WO2000004927A1 - Allogeneic cellular immunogens useful as cancer vaccines - Google Patents

Abstract

A cellular immunogen is provided for immunizing a host against the effects of the product of a target proto-oncogene, where the overexpression of the target proto-oncogene is associated with a malignancy. The cellular immunogen comprises allogeneic (with respect to the host) cells which have been transfected with at least one transgene construct comprising a transgene cognate to the target proto-oncogene and a strong promoter to drive the expression of the transgene in the transfected cells. The transgene encodes a gene product which induces host immunoreactivity to host self-determinants of the product of the target proto-oncogene gene. The transgene may comprise, for example, wild-type or mutant retroviral oncogene DNA cognate to the target proto-oncogene; or wild-type or mutant proto-oncogene DNA of a species different from the host species. The cellular immunogen may be prepared from allogeneic donor cells, e.g. skin fibroblasts, which are stably or transiently transfected with the transgene construct containing the cognate transgene. The donor cells transfected with the cognate transgene constructs are then inserted into the body of the host to obtain expression of the cognate transgene in the host.

Description

ALLOGENEIC CELLULAR IMMUNOGENS USEFUL AS CANCER

VACCINES

Cross-Reference to Related Application

This application claims the benefit of United States Provisional Application 60/093.965 filed July 24, 1998

Field of the Invention

The invention relates to the field of cancer vaccination and lmmunotherapy

Background of the Invention A current goal of cancer research is the identification of host factors that either predispose to tumor formation or serve to enhance tumoi giowth

Genes that confer the ability to convert cells to a tumoπgemc state aie known as oncogenes The transforming ability of a number of retroviiuses has been localized in individual viral oncogenes (generally v-onc) Cellulai oncogenes (generally c-onc) present m many species are related to ural oncogenes It is generall) believed that retroviral oncogenes may represent escaped and or partially metamorphosed cellular genes that are incorporated into the genomes of transmissible, infectious agents, the retroviruses Some c-onc genes mtπnsicalh lack oncogenic properties, but

be converted by mutation into oncogenes whose transforming activity reflects the acquisition of new properties, or loss of old properties Amino acid substitution can convert a cellular proto-oncogene into an oncogene For example, each of the members of the c-ras proto-oncogene family H-ras, N-ras and K-ras) can give rise to a transforming oncogene by a single base mutation Other c-onc genes may be functionally indistinguishable from the corresponding v-onc, but are oncogemc because they are expressed in much greater amounts or in inappropriate cell types These oncogenes are activated by events that change their expression, but which leave their coding sequence unaltered The best characterized example of this type of proto-oncogene is c- myc Changes in MYC protein sequence do not appear to be essential for oncogenicit} Overexpression or altered regulation is responsible for the oncogemc phenotype Activation of c-myc appears to stem from insertion of a retroviral genome within or near the c-m\c gene, or translocation to a new environment A common feature in the translocated loci is an increase in the level of c-m\c expression

Gene amplification provides another mechanism by which oncogene expression may be increased Many tumor cell lines have visible regions of chromosomal amplification For example, a 20-fold c-m\ c amplification has been observed in certain human leukemia and lung carcinoma lines The related oncogene N-myc is five to one thousand fold amplified m human neuroblastoma and retinoblastoma In human acute myeloid leukemia and colon carcinoma lines, the proto-oncogene c-myb is amplified five to ten fold While established cell lines are prone to amplify genes, the presence of known oncogenes in the amplified regions, and the consistent amplification of particular oncogenes in many independent tumors of the same type, strengthens the correlation between increased expression and tumor growth

Immunity has been successfully induced against tumor formation by inoculation with DNA constructs containing v-onc genes, or by inoculation with v-onc proteins or peptides A series of reports describe a form of "homologous" challenge m which an animal test subject is inoculated with either v-src oncoprotein or DNA constructs containing the v-src gene. Protective immunity was induced against tumor formation by subsequent challenge with v- src DNA or v-src-induced tumor cells. See, Kuzumaki et al , JNCI (1988), 80:959-962; Wisner et al. , J. Virol. (1991), 65:7020-7024; Halpern et al. . Virology (1993), 197:480-484: Taylor et al.. Virology (1994), 205:569-573; Plachy et al , Imrnuno genetics (1994), 40:257-265. A challenge is said to be "homologous" where reactivity to the product of a targeted gene is induced by immunization with the same gene, the corresponding gene product thereof, or fragment of the gene product. A challenge is "heterologous" where reactivity to the product of a targeted gene is induced by immunization with a different gene, gene product or fragment thereof.

WO 92/14756 (1992) describes synthetic peptides and oncoprotein fragments which are capable of eliciting T cellular immunity, for use in cancer vaccines. The peptides and fragments have a point mutation or translocation as compared to the corresponding fragment of the proto-oncogene. The aim is to induce immunoreactivity against the mutated proto-oncogene, not the wild-type proto-oncogene. WO 92/14756 thus relates to a form of homologous challenge.

EP 119.702 (1984) describes synthetic peptides having an amino acid sequence corresponding to a determinant of an oncoprotein encoded by an oncogemc virus, which determinant is vicinal to an active site of the oncoprotein. The active site is a region of the oncoprotein required for oncoprotein function. e.g.. catalysis of phosphorylation. The peptides may be used to immunize hosts to elicit antibodies to the oncoprotein active site. EP 119.702 is thus directed to a form of homologous challenge. The protein product encoded by a proto-oncogene constitutes a self antigen and. depending on the pattern of its endogenous expression, would be tolerogenic at the level of T cell recognition of the self peptides of this product. Thus, vaccination against cancers which derive from proto-oncogene overexpression is problematic. Recent attempts have been made to induce immunity in vitro or in vivo to the product of the ΗER-2/neu proto-oncogene. The proto-oncogene encodes a 185-kDa transmembrane protein. The ΗER-2/neu proto-oncogene is overexpressed in certain cancers, most notably breast cancer. In each report discussed below, the immunogen selected to induce immunity comprised a purified peptide of the pl85^HER_2 "^e" protein, and not a cellular immunogen.

Disis et al , Cancer Res. (1994) 54: 16-20 identified several breast cancer patients with antibody immunity and CD4+ helper/inducer T-cell immunity responses to pl85^{HER 2} "^e" protein. Antibodies to pl85^{HER 2/}"^e" were identified in eleven of twenty premenopausal breast cancer patients. It was assumed prior to this work that patients would be immunological ly tolerant to HER-2/new as a self-protein and that immunity would be difficult to generate.

Disis et al . Cancer Res. (1994) 54: 1071-1076 constructed synthetic peptides identical to pl85^{HER 2 w}" protein segments with amino acid motifs similar to the published motif for HLA-A2.1 -binding peptides. Out of four peptides synthesized, two were shown to elicit peptide-specific cytotoxic T- lymphocytes by primary in vitro immunization in a culture system using peripheral blood lymphocytes from a normal individual homozygous for HLA- A2. Thus, it was concluded that the pl85^{HER": ne}" proto-oncogene protein contains immunogenic epitopes capable of generating human CD8^T cytotoxic T- lymphocytes.

The cytotoxic T cells elicited in the latter report were not, however, shown to recognize tumor cells, but only targets that bound the synthesized peptides. Other work (Dahl et al.. J. Immunol. (1996), 157:239- 246) has demonstrated that cytotoxic cells may recognize targets that bind peptide but fail to recognize targets that endogenously synthesize peptide. It is thus unclear whether the cytotoxic cells elicited by Disis et al. would be capable of recognizing rumor cells. In any event, no protection against tumor growth was demonstrated by Disis et al. Peoples et al , Proc Natl Acad Sci USA (1995), 92 432-436. report the identification of antigemc peptides presented on the surface of ovarian and breast cancer cells by HLA class I molecules and recognized by tumor- specific cytotoxic T lymphocytes Both HLA-A2-restπcted breast and ovarian tumor-specific cytotoxic T lymphocytes recognized shared antigemc peptides T cells sensitized against a nme-amino acid sequence of one of the peptides demonstrated significant recognition of HLA-A2 HER2/«ew tumors

It remains unclear whether Peoples et al have successfully attacked proto-oncogene-encoded self as the immunizing peptide which is expressed in the tumor cells contained an isoleucine at position 2, whereas the peptide expressed in normal tissue contains valme residue at this position Moreover, although stimulation of T cells occurred in vitro, this stimulation does not represent a true primary immune response insofar as the starting T cell population represented tumor infiltrating lymphocytes The research accounts of Disis et al and Peoples et al required a form of in vitro stimulation, either priming as described by Disis et al . or restimulation as described by Peoples et al The in vitro protocols of Disis et al and Peoples et al require a mutant cell line to aid m selection of the peptide which will serve to induce reactivity Non-mutant, peptide antigen-presenting cells have their HLA class I molecules already loaded with endogenous peptides. a phenomenon which precludes exogenous loading from without The value of the mutant lines is that they lack the TAP genes (encoding the transporters associated with antigen presentation) Class I binding of mternally-deπved peptides is significantly lowered, and empty class I molecules are present on the cell surface and available for binding of exogenously added peptides This availability of peptide binding sites on membrane-bound class I allows examination of whether a given peptide will (l) even bind to class I, and (n) function as a target m cytotoxic T cell assay s However, the need for a mutant cell line for deduction of candidate immunizing peptide sequences limits the usefulness of peptide-based immunization schemes Fendly et al. , J. Biol Response Modifiers (1990), 9:449-455 ^' present an account of a polypeptide-based immunotherapy. Purified polypeptide corresponding to the extracellular domain of the pl85^HER"2 "^e" protein was obtained from a transfected cell line. The purified peptide was employed in the immunization of guinea pigs. The immunized animals developed a cellular immune response, as monitored by delayed-type hypersensitivity . Antisera derived from immunized animals specifically inhibited the in vitro growth of human breast tumor cells overexpressing pl85^{HER 2/}"™. There is no indication by Fendly et al. of induction of self versus non-self reactivity. It is likely that the guinea pigs were chiefly responding to non-self determinants (as defined in terms of the guinea pig host) on the human polypeptide immunogen.

The use of peptides for immunization is of necessity limited to immunization with a single haplotype. There are approximately thirty HLA types in man. In each case of peptide immunization, one must be careful to select peptides which match the host HLA type. The selected peptide must be immunogenic in the host and be capable of presentation to host immune system cells.

What is needed is an immunization method for immunizing humans and animals against self-encoded proto-oncogenes which are associated with the development of cancer, which dispenses with the need for isolating immunogenic, HLA host-matched peptides for immunization.

Summary of the Invention

It is an object of the invention to induce reactivity to self- determinants of the product of an overexpressed proto-oncogene

It is an object of the invention to provide for a form of therapy or prophylaxis based upon the capacity to induce immune reactivity to proto— oncogene-encoded self as overexpressed in rumor cells

It is an object of the invention to provide a cellular immunogen for use in immunization against self proto-oncogene determinants

It is an object of the invention to provide for a method foi vaccinating a host against disease associated with the overexpression of a proto- oncogene

These and other objects will be apparent from the following disclosure

A method of vaccinating a host against disease associated with the overexpression of a target proto-oncogene is provided The method comprises

(a) transfecting allogeneic donor cells with at least one transgene construct comprising at least one transgene cognate to the target proto-oncogene and a strong promoter to drive the expression of the transgene in the transfected cells, the transgene encoding a gene product which induces host immunoreactivity to host self-determinants of the product of the target proto-oncogene gene,

(b) inserting the cells transfected with the transgene construct into the body of the host to obtain expression of the transgene in the host According to one principal embodiment of the invention, the transgene comprises wild-type or mutant retroviral oncogene DNA According to another principal embodiment of the invention, the transgene comprises wild- type or mutant proto-oncogene DNA of a species different from the host species Where the transgene comprises mutant retroviral oncogene DNA or mutant proto-oncogene DNA, the mutant DNA is preferably nontrans forming. The mutant DNA preferably comprises a deletion mutation in a region of the DNA which is essential for transformation. Preferably, the donor cells are transfected with a plurality, most preferably at least five, different transgene constructs, each construct encoding a different deletion mutation.

In one preferred embodiment of the invention, the mutant DNA has at least about 75 % homology, more preferably at least about 80% homology, most preferably at least about 90% homology, with the corresponding wild-type oncogene or proto-oncogene DNA.

The invention is further directed to a cellular immunogen for immunizing a host against the effects of the product of a target proto-oncogene. the overexpression of which is associated with a cancer. The cellular immunogen comprises the allogeneic donor cells which have been transfected with at least one transgene construct, as described above.

The invention is also directed to a method of preparing the cellular immunogen, by (a) providing allogeneic donor cells, and (b) transfecting the cells with at least one transgene construct, as described above.

The cells transfected with the transgene are preferably rendered non-dividing prior to insertion into the body of the host.

The term "allogeneic" as used herein to describe the genetic relationship between organisms has its ordinary meaning in biology, that is. organisms are allogeneic if they are genetically dissimilar (i.e. , non-identical) members of the same species. Cells comprising an "allogeneic" graft are derived from organisms which are the same species as the host, but are not genetically identical .

The term "corresponds to" is used herein to mean that a polynucleotide sequence is homologous (i.e.. is identical, not strictly evolutionarily related) to all or a portion of a reference polynucleotide sequence. or that a polypeptide sequence is identical to a reference polypeptide sequence. The term "cognate" as used herein refers to a gene sequence that is evolutionarily and functionally related between species For example but not limitation, in the human genome, the human c-myc gene is the cognate gene to the mouse c-myc gene, since the sequences and structures of these two genes indicate that they are highly homologous and both genes encode proteins which are functionally equivalent

By "donor cell" is meant a cell which is used herein for immunization of a host The expression "donor cell" is meant to include not only cells freshly harvested from donor individuals, but also cultured cells which may have been obtained from a cell or tissue bank or other biological repository By "allogeneic donor cell" is meant a cell which is allogeneic with respect to the host, l e . the donor cell, upon transplantation into the host, would compromise an allogeneic graft

By "homology" is meant the degree of sequence similarity between two different amino acid sequences, as that degree of sequence similarity is derived by the FASTA program of Pearson and Lipman, Proc Natl Acad Sci USA (1988). 85 2444-2448, the entire disclosure of which is incorporated herein by reference

As used herein, the term "operably linked" refers to a linkage of polynucleotide elements in a functional relationship A nucleic acid is "operably linked' when it is placed into a functional relationship with another nucleic acid sequence For instance, a promoter or enhancer is operably linked to a coding sequence if it affects the transcription of the coding sequence Operably linked means that the DNA sequences being linked are typically contiguous and, where necessary to join two protein coding regions, contiguous and m reading frame

The word "transfection is meant to have its ordinary meaning that is. the introduction of foreign DNA into eukaryotic cells

By "transgene" is meant a foreign gene that is introduced into one or more donor cells By "transgene construct" is meant DNA containing a transgene and ^• additional regulatory DNA, such as promoter elements, necessary for the expression of the transgene in the donor cells used to innoculate the host.

Description of the Figures Figs. 1A and IB are plots of the mean tumor diameter over time following subcutaneous wing web inoculation of 1-day-old line TK (Fig. 1A) and line SC (Fig. IB) chickens with 100 ug< of tumorigenic plasmids pcsrc527 (—*^•—), pVSRC-Cl (— •— ) or pMvsrc ( — ■ — ). The mean tumor diameter (mm) at a particular time point and for any one group of TK or SC line chickens inoculated was computed as the sum of the diameters of the primary tumors divided by the number of chickens surviving to that point. The ratios at each time point show, for a particular group, the number of chickens bearing palpable tumors to the total number of survivors to that point (standard typeface for pcsrc527, italics for pVSRC-Cl , bold typeface for pMVsrc). Error bars (unless obscured by the symbol) indicate standard error.

Figs. 2A and 2B are plots of the growth of challenge (wing web) tumors in test and control line TK chickens under conditions of (i) priming and homologous challenge with plasmid pαsrc527 (Fig. 2A: — Δ — . test; — * — . control), or (ii) priming and homologous challenge with plasmid pVSRC-Cl (Fig. 2B: — O — , test; — • — , control). Test chickens were primed at 1 day posthatch with 100 μg of construct; test and control chickens were challenged at five weeks posthatch with 200 μg of construct. The mean challenge diameter was computed as in Figs. 1A and IB. At each time point the ratio of chickens bearing palpable challenge tumors to total number of survivors to that point is indicated (standard typeface for control group . bold typeface for test group) . The statistical comparison between the mean challenge tumor diameters of the test versus the control group at a particular time point was made using a two-tailed student's t test, *(p < 0.05). **(p < 0.01). ***(p < 0.001). The statistical comparison between the ratios of chickens bearing palpable challenge tumors to total number of survivors of the test versus the control group at a particular time ^• point was made using a chi-squared test; the paired ratios are underlined for only those time points where p < 0.05. Error bars indicate standard error.

Figs. 3 A and 3B are plots of the growth of challenge (wing web) tumors in TK chickens under conditions of (i) priming with plasmid pVSRC-C 1 and heterologous challenge with plasmid pcsrc527 (Fig. 3 A: — Δ — , test; — * — . control) or (ii) priming with pαsrc527 and heterologous challenge with pVSRC-Cl (Fig. 3B: — O — , test; — • — . control). Test chickens were primed at 1 day posthatch with 100 μg of construct; test and control chickens were challenged at five weeks posthatch with 200 μg of construct. The mean challenge tumor diameter was computed as in Figs. 1A and IB. At each time point the ratio of chickens bearing palpable challenge tumors to total number of survivors to that point is indicated (standard typeface for control group, bold typeface for test group). Statistical comparisons were made between test and control groups at a particular time point as described for Figs. 2 A and 2B. [*(p < 0.05). **(p < 0.01). ***(p < 0.001), for the student's t test] , and the paired ratios are underlined for only those time points where, in the chi-squared test. p < 0.05. Error bars indicate standard error.

Detailed Description of the Invention A vaccination strategy is provided to prevent development of cancers. The vaccination method may be carried out on a subject at risk for a particular cancer, but before the development of the cancer. The practice of the invention may serve for the immunoprevention of prevalent human cancers, such as colon carcinoma, breast carcinoma, and various lymphomas whose progress is accompanied by the overexpression of a cellular proto-oncogene.

The vaccination strategy of the present invention relies on the induction of an immune response that targets tumor cells by virtue of the recognition of the proto-oncogene-specific antigenicity . The aim of the vaccine protocol is to induce reactivity to self-determinants of an overexpressed proto- oncogene product The strategy exploits the structural relatedness between the product of the cellular proto-oncogene and that of the product of genes cognate to the target proto-oncogene The cognate gene may comprise a wild-type or mutant cognate retroviral oncogene or a wild-type or mutant proto-oncogene of a species different from the host species The starting point of the vaccine strategy is the high degree of primary sequence homology that exists between the protein product of a targeted proto-oncogene and that of its cognate retroviral oncogene. or between the proto-oncogene product and the product of a cognate proto-oncogene from a different species However, in contrast to other proposed vaccine strategies, the present invention is not based on the immune recognition of a determinant defined by a cancer specific mutation

For those tumors showing proto-oncogene overexpression, this sequence homology permits application of the following strategy, which can be employed either prophylactically or therapeutically under conditions of cell- surface expression, oi other forms of adjuvamcity , as chosen to enhance immunogenicity (a) transfection of allogeneic donor cells with a DNA construct comprising a transgene cognate to the target proto-oncogene, which transgene encodes a gene product which induces host lmmunoreactivity to host self- determmants of the product of the target proto-oncogene, (b) transplantation of the transfected cells into the body of the host to obtain expression of the transgene in the host, and thus immunity against the pioto-oncogene product The invention relies on the targeting of a self-determinant found on an overexpressed or overabundant proto-oncogene-encoded product The foreign peptide elements of the immunizing oncogene product will trigger peripheral lymphocytes exhibiting a weak cross reactivity for the self peptides of the targeted proto-oncogene product Although such self peptides would be present in normal cells expressing the proto-oncogene targeting of the tumoi cells is favored in view of their overexpression of the proto-oncogene

The immune strategy exploits the antigemcity of two alternative types of determinants (1) tumor-associated antigemc determmant(s) induced as a consequence of the activity of the oncogene product, e.g. , an enzymatic ^• modification of a cellular protein effected by the oncogene product, or (2) tumor associated antigenic determinant(s) intrinsic to the oncogene-encoded product itself. The difficulty in exploiting the first alternative by traditional means, i.e. , antigen purification, is that at present little or no systematic information exists bearing on the properties of an antigen that, though oncogene-induced, is not oncogene-encoded. This situation makes purification of any such antigen problematic. However, this problem is obviated from the outset by the present invention which utilizes alogeneic donor cells which, as transfected in culture by the cognate retroviral oncogene, would express the relevant antigenicity .

In terms of exploiting the second alternative, that of an antigenicity intrinsic to the proto-oncogene product, a relevant consideration is that the protocol of immunization according to the present invention primes the host to determinants of the oncogene product itself. A consequence of this immunization is induction of T-cell reactivity to the divergent, i.e foreign, peptide determinants of the retroviral oncogene product, i.e.. those peptide determinants that show sequence differences with the positionally homologous determinants of the cellular proto-oncogene product. The induction of this reactivity does not in itself have vaccine potential, since the foreign determinants specific to the retroviral oncogene product are normally absent from the cellular proto-oncogene product. Nevertheless, the foreign peptide elements, notably those that differ by only a single amino acid from the positionally homologous self peptides, trigger peripheral T-lymphocytes exhibiting a weak cross-reactivity for the self peptides. Although such self peptides are present in normal cells expressing the proto- oncogene, targeting of the tumor cells is favored in view of their overexpression of the proto-oncogene .

It is possible that many tumor-associated and overexpressed proto- oncogenes might possess mutations. In some cases, overexpression may very well arise as a direct consequence of one or more of the mutations. However, the present vaccination method does not have as its object the deliberate targeting of non-self determinants generated by proto-oncogene mutations. Unlike prior ^• vaccination methods designed to target such mutation-driven non-self determinants, it is the aim of the present invention to induce reactivity for self- determinants in the overexpressed product of tumor associated and overexpressed proto-oncogenes.

Prior efforts attempting to elicit reactivity to proto-oncogene self determinants have relied on in vitro protocols utilizing mutant cell lines to identify individual self peptide immunogens (Disis et al. , Cancer Res. (1994) 54: 1071-1076; Peoples et al , Proc. Natl. Acad. Sci USA (1995), 92:432-436). According to the present invention, the host immune system is presented with the full array of naturally-derived class I binding peptides. The vaccine strategy of the present invention obviates the need for any a priori assessment of the immunogenicity of individual peptides.

While the cellular immunogens of the invention display self peptides. non-self peptides would also be presented which may serve as more effective tolerance breakers. The value of a non-self, but closely related to self, peptide is that it may more readily activate those T cells that have both a weak cross reactivity for the cognate self peptide and an activation threshold (determined by the tightness of binding to the T cell receptor) too high to be triggered by the self peptide. Moreover, cognate non-self is inductive of a good immune response, simply because it does in fact constitute nonself. The non-self immune response is expected to predispose the induction of the inevitably weaker response to the self determinants on the same protein product, since the resultant cytokine release provides local help to initiate the weaker anti-self response. As hereinafter exemplified in a model of 5rc-oncogene -based tumor formation, immunization with cells transfected with a transgene construct expressing the v-src oncogene product induces reactivity to the product of the c- src proto-oncogene, thereby conferring protection against the growth of tumors displaying overexpression of the c-src proto-oncogene. We have described a similar vaccination strategy in ^■ PCT/US97/00582. However, the methodology described therein utilizes excised host cells for preparing the cellular immunogen. According to the present invention, the cellular immunogen is prepared from cells obtained from a donor other than the patient, and other than an identical twin of the patient. Hence, while PCT/US97/00582 describes an inoculation constituting a syngeneic transfer, the present invention relies upon an allogeneic transfer. Given the outbredness of the human population, there will inevitably be allelic differences between the donor and the patient. These differences do not impede the induction of immunity to the cognate proto-oncogene encoded product overexpressed in the allogeneic transfectants used as immunogen.

Until recently, it was believed that antigen presentation was restricted to those class I alleles of the major histocompatibility complex (MHC) that were shared between antigen presenting cells and T cells, a concept that would imply that fibroblasts mismatched at a HLA (human leukocyte antigen) class 1 allele of the human MHC could not present antigen in the context of this allele. It is now known that this view is too narrow. In fact, antigen released from transplanted donor cells or cell debris will be internalized by a recipient^'s bone marrow-derived antigen presenting cells, then processed by these cells for presentation in the context of the class I alleles of the recipient. See Huang el al . Severn* 264:961-964 ( 1994).

This pathway of antigen transfer from the donor cells to the bone marrow-derived antigen presenting cells of the recipient exemplifies an in vivo process designated the "exogenous pathway , to distinguish it from the "classical pathway": the latter as based on processing of antigen-presenting cells themselves.

The operation of the exogenous pathway permits recipient priming to a cellular antigen introduced in the context of a non-recipient class I allele. thereby permitting the use of allogeneic donor cells, mismatched with reference to the recipient at none or at all of the diploid 6 human class I loci, for priming to an overexpressed cognate proto-oncogene. Target Proto-Oncogenes

According to the present invention, patients with a family history of a cancer characterized by the overexpression of a particular proto-oncogene are selected for immunization. Alternatively, patients whose tumors can be shown to overexpress the proto-oncogene are selected. Overexpression of a proto-oncogene may derive from an increase over a basal level of transcription. Overexpression may also derive from gene amplification, that is, an increase in gene copy number, coupled with a basal or elevated level of transcription. Proto- oncogene overexpression may be assayed by conventional probing techniques. such as described in Molecular Cloning: A Laboratory Manual J. Sambrook et al . eds.. Cold Spring Harbor Laboratory Press, 2nd ed. 1989. The level of target proto-oncogene expression may be determined by probing total cellular RNA from patient cells with a complementary probe for the relevant mRNA. Total RNA from the patient cells is fractionated in a glyoxal/agarose gel, trans- ferred to nylon and hybridized to an appropriately labelled nucleic acid probe for the target mRNA. The number of relevant mRNA transcripts found in the patient cells is compared to that found in cells taken from the same tissue of a normal control subject.

As an alternative to measuring mRNA transcripts, the expression level of a target proto-oncogene may be assessed by assaying the amount of encoded protein which is formed. Western blotting is a standard protocol in routine use for the determination of protein levels. See Molecular Cloning, supra. Chapter 18, incorporated herein by reference. Accordingly, a cell lysate or other cell fraction containing protein is electrophoresed on a polyacrylamide gel. followed by protein transfer to nitrocellulose, and probing of the gel with an antibody specific for the protein in question. The probe step permits resolution of the desired protein from all other proteins in the starting mixture. The bound antibody may be prelabeled, e.g. , by a radioisotope such as ¹²⁵I, so as to permit its detection on the gel. Alternatively, a secondary reagent (usually an anti- immunoglobin or protein A) may be radiolabeled or covalently coupled to an enzyme such as horseradish peroxidase or alkaline phosphatase The strength of the signal is proportional to the amount of the target protein The strength of the signal is compared with the signal from a sample analyzed in the same manner but taken from normal as opposed to tumor tissue

A description of the methodology and use of Western blotting to determine the levels of the c-src-encoded protein pp60^{c m} in adenomatous polyps

(colomc epitheha) is provided by Cartwπght et al , Proc Natl Acad Sci USA

(1990), 87 558-562, the entire disclosure of which is incorporated herein by reference

An at least about eight-fold increase in that gene's expression in the patient cells compared to expression in normal control cells from the same tissue would indicate candidacy for vaccination

Table 1 includes a partial list of representative proto-oncogenes. the overexpression of which has been associated with one or more malignancies Each listed proto-oncogene is a target proto-oncogene according to the present invention The corresponding oncogene, of which the target proto-oncogene is the normal cellular homolog, is also identified This list of target proto- oncogenes is intended to be representative, and not a complete list

Table 1

Representative List of Target Proto-Oncogenes

Proto-

Oncogene Tumor Comments/References

AKT-2 ovarian v-Akt is the oncogene of the AKT8 virus, which induces lymphomas in mice

1 Bellacosa et al , (1995) Int J Cancel 64(4) 280-5 Southern-blot analysis has shown AKT-2 amplification m 12 1 % of ovarian carcinomas, while Northern bot analysis has revealed overexpression of AKT-2 in 3 of 25 fresh ovarian carcinomas which were negative for AKT- 2 amplification. 2. Cheng et al. , (1996) Proc. Natl. Acad. Sci.

USA 89(19): 9267-71): Amplification of AKT-2 has been detected in 10% of pancreatic carcinomas.

AKT-2 pancreatic Cheng et al , (1996) Proc. Natl. Acad. Sci. USA 93(8):3636-41 : Amplification of AKT-2 has been detected in 10% of pancreatic carcinomas.

c-crbB-2 bladder c-Erb -2 is also known as ΗER2/neu. V-erbB is the oncogene of the avian erythroblastosis virus.

1. Underwood et al . (1995) Cancer Res. 55(l l):2422-30: Protein overexpression was observed in 45 % of patients with non-recurrent disease and 50% of patients with recurrent disease;

9% of bladder tumors analyzed shoed gene amplification. 2. Coombs et al. , (1993) Pathology 169(l):35-42: c-ErbB-2 gene amplification was observed in 14% of bladder tumors analyzed.

3. Gardiner et al. , (1992) Urolog. Res. 20(2): 17-

20: Nineteen percent of primary transitional cell bladder carcinomas showed c-erbB-2 gene amplification. c-erbB-2 breast 1. Molina et al. , (1966) Anticancer Research ^■

16(4B):2295-300: Abnormal c-erbB-2 levels were found in 9.2% of patients with locoregional breast carcinoma, and in 45.4% of patients with advanced disease. 2. DePotter et al , (1995) Virchows

Arch. 426(2): 107-15: Overexpression of the oncoprotein is observed in about 20% of invasive duct cell carcinomas of the breast. 3. Bandyopadhyay et al , (1994) Acta Oncol. 33(5):493-8: 35.4% of breast tumors showed c- erbB-2 overexpression; 17.4% showed gene amplification. 4. Fontana et al , (1994) Anticancer Res. 14(5B):2099-104: 26% of samples showed c-erbB-2 amplification. 5. Press et al. , (1993) Cancer Research 53(20):4960-70:

Amplified overexpression was identified in 38% of primary breast cancers. 6. Berns et al , (1992) Cancer Res. 52(5): 1107-13 : 23 % of primary breast cancer tissues exhibited amplification. 7. Delvenne et al. . (1992) Eur. J. of Cancer 28(2-

3):700-5: c-erbB-2 mRNA was overexpressed in 34% of breast tumor samples. 8. Inglehart, (1990) Cancer Res. 50(20):6701-7: Two to thirty- two-fold gene amplification was found in multiple stages of tumor progression. 9. Slamon et al. ,

(1989) Science 244:707-12: A 28% incidence of amplification of c-erbB-2 was found in 189 primary breast cancers. 10. Kraus et al. , (1987) EMBO J. 6(3):605-10: Eight cell lines demonstrated c-erbB-2 mRNA levels ranging from 4 to 128-fold overexpression 60% of all tumors analyzed showed elevated levels of c-erbB-2 mRNA

c-erbB-2 lung 1 Osaki et al , (1995) Chest 108(1) 157-62 Lung tissue overexpression of c-crbB-2 was discovered in 42 5 % of samples 2 Lorenz et al , (1994) Clin Invest 72(2) 156-63 A 64-fold increase in the amount of c-erbB-2 mRNA was observed, 33 % of lung tumors showed overexpression of c-erbB-2

c-erbB-2 ovarian 1 Katsaros et al , (1995) Anticancer Res

15(4) 1501-10 Abnormally high expression of c- erbB-2 was found in 31 % of tumor samples 2 Fehp et al , (1995) Cancer 75(8) 2147-52 21 7% of ovarian tumors showed overexpression of c- erbB-2 3 Fan et al , (1994) Chin Med J 107(8) 589-93 c-erbB-2 amplification was found in 30 8% (8 of 26) of human ovarian cancers 4 vanDam et al , (1994) J of Clin Path 47(10) 914-9 24% of ovarian tumors showed c erbB-2 overexpression 5 Csokay et al , (1993) Eur J of Surg Oncology 19(6) 593-9 c-erbB-2 amplification w as found in 34 % of fresh ovarian tumor samples 6 McKenzie et al , (1993) Cancer 71(12) 3942-5 30% of ovarian tumoi samples indicated c-erbB-2 overexpression 7 Hung et al , (1992) Cancer Letters 61(2) 95-103 A 100-fold c-erbB-2 overexpression was discovered in one human cell line Two to four fold amplification was also discovered

MDM-2 leukemia MDM-2 is the muπne double mιnute-2 oncogene

1 Bueso-Ramos et al , (1993) Blood 82(9) 2617- 23 53 % of cases showed overexpression of

MDM-2 mRNA The level of MDM-2 mRNA overexpression in some cases of leukemias was comparable to that observed in some sarcomas which demonstrate more than 50-fold MDM-2 gene amplification o evidence of gene amplification was observed 2 Watanabe et al , (1994) Blood 84(9) 3158-65 28% of patients with B-cell chronic lymphocytic leukemia or non-Hodgkin's lymphoma had 10-fold higher levels of MDM-2 gene expression MDM-2 overexpression was found more frequently in patients at advanced clinical stages

c-mλ b colon V- vb is the oncogene of the avian myeloblastoma virus 1 Ramsay et al , (1992) Cell Growth and Diff 3(10) 723-30 c-myb levels were alway s higher in colon cancer samples than normal tissue

2 Ahtalo et al (1984) Proc Natl Acad Sci 81(14) 4534-8 c-myb levels were always highei m colon cancer samples than normal tissue

c-m\ c breast V-myc is the oncogene of the avian myelocytoma virus 1 Lonn et al , (1995) Cancer 75(11) 2681- 7 Amplification of c-myb occurs in 16% of patients with breast cancer. 2. Hehir et al , ^■ (1993) J of Surg. Oncology 54(4): 207-9: c-m c overexpression was found in 60% of breast carcinoma samples. 3. Kreipe et al , (1993) Cancer Research 53(8): 1956-61 : Amplification of c-myc was found in 52.6% of samples that displayed a Ki-Sl labelling index exceeding 30% . 4. Watson et al , (1993) J. Nat. Cancer Inst. 85(l l):902-7: Amplification of c-myc occurs in up to 20 - 30% of breast cancers. 5. Berns et al .

(1992) Cancer Research 52(5): 1107-13 : Amplification was found in 20 % of primary breast cancer patients; the range was 3-14 gene copies. 6. Watanabe et al , (1992) Cancer Research 52(19) :5178-82: Expression of c-myc was increased by 10-fold.

c-myc gastric/ 1. Rigas, (1990) Clin. Gastroent. 12(5):494-9: colorectal Overexpression of c-myc is found in 80 of colon cancers. 2. Erisman et al , (1988) Oncogene 2(4):367-78: Adenocarcinoma cell lines express 5-

10-fold elevated levels of c-myc mRNA. Eight to thirty-seven- fold higher levels of c-myc protein was found in tumor cell lines compared to normal cells. 3. Sikora et al . (1987) Cancer 59(7): 1289-95: Up to 32-fold overexpression of c-myc mRNA was observed in 12 to 15 tumors. 4. Tsuboi et al . (1987) Biochem. and Biophys. Res. Comm. 146(2):705-10: Gastric Cancer: A 2-3-fold overexpression was observed in gastric cancer. A 2- 10-fold overexpression was observed in colorectal cancer.

c-myc lung 1. Lorenz et al. , (1994) Clin. Invest. 72(2): 156- 63 : A 57-fold increase in c-myc mRNA levels was observed. 23 % of samples indicated strong expression of c-myc. 2. Kato et al. , (1993) Jap. J. of Cancer Res. 84(4): 355-9: Liver tissue metastases from human small cell lung carcinoma revealed 30-fold amplification of c-myc.

c-myc naso- Porter et al , (1994) Ada Oto-Laryng. 114(1): pharn- 1105-9: 22% of samples showed intense staining geal for c-myc.

c-myc ovarian 1. Bian et al , (1995) Chin. J. of Ob. Gyn.

30(7): 406-9: 50% of samples showed amplification of c-myc. 2. Katsaros et al. , (1995)

Anticancer Res. 15(4): 1501-10: 26% of samples exhibited c-mvc amplification. 3. van Dam et al. . (1994) J. Clin. Path. 47( 10) : 914-9 : Overexpression of c-myc was found in 35 % of ovarian carcinomas. 4. Xin et al , (1993) Chin.

J. of Ob. Gyn. 28(7):405-7: 54.5 % of samples showed amplification of c-myc. 5. Tashiro et al. . (1992) Int. J. of Cancer 50(5): 828-33: Overexpression was found in 63.5 % of all serous adenocarcinoma tissues and 37.3% of all ovarian carcinoma tissues. Significant overexpression of c- myc was observed at Stage III compared with other stages.

c-myc prostate Nag et al. , (1989) Prostate 15(2): 115-22: A 10- fold amplification of c-myc was observed. Fifty- fold higher levels of mRNA transcripts of c-myc were found.

c-ras lung Ras oncogenes were first recognized as the transforming genes of Harvey and Kirsten murine sarcoma viruses. Lorenz et al , (1994) Clin. Invest. 72(2): 156-63: a 13-fold increase in overexpression of c-Ki-ras was observed . 18 % of tumors displayed strong overexpression of c-Ki- ras.

c-ras ovarian 1. Katsaros et al . (1995) Anticancer Res.

15(4): 1501-10: Higher levels of ras protein than in normal or benign ovarian tumors were found in 45 % of tumor samples. 2. vanDam et al. , (1994) J. of Clin. Path. 47(10):914-9: 20% of ovarian tumors exhibited c-ras overexpression. The levels of expression of c-ras were much higher in tumors of patients with recurrent or persistent disease after chemotherapy, than in the tumors of patients at initial presentation.

c-src breast V-src is the oncogene of the Rous sarcoma virus. which induces sarcomas in chickens . Muthuswamy et al , (1994) Mol. and Cell. Biol 14(1) 735-43 c-erbB-2-mduced mammary tumors possessed 6-8-fold higher c-src kinase activity than adjacent epithelium

C-Si C colon/ 1 Cartwπght et al , (1994) J of Clin Invest colorectal 93(2) 509-15 c-src activity is 6-10-fold higher in mildly dysplastic ulcerative colitis (a chromic inflammatory disease of the colon with a high on incidence of colon cancer) than in non-dysplastic epitheha This data suggests that activation of c src is an early event in the genesis of UC colon cancel 2 Talamonti et al , (1993) J of Clin Invest 91(1) 53-60 High level of c-src activity from colorectal cancer is found in liver metastases 3 Termuhlen et al , (1993) J of Surg Res 54(4) 293-8 Colon carcinoma metastases to the liver had significantly increased activity of c-sn with an average 2 2-fold increase Extrahepatic colorectal metastases demonstrated an aveiage 12 7-fold increase in c-s/c activity over normal mucosa

c-λes colon V-ves is the oncogene of two avian sarcoma viruses. Esh sarcoma virus and Y73 1 Pena et al , (1995) Gastroent 108(1) 117-24 Twelve to fourteen-fold higher expression of c-ves was found m colomc transforming oncogene adenomas compared to normal mucosa Activity of c-yes was elevated in adenomas that are at greatest risk foi developing cancer 2 Park et al . (1993) Oncogene 8(10):2627-35: A ten to 20-fold higher ■ than normal activity of c-yes was observed in 3 out of 5 colon carcinoma cell lines. A 5-fold higher than normal activity was found in 10 out of 21 primary colon cancers, compared to normal colonic cells.

Selection of Cognate Transgene for Preparation of Cellular Immunogen

According to the present invention, a transgene construct is engineered comprising a transgene which is cognate to the target proto-oncogene (hereinafter "cognate transgene" or "CTG"). The transgene is selected such that it encodes a gene product which induces host immunoreactivity to host self- determinants of the product of the target proto-oncogene. The transgene should be expressed to very high levels in the transfectants. Thus, the construct should contain a strong promoter. The product encoded by the cognate gene must have a high degree of sequence homology with the product of the target proto-oncogene. but also must display some amino acid differences with the target proto-oncogene product. Thus, there must be a subset of one or more amino acid differences between the target proto-oncogene and its cognate in order to provide immunogenic stimulus . Two classes of genes that satisfy these criteria are retroviral oncogenes and xenogenic proto-oncogenes. The word "xenogenic" is intended to have its normal biological meaning, that is. a property or characteristic referring or relating to a different species. Thus, a xenogenic proto-oncogene is meant to include the a homologous proto-oncogene of a species other than the host organism species. It may be appreciated that in the case of a target proto- oncogene. e.g. MDM2, for which no retroviral homolog is yet known, a xenogenic homologue is advantageously utilized as the source of the DNA for the cognate transgene. In principle, a more effective immunogenic stimulus would depend ■ on the particular sequence, and not on the distinction between a retroviral oncogene and a xenogenic proto-oncogene in terms of their relative transforming capacity. Thus, in certain cases, a retroviral oncogene may be better at providing a tolerance-breaking immunogenic stimulus, and in other cases, a xenogenic proto-oncogene may be more effective.

The retroviral oncogene or xenogenic proto-oncogene DNA forming the CTG may comprise the wild type oncogene or proto-oncogene DNA. More preferably, a mutant DNA is utilized, which is engineered so as to be non- transforming in the host. The DNA is mutated to include one or more nucleotide insertions, deletions or substitutions which will encode an oncogene product which is nontransforming in the host, but retains the requisite degree of sequence homology with respect to the target proto-oncogene. A cognate transgene deletion mutant (hereinafter "dCTG") is preferred. A protein sequence is generally considered " cognate " with respect to the target proto-oncogene-encoded protein if it is evolutionarily and functionally related between species. A more precise view of cognation is based upon the following sequence comparison carried out utilizing the FASTA program of Pearson and Lipman, Proc. Natl. Acad. Sci. USA (1988), 85:2444- 2448. the entire disclosure of which is incorporated herein by reference. Cognation is attained upon satisfying two criteria imposed by FASTA; (i) alignment of segments corresponding to at least 75 % of the target proto- oncogene 's encoded amino acid sequence; (ii) at least 80% amino acid identity within the aligned sequences. The segments of the target proto-oncogene protein sequence and protein test sequence satisfying the two criteria are referred to as "homology regions" . Accordingly, at least 75 % of the target proto-oncogene protein sequence is alignable with the test sequence. The alignable segments or homology regions may. however, represent less than 75 % of the total test polypeptide chain for the case of test sequences that may significantly exceed the target proto-oncogene protein in length. One skilled in the art, armed with the FASTA program, may ■ survey existing sequence data bases (either protein sequences or DNA sequences, insofar as the amino acid sequence is determined by FASTA for all reading frames) for test sequences which are cognate with respect to the target proto- oncogene. At the same time, one can isolate and then sequence what are very likely to be cognate test sequences (e.g. feline MDM-2, as likely to be cognate to human MDM-2) and use FASTA to verify the presumed cognation, according to the criteria set forth above. One may obtain the sequences of presumptive cognate proto-oncogenes from a large number of mammalian sequences and screen these sequences with FASTA according to the aforesaid formulation of cognation.

Because the product encoded by a CTG differs at a small number of amino acid positions from the product encoded by the target proto-oncogene. an immunogenic stimulus is provided that (i) is directed against the foreign protein and (ii) with a lower probability, induce an anti-self response. The CTG is selected such that the gene product will yield the greatest immunogenic stimulus to induce anti-self reactivity. Provided that overall sequence homology (preferably greater than about 75 %) is maintained, the presence of scattered amino acid differences is desired, since any one residue would likely have a relatively low probability of inducing self- reactivity. Moreover, the greatest number of residue differences would be advantageous, consistent with maintaining the requisite degree of general sequence homology.

The selection of amino acid modifications for the CTG may be facilitated by resort to available computer-based models used to identify immunogenic peptide fragments of polypeptides. These models could be employed to select CTGs which would possess the maximum number of immunogenic peptides for a given HLA haplotype. Screening Procedure for CTG Selection

Notwithstanding the availability of computer-based algorithms which have some predictive value, it is desirable to design CTGs with resort to a screening procedure based on an actual experimental assay that can be HLA- haplotype specific Accordingly, cells are biopsied from a normal volunteer of particular haplotype The cells are transfected with a CTG construct, preferably a dCTG construct, satisfying the criteria set for cognition More preferably, the cells are transfected with multiple dCTGs preferably at least five dCTGs, satisfying the criteria for cognition The at least five dCTGs are selected to display ammo acid differences that essentially extend throughout the polypeptide chains of the encoded sequences The transfected cells are then used to immunize the volunteer in accordance with the immunization method of the present invention After immunization, the human subject is tested m a standard delayed hypersensitivity (DH) reaction with 10⁴-10⁶ irradiated, autologous fibroblasts as transfected with the same dCTG (or series of dCTGs) as used for the immunizing preparation A positive DH reaction (induration) would verify the induction of reactivity The induction of reactivity in this assay is readily demonstrable because of the priming to the non-self determinants on the dCTG-encoded protein and the readout in the DH reaction of the same nonself determinants Once DH reactivity is demonstrated in a DH reaction that diiectly tests the antigenicity of the non-self determinants encoded by the dCTG (/ e , priming with a non-self construct, DH testing with the same non-self construct), the subject can be then tested in a DH reaction based on testing with the autologous cells transfected with a dCTG derived from the human proto-oncogene itself (i e , priming with a non- self construct, testing with the human self construct) Testing of a battery of human volunteers will lead to a catalogue of HLA-matched dCTGs. such that, foi individuals of the same HLA haplotype. the use of the particular dCTG would be inductive of reactivity to proto-oncogene-encoded self Different CTGs may thus be tested so as to correlate maximal secondary stimulation with a particular HLA haplotype At the same time, this procedure may be used with patients . undergoing tumor resection (if post-operative immuno-suppressive protocols are not mandatory), such that prior to resection, a course of immunization would have been initiated, the endpoint of which would represent the development of a DH reaction.

Any given amino acid difference between the CTG-encoded product and the proto-oncogene-encoded product has a low probability of being a "tolerance-breaker" . Thus, it is preferable to transfect the host cells with a mixture of multiple different CTGs, preferably dCTGs. The number of different dCTGs is preferably five or more. Moreover, it is preferred that, among themselves, the multiple dCTGs show amino acid differences that essentially extend throughout the polypeptide chains of the encoded sequences. The dCTGs would be selected to maximize amino acid differences and, at the same time, make sure that differences are found all along the polypeptide chain. It would thus not be preferable to select a battery of deletions all from within the same domain of the polypeptide chain.

According to a protocol which utilizes 10⁷ irradiated cells for immunization containing five separate dCTGs. five groups of 2 X 10° cells are included in one inoculate, each group of 2 X 10⁶ having been transfected with a separate dCTG from the total set of five CTGs that are cognate to a particular proto-oncogene.

Selection of Non-Transforming Cognate Transgenes

Non-transforming cognate transgene variants are most advantageously derived via deletion of a sequence essential for transformation. Unlike point mutations which are potentially reversible due to back mutations, deletion mutations are irreversible. Furthermore, deletion mutations do not possess the inherent disadvantage attaching to point mutations, namely, even though the requirement for generation of an acceptable cognate transgene is for a qualitative difference with the wild type, i.e. , non-transforming versus transforming, any given point mutation may be neutral or else quantitative in its effect, that is. the mutation may reduce but not totally eliminate transformability Thus, according to a preferred embodiment of the invention, a deletion is created in a region of the cognate transgene which encodes an amino acid sequence required for transformation Consonant with non-transformability, the smallest deletion possible so as to leave intact the bulk of the antigenicity of the transgene product is selected

The engineering of a cognate transgene deletion mutant that satisfies these criteria is facilitated by reports of structure-function relationship in oncogene-encoded proteins Such reports serve to identify regions of oncoprotems that are essential for transformation, as opposed to regions which aie either neutral or serve merely to modulate transformability Although such reports are usually based on in vitro transformation assays, and are therefore independent of immune effects, these studies can be exploited to aid in the construction of non-transforming dCTGs for use in the practice of the present invention

The deletion mutant is engineered to include at least a part of the region identified as critical for transformation In those cases where essential amino acids have been identified, the deletion will span these residues The engineering of any desired deletion can be readily accomplished by polymerase chain reaction (PCR) according to conventional PCR techniques, based upon the known nucleotide sequence of the unmutated cognate transgene

The following describes a representative protocol for deriving a non-transforming dCTG of the smallest possible deletion, for use m the practice of the present invention A test dCTG. engineered on the basis of known oi ascertained transformation-specific domains, and driven by the strongest possible promoter, is used to transfect muπne 3T3 cells A sister culture of 3T3 cells is also transfected. with non-deleted CTG Each CTG or dCTG cell culture is inoculated into nude mice in the absence of any treatment to render the cells non-dividing Those dCTGs which do not \ leld tumors in the mice even after prolonged observation are then utilized as transgenes for the biopsied human cells which, upon transfection with the transgene. will serve as a cellular vaccine according to the practice of the present invention The dCTGs are selected with the smallest deletion mutant consonant with non-transformability Some CTGs representing xenogenic proto-oncogenes may not be tumoπgenic in the 3T3/nude mouse assay For any such non-transforming CTG. it is not essential to generate a dCTG Howevei , even given non-tumoπgenicity in nude mice, it may be desirable to opt for generation of a deletion mutant when the transgene is based upon a xenogenic proto-oncogene In such cases, the deletion would be engineered so as to remove the homologous region to that deleted in the particular dCTG that corresponds to the deletion in the corresponding retroviral oncogene dCTG

Even though the transgene construct may comprise mutant oncogene or proto-oncogene DNA which is nontransforming, it is nevertheless preferable, as a safety measure, to treat the transfected donor cells to render them non-dividing before inoculation into the host The cells are irradiated with a radiation dosage sufficient to render them non-dividmg

Oncogemcity Assay of Cognate Transgenes

As a further safety measure, the oncogemcity of a given dCTG is preferably thoroughly tested prior to infection of the donor cells which are used as cellular immunogens according to the practice of the present invention Foi example, an oncogemcity testing regimen may take the form of three separate assays (I) dCTG transfection of NIH 3T3 cells, followed by inoculation into nude mice, (n) dCTG transfection of human fibroblasts, followed by inoculation into nude mice, and (in) dCTG transfection of human fibroblasts, followed by an in vitro test of anchorage-dependent growth In principle, all three should be negative to \ ahdate the use of any given dCTG in the vaccination method of the present m\ ention According to the oncogemcity assay (1), after stable transfection of NIH 3T3 cells with the test dCTG, the transfectants are inoculated into nude mice Tumoπgenicity of the transfectants in the mice is then evaluated according to standard protocols According to oncogemcity assay (n), human fibroblasts are transfected with the test dCTG as proposed in the above human immunization protocol After stable dCTG transfection of human fibroblasts, however, rather than carrying out X-irradiation of the transfectants to render them non-dividmg followed by inoculation of the irradiated transfectants into the human host, the transfectants are directly inoculated into nude mice as a direct test of tumoπgenicity Given the gi eater susceptibility of muπne 3T3 cells to oncogemc transformation, vis a vis primary human or muπne transfectants fibroblasts, assay (n) is probably much less sensitive than assay (I), but does have the advantage of offering a direct test of dCTG oncogemcity in human cells According to oncogemcity assay (in), non-irradiated dCTG- transfected human fibroblasts are assayed for anchorage-dependent growth. / e colony formation in soft agar, as a test of dCTG transforming potential in human cells Anchorage independence, as defined by the ability of cells to grow when suspended in semisohd medium, is a common phenotype acquired by human tumor cells particularly those tumor cells of mesenchymal ongm, such as fibrosarcomas While assay (in) has no ιn

o readout, it offers an independent test of the critical issue of dCTG oncogemcity m human cells

The oncogemcity assays are performed according to published protocols Assay (I). comprising dCTG transfection of NIH 3T3 cells followed by inoculation into nude mice, may be performed according to the protocol of Stevens etal Proc Natl Acad Sci USA (1988), 85 3875-3879, including DNA transfection by the calcium phosphate coprecipitation method of Manohaven et al Carcinogenesis (1985). 6 1295-1301 Accordingly, NIH 3T3 cells (7 5 λ 10" cells per 100-mm dish) are exposed to a calcium phosphate-DNA coprecipitate (40 μg of genomic DNA plus 3 g of pSV2neo per dish) foi 4 hours. Two days later, each dish is trypsinized and reseeded into a 175-cπr ^■ flask. For the next 10 days, cultures are selected in G418 (400 μg/ml), and the flasks are then trypsinized and cells are replated in the same flask to disperse the G418-resistant colonies into a diffuse lawn of cells. Two days later, the cells are harvested and washed with serum-free medium prior to injection. One injection of 5 X 10⁶ cells into the right flank and one injection of 1 X 10⁷ cells into the left flank, each in a volume of 200 μl, are done on each nude mouse. Injection sites are monitored at 3- or 4-day intervals for 100 days. The sites are scored for the number of tumors induced per injection site. Oncogenicity assay (ii). whereby dCTG transfection of human fibroblasts followed by inoculation into nude mice, is carried out in the same manner as assay (i) except that for assay (ii) the human fibroblast transfectants are substituted for the murine 3T3 transfectants.

Assay (iii), involves a test of the in vitro anchorage-dependent growth of dCTG-transfected human fibroblasts. The assay is carried out as described in Stevens et al. , J. CancerRes. and Clin. Oncol. 1989, 115: 118-128. 1 x 10^s cells are seeded per 60-mm dish into 0.33 % Noble agar over a 6-ml 0.5 % agar base layer in Hams F10 supplemented with 6% fetal bovine serum. A portion of the agar suspension is diluted with Hams F10 plus 6% fetal calf serum to 200 cells '5 ml to determine the cloning efficiency of these cells when seeded into plastic 60-mm dishes. Agar dishes are fed with 1 ml Hams F10 supplemented with 6% fetal bovine serum on the 1st and 15th day after seeding. Four weeks after seeding, all agar colonies > 75 μm in diameter are counted and the colony counts are normalized to the plating efficiencies which aliquots of the initially seeded cells showed on plastic. This comparison, or normalization, of the agar colony counts to the plastic dish colony counts is useful in identifying and correcting for any mechanical artifacts which might result from the seeding into agar of dead cells that had persisted from the initial transfection treatment or from heat-induced cell death, which might have occurred while suspending cells in molten agar during the process of seeding the agar dishes. The following is a partial list of various deletions which, based upon published accounts of experiments with human or animal cells, are believed to render the identified CTG non-tumorigenic.

Table 2 Deletion Mutations Rendering Indicated Gene Non-Transforming

Engineering of Vectors for Donor Cell Transfection

The engineering of vectors for expression of a particular CTG. preferably a dCTG. is based on standard methods of recombinant DNA technology, i.e. insertion of the dCTG via the polylinker of standard or commercially available expression vectors. The dCTG is operably linked to a strong promoter. Generally speaking, a "strong" promoter is a promoter which achieves constitutively high expression of the dCTG in the transfected cells. Each promoter should include all of the signals necessary for initiating transcription of the relevant downstream sequence. These conditions are fulfilled, for example, by the pBK-CMV expression vector available from Stratagene Cloning Systems. La Jolla. CA (catalog no. 212209). The pBK-CMλ⁷ vector contains the cytomegalovirus (CMV) immediate early promoter. dCTGs xenogenic with respect to a particular target proto-oncogene may be isolated by conventional nucleic acid probing techniques, given the availability of a highly homologous probe represented by the cognate retroviral oncogene and/or the human proto-oncogene itself.

Donor Cells for Transfection

The allogeneic donor cells which may be transfected to derive the cellular immunogens of the present invention must express class I MHC and be susceptible to isolation and culture. Fibroblasts express class I MHC and may be cultured. Other preferred donor cells are bone marrow-derived antigen- • presenting cells such as macrophages, follicular dendritic cells, and Langerhans cells, for example. The cells may comprise primary explants or established cell lines. Primary skin fibroblasts may be obtained as follows. Punch biopsies can be performed by a competent physician as a standard clinical procedure. Each biopsy yields a starting population of 1-2 X 10⁷ cells that would proliferate in culture. Methods for the preparation of tissue cultures of human fibroblasts are well developed and widely used. See, Cristofalo and Carpenter, J. Tissue Culture Methods (1980), 6: 117-121. the entire disclosure of which is incorporated herein by reference. Essentially, skin obtained by punch biopsy is washed using an appropriate wash medium, finely minced and cultured in a suitable culture medium, such as Dulbecco's Modified Eagle Medium (DMEM), under C0₂ at 37 °C. The cells are trypsinized with a trypsin solution and transferred to a larger vessel and incubated at 37° C in culture fluid. Donor Cell Transfection

The expression vector carrying the dCTG is used to transfect allogeneic donor cells according to conventional transfection methods. One method of transfection involves the addition of DEAE-dextran to increase the uptake of the naked DNA molecules by a recipient cell. See McCutchin and Pagano, J. Natl. Cancer Inst. (1968) 41 :351-7. Another method of transfection is the calcium phosphate precipitation technique which depends upon the addition of Ca" to a phosphate-containing DNA solution. The resulting precipitate apparently includes DNA in association with calcium phosphate crystals. These crystals settle onto a cell monolayer; the resulting apposition of crystals and cell surface appears to lead to uptake of the DNA. A small proportion of the DNA taken up becomes expressed in a transfectant. as well as in its clonal descendants. See Graham et al , Virology (1973). 52:456-467 and Virology (1974), 54:536- Preferably, transfection is carried out by cationic phospholipid- ^• mediated delivery. In particular, polycationic liposomes can be formed from N- [l-(2,3-dioleyloxy)propyl]-N,N,N-trimethylammonium chloride (DOTMA) or related liposome-forming materials. See Feigner et al. , Proc. Natl. Acad. Sci. USA (1987) 84:7413-7417 (DNA-transfection); Malone et al. , Proc. Natl. Acad. Sci. USA (1989), 86:6077-6081) (RNA-transfection). One preferred technique utilizes the LipofectAMINE™ Reagent (Cat. No. 18324-012, Life Technologies, Inc. , Gaithersburg, MD) which is a 3: 1 (w/w) liposome formulation of the polycationic lipid 2, 3-dioleyloxy-N-[2(sperminecarboxamido)ethyl-N,N-dimethyl- 1 -propanaminium trifluoroacetate (DOSP A) (Chemical Abstracts Registry name : N-[2-({2.5-bis[(3-aminopropyl)amino]-l-oxypentyl}amino)ethyl]-N,N-dimethyl- 2,3-bis(9-octadecenyloxy)-l-propanaminium trifluoroacetate), and the neutral lipid dioleoyl phosphatidylethanolamine (DOPE) in membrane filtered water. Transfection utilizing the LipofectAMINE™ Reagent is carried out according to the manufacturer's published protocol. The protocol (for Cat. No. 18324-012) provides for either transient or stable transfection, as desired.

The advantage of transient expression is its rapidity, i.e. there is no requirement for cellular proliferation to select for stable integration events . This rapidity could conceivably be of major clinical importance, in cases of an already metastatic tumor burden, wherein the weeks required for selection of stable transfectants may simply not be available to the clinician.

There are, nonetheless, two general disadvantages to the use of transient transfection. The first is that expression usually peters out after a few days, in contrast to the continual expression in the case of stable transfection. This is not particularly crippling in terms of our immunization protocol. The inoculated, irradiated cells used for immunization would likely not survive in vivo for more than 4 or 5 days, in any case. (For the same reason, immunological rejection of the allogeneic donor cells by the host is not a concern). Thus the nominal advantage accruing to stable transfection. that of a long-duration expression by the progeny of the parental inoculated cell, is not of particular relevance in the case of the immunizing regime described herein, which is based ^• on the use of non-dividing, probably short-lived cells.

A second disadvantage of transient transfection resides in the fact that it yields a cell population, only a subset of which has actually been transfected and thus expresses the protein encoded by the transgene. This problem is obviated in the case of stable transfection, wherein over time one can develop a pure population of transfectants via selection for a resistance marker, such as neo, under conditions of clonal proliferation of the initial stable transfectants. i. e. daughter cells of transiently transfected cells lack the transgene. in contrast to the case with stable transfectants. In the situation where there is sufficient time to effect immunization based on stably transfected cells, the progeny of all transfected clones would be utilized, not just the progeny of a single clone, as is sometimes done for detailed biochemical and molecular analyses of gene expression. Clearly the more clones utilized, the more quickly one can arrive at the requisite number of cells to be used for immunization.

Percentage of Cells Exhibiting dCTG Expression

The percentage of cells exhibiting dCTG expression may be determined by an immunohistology assay. In this procedure, a small number of cells (-500) from the harvested pellet following centrifugation of transfected cells are deposited on a cover slip and fixed with cold acetone. At this point, a standard immunohistological assay is carried out with the cells on the cover slip. i.e. addition of a primary monoclonal antibody reactive to the dCTG-encoded protein, followed by the addition of a developing antibody, e.g. a fluorescent tagged antibody reactive to the primary monoclonal antibody. Measurement of the percentage of cells scoring as dCTG-positive in the fluorescent assay allows a determination of the number of positive transfectants in the starting culture, and thus the number of total cells to be used for immunization to arrive at the desired number of dCTG-positive cells to be inoculated in the patient. If, as would be almost certain, the percentage of cells scoring as ^• dCTG-positive is less than one hundred percent, one can simply increase the number of cells to be used for immunization, so as to include the desired number of transfectants. The non-transfected cells in the immunizing population would simply represent x-irradiated, autologous fibroblasts that would constitute no danger to the patient.

Transfectant Irradiation

Prior to innoculation into the host, the transfected cells are preferably irradiated. The transfectants are irradiated with a radiation dose sufficient to render them non-dividing, such as a dose of 25 By or 2500R. The cells are then counted by trypan blue exclusion, and about 2 X 10⁷ irradiated transfectants are resuspended in a volume of 0.2-0.4 ml of Hanks Balanced Salt Solution.

Vaccination Procedure The transfected cells are inserted into the host to achieve vaccination.

It is the object of the present invention to generate a systemic tumor immune response, so as to fight metastasis formation wherever any metastases are found. Intramuscular, intraperitoneal or subcutaneous inoculation will suffice to yield a systemic response. Thus, patients are vaccinated with the transfected cells accordingly.

For s-crc overexpression associated with colon carcinoma, partial venous inoculation is preferred, as the liver is a frequent site of metastases. For vaccinating against breast cancers and lymphomas. systemic immunization is preferred.

As a general rule, it is desirable to generate the strongest immune response consistent with clinical monitoring of no adverse side effects, i.e. multiple rounds of inoculation with, for example 10⁷ cells, at each round. The number of rounds of inoculation is selected accordingly . The efficacy of the ^• inoculation schedule may be monitored by a delayed hypersensitivity reaction administered to the patient. A course of about up to 10 inoculations, at 2-3 week intervals, may be utilized. It may be appreciated that the inoculation schedule - may be modified in view of the immunologic response of the individual patient, as determined with resort to the delayed-type hypersensitivity (DTH) reaction.

Patient Response Monitoring by Delayed-type Hypersensitivity Reaction

Patients are assessed for reactivity to the irradiated transfectants by a test of skin reactivity in a DTH reaction. DTH has been used clinically (Chang et al. (1993). Cancer Research 53: 1043-1050). To measure reactivity to the autologous irradiated transfectants, 10⁴ - 10⁶ cells in a volume of 0.1 ml Hanks buffered saline solution (HBSS) are inoculated intradermally into the host. Induration is measured 48 hours later, as an average of two perpendicular diameters (responses of greater than >2 mm is considered positive). One advantage to the DTH assay is that it can independently assess the induction of T cell reactivity to (i) the transfectants used for immunization (i.e. the set of 5 or more dCTGs chosen for immunization purposes, each containing non-self determinants) and (ii) transfectants, as transfected with the human dCTG itself containing only self determinants. Thus, the induction of reactivity to the transfectants used for immunization establishes that the immunizing transfectants are in fact immunogenic, that is, the patient has not exhibiting a much weakened capacity for immune response. If the patient is demonstrably capable of response to the immunizing transfectants, then skin testing with the dCTG (human) transfectants would establish whether or not reactivity to the human proto-oncogene encoded product had been induced. According to the practice of the invention, inoculation of the immunizing transfectants would continue for at least as long as the induction of reactivity to the human proto-oncogene-encoded protein occurs . The practice of the invention is illustrated by the following ^■ nonlimiting examples.

Example 1 Immunization of Chickens Against c-src(527)-Induced Tumors By Vaccination with \-src DNA

A. Genes

The oncogene c-src(527) is an activated form of chicken c-src. Its protein product pp60^{c srL(5:7)} differs from the protein product of c-src, pp60^{c src}, by only a single amino acid substitution, phenylalanine for tyrosine at residue 527 (Kmiecik and Shalloway , ( 1987) Cell 49 , 65-73) . This substitution eliminates the negative regulatory influence exerted on pp60^{c src} phosphokinase activity by the enzymatic phosphorylation of the position 527 tyrosine. The protein product of v-src, pp60'^"src, shows a number of sequence differences with pp60^{c src} (Takeya and Hanafusa, (1983) Cell 32, 881-890), including scattered single amino acid substitutions within the first 514 residues and a novel C terminus of 12 amino acids (residues 515-526). in place of the nineteen C terminal amino acids of pp60^{c src} (residues 515-533). Both the v-src-positive plasmid, pMvsrc, and the c-src(527)-positive plasmid, pcsrc527. were originally shown (Kmiecik and Shalloway. (1987) Cell 49. 65-73) to transform murine NIH 3T3 cells in culture However, the v-src-induced transformants exhibited a more rapid or more extensive colony growth in soft agarose than the c-src(527)-induced transformants. as well as a usually shorter latency of tumor formation in nude mice (id.). B. Plasmids

1. pySRC-Cl

The pVSRC-Cl plasmid was prepared as described by Halpern et al. , (1991) Virology 180, 857-86. Essentially, the plasmid was derived from the pRU -src plasmid (Halpern et al. , (1990) Virology 175, 328-331) by subcloning the v-src( + ) Xhol-EcoRl fragment of the latter into the multiple cloning sequence of pSP65 (Melton et al. , (1984) Nucleic Acids Res. 12, 7035-7056) which had been cleaved with Sail and EcoRI; since ligation of the Xhol overhang at the Sail site destroys both recognition sequences, subsequent removal of the v-src( + ) insert from the vector was achieved by digestion with EcoRI and with H dIII, which cleaves at a position in the multiple cloning sequence adjacent to the Sail site. The pVSRC-Cl plasmid was restricted with EcoRI and Hindlll, so as to liberate the tumorigenic insert. This insert included the v-src oncogene of the subgroup A strain of Prague RSV, as flanked downstream by a portion of the long terminal repeat (LTR) of RSV (from the 5' start of the LTR, to the single EcoRI site).

2. pMvsrc

The pM vsrc plasmid was generously provided by Dr. David Shalloway. Cornell University, Ithaca, NY. The plasmid is prepared according to Johnson et al , (1985) Mol. Cell. Biol. 5. 1073-1083. Briefly, the 3.1- b BamHl-Bg/ll Schmidt Ruppin A v-src fragment from plasmid pN4 (Iba et al. , (1984) Proc. Nat. Acad. Sci. USA 81. 4424-4428) is inserted into the pΕVX plasmid (Kriegler et al. , (1984) Cell 38,483-491) at a Bg/ll site lying between two Moloney murine leukemia virus (MoMLV) long terminal repeats (LTRs). This fragment contains 276 bp of pBR322 DNA from the pBR322 BamHI to Sail sites followed by 2.8 kb of Rous sarcoma virus (RSV) DNA from the Sail site that is about 750 bp upstream of the env termination codon down to the Nrul site that is about 90 bp downstream of the v-src termination codon. (The Nrul site is converted to a Bglll site in the construction of pN4.) Ligation is performed ^■ by using a 10: 1 insert-vector DNA fragment molar ratio.

The pM src plasmid was restricted with Nhel, so as to liberate a tumorigenic fragment. The fragment included the v-src oncogene of the subgroup A strain of Schmidt-Ruppin RSV, as flanked upstream by most of the Moloney murine leukemia virus (MoMLV) LTR (from the Nhel site near the 5 ' start of the LTR, to the 3' end of this LTR) and downstream by a small portion of the MoMLV LTR (from the 5' start to the Nhel site).

3. pcsrc527 The pcsrc527 plasmid is prepared according to Kmiecik and

Shalloway. (1987) Cell 49, 65-73. Briefly, a plasmid is constructed by cleaving expression vector pEVX (Kriegler et al , (1984) Cell 38,483-491 at its unique 5gIII site lying between two MoMLV LTRs and inserting the 3.2 kilobase (kb) pair 5<3mHI-5gIII hybrid src fragment from plasmid pHB5 in the proper orientation. This fragment contains sequences from pBR322, the SRA env 3 ' region, SRA v-src, src from recovered ASV, and chicken c-src. The Bglll site is generated by insertion of a linker at the Sacl site about 20 bp downstream from the c-src termination codon. The restriction map of pMHB5 contains the MoMLV splice donor about 60 bp downstream from the 3 'end of the upstream LTR and the v-src splice acceptor about 75 bp upstream from the src ATG.

Plasmid pMHB5527 is constructed by inserting the synthetic double-stranded DΝA oligomer

5' CCAGTTCCAGCCTGGAGAGAACCTATA (SEQ ID NO : 1 ) 3^'

3' TCGGGGTCAAGGTCGGACCTCTCTTGGATATCTAG (SEQ ID NO : 2 ) 5

into pMHB5 between the Baήll site at c-src codon 524 and the downstream unique Bglll site. This alters the TAC Tyr 527 codon to a TTC Phe codon while preserving the remaining c-src coding region. Equimolar amounts of the double- stranded oligomer and three gel-purified tandem restriction fragments from pMHB5 are ligated in one reaction, which contains the following the ohgomei with Banll and Bglll complementary ends, the 3 kb Bglll-Bgll (Bgll in the pEVX ampicilhn resistance gene) partial digest fragment, the adjacent 6 1 kb Bgll-Bgll (downstream Bg/7 m c-s/c) fragment, and the 0 38 kb Bgll-Baril (Banll at c-src codon 524) fragment

Plasmid pcsrc527 is constructed by replacing the 2 kb Sail (in env)-Mlul (in c-src) fragment in plasmid pMHB5527, with the homologous fragment from plasmid p5H This fragment contains the coding sequence for the c-s/ c amino region (codons 1 to 257) that have been isolated by molecular cloning of a c-src provirus and previously shown by sequencing to contain authentic c-src sequence without the mutation at codon 63 (Levy et al , (1986) Proc Natl Acad Sci USA 83, 4228-4232) Equimolar amounts of complementary gel-purified Sall-Mlul fragments from p5H and the other plasmids are ligated The pcsrc527 plasmid was restricted with Nhel, so as to liberate a tumoπgenic fragment The tumoπgenic fragment included the c-src(527) oncogene. as flanked by the same LTR complement as in pM vsrc

C Animals

Chickens of two closed lines SC and TK, were utilized These lines differ at the major histocompatibility (B) complex (B²IB² for the SC line,

B^lVB²¹ for the TK line) Embryonated eggs were obtained from Hyline

International (Dallas Center. IA) All chickens were hatched at the University of New Hampshire Poultry Research Farm and housed in isolation

D Tumor Induction by Plasmid DNA Tumors were induced by subcutaneous inoculation in the wmg web of a src-positive plasmid according to the technique described by Fung et al (1983) Proc Natl Acad Sci USA 80, 353-357 and Halpern et al , (1990) Virology 175, 328-331 Of the three tumoπgenic plasmids utilized here, all were adjusted, prior to inoculation, to a concentration of 100 μg of enzyme-restricted ^• DNA per 100 μl of phosphate-buffered saline. The conditions of inoculation used for particular experiments (age of chicken at time of inoculation, amount of plasmid, etc.) are indicated below.

E. Growth of Primary (wing web) Tumors in TK or SC Chickens

Inoculated with pVSRC-C 1 , pMvsrc or pcsrc527

Individual 1 -day-old chickens of line TK or of line SC were inoculated with 100 μg of either pVSRC-Cl , pMvsrc or pcsrc527. The mean tumor diameter (mm) at a particular time point and for any one group of TK or SC line chickens inoculated with an individual src-positive construct was computed as the sum of the diameters of the primary tumors divided by the number of chickens surviving to that point. The results are shown in Fig. 1A

(line TK) and Fig. IB (line SC). The ratios at each time point show, for a particular group, the number of chickens bearing palpable tumors to the total number of survivors to that point (standard typeface for pcsrc527, italics for pVSRC-Cl . bold typeface for pMVsrc). Error bars (unless obscured by the symbol) indicate standard error.

F. Growth of Challenge (wing web) Tumors in Test and Control

Line TK Chickens Under Conditions of Priming and Homologous Challenge with pcsrc527. or Priming and Homologous Challenge with pVSRC-Cl

Growth of challenge (wing web) tumors in test and control line TK chickens was determined under conditions of (i) priming and homologous challenge with pcsrc527, or (ii) priming and homologous challenge with pVSRC- Cl. Test chickens were primed at 1 day posthatch with 100 μg of construct; test and control chickens were challenged at five weeks posthatch with 200 μg of construct. The mean challenge tumor diameter was computed as described in the preceding section. At each time point the ratio of chickens bearing palpable challenge mmors to total number of survivors to that point is indicated for ^• priming and homologous challenge with pcsrc527 (Fig. 2A) and priming and homologous challenge with pVSRC-Cl (Fig. 2B) (standard typeface for control group, bold typeface for test group). The statistical comparison between the mean challenge tumor diameters of the test versus the control group at a particular time point was made using a two-tailed student's t test, *(p < 0.05), **(p < 0.01), ***(p < 0.001). The statistical comparison between the ratios of chickens bearing palpable challenge mmors to total number of survivors of the test versus the control group at a particular time point was made using a chi- squared test; the paired ratios are underlined for only those time points where p < 0.05. Error bars indicate standard error.

G. Growth of Challenge (wing web) Tumors in Test and Control line

TK chickens under Conditions of Priming with pVSRC-Cl and Heterologous Challenge with pcsrc527, or Priming with pcsrc527 and Heterologous Challenge with pVSRC-Cl

Growth of challenge (wing web) mmors in test and control line TK chickens, was determined under conditions of (i) priming with pVSRC-Cl and heterologous challenge with pcsrc527. or (ii) priming with pcsrc527 and heterologous challenge with pVSRC-Cl. Test chickens were primed at 1 day posthatch with 100 μg of construct; test and control chickens were challenged at five weeks posthatch with 200 μg of construct. The mean challenge tumor diameter was computed as described in Section E. At each time point the ratio of chickens bearing palpable challenge tumors to total number of survivors to that point is indicated for priming with pVSRC-Cl and heterologous challenge with pcsrc527 (Fig. 3 A) and priming with pcsrc527 and heterologous challenge with pVSRC-Cl (Fig. 3B) (standard typeface for control group, bold typeface for test group) . Statistical comparisons were made between test and control groups at a particular time point as described in the preceding section [*(p < 0.05), **(p < 0.01). ***(p < 0.001), for the student's t test], and the paired ratios are underlined for only those time points where, in the chi-squared test, p < 0.05. ^• Error bars indicate standard error.

H. Discussion

In a direct comparison of the growth of mmors induced in line TK by either pMvsrc or pVSRC-Cl, a similar pattern of relatively rapid regression was observed. This result established that the difference in LTR complement between these two v-src positive constructs did not exert a major influence on the tumor growth pattern in the TK line (Fig. 1A). By contrast, much more extensive and persistent tumor growth resulted from inoculation of TK chickens with the pcsrc527 construct (Fig. 1A). The relatively greater growth capacity of mmors induced by this construct indicated that in the TK line, the c-src(527) oncogene is much more highly tumorigenic than the v-src oncogene. This difference did not, however, generalize to the SC line (Fig. IB). The SC line was chosen for comparison with the TK line on the basis of earlier observations (Halpern et al. , (1993) Virology 197, 480-484) that v-src DNA-induced tumors engender a much weaker tumor immune response in line SC than in line TK. Whereas the growth of pcsrc527-induced primary mmors was virtually indistinguishable in the two lines, the growth of the v-src-induced tumors was considerably greater in the SC than in the TK line (Figs. 1A and IB). Thus v- src, but not c-src(527), gives rise to primary mmors whose growth patterns differ in the two lines analyzed here.

Only minimal protection against homologous challenge was observed under conditions of priming to c-src(527) DNA, indicative of the induction of a relatively weak tumor immune response (Fig. 2A; a statistically significant lowering of challenge tumor growth in the test versus the control chickens was observed at only one time point). By contrast, the v-src DNA- primed chickens showed excellent protection against the homologous tumor challenge (Fig. 2B). Priming with v-src DNA engenders a relatively greater degree of ^■ protection against challenge with c-src(527) DNA, than that afforded by priming with c-src(527) DNA itself (Fig. 3A). The degree of protection was weaker than that determined (Fig. 2B) for the case of priming and homologous challenge with v-src DNA. Only marginal protection was observed, however, when the heterologous challenge protocol was carried out in the reverse order (Fig. 3B). These results demonstrate that induction of reactivity to an antigenicity specified in tumor cells by an overexpressed proto-oncogene can confer tumor immunity.

Example 2 Vaccination Protocol

The following is a representative vaccination protocol according to the present invention. While the protocol includes transfection of cells freshly harvested from appropriate donors, the donor cells could as easily be obtained from cultures housed in cell banks. In such a case, the protocol would commence directly with part C, Fibroblast Transfection. A. Skin Punch Biopsy

A punch biopsy of skin is obtained from a donor (other than the patient-to-be-treated or identical twin thereof) by a trained physician following standard medical practice. B. Preparation of Fibroblast Culture

Under sterile conditions, the skin obtained by punch biopsy is put in a tube with 10 ml of the following wash medium: Dulbecco's Modified Eagle Medium (DMEM), containing sodium bicarbonate (30 ml/liter of a 5.6% solution) and penicillin/streptomycin (2 ml/liter of a pen-strep stock solution containing 5000 units penicillin and 5000 g of streptomycin/ml, pH 7.2-7.4.). In a sterile hood, the skin biopsy is added to a Petri dish, and then transferred several times to new Petri dishes containing the same wash medium. The biopsy is then finely minced with two scalpels, and 2-4 pieces ( < 1 mm³) of the minced biopsied are placed in the middle part of one or more T25 flasks. The flask is placed in a tissue culture incubator at 37 °C for one half hour with the cap firmly ^■ closed, then opened for 10 minutes. The following culture medium is prepared: DMEM containing sodium bicarbonate; antibiotics; and 10% fetal calf serum containing 2.5 μg/ml fungizone, 40 μg/ml gentamicin, and 1 % glutamine( 3 % W/V). Two ml of the culmre medium is then added to the flask, and the flask is incubated at 37°C (5 % C0₂), with the cap lightly unscrewed. The flask is left for three days without moving so as to obtain adhesion of the separate pieces of skin to the plastic. Afterwards, the medium is changed two times per week over a 3-4 week period always adding 2-3 ml of medium. To trypsinize the skin cell culture, one needs zones of confluence. After aspirating the culture medium. 5 ml of the Puck's Saline A/EDTA solution (0.4 g EDTA to 1 liter of Puck's Solution A) is added and immediately aspirated. Then 1 ml of trypsin solution (0.05/0.02 % trypsin in PBS. without Ca+ + or Mg+ +) is added and incubated for 5 min at 37°C, at which time 2 ml of culmre fluid is added to stop the action of the trypsin. The cells are then transferred to a larger flask (T75) and incubated at 37°C in 15 ml of culture fluid, which is changed every 2 days.

C. Fibroblast Transfection

The fibroblasts (2 X 10' cells) are washed twice in DMEM without serum or antibiotics. A LipofectAMINE™-DNA solution is prepared by mixing in tube #1 mix 400μl DMEM and 10 \ of dCTG vector DNA (lμg/ul). In tube

#2. 400 .-1 DMEM and 25 Ml of LipofectAMINE Reagent (Life Technologies. cat. no. 18324-012) are mixed. The contents of tube #1 and #2 are mixed together and are then left sitting at room temperature for 30 hours. Then, 3.2 ml of the LipofectAMINE™-DNA solution is added to the cells. The cells are incubated for six hours at 37°C, washed once with Hank's Balanced Salt

Solution, and then refed with growth medium and incubated for an additional 24 hours at 37 ^:C D. Transfectant Irradiation

Transfectants are irradiated to a dose of 25 By or 2500R. the cells are then counted by trypan blue exclusion. 2 X 10⁷ irradiated transfectants are resuspended in a volume of 0.2-0.4 ml of Hanks Balanced Salt Solution.

E. Vaccination

Patients are vaccinated by subcutaneous inoculation of 2 X 10 irradiated cells at 2-3 week intervals. A shorter or longer regimen is used, depending upon the results of delayed type hypersensitivity (DTH) reaction monitoring (described below).

F. Patient Assessment by DTH Monitoring

Patients are assessed for reactivity to the irradiated transfectants by a test of skin reactivity in a DTH reaction, as described by Chang et al.

(1993). Cancer Research 53: 1043-1050. To measure reactivity to the autologous irradiated transfectants, 10⁴ - 10⁶ transfected irradiated cells in a volume of 0.1 ml HBSS are inoculated intradermally. Induration is measured 48 hours later. as an average of two perpendicular diameters . Responses of greater than 2 mm are considered positive.

Example 3 v-myc Transfection of Murine Fibroblasts

A. Vector Preparation

The v-myc retroviral oncogene of avian myelocytomatosis virus MC29 (Land etal. (1983). Nature 304:596-602) was obtained from the American Type Culmre Collection, Rockville. MD. 20852. as the pSVv-myc vector (ATCC No. 45014). The v-myc-positive EcoRl-Kpnl fragment of pSVv-myc was ligated into the polylinker sites of the pBK-CMV plasmid (Stratagene Cloning Systems. La Jolla, CA). B. Cell Transfection

Stable transfection using the pBK-CMV-v-myc vector was carried out on a line of A31 fibroblasts (Balb/c origin), obtained from the ATCC. 2 X lO^"1 cells were seeded in a 100 mm/dish and allowed to grow for 18-20 h (RPMI 1640 medium and 10% fetal bovine serum), at which time the cells reached 50- 70% confluence. The cells were then washed twice in Dulbecco's Modified Eagles Medium (without serum or antibiotics). A LipofectAMINE™-DNA solution was prepared according to Example 2.C. , with the pBK-CMV-v-myc vector DNA. and 3.2 ml of the LipofectAMINE™-DNA solution added to the cells. The cells were then incubated for 6 hours at 37 °C, washed once with Hank's Balanced Salt Solution, and then refed with the growth medium and incubated for an additional 24 hour at 37 °C. Thereafter, the cells were fed once every two days with growth medium containing 250 μg/ml geneticin (G418; Gibco BRL cat. no. 11811) as the selective marker. Within two weeks, colonies were picked and expanded into permanent cell lines. The cells were then washed and collected by centrifugation.

It should be noted that the procedure for transient transfection is the same, through the point of incubation with the Lipofectamine™-DNA solution. Thereafter, the cells are washed and incubated for 72 hours in growth medium.

All references cited with respect to synthetic, preparative and analytical procedures are incorporated herein by reference.

The present invention may be embodied in other specific forms without departing from the spirit or essential attributes thereof and. accordingly. reference should be made to the appended claims, rather than to the foregoing specification, as indication the scope of the invention.

Claims

Claims

1. A cellular immunogen for immunizing a host against the effects of the product of a target proto-oncogene. the overexpression of which target proto-oncogene is associated with a cancer, which cellular immunogen comprises allogeneic donor cells which have been transfected with at least one transgene construct comprising at least one transgene cognate to the target proto-oncogene and a strong promoter to drive the expression of the transgene in the transfected cells, the transgene encoding a gene product which induces host immunoreactivity to host self-determinants of the product of the target proto- oncogene gene.

2. An immunogen according to claim 1 wherein the transgene comprises wild-type or mutant retroviral oncogene DNA; or wild-type or mutant proto-oncogene DNA of a species different from the host species .

3. An immunogen according to claim 2 wherein the transfected cells are non-dividing.

4. An immunogen according to claim 2 wherein the transgene comprises mutant retroviral oncogene DNA or mutant proto-oncogene DNA.

5. An immunogen according to claim 4 wherein the mutant DNA is nontransforming.

6. An immunogen according to claim 5 wherein the mutant DNA ■ comprises a deletion mutation in a region of said DNA which is essential for transformation.

7. A cellular immunogen according to claim 6 wherein the donor cells have been transfected with a plurality of transgene constructs , each construct encoding a different deletion mutation.

8. An immunogen according to claim 1 wherein the donor cells have been transfected with a transgene cognate to a target proto-oncogene selected from the group of proto-oncogenes consisting of AKT-2, c-erbB-2. MDM-2. c-myc. c-myb, c-ras, c-src and c-yes.

9. An immunogen according to claim 1 wherein the donor cells comprise fibroblasts or bone marrow-derived antigen-presenting cells.

10. A method for preparing a cellular immunogen for immunizing a host against the effects of the product of a target proto-oncogene. the overexpression of which target proto-oncogene is associated with a cancer, the method comprising: transfecting allogeneic donor cells with at least one transgene construct comprising at least one transgene cognate to the target proto-oncogene and a strong promoter to drive the expression of the transgene in the transfected cells, the transgene encoding a gene product which induces host immunoreactivity to host self-determinants of the product of the target proto-oncogene gene.

11. A method according to claim 11 wherein the transgene ^• comprises wild-type or mutant retroviral oncogene DNA; or wild-type or mutant proto-oncogene DNA of a species different from the host species.

12. A method according to claim 11 wherein the transfected cells are non-dividing.

13. A method according to claim 11 wherein the transgene comprises mutant retroviral oncogene DNA or mutant proto-oncogene DNA.

14. A method according to claim 13 wherein the mutant DNA is nontransforming .

15. A method according to claim 14 wherein the mutant DNA comprises a deletion mutation in a region of said DNA which is essential for transformation.

16. A method according to claim 15 wherein the donor cells are transfected with a plurality of transgene constructs, each construct encoding a different deletion mutation.

17. A method according to claim 11 wherein the transgene is cognate to a target proto-oncogene selected from the group of proto-oncogenes consisting of AKT-2. c-erbB-2, MDM-2. c-myc, c-myb, c-ras, c-src and c-yes .

18. A method according to claim 1 wherein the donor cells comprise fibroblasts or bone marrow-derived antigen-presenting cells.

19 A method of vaccinating a host against disease associated with the overexpression of a target proto-oncogene comprising

(a) transfecting allogeneic donor cells with at least one transgene construct comprising at least one transgene cognate to the target proto-oncogene and a strong promoter to drive the expression of the transgene in the transfected cells, the transgene encoding a gene product which induces host immunoreactivity to host self-determinants of the product of the target proto-oncogene gene,

(b) inserting the cells transfected with the transgene construct into the body of the host to obtain expression of the transgene in the host

20 A method according to claim 19 wherein the transgene comprises wild-type or mutant retrovnal oncogene DNA, or wild-type or mutant proto-oncogene DNA of a species different from the host species

21 A method according to claim 20 wherein the transfected cells aie rendered non-dividmg prior to insertion into the body of the host

22 A method according to claim 20 wherein the transgene comprises mutant retroviral oncogene DNA or mutant proto-oncogene DNA

23 A method according to claim 22 wherein the mutant DNA is nontransforming

24. A method according to claim 23 wherein the mutant DNA ^■ comprises a deletion mutation in a region of said DNA which is essential for transformation.

25. A method according to claim 24 wherein the donor cells are transfected with a plurality of transgene constructs, each construct encoding a different deletion mutation.

26. A method according to claim 19 wherein the transgene is cognate to a target proto-oncogene selected from the group of proto-oncogenes consisting of AKT-2. c-erbB-2, MDM-2. c-myc, c-myb. c-ras, c-src and c-yes .

21. A method according to claim 19 wherein the donor host cells comprise fibroblasts or bone marrow-derived antigen-presenting cells..

28. A method of vaccinating a host against disease associated with the overexpression of a targeted proto-oncogene comprising

(a) transfecting allogeneic donor cells with at least one transgene construct comprising at least transgene and a strong promoter to drive the expression of the transgene in the transfected cells, wherein the transgene comprises

(1) wild-type or mutant cognate retroviral oncogene DNA: or

(2) wild-type or mutant cognate proto- oncogene DNA of a species different from the host species;

(b) inserting the cells transfected with the transgene construct into the body of the host to obtain expression of the transgene in the host.