor cells with a therapeutically effective amount of a pharmaceutical composition that acts to cause FasL expression by a mechanism selected from the group consisting of induction, and upregulation, and wherein the pharmaceutical composition is selected from the group consisting of a biological modifier, an expression vector for expressing FasL, and a combination thereof.

2. The method accordmg to claim 1, wherein the composition further comprises an affinity ligand.

3. The method accordmg to claim 1, wherein the composition further comprises a pharmaceutically acceptable carrier .

4. The method accordmg to claim 2, wherein the composition further comprises a pharmaceutically acceptable carrier.

5. The method accordmg to claim 1, wherein the composition is a biological modifier.

6. The method accordmg to claim 5, wherein the composition further comprises a affinity ligand linked to the biological modifier m forming a conjugate.

7. The use of a biological modifier consisting of a compound that causes FasL expression m FasL(-) tumors in the manufacture of a pharmaceutical composition for treating FasL(-) tumor cells to become FasL+, wherein the FasL+ tumor cells can be used in nonadherent conditions to contact and induce Fas mediated cytotoxicity of Fas expressing cells selected from the group consisting of tumor cells, B cells involved in a pro-tumor immune response, and a combination thereof, and wherein by contacting the FasL(-) tumor cells with a therapeutically effective amount of the pharmaceutical composition causes FasL expression by a mechanism selected from the group consisting of induction, and upregulation.

8. The use of a biological modifier, consisting of a compound that causes FasL expression in FasL(-) tumors, coupled to an affinity ligand in the manufacture of a pharmaceutical composition for treating FasL(-) tumor cells to become FasL+, wherein the FasL+ tumor cells can be used in nonadherent conditions to contact and induce Fas mediated cytotoxicity of Fas expressing cells selected from the group consisting of tumor cells, B cells involved in a pro-tumor immune response, and a combination thereof, and wherein by contacting the FasL(-) tumor cells with a therapeutically effective amount of the pharmaceutical composition that causes FasL expression by a mechanism selected from the group consisting of induction, and upregulation.

9. The use of an expression vector encoding FasL that causes FasL expression in FasL(-) tumors in the manufacture of a pharmaceutical composition for treating FasL(-) tumor cells to become FasL+, wherein the FasL÷ tumor cells can be used in nonadherent conditions to contact and induce Fas mediated cytotoxicity of Fas expressing cells selected from the group consisting of tumor cells, B cells involved in a pro-tumor immune response, and a combination thereof, wherein the method comprises contacting the FasL(-) tumor cells with a therapeutically effective amount of a pharmaceutical composition that acts to cause FasL expression.

10. The use of an expression vector, encoding FasL that causes FasL expression in FasL(-) tumors and coupled to an affinity ligand in the manufacture of a pharmaceutical composition for treating FasL(-) tumor cells to become FasL+, wherein the FasL+ tumor cells can be used in nonadherent conditions to contact and induce Fas mediated cytotoxicity of Fas expressing cells selected from the group consisting of tumor cells, B cells involved in a pro-tumor immune response, and a combination thereof, and wherein by contacting the FasL(-) tumor cells with a therapeutically effective amount of the pharmaceutical composition acts to cause FasL expression.

11. A method for impairing metastasis and tumor growth in an individual having solid nonlymphoid tumor that comprises FasL(-) tumor cells, said method comprises administering to the individual a therapeutically effective amount of a pharmaceutical composition that contacts and causes FasL expression by the tumor cells; wherein the composition is selected from the group consisting a biological modifier, an expression vector for expressing FasL, and a combination thereof; and wherein the FasL+ tumor cells, in nonadherent conditions, impair metastasis and tumor growth by contacting and inducing Fas mediated cytotoxicity of Fas expressing cells selected from the group consisting of tumor cells, B cells involved in a pro-tumor immune response, and a combination thereof.

12. The method for impairing metastasis and tumor growth according to claim 11, wherein the composition is a biological modifier.

13. The method for impairing metastasis and tumor growth according to claim 12, wherein the composition further comprises a affinity ligand linked to the biological modifier in forming a conjugate.

14. The method for impairing metastasis and tumor growth according to claim 12, wherein the composition further comprises a pharmaceutically acceptable carrier.

15. The method for impairing metastasis and tumor growth according to claim 12, wherein the composition is administered to the individual by using a catheter.

16. The method for impairing metastasis and tumor growth according to claim 11, wherein the composition is an expression vector for expressing FasL in tumor cells which contain the vector.

17. The method for impairing metastasis and tumor growth according to claim 16, wherein the composition further comprises a affinity ligand linked to the expression vector in forming a conjugate.

18. The method for impairing metastasis and tumor growth according to claim 16, wherein the composition further comprises a pharmaceutically acceptable carrier.

19. The method for impairing metastasis and tumor growth according to claim 16, wherein the composition is administered to the individual by using a catheter.

20. A pharmaceutical composition that contacts FasL(-) tumor cells and causes FasL expression by the contacted tumor cells, wherein the composition is selected from the group consisting of a biological modifier coupled to an affinity ligand, and an expression vector coupled to an affinity ligand wherein the expression vector encodes FasL.