WO1999053931A9 - Tumor cells with increased immunogenicity and uses thereof - Google Patents

Abstract

A tumor cell which co-express SEB with MHC class II alone, or also with B7 (B7.1 and/or B7.2), wherein the tumor cell is preferably a mammary tumor cell, and use of the same to treat tumor, particularly, metastatic tumors.

Description

TUMOR CELLS WITH INCREASED IMMUNOGENICITY AND USES THEREOF

The invention described herein was developed with support from the National Institutes of Health under Grant No. 1 R01 CA 52527; and the Department of Defense, U.S. Army Research and Development Command, DAMD 17-94. The government has certain rights to this invention.

FIELD OF THE INVENTION

The present invention relates to tumor cells which co-express SEB with MHC class II alone, or also with B7 (B7.1 and/or B7.2), wherein the tumor cells are preferably mammary tumor cells, and use of the same to treat tumors, particularly, metastatic tumors.

BACKGROUND OF THE INVENTION

I . Stimulating Immune Responses to Tumors

Tumor cells are not very potent immunogenic agents, as they typically lack the surface receptors needed to induce immune responses. For example, tumor cells can lack class I (Elliott et al. Adv. Cancer

Res . . ^■ 53:181-245 (1989); Doyle et al, J. Exp . Med. , 161:1135-1151 (1985); and Lassa et al, J. Immunol . , 143:3792-3797 (1989)) and/or class II (Kern et al, J^". Immunol . , 136 :4303-4310 (1986)) major histocompatibility complex (MHC) gene products which are needed for presentation of antigen to CD8^* and CD4⁺ T-cells, respectively (Hood et al, Ann. .Rev. Immunol . , 1:529-568 (1983); and Lancki et al, Immunol. Rev. , _.81:65-94 (1984)). Likewise, co-stimulatory molecules, which have been shown to be important for induction of tumor-specific immunity, are often not expressed on tumor cells (Chen et al, Cell , 71:1093-1102 (1992)). Thus, it has been postulated that poor immune responses to tumors are caused by immunological "ignorance", i.e., that the immune system has the capacity to respond to tumor antigens presented by tumor cells, but fails to respond under normal conditions (Miller et al, Immunol. Rev. ,

133:131-150 (1993)).

II . Recent Advances in Immunotherapy Have Led to the Generation of Protective T-Cell-Mediated Immunity Against Primary Tumors

The stimulation of CD8^* cytotoxic T-lymphocytes (CTL) has been the primary focus of many studies, as these effector cells are capable of specifically and directly destroying malignant tumor cells. With the increasing knowledge regarding immune response mechanisms, and with the isolation/identification of genes involved in CTL activation, those in the art have begun to manipulate the immune system to their advantage by designing tumor cells to act as better antigen presenting cells (APC) . For example, various cytokine genes have been transfected or transduced into tumors, and have enhanced immune responses to primary tumors (Colombo et al, Immunol . Today, 15_:48-51 (1994); and Pardoll, Immunol . Today, 14:310-316 (1993)). While it was intended to circumvent the need for CD4^* T-cell help by allowing the tumor cells to directly supply cytokines to CTL, some studies demonstrated that both CD4⁺ and CD8⁺ T-cells were still necessary (Asher et al, J. Immunol . , 146:3227-3234 (1991); Dranoff et al, Proc. JVatl. Acad. Sci . , USA, 90:3539-3543 (1993); and Pulaski et al, Cancer Res . , 53:2112-2117 (1993)).

III. Methods for Effective Treatment of Metastatic Disease Still Need to be Developed

As discussed above, advances have been made to increase the immunogenicity of tumor cells. However, work still needs to be done to create effective tumor vaccines that eliminate tumor cells found in sites other than at the primary tumor. In some cases, such as in human breast cancer, where the primary tumor has not metastasized, surgical removal of primary tumors can lead to full recovery of the patient. However, when the primary breast tumors have metastasized, other therapies, such as chemotherapy, are needed to eliminate metastatic tumor cells, and these therapies may not always be successful. Therefore, by designing tumor cells to act as the vaccination vehicle for stimulating both CD4^* and CD8⁺ T-cells, it should be possible to induce long-lived, tumor-specific responses that can be used in combination with established methods.

IV. Optimal T-cell Activation Requires an Antigen-Specific Signal Plus a Second Co-stimulatory Simal

T-cells recognize peptide/MHC complexes through their T-cell receptor (TCR) (Collins et al, Curr. Opin . Immunol . , 6:385-393 (1994); and Janeway et al, Ann. Rev. Immunol . , 10:645-674 (1992)). However, to achieve maximum activation of CD4* or CD8* T-cells, a second TCR-independent signal (co-stimulation) is required (Macatonia et al, J. Exp. Med. , 169:1255-1264 (1989) ) . Several studies have demonstrated the role of B7.1 and B7.2 in co-stimulation (Allison, Curr. Opin. Immunol., 6_:414-419 (1994)). However, other molecules, such as ICAM-1, VCAM-1, and heat stable antigen, can also function in a co-stimulatory fashion (Liu et al, J^". Exp. Med. , 17_5:437-445 (1992); Lui et al, Eur. J. Immunol . , 22_.:2855-2859 (1992); and Damle et al, J. Immunol . , 148:1985-1992 (1992)). Also, 4-1BB ligand (4-1BB-L) , a newly described molecule found on activated macrophages and mature B cells, has been found to enhance proliferative T-cell responses and to synergize with B7 (Goodwin et al, Eur. J. Immunol . , 23.: 2631-2641 (1993) DeBenedette et al, J. Exp. Med. , 181:985-992 (1995) Pollack et al, Eur. J. Immunol . , 24:367-374 (1994) and Hurtado et al , J.^' Immunol . , 155:3360-3367 (1995)). Failure of T-cells to receive both signals from APC will result in T-cell anergy (unresponsiveness) or death.

V. Designing Tumor Cells to Act as the APC is a Feasible Rationale

Using tumor cells as APC is advantageous as the relevant tumor antigen will be presented to T-cells.

With the right combination of genes expressed by tumor cells, these APC can possess an increased immunogenicity and subsequently stimulate immune responses. For example, it has been shown that the transfection of MHC class II into sarcoma and melanoma cells enhanced tumor rejection and reduced tnetastatic potential (Ostrand-Rosenberg et al, Tissue Antigens, 4_.7:414-421 (1996)), respectively. Furthermore, expression of either B7.1 or B7.2 in addition to MHC class II increased these effects (Ostrand-Rosenberg et al, supra; Baskar et al, J^". Exp. Med. , 181:619-628 (1995); and U.S. Patent Application Serial No. 08/147,772, filed November 3, 1993; which are incorporated by reference herein in their entirety) .

In the present invention, additional genes, i.e., S. aureus enterotoxin B (SEB) , are postulated to synergize with MHC class II and/or B7.1 or B7.2 in enhancing CD4* tumor-specific response. VI. SEB or Other Superantigens (sAg) can Activate CD4* T-Cells in a Polyclonal Manner

The toxin SEB is a well-known sAg that has been established as a potent stimulator of T-cells when complexed with MHC class II (Herman et al, Ann. Rev.

Immunol., 9.:745-772 (1991)). Therefore, tumor cells expressing both MHC class II and SEB are postulated in the present invention to stimulate tumor-specific

T-cells. SEB activates T-cells which express TCR containing V_βl or V_e8 gene segments by associating with

MHC class II found on the surface of APC without prior intracellular processing (Marrack et al, J. Exp. Med. ,

171:455-464 (1990)). This is not believed to be problem in the present invention as administration of SEB either in vivo or in vitro has been shown to activate T-cells against the 1591 skin tumor

(Newell et al, Proc . Natl . Acad. Sci . , USA,

88.= 1074-1078 (1991)), and the MCA 205/207 sarcomas

(Shu et al, J^". Immunol . , 151:1277-1288 (1994)), respectively. Recently, it was also demonstrated that pulmonary metastases formed by CL-62 melanoma cells injected intravenously can be neutralized using soluble SEB (Penna et al, Cancer Res . , 54:2738-2743

(1994) ) . In the present invention, where the tumor cells directly express SEB, activation of T-cells is postulated to occur in a more efficient manner

(smaller amounts of SEB would be needed to stimulate its effects) , and would less likely cause adverse side-effects sometimes associated with large systemic treatments.

VII . Breast Cancer Patients

In breast cancer patients, primary tumor burden seldom causes the patient's demise as primary tumor can be surgically removed. In fact, it is spontaneous metastases that arise during the tumor growth or long after the primary tumor has been removed to which the patient eventually succumbs. Therefore, in one embodiment, the present invention provides a mammary tumor vaccine using the tumor cells themselves to directly activate "the immune system to combat spontaneous metastatic disease either before or after it becomes established. In particular, the present invention provides mammary tumor cells which co-express SEB with MHC class II, as it is postulated in the present invention that these cells will activate CD4^* T-cells, and subsequently CD8^* T-cells, immune ef ectors capable of circulating throughout the body so as to provide long-lived protection against tumor cells at the primary tumor site, and as well as at the site of metastatic disease.

SUMMARY ON THE INVENTION Accordingly, an object of the present invention is to provide tumor cells which have increased immunogenicity.

Another object of the present invention is to provide tumor cells that are capable of inducing an immune response which eliminates tumor cells found in sites other than at the primary tumor. Still another object of the present invention it to provide tumor cells which will activate CD4* T-cells, and subsequently CD8^* T-cells, and other relevant effector cells.

Yet another object of the present invention is to use the tumor cells to treat tumors, particularly metastatic tumors .

These and other objects of the present invention, which will be apparent from the detailed description of the invention provided below, have been met, in one embodiment, by a tumor cell which has been transfected so as to co-express SEB with MHC class II, and which is capable of enhancing a CD4⁺ tumor-specific response.

In another embodiment, the above-described objects of the present invention have been met by tumor cells which further also co-express B7 (B7.1 and/or B7.2) .

In yet another embodiment, the above-described objects of the present invention have been met by a pharmaceutical composition comprising a pharmaceutically effective amount of said tumor cells, and a pharmaceutically acceptable carrier or diluent.

In still another embodiment, the above-described objects of the present invention have been met by a method of treatment comprising administering to s tumor-bearing subject, a pharmaceutically effective amount of said tumor cells .

BRIEF DESCRIPTION OF THE DRAWINGS

Figures 1A-1B show the tumorigenicity of 4T1 cells in synergenic BALB/c mice injected subcutaneously with either 5 x 10³ 4T1 cells (Figure 1A) , or 10⁴ (Figure IB) 4T1 cells. In Figures 1A-1B, each line represents an individual mouse.

Figure 2 schematically shows the pH -Apr-l-neo expression vector.

Figures 3A-3C show the results of immunofluorescence assays for B7.1 expression in parental 4T1 cells (Figure 3A) ; and 4T1 cells transfected with pH_/3-mB7.1-1-neσ (Figure 3B transfectant #1; and Figure 3C - transfectant #6) .

B7.1 immunofluorescence staining is shown as solid lines, with the second step in staining alone as the dotted line. Figure 4 shows the ability of 4T1 cells transfected with SEB to secrete a molecule capable of stimulating T lymphocyte proliferation in BALB/c mice. Figure 5 shows the specificity of the superantigen produced by 4T1 cells transfected with SEB by an assay for inhibition of T-cell proliferation using SEB-specific polyclonal antibodies.

Figures 6A-6E show the size of clonogenic lung metastases following immunotherapy with parental 4T1 cells (Figure 6A) ; 4T1 cells transfected with SEB

(transfectant #12 - Figure 6B, and transfectant # 14 -

Figure 6C) ; and 4T1 cells transfected with B7.1 and

MHC class II (Figure 6D) ; and 4T1 cells transfected with SEB, MHC class II and B7.1 (Figure 6E) . Figures 7A-7F show the size of liver metastases following immunotherapy with parental 4T1 cells

(Figure 7A) ; 4T1 cell transfected with MHC class I and

B7.1 (Figure 7B) ; and 4T1 cells transfected with SEB,

B7.1 and MHC class II (Figure 7C) ; and show the size of brain metastases following immunotherapy with parental 4T1 cells (Figure 7D) ; 4T1 cells transfected with MHC class I and B7.1 (Figure 7E) ; and 4T1 cells transfected with SEB, B7.1 and MHC class II

(Figure 7F) .

DETAILED DESCRIPTION OF THE INVENTION As discussed above, in one embodiment, the above-described objects of the present invention have been met by a tumor cell which has been transfected so as to co-express SEB with MHC class II, and which is capable of enhancing a CD4⁺ tumor-specific response.

As discussed above, in another embodiment of the present invention, the tumor cells further also co-express B7 (B7.1 and/or B7.2) . The tumor cells of the present invention are obtained by "transfection" , i.e. a mammalian cell in which exogenous nucleic acid has been introduced. The particular method by which transfection is achieved is not critical to the present invention. Examples of such methods include electroporat ion, calcium-phosphate precipitation, DEAE-dextran treatment, lipofection, microinjection and infection with viral vectors. These methods of transfection of mammalian cells are well-known in the art, and are described, e.g., in Sambrook et al, Molecular Cloning: A Laboratory Manual , 2nd Edition, Cold Spring Harbor Laboratory press (1989) .

As used herein, "desired molecule" refers to SEB and/or B7.

A preferred approach for introducing nucleic acid encoding the desired molecule into tumor cells is by use of a viral vector containing nucleic acid encoding the desired molecule. The particular viral vector employed is not critical to the present invention.

Examples of viral vectors which can be employed in the present invention include retroviral vectors

(Eglitis et al, Science, 230:1395-1398 (1985);

Danos et al, Proc. Natl . Acad. Sci . , USA, 85:6460-6464 (1988); and Markowitz et al, J. Virol . , 62:1120-1124

(1988)), adenoviral vectors (Rosenfeld et al, Cell ,

6j3.:143-155 (1992)) , and adeno-associated viral vectors

(Tratschin et al, Mol . Cell . Biol . , 5:3251-3260

(1985) ; Rosenfeld, supra; Anderson, Science, 226:401-409 (1984); and Friedman, Science,

244:1275-1281 (1989)). Infection of tumor cells with a viral vector has the advantage that a large proportion of cells will receive the nucleic acid, thereby obviating the need for selection of cells which have received the nucleic acid. Also, molecules encoded within the viral vector are expressed efficiently in cells which have taken up viral vector nucleic acid.

Alternatively, the desired molecule can be introduced into the tumor cell using a plasmid expression vector which contains the nucleic acid encoding the desired molecule. The particular plasmid expression vector employed is not critical to the present invention. Examples of such plasmid expression vectors include CDM8 (Seed, Nature, 329.: 840 (1987)) ; and pMT2PC (Kaufman et al, EMBO J. , 6:187-195 (1987) ) .

Alternatively, nucleic acid can be delivered to tumor cells in vivo by direct injection of naked nucleic acid into tumor cells (Acsadi et al, Nature, 3_32: 815-818 (1991)). A delivery apparatus is commercially available (BioRad) . Optionally, to be suitable for injection, the nucleic acid can be complexed with a carrier such as a liposome. Nucleic acid encoding an MHC class I complexed with a liposome has been directly injected into tumors of melanoma patients (Hoffman, Science, 256:305-309 (1992)).

The type of nucleic acid to be introduced is not critical to the present invention. For example, the nucleic acid may be a DNA molecule comprising a gene encoding the desired molecule, a sense-strand RNA encoding the desired molecule or a recombinant expression vector containing a DNA molecule encoding the desired molecule. Further, the DNA molecule is preferably cDNA.

A preferred cDNA encoding SEB is described in Ranelli et al, Proc. Natl . Acad. Sci . , USA, 82_.: 5850-5854 (1985), which is incorporated by reference herein in its entirety. A preferred cDNA encoding MHC class II is described in Hood et al, Ann. Rev. Immπunol . ,

1:529-568 (1983) ; and Auffray et al, Advances in Human

Genetics, 15:197-247 (1987) , which are incorporated by reference herein in' its entirety.

A preferred cDNA encoding human B7 is described in Freeman et al, J. Exp. Med. , 174:625-631 (1991), which is incorporated by reference herein in its entirety.. The nucleic acid is in a form suitable for expression of the desired molecule, i.e., the nucleic acid contains all of the coding and regulatory sequences required for transcription and translation of a gene, which may include promoters, enhancers and polyadenylation signals, and sequences necessary for transport of the molecule to the surface of the tumor cell, including N-terminal signal sequences. When the nucleic acid is a cDNA in a recombinant expression vector, the regulatory functions responsible for transcription and/or translation of the cDNA are often provided by viral sequences . Examples of commonly used viral promoters include those derived from polyoma, Adenovirus 2, cytomegalovirus and Simian Virus 40, and retroviral LTRs. Regulatory sequences linked to the cDNA can be selected to provide constitutive or inducible transcription, by, for example, use of an inducible promoter, such as the metallothienin promoter or a glucocorticoid-responsive promoter. Expression of the desired molecule on the surface of the tumor cell can be accomplished, for example, by including a native transmembrane coding sequence of the molecule in the nucleic acid sequence, or by including signals which lead to modification of the protein, such as a C-terminal inositol-phosphate linkage, that allows for association of the molecule with the outer surface of the cell membrane.

Expression of the desired molecule on the surface of the tumor cell can be confirmed by immunofluorescent staining of the cells. For example, cells may be stained with a fluorescently labeled monoclonal antibody reactive against the desired molecule, or with a fluorescently labeled protein which binds the desired molecule. More specifically, expression of SEB can be determined using a polyclonal antibody, which recognizes SEB. Expression of B7 can be determined using a monoclonal antibody, 133, which recognizes B7 (Freedman et al, J. Immunol . , 139:3260-3267 (1987)). Alternatively, a labeled protein or fusion protein known in the art which binds to SEB can be used to detect expression of SEB; and a labeled soluble CD28 or CTLA4 protein or fusion protein which binds to B7 can be used to detect expression of B7. Since only a small fraction of cells (about 1 out of 10^s) typically integrate transfected plasmid DNA into their genomes, it is advantageous to transfect a nucleic acid encoding a selectable marker into the tumor cell along with the nucleic acid(s) of interest. The particular selectable marker employed in not critical to the present invention. Examples of such selectable markers include those which confer resistance to drugs, such as G418, hygromycin and methotrexate . Selectable markers may be introduced on the same plasmid as the gene(s) of interest or may be introduced on a separate plasmid. Following selection of transfected tumor cells using the appropriate selectable marker (s), expression of the desired molecule on the surface of the tumor cell can be confirmed by immuno luorescent staining of the cells as discussed above.

When transfection of tumor cells leads to modification of a large proportion of the tumor cells, and efficient expression of the desired molecule on the surface of tumor cells, e.g., when using a viral expression vector, tumor cells may be used without further isolation or subcloning. Alternatively, a homogenous population of transfected tumor cells can be prepared by isolating a single transfected tumor cell by limiting dilution cloning, followed by expansion of the single tumor cell into a clonal population of cells by standard techniques.

In the present invention, the starting material tumor cell may be capable of expressing the desired molecule, but fails to do so, or the starting material tumor cell may express insufficient amounts of the desired molecule thereby failing to activate T-cells. Thus, in the present invention the starting material tumor cell may be induced to express the desired molecule or the level of expression of the desired molecule may be increased. An agent which stimulates expression of the desired molecule can be used in order to induce or increase expression of the desired molecule on the tumor cell surface. For example, tumor cells can be contacted with the agent in vitro in a culture medium. The agent which stimulates expression of the desired molecule may act, for example, by increasing transcription of the gene of the desired molecule, by increasing translation of mRNA encoding the desired molecule, or by increasing the stability or transport of the desired molecule to the cell surface. For example, expression of SEB can be upregulated in a cell by bacterial infection. In addition, expression of B7 can be upregulated in a cell by a second messenger pathway involving cAMP (Nabavi et al, Nature, 360:266-268 (1992)). Thus, a tumor cell can be contacted with an agent, which increases intracellular cAMP levels or which mimics cAMP, such as a cAMP analogue, e.g., dibutyryl cAMP, to stimulate expression of B7 on the tumor cell surface. Expression of B7 can also be induced on normal resting B cells by crosslinking cell-surface MHC class II on the B cells with an antibody against the MHC class II molecules (Kuolova et al, J^". Exp. Med. , 173:759-762 (1991)). Similarly, a tumor cell which expresses MHC class II its surface can be treated with anti-MHC class II antibodies to induce or increase B7 expression on the tumor cell surface. Another agent which can be used to induce or increase expression of the desired molecule on a tumor cell surface is a nucleic acid encoding a transcription factor which upregulates transcription of the gene encoding the desired molecule. This nucleic acid can be transfected into the tumor cell to cause" increased transcription of the gene of the desired molecule, resulting in increased cell-surface levels of the desired molecule.

In the present invention, the tumor cells may express one or more MHC molecules on their surface to trigger both a costimulatory signal and a primary, antigen-specific, signal in T-cells. Before modification, tumor cells may be unable to express MHC molecules, may fail to express MHC molecules, although they are capable of expressing such molecules, or may express insufficient amounts of MHC molecules on the tumor cell surface to cause T-cell activation. Tumor cells can be modified to express either MHC class I or MHC class II molecules, or both. One approach to modifying tumor cells to express MHC molecules is to transfect the tumor cell with one or more nucleic acids encoding one or more MHC molecules .

Alternatively, an agent which induces or increases expression of one or more MHC molecules on tumor cells can be used to modify tumor cells . Inducing or increasing expression of MHC class II on a tumor cell can be particularly beneficial for activating CD4⁺ T-cells against the tumor. This is because the ability of MHC class II^* tumor cells to directly present tumor peptides to CD4^* T-cells bypasses the need for professional MHC class II⁺ APCs. This can improve tumor immunogenicity, because soluble tumor antigen (in the form of tumor cell debris or secreted protein) may not be available for uptake by professional MHC class II^* APCs.

MHC class II molecules are cell-surface /β heterodimers which structurally contain a cleft into which antigenic peptides bind and which function to present bound peptides to the antigen-specific TcR. Multiple, different MHC class'^' II proteins are expressed on professional APCs and different MHC class II proteins bind different antigenic peptides . Expression of multiple MHC class II molecules, therefore, increases the spectrum of antigenic peptides that can be presented by an APC or by a modified tumor cell . The and β chains of MHC class II molecules are encoded by different genes. For instance, the human MHC class II protein HLA-DR is encoded by the HLA-DRα and HLA-DR/3 genes . Additionally, many polymorphic alleles of MHC class II genes exist in human and other species. T-cells of a particular individual respond to stimulation by antigenic peptides in conjunction with self MHC molecules, a phenomenon termed MHC restriction. In addition, certain T-cells can also respond to stimulation by polymorphic alleles of MHC molecules found on the cells of other individuals, a phenomenon termed allogenicity. A review of MHC class II structure and function is provided in Germain et al, Ann. Rev. Immunol., ^'11:403-450 (1993).

The tumor cells of the present invention can be modified ex vivo to express MHC class II by transfection of isolated tumor cells with one or more nucleic acids encoding one or more MHC class II chains and one or more MHC class II β chains in a form suitable for expression of the MHC class II molecule (s) on the surface of the tumor cell. Both an α and a β chain protein must be present in the tumor cell to form a surface heterodimer, and neither chain will be expressed on the cell surface alone. The nucleic acid sequences of many murine and human class II genes are known (Hood et al, Ann. Rev. Immnunol . , 1:529-568 (1983); and Auffray et al, Advances in Human Genetics, 15.:197-247 (1987)). Preferably, the introduced MHC class II molecule is a self MHC class II molecule. Alternatively, the MHC class II molecule could be a foreign, allogeneic, MHC class II molecule. A particular foreign MHC class II molecule to be introduced into tumor cells can be selected by its ability to induce T-cells from a tumor-bearing subject to proliferate and/or secrete cytokines when stimulated by cells expressing the foreign MHC class II molecule, i.e., by its ability to induce an allogeneic response. The tumor cells to be transfected may not express MHC class II molecules on their surface prior to transfection or may express amounts insufficient to stimulate a T-cell response. Alternatively, tumor cells which express MHC class II molecules prior to transfection can be further transfected with additional, different MHC class II genes or with other polymorphic alleles of MHC class II genes to increase the spectrum of antigenic fragments that the tumor cells can present to T-cells . Fragments, mutants or variants of MHC class II molecules that retain the ability to bind peptide antigens and activate T-cell responses, as evidenced by proliferation and/or lymphokine production by T-cells, are considered within the scope of the invention. A preferred variant is an MHC class II molecule in which the cytoplasmic domain of either one or both of the and β chains is truncated. Truncation of the cytoplasmic domains allows peptide binding by and cell surface expression of MHC class II molecules, but prevents the induction of endogenous B7 expression, which is triggered by an intracellular signal generated by the cytoplasmic domains of the MHC class II protein chains upon crosslinking of cell surface MHC class II molecules (Kuolova et al, supra; and Nabavi et al (1992) , supra) . In tumor cells transfected to constitutively express B7 and/or SEB, it may be desirable to inhibit the expression of endogenous B7 and/or SEB, for instance to restrain potential downregulatory feedback mechanisms . Transfection of a tumor cell with a nucleic acid(s) encoding a cytoplasmic domain-truncated form of MHC class II and β chain proteins would inhibit endogenous B7 and/or SEB expression. Such variants can be produced by, for example, introducing a stop codon in the MHC class II chain gene(s) after the nucleotides encoding the transmembrane spanning region. The cytoplasmic domain of either the α chain or the β chain protein can be truncated, or, for more complete inhibition of B7 and/or SEB induction, both the and β chains can be truncated (Griffith et al, Proc. Na tl . Acad Sci . , USA, 85:4847-4852 (1988); and Νabavi et al, J. Immnxmol . , 142:1444-1447 (1989)) .

When a tumor cell is transfected with nucleic acid which encodes more than one molecule, the transfections can be performed simultaneously or sequentially. If the transfections are performed simultaneously, the molecules can be introduced on the same nucleic acid, so long as the encoded sequences do not exceed a carrying capacity for a particular vector used. Alternatively, the molecules can be encoded by separate nucleic acids. If the transfections are conducted sequentially, and tumor cells are selected using a selectable marker, one selectable marker can be used in conjunction with the first introduced nucleic acid while a different selectable marker can be used in conjunction with the next introduced nucleic acid. Again, the particular selectable marker employed is not critical to the present invention; examples of which are set forth above. The expression of MHC class II on the cell surface of a tumor cell can be determined, for example, by immunofluorescence of tumor cells using fluorescently labeled monoclonal antibodies directed against different MHC molecules . Monoclonal antibodies which recognize either non-polymorphic regions of a particular MHC molecule (non-allele specific) or polymorphic regions of a particular MHC molecule (allele-specific) can be used are known to those skilled in the art. Another approach to modifying a tumor cell ex vivo to express MHC molecules on the surface of a tumor cell is to use an agent which stimulates expression of MHC molecules in order to induce or increase expression of MHC molecules on the tumor cell surface. For example, tumor cells can be contacted with the agent in vitro in a culture medium. An agent which stimulates expression of MHC molecules may act, for instance, by increasing transcription of MHC class II genes, by increasing translation of MHC class II mRNAs or by^' increasing stability or transport of MHC class II proteins to the cell surface. The particular agent employed is not critical to the present invention. Examples of such agents which increase the level of cell-surface expression of MHC class II are described in Cockfield et al, J. Immunol . , 144:2967-2974 (1990); Noelle et al, J^". Imnmunol . , 137:1718-1723 (1986); Mond et al, J. Immunol . , 127:881-888 (1981); Willman et al, J. Exp. Med. , 170:1559-1567 (1989); Celada et al, J^". Immunol., 146_.: 114-120 (1991); and Glimeher et al,

Ann. Rev. Immunol . , 10:13-49 (1992). These agents include cytokines, antibodies to other cell surface molecules and phorbol esters . One agent which upregulates MHC class II on a wide variety of cell types is the cytokine interfe^'ron-γ. Thus, for example, tumor cells modified to express a costimulatory molecule can be further modified to increase expression of MHC molecules by contact with interferon-γ. Another agent which can be used to induce or increase expression of an MHC molecule on a tumor cell surface is a nucleic acid encoding a transcription factor which upregulates transcription of MHC class II gene. Such a nucleic acid can be transfected into the tumor cell to cause increased transcription of MHC gene, resulting in increased cell-surface levels of MHC proteins. MHC class II genes are regulated by different transcription factors. However, the multiple MHC class II genes are regulated coordinately. Therefore, transfection of a tumor cell with a nucleic acid encoding a transcription factor which regulates MHC gene expression may increase expression of several different MHC molecules on the tumor cell surface. Several transcription factors which regulate the expression of MHC genes, and which can be employed in the present invention, have been identified, cloned and characterized (Reith et al, Genes Dev. , 4:1528-1540 (1990); Liou et al, Science, 247:1581-1584 (1988); and Didier et al, Prσc. Natl . Acad. Sci . , USA, 85:7322-7326 (1988)).

The tumor cells of the present invention can be transfected or treated by one or more of the approaches encompassed by the present invention to express the desired molecules. The particular starting material tumor cell employed in the present invention is not critical thereto . Tumor cells can be obtained can be obtained from a spontaneously tumor which has arisen, e.g., in a human subject, i.e., an autologous tumor, or may be obtained from experimentally derived or induced tumor, e.g., "in an animal subject, or may be an established tumor cell line having an identical tissue type as the tumor of said tumor-bearing subject and is HLA class II matched to said subject. Further, the tumor cells can be obtained, for example, from a solid tumor of an organ, such as a tumor of the lung, liver, breast, colon, bone, etc. Malignancies of solid organs include carcinomas, sarcomas, melanomas and neuroblastomas . The tumor cells can also be obtained from a blood-borne (i.e., dispersed) malignancy, such as a lymphoma, a myeloma or a leukemia.

The tumor cells to be modified include those that express MHC molecules on their cell surface prior to transfection, and those that express no or low levels of MHC class I and/or class II molecules. A minority of normal cell types express MHC class II molecules .

It is therefore expected that many tumor cells will not express MHC class II molecules naturally. These tumors can be modified to express the desired molecule. Several types of tumors have been found to naturally express surface MHC class II, such as melanomas (van Duinen et al, Cancer Res., 48 -.1019-1025

(1988)), diffuse large cell lymphomas (O'Keane et al,

Cancer, 66.:1147-1153 (1990)), squamous cell carcinomas of the head and neck (Mattijssen et al, Int. J.

Cancer, 6.:95-100 (1991)), and colorectal carcinomas

(Moller et al, Int. J. Cancer, 6:155-162 (1991)).

Tumor cells which naturally express MHC class II can be modified to express the desired molecule, and, in addition, other MHC class II which can increase the spectrum of TAA peptides which can be presented by the tumor cell .

The tumor cells can also be modified in vivo by use of an agent which induces or increases expression of a costimulatory molecule (and, if necessary, MHC molecules as described herein) . The agent may be administered systemically, e.g., by intravenous injection, or, preferably, locally to the tumor cells.

The tumor cells of the invention are useful for stimulating an anti-tumor T-cell-mediated immune response by triggering an antigen-specific signal and a costimulatory signal in tumor-specific T-cells.

Following this inductive, or afferent, phase of an immune response, effector populations of T-cells are generated. These effector T-cell populations can include both CD4⁺ T-cells and CD8* T-cells. The effector populations are responsible for elimination of tumors cell, by, for example, cytolysis of the tumor cell. Once T-cells are activated, expression of a costimulatory molecule is not required on a target cell for recognition of the target cell by effector T-cells, or for the effector functions of the T cells (Harding et al, J. Exp. Med. , 177:1791-1796 (1993)). Therefore, the antitumor T-cell-mediated immune response induced by^' the tumor cells of the present invention is effective against both the tumor cells of the present invention and native tumor cells which do not express the desired molecules .

Additionally, the density and/or type of MHC on the cell surface required for the afferent and efferent phases of a T-cell-mediated immune response can differ. Fewer MHC molecules, or only certain types of MHC molecules (e.g., MHC class I but not MHC class II) may be needed on a tumor cell for recognition by effector T-cells than is needed for the initial activation of T-cells. Therefore, tumor cells which naturally express low amounts of MHC molecules, but are modified to express increased amounts of MHC molecules, can induce a T-cell-mediated immune response which is effective against the unmodified tumor- cells. Alternatively, tumor cells which naturally express MHC class I, but not MHC class II, which are then modified to express MHC class II induce a T-cell-mediated immune response which includes effector T-cell populations which can eliminate the parental MHC class I⁺, class II—tumor cells.

As discussed above, the above-described objects of the present invention have been met by a pharmaceutical composition comprising a pharmaceutically effective amount of the tumor cells of the present invention, and a pharmaceutically acceptable carrier or diluent .

A pharmaceutically acceptable carrier or diluent is one which is biologically compatible with the subject. The particular pharmaceutically acceptable carrier or diluent employed is not critical to the present invention. Examples of acceptable carriers or diluents include saline and aqueous buffer solutions. In all cases, the compositions must be sterile and must be fluid to the extent that easy syringability exists .

In still another embodiment, the above-described objects of the present invention have been met by a method of treatment comprising administering to a tumor-bearing subject, a pharmaceutically effective amount of said tumor cells .

The term "tumor-bearing subject" is intended to include living organisms in which tumors can arise or be experimentally induced. The particular tumor-bearing subject employed is not critical to the present invention. Examples of tumor-bearing subjects include tumor-bearing humans, dogs, cats, mice, rats, and transgenic species thereof .

Administration of the tumor cells of the present invention can be carried out using known procedures, at dosages and for periods of time effective to achieve the desired result. For example, a pharmaceutically effective dose of the tumor cells may vary according to such factors as age, sex and weight of the individual, the type of tumor cell and degree of tumor burden, and the immunological competency of the subject. Dosage regimens may be adjusted to provide optimum therapeutic responses. For example, a single dose of the tumor cells may be administered or several doses may be administered over time.

The mode of administration is not critical to the present invention. Examples of such modes of administration include intravenous, intramuscular, intraperitoneal and subcutaneous injections. Another approach to inducing or enhancing an anti-tumor T-cell-mediated immune response by triggering a costimulatory signal in T-cells is to obtain T lymphocytes from a tumor-bearing subject and activate the cells ^'in vitro by contacting them with the tumor cells of the present invention, and a stimulatory form of a costimulatory molecule.

In this embodiment, the T-cells can be obtained from a subject, for example, from peripheral blood. Peripheral blood can be further fractionated to remove red blood cells and enrich for or isolate T lymphocytes or T lymphocyte subpopulations .

The T-cells can be activated in vi tro by culturing the T-cells with tumor cells obtained from the subject (e.g., from a biopsy or from peripheral blood in the case of blood-borne malignancies) together with a stimulatory form of a costimulatory molecule and/or by exposure to the tumor cells of the present invention. The term "stimulatory form" means that the costimulatory molecule is capable of crosslinking its receptor on a T-cell and triggering a costimulatory signal in T-cells. The stimulatory form of the costimulatory molecule can be, for example, a soluble ultivalent molecule or an immobilized form of the costimulatory molecule, for instance coupled to a solid support. Fragments, mutants or variants (e.g., fusion proteins) of costimulatory molecules which retain the ability to trigger a costimulatory signal in T-cells can also be used. In a preferred embodiment, a soluble extracellular portion of SEB and/or B7 is used to provide costimulation to the T-cells . Following culturing of the T-cells in vitro with tumor cells and a costimulatory molecule, or the tumor cell of the present invention, to activate tumor-specific T-cells, the T-cells can be administered to the subject, for example, by intravenous injection.

The tumor cells of the present invention can be used to increase tumor immunogenicity, and therefore can be used therapeutically for inducing or enhancing T lymphocyte-mediated anti-tumor immunity in a subject with a tumor or at risk of developing a tumor. A method for treating a subject with a tumor involves obtaining tumor cells from the subject, modifying the tumor cells ex vivo to express a T-cell costimulatory molecule, for example by transfecting them with an appropriate nucleic acid, and administering a therapeutically effective dose of the modified tumor cells to the subject. Appropriate nucleic acids to be introduced into a tumor cell include a nucleic acid encoding a T-cell costimulatory molecule, for example SEB and/or a CD28 and/or CTLA4 ligand such as B7, alone or together with nucleic acids encoding MHC class II as described herein. Alternatively, after tumor cells are obtained from a subject, they can be modified ex vivo using an agent which induces or increases expression of a costimulatory molecule (and possibly also using agent (s) which induce or increase MHC molecules) . Tumor cells can be obtained from a subject by, for example, surgical removal of tumor cells, e.g., a biopsy of the tumor, or from a blood sample from the subject in cases of blood-borne malignancies. In the case of an experimentally induced tumor, the cells used to induce the tumor can be used, e.g., cells of a tumor cell line. Samples of solid tumors may be treated prior to modification to produce a single-cell suspension of tumor cells for maximal efficiency of transfection. Possible treatments include manual dispersion of cells or enzymatic digestion of connective tissue fibers, e.g., by collagenase.

The tumor cells can be transfected immediately after being obtained from the subject or can be cultured in vi tro prior to transfection to allow for further characterization of the tumor cells (e.g., determination of the expression of cell surface molecules) . The nucleic acids chosen for transfection can be determined following characterization of the proteins expressed by the tumor cell. For instance, expression of MHC proteins on the cell surface of the tumor cells in the tumor cell can be assessed. Tumors which express no, or limited amounts of or types of MHC class II can be transfected with nucleic acids encoding MHC proteins. If necessary, following transfection, tumor cells can be screened for introduction of the nucleic acid by using a selectable marker (e.g. drug resistance) which is introduced into the tumor cells together with the nucleic acid of interest.

Prior to administration to the subject, the modified tumor cells can be treated to render them incapable of further proliferation in the subject, thereby preventing any possible outgrowth of the modified tumor cells . Possible treatments include irradiation or mitomycin C treatment, which abrogate the proliferative capacity of the tumor cells while maintaining the ability of the tumor cells to trigger antigen-specific and costimulatory signals in T-cells, and thus to stimulate an immune response.

The tumor cells of the present invention can be administered to the subject by injection of the tumor cells into the subject. The route of injection can be, for example, intravenous, intramuscular, intraperitoneal or subcutaneous. Administration of the tumor cells of the present invention at the site of the original tumor may be beneficial for inducing local T-cell-mediated immune responses against the original tumor. Administration of the tumor cells of the present inventio in a disseminated manner, e.g., by intravenous injection, may provide systemic anti-tumor immunity and, furthermore, may protect against metastatic spread of tumor cells from the original site. The tumor cells of the present invention can be administered to a subject prior to or in conjunction with other forms of therapy, or can be administered after other treatments, such as chemotherapy or surgical intervention.

Another method for treating a subject with a tumor is to modify tumor cells in vivo to express the desired molecule, alone or in conjunction with MHC molecules . This method can involve modifying tumor cells in vivo by providing nucleic acid encoding the protein (s) to be expressed using vectors and delivery methods effective for in vivo gene therapy as described herein. Alternatively, one or more agents which induce or increase expression of the desired molecule, and possibly MHC molecules, can be administered to a subject with a tumor. The tumor cells of the present invention may also be used in a method for treating metastatic spread of a tumor to and in, e.g., the lungs, liver or brain, or preventing, or treating recurrence of a tumor, particularly mammary tumors. The following examples are provided for illustrative purposes only, and are in no way intended to limit the scope of the present invention. EXAMPLE 1

Generation of Mammary Tumor Transfectants

Which Can More Effectively and Directly

Present Tumor Antiσen to CD4* T-cells A. Mouse 4T1 Mammary Carcinoma

The mouse 4T1 mammary carcinoma is used, as such is an excellent tumor for studying metastatic disease.

4T1 is a poorly immunogenic mammary tumor

(Dexter et al, Cancer Res. , 38:3174-3181 (1978); Miller et al, Invasion Metastasis, 3 :22-31 (1983) ; and

Aslakson et al, Cancer Res. , 52_.:1399-1405 (1992)).

This tumor expresses adequate levels of MHC class I, making it a suitable target for CTL. However, there is no detectable MHC class II expression. Preliminary data confirm that 4T1 is highly tumorigenic, in that progressive tumors develop in 100% of naive syngeneic

BALB/c mice challenged with as little as 5 x 10³ cells

(see Figures 1A-1B, and Table 1 below) .

TABLE 1

Tumor Growth Analysis of 4T1 Cells in Syngeneic BALB/c Mice

BALB/c mice were challenged with the indicated number of parental 4T1 mammary tumor cells . Tumor incidence indicates the percentage of animals in which the thigh diameter reached 14-16 mm, leading to sacrifice of the animal.

The numbers in parentheses indicate the number of animals in which the thigh diameter reached 14-16 mm, leading to sacrifice of the animal, from the total number of animals tested at the indicated tumor dose .

Also, using the method described by Aslakson et al, supra, spontaneous metastases can be measured in a variety of organs isolated from BALE/c mice challenged in the mammary gland with 10⁶ live tumor cells (see Table 2 below) .

TABLE 2 Kinetics of Spontaneous Metastases Development in BALB/c Mice

Naive syngeneic BALB/c mice were challenged s.c. in the abdominal mammary gland with 10^δ live 4T1 cells. Indicated organs were removed and analyzed for spontaneous metastatic tumor cells as described in the Methods section. ND, no data

The above data indicate that 4T1 is a good model for human disease, where at time of surgery, lymph nodes can be positive for metastases, but metastases are not obvious in other parts of the body. Animals challenged with 5 x 10³ tumor cells exhibit metastases with a similar kinetics delayed by approximately 10 days. Therefore, even small protective advantages are capable of being detected as a range of tumor challenge doses is used.

B. cDNAs

The MHC class II cDNAs (I-A , I-A„^d) were generated by PCR as described by Pulaski et al, Cancer Res . , 58_.:1486-1493 (1998); which is incorporated by reference herein in its entirety.

B7.1 cDNA was also generated using PCR as described by Yeh et al, Cell . Immunol . , 165_.:217-224 (1995) ; which is incorporated by reference herein in its entirety. The sAg cDNAs for SEB and detection polyclonal antibodies have been provided by Dr. Saleem Khan (University of Pittsburgh) and are described in Ranelli et al, Proc. Natl . Acad. Sci . , USA, 82_.:5850-5854 (1985); which is incorporated by reference herein in its entirety.

C. cDNA Expression Constructs

Each cDNA is subcloned separately into the pH/3-Apr-l-neo expression vector (Figure 2) . This vector contains a human 13-actin promoter and a neomycin (neo^R) selectable marker (Gunning et al, Proc.

Natl. Acad. Sci . , USA, 84=4831-4835 (1987)).

Immunofluorescence data displaying the surface expression of B7.1 by 4T1 cells transfected with pH/3-mB7.1-neo (Yeh et al, supra) shown in Figures 3A-3C, demonstrates that the human 13-actin promoter allows high and stable cDNA expression.

D. Transfections The 4T1 mammary tumor transfectants are generated using either cationic lipids (such as lipofectin or lipofectamine) or calcium phosphate methods

(Pulaski et al, Cancer Res . , 52:2112-2117 (1993)).

Transfectants are cloned 1-2 times by limiting dilution in selective media, and cDNA expression is determined by immunofluorescence , using antibodies from MKD6 hybridoma (anti-MHC class II (Ad) ) (Kappler et al, J. Exp. Med. , 153:1198-1214 (1981)). This hybridoma was purchased from the American Type Culture Collection, ATCC No. HB3.

Double transfectants are obtained by co-transfection of, e.g., MHC class II^* transfectants with plasmid PMT3 , which contains an amethopterin selectable marker (Oprian et al, Biochemistry, 20:11367-11372 (1991)), plus the appropriate second cDNA construct encoding SEB and/or B7, as these clones are already neo^R. Other tumor cell lines have previously been transfected with multiple genes (Lui et al, supra; and Damle et al, supra) . Hence, a variety of cDNAs can be expressed in the 4T1 cells. Only those transfectants which stably express the genes are used, and several clones of each transfection are used.

EXAMPLE 2

Characterize the Immunogenicity of the Mammary Tumor Transfectants

If the transfectants are an effective cell-based vaccine they should be capable of inducing an immune response. Therefore, as described in detail below, the tumorigenicity and spontaneous metastases development of the transfectants is examined in comparison to parental cells to screen for potential immunotherapeutic agents.

A. In Vivo Tumor Growth

Female BALB/c mice (5-10/group) are challenged subcutaneously (s.c.) in the abdominal mammary gland with varying doses (starting at 5 x 10³ and up to 10⁶) of parental or transfected 4T1 tumor cells . As standard controls, a wild-type tumor or an irrelevant BALB/c derived wild-type tumor, such as EMT6 , a mouse mammary tumor (Rockwell et al, J^". Natl. Cancer Inst . , 149:735-747 (1972)), or line 1, a lung carcinoma (Yuhas et al, Cancer Res . , 34 : 722-728 (1974) ) , or tumor cells transfected with vector alone, are used.

Measurements of tumor size are taken every 3-4 days following tumor challenge by using vernier calipers. Tumor size is calculated as the square root of the product of two perpendicular diameters. Animals are sacrificed when the mean tumor diameter reaches 15 mm or when the mice become moribund.

B. Metastases Assay Spontaneous metastases are measured using the methods described by Aslakson et al, supra. Briefly, female BALB/c mice (5-10/group) are challenged in the abdominal mammary gland s.c. with 10⁶ parental or transfected 6-thioguanine-resistant tumor cells, and sacrificed in groups of 5 at weekly intervals (up to 4 weeks) . Similar to human breast cancer, 4T1 cells metastasize prior to removal of the primary tumor. Hence, metastases formation is assessed in the presence of primary tumor. Organs are removed (blood, lymph node, liver, lung, and possibly brain) , digested in enzyme cocktails, and plated in selective media containing 6-thioguanine for outgrowth of metastatic cells . Thus, this strategy measures clonogenic cells instead of metastatic nodules. One disadvantage is that it is not be possible to distinguish between the clonogenic cells which metastasize from the primary tumor, and those clonogenic cells which are due to the doubling of the original metastatic tumor cells in the organ. However, this type of analysis is better because :

(1) it allows one to follow the progression of metastases throughout the whole body

(i.e. , blood, an organ in which nodules can not form) , and (2) one can detect metastatic cells not measurable by sight or weight differences (i.e., those which are too far inside organs or too small) . Also, this type of analysis is very informative for determining the kinetics of metastases development using -a lower challenge dose (5 x 10³ ceils instead of 10⁶) so as to demonstrate whether metastases formation is dos -dependent or time-dependent or both, as these issues are important for determining who are good candidates for the immunotherapy strategy developed in this proposal .

C. Expectations

If the transfectants are more immunogenic than the parental tumor, then they should show reduced tumorigenicity when compared to parental tumor. Likewise, the metastases assays should show either a delay or elimination of metastatic outgrowth by the transfectants as compared to the parental tumor. Appropriate comparisons between the parental, single, and/or double transfectants are made to determine the efficacy of each type of transfectant so that the most immunogenic agents are used.

^' EXAMPLE 3

Generation of Tumor Cells Expressing Suoerantigens Tumor cells expressing superantigen (SEB, Staphylococcal enterotoxin B) were generated as described in Example 1, and tested for superantigen expressing by the ability of their supernatants to secrete a molecule capable of stimulating T lymphocyte proliferation by conventional techniques . The results are shown in Figure 4.

As shown in Figure 4, the supernatants of two SEB transfectant lines, i.e., # 14 and #12, were potent stimulators of T-cell proliferation, more potent than soluble, purified SEB used at 2.0 μg/ml . The specificity of the superantigen produced by the 4T1 transfectants was demonstrated by the ability of SEB-specific polyclonal antibodies to inhibit T-cell proliferation (Figure 5) .

EXAMPLE 4

Immunogenic Transfectants Prevent Metastatic Disease and/or Eliminate and/or Reduce Established Metastatses

Only those immunogenic transfectants showing reduced tumorigenicity and/or metastatic potential are tested as putative vaccines . The following regimens are designed to determine their therapeutic efficacy.

A. Wild-type Tumor Challenge After

Immunization with Cell-Based Vaccine

Female BALB/c mice (5-10/group) are immunized with irradiated parental 4T1 cells or irradiated transfectants 28 days prior to challenge with live parental 4T1 cells. Spontaneous metastases are then be measured as described above . Therefore, it can be determined if these immunotherapeutic agents can prevent or delay metastatic disease. A starting immunization dose ' of 10^β cells is given intraperitoneally. However, other immunization doses or routes can be used to elicit maximum protection. In order to demonstrate the specificity of the immune response, immunized mice are challenged with an irrelevant tumor. Lastly, the longevity of the immune response is determined by increasing the amount of time between the day of immunization and day of tumor challenge .

B. Wild-type Tumor Challenge on the

Same Day as Immunization with Cell-Based Vaccine

The immunogenic transfectants exhibiting the best potential from the first regimen are used to combat established metastases. In this scheme, naive mice are immunized and challenged at the same time, and examined for spontaneous metastases in comparison to controls .

C. Wild-type Tumor Challenge Before

Immunization with Cell-Based Vaccine

Ultimately, these tests lead to the expectation of eliminating metastases that have established themselves prior to immunization, as this scenario best mimics the clinical case. Mice are challenged with appropriate wild-type tumors first, and then at a later time point these same mice are immunized with irradiated transfectants or controls. Again, multiple immunizations can be used. As shown in Figure 6A-6E and 7A-7F, mice with large established primary tumors (>l-2 mm) , and established metastases have significant reduction or elimination of metastases following therapy with transfectants . Therapy with 4T1 transfectants expressing superantigen (SEB) alone (Figure 5) show some reduction in metastases, while therapy with 4T1 transfectants expressing both MHC class II and superantigen (SEB, Figure 6E) show maximum reduction in metastases .

D. Treatment of Mice Following Surgical

Removal of the Primary Tumor

This scheme closely models the human disease setting in that mice having established primary breast tumors also have established metastases. The primary tumor is surgically ressected from the mice, and immunotherapy with the tumor cells of the present invention is then initiated as described above.

While the invention has been described in detail, and with reference to specific embodiments thereof, it will be apparent to one of ordinary skill in the art that various changes and modifications can be made therein without departing from the spirit and scope thereof .

Claims

WHAT IS CLAIMED;

Claim 1. A tumor cell which has been transfected so as to co-express SEB with MHC class II, and which is capable of enhancing a CD4* tumor-specific response.

Claim 2. The tumor cell of Claim 1, wherein said tumor cell also co-expresses B7.

Claim 3. The tumor cell of Claims 1 or 2 , wherein said tumor cell is obtained from a tumor selected from the group consisting of a carcinoma, sarcoma, melanoma, neuroblastoma, lymphoma, myeloma and leukemia.

Claim 4. The tumor cell of Claims 1 or 2, wherein said tumor cell is a mammary tumor cell.

Claim 5. A pharmaceutical composition comprising a pharmaceutically effective amount of a tumor cell of Claims 1 or 2, and a pharmaceutically acceptable carrier or diluent .

Claim 6. The pharmaceutical composition of Claim 5 , wherein said tumor cell is obtained from a tumor selected from the group consisting of a carcinoma, sarcoma, melanoma, neuroblastoma, lymphoma, myeloma and leukemia.

Claim 7. The pharmaceutical composition of Claim 5 , wherein said tumor cell is a mammary tumor cell.

Claim 8. A method for treatment comprising administering to a tumor-bearing subject, a pharmaceutically effective amount of a tumor cell of Claims 1 or 2.

Claim 9. The method of Claim 8, wherein said tumor cell is obtained from a tumor selected from the group consisting of a carcinoma, sarcoma, melanoma, neuroblastoma, lymphoma, myeloma and leukemia.

Claim 10. The method of Claim 8, wherein said tumor cell is a mammary tumor cell.

Claim 11. The method of Claim 8 , wherein said treatment is treatment of metastatic disease, and said tumor-bearing subject has metastases.

Claim 12. The method of Claim 11, wherein said metastases are located in at least one of the lung, liver or brain.

Claim 13. The method of Claim 8 , wherein said tumor cell is an autologous tumor cell or an established tumor cell line having an identical tissue type as the tumor of said tumor-bearing subject and is

HLA class II matched to said subject.