CA1341566C - Immune response to tumors induced by anti-idiotype antibodies - Google Patents

Abstract

Anti-idiotype antibodies, their production and their use in inducing immunological response to tumors are described.
The antibody bears an internal image of an epitope present on a solid tumor cell, wherein said anti-idiotypic antibody: (a) immunoreacts through its parotope with a monoclonal antibody specific for said solid tumor cell epitope, and (b) will induce formation of anti-anti-idiotypic antibodies specific for said solid tumor cell epitope upon administration to a human.

Description

tt 1 IMMUNE RESPONSE TO TUMORS INDUCED BY
ANTI-IDIOTYPE ANTIBODIES
TECHNICAL FIELD
The present invention is directed to the induction of immunological responses to antigens, in particular tumors.
More specifically the present invention is directed to the use of anti-idiotype antibodies to induce such an immunological response, as well as the antibodies and cell lines that produce them and is a divisional of Canadian Application Serial No. 467,155, filed November 6, 1984.
BACKGROUND OF THE INVENTION
The sequence of amino acids in the variable regions of both heavy (VH) and light (VL) chains of immunoglobulin (Ig) produces a conformation in the antigen binding site (i.e., parotope) allowing interaction of that antibody with a specific antigen. Injection of Ig into a heterologous host animal will give rise to anti-xenotypic (specific for species), anti-isotypic (specific for Ig class), and anti-idiotypic (specific for antibody variable region) antibodies.
Two functional classes of anti-idiotypic antibodies can exist, one of which reacts with the parotope, and another which reacts with the VH and/or VL framework (framework determinants). See generally, Geha, (1981) N. Encrl. J. Med.
305:25-28; Jerne, (1974) Ann. Immunol. (Paris) 125C: 373-389.
SUMMARY OF THE INVENTION
It is an object of the present invention to provide a method of inducing an immunological response to tumors in particular, solid tumors.
It is also an object of the present invention to employ anti-idiotype antibodies to produce such an immunological response.
Yet another object of the present invention is to provide 13 41 5 6~

monoclonal anti-idiotype antibodies, and immortal B lymphocyte sources for such antibodies, that are useful in the induction of immunological responses to tumors, particularly solid tumors.
These and other objects of the present invention are achieved by one or more of the following embodiments.
In one embodiment, the present invention provides a method of inducing an immunological response to a tumor comprising:
(a) providing an anti-idiotype antibody, an epitope identified by said anti-idiotype antibody being the parotope of an anti-tumor antibody; and (b) stimulating in a subject the production of anti-(anti-idiotype) antibody that identifies an epitope on a tumor cell by administering said anti-idiotype antibody to said human.
The present invention also provides polyclonal anti-idiotype antibodies, an epitope identified by said anti-idiotype antibodies being the parotope of an anti-tumor antibody, substantially free of anti-isotypic antibodies.
In another embodiment, the present invention provides an immortal B lymphocyte that produces an anti-idiotype antibody, an epitope identified by said anti-idiotype antibody being the parotope of an anti-tumor antibody. The present invention also provides the monoclonal antibodies produced by the above immortal B lymphocyte substantially free of other antibodies.
Suitable tumors are solid tumors such as solid gastrointestinal tumors. On the administration of the anti-idiotype antibodies in accordance with the invention, the subject, in particular mammal, more particularly human, is stimulated to produce anti-(anti-idiotype) antibody that identifies an epitope on a tumor cell.
DETAILED DESCRIPTION OF THE INVENTION
The following detailed description refers to both tumor cells and viral particles, but it will be understood that only the tumor information is relevant to the present invention.
The present invention provides a unique approach to cancer therapy and viral immunisation. Traditionally ~

approaches in tumor therapy have involved administering anti-tumor antibodies (i.e., antibodies that identify an epitope on a solid tumor cell) to patients in an effort to destroy the tumor. Traditionally, anti-viral therapy has involved immunisation with conventional vaccines. Applicants, however, have discovered that an immunological response to a patient's own tumors or to viruses can be induced with an antibody that is anti-idiotypic to an antibody that recognizes a tumor or virus antigen. The induction of this immunological response has utility as a therapeutic, and at least in the case of viruses, preventative treatment.
Although applicants do not wish to be bound by any parti-cular theory of operability, it is believed that the observed immunological response achieved by the present invention is attributable to an interaction between the anti-idiotype anti-body molecules and the human patient's immune system. The idiotypic (i.e., variable) region of the anti-idiotype anti-body molecule contains antigenic determinants (i.e, epitopes) seen as an antigen by the subject. This induces the produc-tion of anti-(anti-idiotype) antibodies by the subject.
Within this set of anti-(anti-idiotype) antibodies are those that are directly complimentary to the parotope of the anti-idiotype antibody. It is further believed that the parotope of the anti-idiotype antibody presents an "internal" image of the tumor cell or virus epitope identified (i.e., selectively bound) by the idiotype antibody and, therefore, the anti-(anti-idiotype) antibodies will also bind the tumor or virus antigen. In effect, the present method induces an immunolo-gical response to the tumor or virus by presenting an antigen (the parotype of the anti-idiotype antibody) which is essentially indistinguishable from the tumor or virus antigen to a portion of the patient's resulting antibodies.

Surprisingly, the above method is an effective procedure for controlling tumor growth or inducing an irrmunological response to a virus. Furthermore, it has several advantages over the more traditional approaches. First, much less foreign antibody need be administered to a patient. Second, the patient's anti-idiotype response is beneficial rather detrimental to the intended effect. Third, the patient's own antibodies are the anti-tumor or anti-viral antibodies, thus eliminating the necessity of repeated administration of exogenous anti-tumor anti-bodies. Other advantages will be readily apparent to those skilled in the art.

The idiotype of an antibody is defined by individually distinctive antigenic determinants in the variable or idiotypic region of the anti-body molecule. A portion of these idiotypic determinants will be on or closely associated with the parotope of the antibody, while others will be in the framework of the variable region. While each antibody has its own idiotype, particular antibodies will be referred to below by the following terms. "Idiotype antibody" or "Id Ab" refers to an anti-tumor or anti-viral antibody (i.e., the epitope identified by the idiotype antibody is on a cell of a tumor or a virus particle. "Anti-idiotype antibody" or "anti-Id Ab" refers to an antibody which identifies an epitope in the variable region of an idiotype antibody. A portion of such antibodies will identify an epitope that is the parotope of the idiotype antibody, thus presenting an "internal" image of the epitope identified by the idiotype antibody on the tumor cell. "Anti-(anti-idiotype) antibody" or "anti-(anti-Id) Ab" is an antibody that identifies an epitope in the variable region of the anti-idiotype antibody. A
portion of the anti-(anti-idiotype) antibodies will identify an epitope that corresponds to (i) the parotope of the anti-idiotype antibody, and (ii) an epitope on a tumor cell.

At stated below, the method of the present invention contemplates administering anti-idiotype antibody to a host. The anti-idiotype anti-body is administered to the host in any physiologically suitable carrier (e.g., sterile, pyrogen-free physiological saline), the formulations of which are within the skill of the art. The selection of carrier is not critical and the antibody can be administered by any method that introduces the antibody into the circulatory system (e.g., intravenous, intramuscular or subcutaneous injection).

The host may be any mammal, most commonly a human plus in the case of viruses a cat or dog. The amount of antibody administered to a host can vary widely dependent, for example, upon the particular antibody employed and the patient inoculated. It is only necessary that sufficient anti-idiotype antibody be administered to stimulate the production of anti-(anti-idiotype) antibodies by the patient's immune system. The amounts of antibody employed, however, need not be very great because only very small amounts are necessary to induce an immunological response.
In many cases, a dosage of antibody within the range of a few micrograms to a few milligrams should be sufficient, (e.g., about 50-200 ug to about 1-5 mg). The determination of an appropriate dosage is readily within the skill of the art.

In one embodiment, the present invention contemplates administering a formulation containing anti-idiotype antibody to a human patient to produce an immunological response to a tumor e.g. a solid tumor (i.e. a solid mass of malignant cells such as produced by carcinomas, sarcomas, melanomas etc., as opposed to disperse, circulating malignant cells such as leukemias).
In a preferred embodiment, the tumor is a gastrointestinal tumor. As defined above, a subclass of the anti-idiotype antibody selectively binds to (i.e., identifies) the parotope of an anti-tumor antibody (the idiotype antibody). A formulation containing an anti-idiotype may also be administered to a host, wherein the parotype of the anti-idiotype antibody is an internal image of a virus antigen. Such an antibody recognises an epitope that is the parotype of the corresponding idiotype antibody.

Anti-idiotype antibodies, which present internal images of the tumor or, virus antigen, can be distinguished from anti-idiotype antibodies that recognize framework determinants in the variable region of the idiotype antibody by any of several methods. One method of identifying the desired anti-idiotype antibodies is a competitive binding assay between the tumor or viral antigen (or hapten if available), the idiotype antibody and the anti-idiotype antibody. If the antigen blocks binding of the anti-idiotype antibody to the idiotype antibody, the epitope identified by the anti-idiotype antibody is closely associated with the idiotype antibody's parotope. Another test is to determine if anti-sera to the anti-idiotype antibody is also anti-tumor or -virus. These and other methods of identifying the appropriate anti-idiotype antibody are within the skill of the art. In the formulation administered to a host, the inclusion of anti-idiotype antibodies directed to framework determinants along with the subclass directed to the idiotype antibody's parotope is acceptable. It is only necessary that the formulation contain the sub-class directed to the idiotype antibody's parotope.

The anti-idiotype antibody employed can be homologous or hetero-logous to the host. The preferred antibody for a human is however a human antibody to minimize immunological response to the constant region to the antibody molecule. However, since relatively small doses of anti-idiotype antibody are required in the present invention, heterologous antibody can be employed (e.g., mouse, rat, goat, rabbit, etc.). In the absence of any serious reaction to heterologous anti-idiotype antibody, however, such antibody may be preferred due to ease and cost of prepara-tion. Furthermore, polyclonal anti-idiotype antibodies can be employed as well as monoclonal anti-idiotype antibodies.

Polyclonal anti-idiotype antibody can be prepared by conventional methods known in the art. For example, polyclonal anti-Id Ab can be produced by immunizing an animal with a monoclonal anti-tumor or virus antibody (i.e., Id Ab). The immunized animal will produce anti-Id Ab.
A subclass of this anti-idiotype antibody in the anti-sera will identify an epitope that is the parotope of the anti-tumor or -virus antibody.

Anti-sera collected from the animal can be purified, for example, by sequential absorbtion with (i) an immobilized antibody of the same isotype as the monoclonal Id Ab, but different idiotype, to remove anti-isotypic antibodies from the anti-sera, and (ii) the immobilized monoclonal Id Ab to remove the anti-Id Ab, a subclass of which will present internal images of the tumor or viral antigen. The anti-Id Ab can then be eluted from the bound monoclonal anti-tumor or viral antibody to provide a solution substantially free of anti-isotype antibodies. This solution can then be tested for the presence of anti-Id Ab that identifies the parotope of the Id Ab.

Monoclonal anti-idiotype antibodies substantially free of other antibodies can be isolated from the supernatant of substantially pure cultures of immortal B lymphocytes. The term "immortal B lymphocyte"
encompasses any relatively stable, continuous antibody-producing cell that can be maintained in culture for several months (preferably indefinitely), such as hybridomas (somatic cell hybrids of normal and malignant lymphocytes) and normal lymphocytes transformed by virus (e.g., Epstein-Barr virus) or oncogenic DNA. The production of immortal B
lymphocytes from normal B lymphocytes that product anti-idiotype anti-body is within the skill of the art. See, e.g., Monoclonal Antibodies (R.H. Kennett, T.J. McKearn & K.B. Bechtol 1980); M. Schreier et al., Hybridoma Techniques (Cold Spring Harbor Laboratory (1980); Monoclonal Antibodies and T-Cell Hybridomas (G.J. Hammerling, U. Hammerling &
J.F. Kearney 1981); Kozbor et al., (1982) Proc. Natl. Acad. Sci. U.S.A.
79:6651-6655; Jonak et al., (1983) Hybridoma 2:124; Monoclonal Antibodies and Functional Cell Lines (R. H. Kennett, K.B. Bechtol & T.J. McKearn 1983); Kozbor et al., (1983 Immunology Today 4: 72-79.

Normal B lymphocytes producing anti-Id Ab and suitable for the production of an immortal B lymphocyte can be provided by various methods within the skill of the art. For example, an animal, such as a rat or mouse, can be immunized with a monoclonal anti-tumor or -virus antibody and B lymphocytes producing anti-Id Ab recovered from the animal's spleen. Human B lymphocytes producing anti-Id Ab can be obtained by immunizing a patient with monoclonal anti-tumor or -virus antibody/

13 41~ 6 fi collecting peripheral blood lymphocytes from the patient, and then inducing in vitro the growth of B lymphocytes producing anti-Id Ab by stimulating the culture with the monoclonal anti-tumor or -virus antibody.
See, e.g., DeFreites et al., (1982) Proc. Natl. Acad. Sci. U.S.A. 79:
6646-6650. The animal or human B lymphocytes producing anti-Id Ab can thus be recovered and immortalized by those of skill in the art. Of course it is understood that those lymphocytes producing anti-Id Ab that present internal images of the tumor cell or virus antigen should be distinguished from B lymphocytes producing anti-Id Ab directed to framework determinants in the idiotypic region.

The method of the present invention as applied to tumors ca> also be practiced in conjunction with other methods of inducing immunological responses to tumors such as the use of viral oncolysate vaccines as described in U.S. Patent 4,108,983. As regards viruses, if desired, the immune response produced in the host mammal to a virus can be further heightened by immunization with a conventional virus vaccine in addition to the administration of anti-Id Ab as described above. After the production of anti-(anti-Id) Ab is stimulated in the host mammal by administering anti-Id Ab (e.g., about 2 weeks post-administration), the host is then given one inoculation of a virus vaccine employing conven-tional techniques known in the art (e.g., killed virus vaccines such as HDC rabies vaccine). See, Plotkin et al., (1976) Am. J. Epidemiology 103:75-80; Wiktor et al., Rabies Vaccine for Human Use: Volume 40, pp. 3-9 (1978). The types of vaccine and protocols for their administra-tion are within the skill of the art.

The following examples are illustrative of the present invention and are not intended to limit its scope.

13 41~b6 Example 1 - Tumor Antibodies Monoclonal Idiotype Antibodies The following experiments employed mouse monoclonal antibodies 17-1A, C42032 and C41472, which bind to human gastrointestinal cancer cells, and are described in Herlyn et al., (1979) Proe. Natl. Acad. Sci.
U.S.A. 76: 1438-1442 and Koprowski, "Monoclonal Antibodies In Vivo,"
in Monoclonal Antibodies in Cancer: Proceedings of the Fourth Armand Hammer Cancer Symposium (B. Boss, R. Langman, I. Trowbridge &
Dulbecco 1983). Monoclonal antibody (MAb) C42032 has specificity for colorectal carcinoma (CRC)-associated antigen(s) of Mr 180, 160, 50 and 40 K. MAb C41472 (IgG2a) has specificity for CRC-associated antigen Mr 50 K. The MAb A5C3 against hepatitis virus was also employed and is described in Wands et al., (1981) Proc. Natl. Acad.
Sci. U.S.A. 78: 1214-1218. MAb's 17-1A (IgG2A, kappa light chain) and C42032 (IgG2a kappa light chain) were purified from ascites obtained from hybridoma-bearing mice by affinity chromotography on protein A-sepharose column (Pharmacia, Piscataway, NJ) as described by Ey et al., (1978) Immunochemistry 15: 429-436.
Patients All patients had metastatic or recurrent gastrointestinal adenocarcinoma and were injected systematically with one dose of a purified sterile, pyrogen-free preparation of MAb 17-1A concentrated from ascites fluid of Balb/c mice per the method disclosed in Sears et al., (1982) Lancet 762-765. Of nine patients who received 192 mg or less of MAb 17-1A seven developed anti-mouse globulin antibodies. Of the 20 patients who received 200-1000 mg of monoclonal antibody, three developed anti-mouse globulin antibodies. Sera of three patients of the first group who developed anti-mouse globulin molecules (patient Nos. 07, 08 and 09) and of two patients of the second group (Nos. 14 and 23), were either screened or processed for isolation of anti-idiotype 134156fi antibodies. Sera used for isolation of anti-idiotype antibodies from subject Nos. 07, 08 and 23 were obtained at the time when all three showed the highest concentration of anti-mouse globulin antibodies.
Patient No. 08 received a second injection of 130 mg of monoclonal antibody 20 months after the first injection, and serum obtained after this second injection was used in a screening test for the presense of anti-idiotype antibody.
Preparation of Polyclonal Anti-Idiotype AntibodIes New Zealand white rabbits were injected subcutaneously at mul-tiple sites with 300 ug purified MAb 17-IA emulsified in Freund's complete adjuvant and, 30 days later, boosted Intramuscularly with 100 ug of the monoclonal antibody. Sera was collected on day 10 of the secondary response.
Anti-sera was absorbed on immobilized MAb C42032 and MAb i5 17-1A. The purified monoclonal antibodies (30 mg each) were coupled to 2 ml of Affi-Ge~10 (Bio-Rad Laboratories,' Richmond, CA). The anti-sera was then sequentially absorbed on MAb C42032 and MAb 17-IA immunoabsorbents to remove anti-isotypic and anti-idiotypic antibodies, respectively. Absorbed antibodies were eluted with 0.1 M
glycine buffer (pH 2.8), immediately neutralized with phosphate buffer, dialyzed against phosphate-buffered saline, and protein quantitated by absorptivity at 280 nm (E196280=14).
Anti-sera obtained from patients after one injection of MAb 17-1A was also obtained and purified as described above. Sera from subject No. 23, who received 750 mg of MAb 17-1A, and from subjects No. 08 and 07, who received 133 and 125 mg of MAb 17-1A, respec-tively, were collected at various times after the first injection of antibody. Samples shown by radioimmunoassay analysis to contain anti-murine IgG antibody were sequentially absorbed on MAb C42032 and MAb 17-1A immunoabsorbents to remove anti-isotypic and.anti-idiotypic antibodies, respectively, as described above. The anti-idibtype antibody isolated from the sera was shown to be human immunoglobulins by binding to 1251-labeled anti-human F(ab')Z fragments. The yield of 13 415b6 ->>-anti-idiotype protein from serum samples varied: 13 ug/mi from No. 08;
8.9 ug/ml from No. 07; and 43 ug/ml from No. 23. The largest amount was obtained from No. 23 serum which also showed the highest levels of anti-mouse globulin antibodies.
Screening Sera for Presence of Anti-Idiotype Antibodies To screen serum samples for the presense of anti-idiotype antibody, a competition assay was performed using the rabbit-anti-idiotype antibody and four human sera pre-incubated with 1251-labeled MAb 17-1A.
Polystyrene beads of 1/4-inch (6.35 mm) size (Precision Plastic Ball Co., Chicago, IL) were washed three times with 95% ethynol.
The air-dried beads were incubated with a dilution of either rabbit or human anti-idiotype antibody in 0.02 1Vi sodium tetraborate, pH 8.2.
After overnight incubation at 4 C with gentle shaking, the beads were washed three times with phosphate-buffered saline and then incubated for at least three hours at room temperature with phosphate-buffered saline containing 2% bovine serum albumin and 0.04% NaN3. The beads were then exposed to 125I-labeled MAb 17-1A as the reference idiotype that had been preincubated with the potential source of human anti-idiotype antibody, i.e., human sera was diluted to 25% concen-tration in phosphate-buffered saline without Ca++ and Mg*+ and supplemented with 2% bovine serum albumim and 0.04% NaN3. After an additional overnight incubation, the beads were washed and the radi-oactivity bound was measured in a gamma counter.
Three of the sera obtained after one injection with monoclonal antibody, (Nos. 23, 09, and 14) showed inhibition of binding of MAb 17-1A that was higher than that of pre-monoclonal antibody injection samples. Binding inhibition values obtained for post-monoclonal antibody serum of patient No. 14 were low as compared to the other two sera, but higher than that for the pre-monoclonal antibody exposure serum of the same subject. Inhibition values for serum obtained from patient No. 08 seven days after he received a second injection of monoclonal antibody were already high.

Competition Assay for Detection of Anti-Idiotype A competition assay was conducted in a manner similar to that described above to determine the binding of isolated anti-idiotype antibodies to MAb 17-1A, as well as to monoclonal antibodies C42032, C41472 and A5C3.
The results indicated that the binding of anti-idiotype antibodies from three sera (Nos. 08, 07 and 23) to MAb 17-1A was significantly higher than to the three other monoclonal antibodies, two of which (C42032 and C41472) also detect antigenic sites on colorectal carci-noma cells. These sites, however, are different from the sites recog-nized by MAb 17-1A. The immunoglobulin isolated from sera of all three subjects prior to exposure to MAb 17-1A was concentrated to approximately 2.5 ug/ml and coupled to polystyrene beads. These prep-arations, however, did not bind any of the monoclonal antibodies tested, indicating the absense of anti-idiotype antibodies in pre-exposure serum.
Cross-Reactivity Between Human Anti-Idiotype Antibodies A competition assay was conducted to determine whether there was cross-reactivity among the human anti-Id sera.
Results of the competition assay, shown below, indicates signifi-cant cross-reactivity between the anti-idiotype sera of patients No. 07 and 23, and slightly less (but still significant) cross-reac-tivity between the anti-idiotype sere of patients Nos. 08 and 23. Similar cross-reactivity was found between the anti-idiotype sera of patient No.
07 and post-monoclonal antibody serum obtained from patient No. 08 (results not shown). These results indicate that the anti-idiotype antibodies produced by different patients are directed against the same antigenic site.

2nd Ist Antibody cmp % of Inhibition of Antibody Serum of: bound 17-1A MAb Binding None 4297 anti-Id 23 07 Pre 4595 0 Post 1231 71 08 Pre 4097 5 Post 2585 40 Epitope Detected by Anti-Idiotype Antibody Hapten inhabition of binding of human anti-idiotype antibodies to MAb 1?-lA was performed to show that the anti-idiotype antibodies were directed to the parotope of the idiotype antibody.
A 3 M KC1 extract of SW 1222 cells, a cell line derived from colorectal cancer, was prepared as described in Herlyn et al., (1982) J.
Clin. Immunol. 2: 135-140. The preparation bound MAb 17-1A, indicat-ing that the material contained the antigen in its soluble form.
1251-labeled MAb 17-1A was incubated with the CRC cell extract and with a 3 M KCI extract of melanoma cells which do not bind MAb 17-1A. The antibody-antigen mixtures were then added to beads coated with anti-idiotype antibody obtained from patient No. 23 and the binding compared to the binding of the radiolabeled monoclonal antibody alone. These experiments were performed with non-saturating a m ounts of iodinated monoclonal antibody in order to detect changes in binding with small amounts of competitive haptens. For control purposes, the iodinated MAb 17-IA was mixed with an extract from melanoma cells that was known not to bind MAb 17-1A. The CRC
cell extracts in= concentrations of 0.1 or 0.5 mg/ml were found to inhibit the binding of the anti-idiotype antibody from patient No. 23 to lodinated MAb 17-1A by 39% and 68%, respectively. The extract from melanoma cells in concentrations up to 0.5 mg/ml did not significantly affect the monoclonal antibody binding.

This hapten inhibition of the binding reaction indicates the presence of an "internal image" of the CRC epitope on the anti-idiotype antibody. This is also supported by the fact that the finding that the extract of CRC cells did not bind to the anti-idiotype antibodies but did bind, as expected, to MAb 17-IA.
Production of Anti-Idiotype and Anti-(Anti-Idiotype) Antibodies by Human B Lymphocytes Buffy coat cells were obtained from patient Nos. 08 and 23, twenty and five months, respectively, after injection with MAb 17-1A.
The cells were stimulated with 10 ng/ml F(ab')2 fragments of 17-lA in vitro as described in DeFreitas et al., (1982), Proc. Natl. Acad. Sci.
U.S.A. 79: 6646-6650. During the following seven days, aliquots of cells were separated into T and B cell populations by rosetting with sheep erythrocytes treated with 2-amino ethylisouronium bromide. See, Pellogrino et al., (1975) Clin. Immunol. & Immunopathol. 3: 324-333.
Both cell populations were stained with F(ab')2 fragments of 17-1A or anti-influenza monoclonal antibody, and goat anti-mouse Ig-FITC. The cell populations were then subsequently analyzed in a cytofluorograph.
In addition, peripheral blood mononuclear cells from the same patient were stimulated with F(ab')2 fragments of 17-1A or anti-influenza monoclonal antibodies for nine days in a modified Mishell-Dutton culture for specific human Ig production as described in DeFreitas et al., supra. Supernatants from these cultures were assayed in a solid-phase enzyme-linked immunoabsorbent assay for specific human IgG
(KPL Laboratories, Gaithersburg, MD).
The percentage of lymphocytes that initially bound 17-1A F(ab')2 of patient No. 08 was 1.2%, and of patient No. 23 was 0.2%. During seven days in culture with MAb 17-1A, the percentage of lymphocytes of patient No. 23 that specifically bound 17-1A F(ab')2 increased from 0.2 to 13%. All the 17-1A binding cells were present in the B cell population. In addition, after nine days, human anti-MAb 17-1A IgG
was detected. Incubation of lymphocytes from the same patient with anti-influenza monoclonal antibody under identical conditions produce no detectable human Ig to either MAb 17-1A or anti-influenza monoclonal antibodies.
In another experiment, human B lymphocytes were stimulated to produce anti-(anti-idiotype) antibody. B lymphocytes were collected and stimulated in vitro as described above, except that the cells were stimulated with autologous anti-idiotype antibody rather than idiotype antibody. Anti-(anti-idiotype) antibodies were produced by stimulated B
lymphocytes and these anti-(anti-Id) Ab were shown to identify as epitope on CRC cell extract and on whole cells. Thus, the human immune system will produce anti-tumor antibodies in response to stimulation with anti-idiotype antibodies.
Immortal B Lymphocytes Producing Anti-Idiotype Antibody Various methods of producing immortal B lymphocytes secreting monoclonal antibodies are known in the art. See Kozbor et al., (1983) Immunology Today 4: 72-79. Human B lymphocytes secreting anti-(anti-idiotype) antibody, obtained from peripheral blood lymphocytes as described above, can be immortalized, therefore, by one of skill in the art.
One method that can be readily employed is immortalization with Epstein-barr virus (EBV). In this method, the normal lymphocytes described above are infected with EBV in vitro and immortal cell lines then establish, for example, by limiting dilution on a feeder layer.
See, e=Q=, Kozbor, et al., (1983), supra, and references 51-60 cited therein.
Another approach is to fuse either the above described anti-Id Ab secreting lymphocytes or an EBV-transformed lymphocyte with a human plasmacytoma or lymphoblastoid fusion partner. For example, and EBV-transformed B lymphocyte secreting anti-Id Ab can be fused with, for example, the human Iymphoblastoid cell line KR-4. The desired hybridomas would then be selected for in hypoxanthine-aminopterin-thymidine medium containing ouabain, which eliminates the parental cells. Hybridomas are tested for specific antibody production. Positive hybrids are then cloned, recloned and then propagated in bulk culture or in the peritoneal cavity of an immune-suppressed mammal nude mouse). See, e.g., Kozbor et al., (1982) Proc. Natl. Acade. Sci. U.S.A. 79: 6651-6655.

Example 2 - Viral Antibodies Preparation of Anti-Idiotype Antibody The ability of anti-idiotype antibody (anti-Id Ab) made against various monoclonal antibodies (mAb) to reconstruct the epitopes of rabies virus glycoprotein (G) was studied. This G is responsible for many of the important biological properties of rabies virus, including the induction of the virus-neutralizing antibodies (VNA). Precisely which portion of the G is responsible for this function is not known at present. A functional epitope mat for challenge virus standard (CVS) strain of rabies virus G suggesting the existance of at least three major antigenic sites for type-specific VNA has been described in Lafon et al., (1983) J. Gen. Virol. 64:843-851.
Five anti-G mAb were selected from a large panel of hybridomas on the basis of their isotype (enabling purification by pro-tein A-Sepharose chromotography) and binding site on the rabies virus G as defined by the functional epitope mat. Anti-rabies virus G
mAb's 509-6, 101-1, 507-1 and 719-3 have been described previously.
Lafon et al., supra; Wiktor et al., (1980) J. Exp. Med. 152:99-112.
These mAb, which strongly neutralize rabies virus infectivity, have the following characteristics: 509-6 (epitope map site I; IgG2a), 101-1 (epitope map site IIb; IgG2a), 507-1 (epitope map site IIIb; IgGi), and 719-3 (epitope mat site IIc; IgG2a). Anti-G mAb 1104-2 (IgGl) was obtained from an additional fusion of splenocytes from rabies virus immunized BALB/c mice with the 653 variant of P3x63Ag8 mouse myeloma cells, Kearney et al., (1979) J. Immunol. 123:1548-1550, as described In Wiktor et al., (1978) Proc. Natl. Acad. Sci. USA
75:3938-3942. The anti-G-secreting hybridoma cells were selected, cloned by limiting dilution, and ascites fluids prepared as also described in Wiktor et al., (1978), supra. The 1104-2 mAb was selected because it demonstrated excellent binding to ERA virus, but poor neutralization.
Anti-rabies virus nucleocapsid mAb 515-3 (IgG2a) and 389-1 (IgGi) were isolated by similar techniques from Kelev virus-immunized BALB/c mice.
The stocks of ERA or CVS strains of rabies virus employed were propagated in BHK-21 cells by standard methods. See Wiktor et al., in Laboratory Techniques In Rabies, pp. 101-123 (M. Kaplan & H.
Koprowski 3rd ed. 1973). Rabies antigenic variants, ERA RV194-2 and RV509-6, have been described and represent viruses which are resistant to neutralization by anti-G mAb 194-2 and 509-6, respectively. See Dietzschold et al., (1983) Proc. Natl. Acad. Sci. USA 80: 70-74. Rabies soluble glycoprotein (Gs) was purified from virion-depleted culture fluids by immunoadsorbent chromotography as described in Dietzschold et al., (1983) Virology 124: 330-337.
All anti-G mAb were purified from ascites fluids. Antibodies of the lgG2a isotype were diluted appropximately 1:30 with Britton-Robinson buffer (BRB) at pH 8Ø See Gerhard et al., in NSonoclonal Antibodies, pp. 317-333 (R. Kennett, T. McKearn & K.
Bechtol 1980). The mAb in BRB was passed tt1rough a Nalgene 0.45 u filter, and then over a protein A-SepharosetD4B absorption column (Pharmacia, Piscataway, N.J.). The column was washed with 30 ml of BRB, pH 8.0, before antibody was eluted with BRB at pH 3Ø Anti-G
mAb of IgGl isotype were first precipitated with sodium sulfate at a final concentration of 18% (w/v). The Ig fraction was dissolved and dialyzed in BRB, pH 8.0, and then passed over the protein A-Sepharose column as before. Ig eluted from the column was detected by radioimmunoassay (RIA) using ERA virus as antigen. See, Dietzschold et al., (1982) J. Virot. 44: 595-602. Antibodies were concentrated by vacuum dialysis against phosphate-buffered saline, pH 7.4 (PBS).

-18- ~
Protein concentration was determined using bovine serum albumin as a standard. See Bramhall et a1., (1969) Anal Biochem. 31: 146-148.
Anti-Id Ab were prepared as described in Staudt et al., (1983) J.
. Med. 157: 687-704. Briefly, female New Zealand white rabbits were injected subcutaneously in multiple sites along the mammary chain with 300 ug of protein A-Sepharose-purified mAb emulsified in Freund's complete adjuvant (FCA). Two intramuscular boosters of 100 ug of antibody in PBS were given on days 7 and 30, and sera are collected days later. Each anti-idiotypic anti-serum was made specific for 10 idiotype regions by passage over a Sepharose 4B column to which either mAb 515-3 (IgG2a) or 389-1 (IgGl) were coupled to remove anti-isotype Ab. The effluent from such columns contained the antibodies reactive with the idiotypic determinants. Antibodies to con-stant regions were eluted from the mAb 515-3 and 389-1 immunoadsorbent columns with 0.1 M diethylamine, pH 11.5. IgG from each anti-Id Ab was isolated by protein A-Sepharose chromotography as above. The reactivity of the column effluent with non-idotypic determinants was neglegible.
Characterization of Anti-Idiotype Antibodies The specificity of the anti-Id Ab prepared above and the existence of cross-reactive idiotypes among the various anti-G mAb was determined for each anti-Id Ab preparation in a RIA.
A solid phase RIA was used to measure the bindings of anti-Id Ab to fixed mAb. Individual mAb were diluted depending on the ascites fluid concentration in carbonate-bicarbonate buffer, pH 8.9, as follows: 509-6 (1:3,000), 101-1 (1:6,000), 719-3 (1:6,000), 507-1 (1:6,000), 1104-2 (1:16,000). Then 25 ul of each was added to wells of polyvinyl microtiter plates (Dynatech Laboratories). These antibodies were allowed to dry to the wells by overnight incubation at 37 C.
Free binding sites on the wells were blocked for at least 1 hour with 10% agamma horse serum (GIBCO Laboratories) in PB=S with 0.08%
sodium azide (PBSN). Twnety-five microliters of dilution of anti-Id Ab was added, and after 1 hour at room temperature, the plates were washed extensively. Bound anti-Id Ab was detected by adding 25 ul (30,000 cpm) of 1251 goat anti-rabbit IgG (Cappel Laboratories) labeled by the iodogen method and incubating for an additional hour at room temperature. See Markwell et al., (1978) Biochemistry 17: 4807-4817.
All dilutions of anti-Id Ab or radiolabeled probe were made in 10%
agamma horse serum in PBSN. Plates were washed free of unbound probe and radioactivity bound to individual wells was measured in a gamma counter.
The results of titrating each anti-Id Ab preparation against homologous and heterologous mAb demonstrated that each anti-Id Ab was specific for its homologous Id mAb.
Determination of Anti-Idiotype Antibodies Directed to the Parotope of the Idiotype Antibody A competition RIA was devised to test the ability of rabies virus G to prevent the interaction between Id Ab and anti-Id Ab.
See, Chaflin et al., (1974) J. Immunol. 112:1747-1756; Sher et al., (1972) J. Immunol. 109:176-178.
The rabies Gs was used as the competing antigen. Pretitrated levels of mAb were incubated with serial two-fold dilutions of Gs for 1 hour prior to the addition of a standardized anti-Id Ab dilution.
The amount of bound anti-sera was determined by the binding of a 1251-labeled goat anti-rabbit antibody probe. The binding of three of the five anti-Id Ab to their corresponding anti-G mAb (509-6, 507-1 and 1104-2) was inhibited by Gs. Maximum inhibition varied from 20-50% with as much as 6 ug/mi Gs, but at least a 15% reduction was observed with as little as 0.75 ug/ml Gs. The inability to totally inhibit binding of anti-Id Ab was an indication that both framework and parotope site specificities were present in the three polyclonal anti-Id sera. This was confirmed by a Western blot analysis of anti-Id Ab reactivity with mAb which had been resolved by sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) under reducing and non-reducing conditions. Since the antigen-combining site is elim-inated by reduction of mAb, the reactivity of anti-Id Ab with reduced -2o-mAb showed a specificity for framework determinants. The absence of any significant binding inhibition by G. for anti-Id 101-1 Ab or anti-Id 719-3 Ab, suggests a minor or absent population of parotope spe-cificities in these sera.
Prepartion of Anti-(Anti-Idiotype) Antibody The following examples demonstrates that anti-Id Ab reactive with the antigen-combining site of anti-G mAb contains a subpopulation which mimics the viral epitope recognized by these anti-G mAb and can induce an immunological response to G.
ICR mice in groups of 4 were inoculated subcutaneously with 40 ug/mouse of protein A-Sepharose-purified anti-Id IgG emulsified in FCA.
Subcutaneous booster inoculations were administered on days 7 and 32;
each with 40 ug of anti-Id IgG in Freund's incomplete adjuvant. Five days following the last booster inoculation, animals were bled via the retro-orbital plexus and pooled sera checked for rabies virus-neutralizaing antibodiy (VNA). As a control, another group of mice received an immunization of protein A-Sepharose-purified normal rabbit IgG.
The levels of VNA in immunized mice were determined by a modification of the rapid fluorescence focus inhibition test. See Smith et al, in Laboratory Techniques in Rabies, pp. 354-357 (M. Kaplan ~C
H. Koprowski 3rd ed. 1973).~Serial two-fold dilutions of mouse serum were prepared in ;Vticrottter II plates (Falcon Plastics) (50 ul/well) and incubated for 1 hour at 37 C with an equal volume of virus, containing 104 PFU/50 ul. Following incubation, 50 ul of freshly trypsinized BHK-21 cells (2 X 106 cells/ml) were added to each well, mixed, and 10 ul aliquots of the serum-virus-cell mixture were trans-ferred (in duplicate) into wells of Terasaki plates (Falcon Plastics).
After 20 hour incubation, plates were first rinsed with PBS and then with 80% (v/v) acetone in distilled water, and fixed for 30 min in 80%
acetone at room temperature. Plates were dried and cells were stained for 30 min at 37 C with S ul/well of fluorescein-conjugated anti-rabies nucleocapsid antibody of rabbit origin. See Wiktor, (1974) Symp. Series Immunobiol. Standard 21: 102-118. The control wells (containing virus but no antibody) showed approximately 40% of cells containing rabies virus-specific inclusions. The end point of virus neu-tralization was defined as the reciprocal of the highest serum dilution capable of reducing the number of rabies virus-infected cells by 50%.
The results of the rapid fluorescent focus inhibition test is shown in the table below. Significant VNA titers were generated against ERA strain rabies virus in mice immunized with anti-Id 509-6 Ab, and with anti-Id 1104-2 Ab. The three other anti-(anti-Id) Ab failed to neutralize ERA virus. Control experiments showed that mouse anti-(normal rabbit IgG) Ab as well as the anti-Id Ab used for immunization, had no rabies virus neutralizing activity. Furthermore, preincubation of anti-(anti-Id 509-6) sera with anti-Id 509-6 Ab removed the neutralizing activity. Preincubation with normal rabbit serum, how-ever, did not.
In order to test the specificity of the VNA generated, other rabies viruses were used in the neutralization assay. The CVS strain of rabies virus bound both mAb 509-6 and 1104-2 (data not shown), and the Table shows that CVS was neutralized by both anti-(anti-Id 509-6) sera and anti-(anti-Id 1104-2) sera. On the other hand, a vari-ant of ERA virus, RV 509-6, which possessed a mutation in the 509-6 epitope resulting in the loss of binding or neutralizing activity by mAb 509-6, see Laton et al., (1983) J. Gen. Virot. 64: 843-851, was not neutralized by anti-(anti-Id 509-6) sera. Another neutralization-resistant variant, RV 194-2, was effectively neutralized by anti-(anti-Id 509-6) sera. Previous results demonstrated that the antigenic sites recognized by anti-G mAb 509-6 and 194-2 are totally independent. See Laton et al., su ra. These data show that anti-(anti-Id 509-6) sera reacts solely with the epitope which is recognized by anti-G mAb 509-6, and thus this epitope had been simulated by anti-Id 509-6 Ab.

Neutralization of Rabies Virus Strains by Anti-(anti-Id) Serumt Mouse antiserum to: ERA cvs Parent RV 509-6 RV 194-2 Parent Normal rabbit IgG 4 4 4 4 anti-Id 509-6 32 4 64 64 anti-Id 101-1 4 ND* ND ND
anti-Id 179-3 4 ND ND ND
anti-id 507-1 4 ND ND ND

anti-Id 1104-2 128 4 128 64 RIG** 16 16 32 32 t The reciprocal of the highest dilution of serum capable of reducing the number of infected cells by 50% was taken as the neu-tralization titer.

* Not done.

** Human anti-rabies Ig was diluted to 0.2 international units per ml, and serial two-fold dilutions incubated with rabies virus.
The above examples were presented for illustrative purposes only and are not intended to limit the invention which is defined solely by the claims.

Claims (3)

1. ~A polyclonal antibody preparation comprising anti-idiotypic antibodies bearing the internal image of an epitope present on a solid tumor cell, wherein said anti-idiotypic antibodies (a) ~immunoreact through their paratopes with the paratope of a monoclonal antibody specific for said solid tumor cell epitope; and (b) ~induce formation of anti-anti-idiotype antibodies specific for said solid tumor cell epitope upon administration to a human;

and wherein said polyclonal antibody preparation is substantially free from anti-isotype antibodies and is produced by immunizing a mammalian host with a monoclonal antibody specific for said solid tumor cell epitope.

2. ~Use of the polyclonal antibody preparation of claim 1 for administration to a human to stimulate production of antibodies which bind to an epitope on a solid tumor cell.

3. ~The use of claim 2 wherein said solid tumor cell is a gastrointestinal tumor cell.