Immunotherapy against murine leukemia - Nature

Leukemia (1998) 12, 401–405  1998 Stockton Press All rights reserved 0887-6924/98 $12.00

Immunotherapy against murine leukemia S de Vos1, DB Kohn2, SK Cho1, WH McBride3, JW Said4 and HP Koeffler1 1

Division of Hematology/Oncology, Cedars-Sinai-Medical Center, UCLA School of Medicine; 2Research Immunology/BMT, Children’s Hospital; 3Department of Radiation Oncology, UCLA; and 4Department of Pathology, UCLA, Los Angeles, CA, USA

The central hypothesis underlying specific anti-leukemia immunotherapy is that leukemic cells express antigenic determinants not expressed on their counterpart normal adult cells. We have developed a murine myeloid leukemia/tumor immunization model using the low-immunogenic WEHI3 leukemia in syngeneic mice. Mice preimmunized with irradiated, transduced IL-7-producing WEHI3 cells showed systemic protection and rejection of a lethal dose of intravenously (i.v.) injected parental WEHI3 cells (5 ⴛ 104) with 40% long-term survival. When vaccinated with a mixture of parental WEHI3 cells and IL-2-producing NIH-3T3 fibroblasts (5 ⴛ 105), 60% survival was observed. Vaccination with murine granulocyte–macrophage colony-stimulating factor (GM-CSF)-producing WEHI3 cells resulted in only 20% survival of i.v. challenged mice, and the additional combination of IL-2- and IL-7-producing vaccine did not reveal any additive or synergistic effects. Immunizing mice with a pre-established leukemia burden (injected with 5 ⴛ 104 WEHI3 cells, i.v., 3 days prior to immunization) did not cure or result in a prolongation of survival, indicating that improved methods of immunization are needed. Taken together, we have identified IL-7 and IL-2 as effective cytokines in our leukemia/vaccination model with only marginal activity by GM-CSF. Keywords: immunotherapy; murine leukemia; IL-7; IL-2; GM-CSF

Introduction The central hypothesis underlying specific anti-leukemia immunotherapy is that leukemic cells express antigenic determinants not expressed on their normal counterpart adult cells. Those antigenic determinants must be immunogenic for an effective attack by the immune system.1,2 Anecdotal evidence that the immune system may help to eradicate leukemia is provided by occasional reports of improved survival following post-induction immunization with non-specific immunostimulators,3–5 although convincing evidence is still lacking. The graft versus leukemia (GVL) phenomenon after allogeneic bone marrow transplantation (BMT) is the most striking evidence of a role of the immune system in controlling leukemia.6,7 Alloreactive T cells are not the only mediators of the GVL effect; graft-versus-host disease (GVHD) and GVL may, therefore, be separable.8 Several studies have shown that lymphokine-activated killer cells (LAK) from patients with acute leukemia lyse autologous blast cells.9–13 Contrasting allogeneic anti-leukemia CTLs, only a few reports about allogeneic anti-leukemic cytotoxic T lymphocytes (CTLs) have been published,14,15 suggesting impaired leukemia-specific immune responses in leukemic patients.16,17 Possible targets for such an immune response might be the p210 product of the bcr/abl rearrangement, PML/RAR peptides, p53 mutations, and EBNA-1. Cancer evolves from a series of mutational events within a cell that can result in the production of genetically altered

proteins.18 Peptides from these mutated proteins may bind to major histocompatibility complex (MHC) class I molecules, and in this context, may serve as targets for specific cytotoxic T cells (CTL).19 Tumor-associated antigens that provoke tumor rejection in the host have been demonstrated in experimentally, chemically, and virally induced tumors. But in most experimental, spontaneous tumor systems (resembling the situation found in most patients), tumor-associated antigens usually cannot be detected. Recent experiments have shown that several poorly immunogenic solid tumors can be recognized by MHC-class restricted CD8+ CTL if the tumors are engineered to secrete one of several cytokines or costimulatory molecules.20–29 In addition, secretion of cytokines by the tumor cells stimulated the host’s immune system to identify and kill the untransduced parental cells upon reimplantation; and even more, a rejection of established cancer can occur by inducing a systemic anticancer immune response by vaccination with cytokine-transduced tumor cells.21,22–30 Leukemic cells may escape immune surveillance by not expressing leukemia-specific antigens or by not expressing signals that are essential for activation of the host immune system.31 At the molecular level, the defective signaling of leukemic cells might be attributable to (1) down-regulation of major histocompatibility complex (MHC) molecules; (2) alteration of antigen-presenting pathways, resulting in an inability to present tumor-specific antigens to host T cells; (3) absence of costimulatory or adhesion molecules that are essential for activation of the host immune system; or (4) production of factors that modify host immune responses. Murine models have shown IL-2 to be effective in preventing relapse after BMT in a B cell leukemia (BCL1);32 combined IL-1/IL-2 therapy inhibited metastatic tumor growth of Friend erythroleukemia cells (FLC);33 continuous coadministration of M-CSF and IL-2 protected mice against a lethal dose of T cell leukemic cells (EL4);34 IFN-alpha gene transfer into Friend erythroleukemia cells (FLC) abrogated tumorigenicity, and injections of those IFN-alpha producing cells were effective in inhibiting tumor growth in mice with established metastatic tumors.35 These described murine tumor models used virally or chemically induced cancers, which are already immunogenic to begin with.26 We have developed a model system to study this tumor–vaccination approach using a nonimmunogenic murine myeloid leukemia which spontaneously arose (WEHI3) to test the hypothesis that a non-immunogenic leukemia can be used in the development of a leukemia vaccine.

Materials and methods

Mice Correspondence: S de Vos, Division Hematology/Oncology, CedarsSinai-Medical Center, UCLA School of Medicine, 8700 Beverly Blvd, Los Angeles, CA 90048, USA; Fax: 310 562 8411 Received 22 April 1997; accepted 17 November 1997

Pathogen-free female BALB/c mice, 10–14 weeks old, were obtained from Harlan-Sprague Dawley (Indianapolis, IN, USA).

Immunotherapy against murine leukemia S de Vos et al

402

WEHI3 WEHI3, a rapidly fatal acute myelomonocytic leukemia, spontaneously developed is of Balb/c origin: the cells are nonimmunogenic and express class I MHC molecules.36 The cell line was obtained from ATCC (Rockville, MD, USA) and was maintained in IMDM/10% fetal calf serum (FCS). Cells were washed three times in PBS before being injected into mice.

Retroviral vectors One vector used for transduction contained the LN-based G1Nab CVhIL-2 construct (titer range 1–5 × 104 c.f.u./ml) in which the neoR selectable marker was driven by the Moloney murine leukemia virus long terminal repeat (LTR), and the human IL-2 cDNA was driven by the cytomegalovirus (CV) early enhancer/promoter. The vector was provided as frozen supernatant from Genetic Therapy (Gaithersburg, MD, USA). The JZEN hIL-7/tk neo (titer range 1–10 × 106 c.f.u./ml) was constructed with the neoR gene driven by the thymidine kinase promoter, and the hIL-7 cDNA was under the transcriptional control of the myeloproliferative sarcoma virus LTR (Graeme J Dougherty, Terry Fox Lab., Vancouver, Canada).37 This construct was packaged in the GP + env AM 12 amphotrophic cell line and subcloned to produce high-titer stocks.38 The MGF-mGM-CSF vector was MoMLV-derived, without any selectable marker sequences, encoding murine GM-CSF; a high-titer packaging cell line psi-CRIP was kindly provided by Richard Mulligan (Whitehead Institute, Cambridge, MA, USA).

Table 1 Leukemogenicity of the WEHI3 cell line in the syngeneic host

i.v. injected cells/mouse

Death due to leukemia post-injection (days) Balb/c–WEHI3

107 106 105 5 × 104 104 103 102

11 16 20 28 27 (50% survival) 100% survival 100% survival

The data represent the medium time in days to death due to leukemia for four mice per i.v. inoculum of WEHI3 leukemia cells into syngeneic Balb/c mice.

Vaccination as treatment for leukemia The same vaccination schedule as used for preimmunization studies was started 3 days after i.v. injection of 5 × 104 WEHI3 cells per mouse, with 10 mice per group. Mice were followed for development of leukemia and survival.

Statistical analysis The survival of the different subgroups was expressed as Kaplan–Meier curves, followed by comparisons using the logrank ␹2 test with corresponding P values. A P value below 0.05 indicates significant different survival times of different subgroups.

Gene transfer Results About 1–2 × 106 exponentially growing leukemic target cells were grown in 10 ml supernatant of high-titer retroviralpackaging cell lines in the presence of 4–8 ␮g/ml polybrene for 2–18 h. Following selection in cultures containing the neomycin-analog G418 (0.5–1.0 mg/ml bioactive G418) for 1–2 weeks, the surviving successfully transduced cells were subcloned by methylcellulose soft-gel culture. Individual colonies were plucked and expanded in liquid culture. Using ELISA assays (hIL-2- and hIL-7-Quantikine; R+D Systems, Minneapolis, MN, USA) to measure the protein expression, high/low cytokine-producing clones were identified, expanded, and viably frozen. The MGF vector does not use any selectable marker; therefore, directly after transduction, the leukemic cells were cloned in soft-gel culture and tested for mGM-CSF production using an ELISA assay. Transduction efficacy was 15%.

Vaccination and leukemia challenge Either cytokine-producing leukemic cells or a mixture of cytokine-producing NIH-3T3 fibroblasts with unaltered parental leukemic cells were used for immunization. After irradiation (WEHI3: 1000 rad; NIH-3T3-fibroblasts: 10 000 rad) to inhibit in vivo growth while preserving cytokine production, the vaccine preparation was injected s.c. every 7 days for up to 4 weeks. One week later, mice were challenged s.c. or i.v. with non-transduced, parental leukemic cells. Mice were followed for development of leukemia and survival.

Murine leukemia models Survival of mice injected with increasing i.v. doses of syngeneic leukemia cells is shown in Table 1.

Leukemia vaccines The WEHI3 leukemia line was transduced with the JZEN hIL7/tk neo, only the neoR-encoding control retroviral vectors, or with the MGF-mGM-CSF vector. The cytokine production of 106 cells/24 h from stably transduced subclones and a subclone of NIH-3T3 fibroblasts, transduced with the G1Na CVhIL-2 retroviral vector is shown in Table 2. Characterization of these subclones showed no difference in either their Table 2 Subclones genetically engineered to produce hIL-2, hIL7, or mGM-CSF

Transduced cell line

Retroviral vector

Subclone

Cytokine production of 106 cells/24 h

WEHI3

JZEN hIL-7

2 23 polyclonal 16 5

2680 pg 8250 pg 7860 pg 18 500 pg 21 000 pg

NIH 3T3

MGF mGM-CSF G1NaCV IL-2


morphologies or in vitro growth rate in comparison to the parental cells. The hIL-7, hIL-2, and mGM-CSF production were stable over months. Following irradiation, the cytokine production continued for up to a week at a level of two thirds of the non-irradiated cells (data not shown). These clones were used for further in vivo experiments. In vivo growth of hIL-7transduced leukemia clones was compared to the parental cell line. In the WEHI3 model, no statistically significant differences could be observed in the survival times of mice injected with unirradiated mGM-CSF or hIL-7-transduced subclones in comparison to the unaltered parental cells (data not shown).

Vaccination studies Mice were vaccinated with four s.c. injections (1–5 × 106 cells; one injection weekly) with one of a panel of different vaccine preparations. Subsequently, the mice were challenged with i.v. injections (5 × 104, in a volume of 300 ␮l) of the parental cell line. As shown in the second column of Figure 1, WEHI3 was a low-immunogenic cell line. The Balb/c mice vaccinated with high IL-7-producing WEHI3 cells (clone No. 23 and the polyclonal preparation) showed systemic protection and rejected an i.v. challenge with 5 × 104 parental WEHI3 cells (43% and 40% survival (⬎120 days), respectively). Compared to controls, these prolonged survival times were statistically highly significant. The P values were 0.0045 using the IL-7-producing WEHI3-clone No. 23, and 0.0079 using the IL-7-producing WEHI3-polyclonal preparation. The difference between these two treatment groups were not statistically significant. Vaccination with a low IL-7producing WEHI3 clone (clone No. 2) showed no protection, with all mice becoming moribund by day 24. When vaccinated with a mixture of 106 parental WEHI3 cells and IL-2-

producing NIH-3T3 fibroblasts (5 × 105 (low dose) or 5 × 106 (high dose)), only the preparation with 5 × 105 IL-2 producers showed systemic protection upon a subsequent i.v. challenge with parental WEHI3 cells (60% survival); with a 10-fold higher local IL-2 ‘dose’ (5 × 106 IL-2-producing cells), this protective effect was no longer present. When mixing (5 × 106) IL-2-producing NIH-3T3 cells (no protection when used alone) with 106 IL-7-producing WEHI3 cells (43% protection when used alone) for vaccination, three out of five mice survived after challenge with the leukemia. Vaccination with mGM-CSF-producing WEHI3 cells (clone No. 16) resulted in only 20% survival of i.v. challenged mice. The combination of IL-2-, IL-7- and GM-CSF-producing cells as a vaccine provided no greater protection than the IL-2-producing cells alone. The surviving mice were disease-free as shown by HEstained histologic sections of bone marrow, spleen and liver (data not shown). Attempts to dissect the immune response and attribute it to natural killer (NK) cells, CD4+ or CD8+ cells in vitro was technically hampered by poor uptake of chromium by WEHI3 cells.

Vaccination as treatment for leukemia The same vaccination schedules as used for studies shown in Figure 1 were started 3 days after i.v. injection of 5 × 104 WEHI3 cells per mouse (10 mice per group). No significant difference in development of leukemia/survival was detected between treated and control groups (data not shown). Discussion We have shown that transduction of the non- or low-immunogenic murine leukemia line WEHI3 with the gene for human

Figure 1 Survival of vaccinated and i.v.-challenged Balb/c mice. 106 WEHI3 leukemic cells and 5 × 106 (low dose) or 5 × 106 (high dose) IL-2-producing NIH-3T3 fibroblasts (after irradiation) were given s.c. every week × 4 as a vaccine. Balb/c mice were then challenged with WEHI3 (104, i.v.) and time to death recorded.

403


404

IL-7 elicits a systemic anti-leukemic immune response and causes rejection of leukemia in 43% of vaccinated syngeneic hosts. This effect was dose-dependent; only leukemic cells expressing high levels of IL-7, as compared to lower level producers, stimulated an immune response. To rule out the possibility that the differences in vaccination potential of WEHI3 clones 2 and 23 is due to properties intrinsic to individual clones rather than the secreted IL-7, we vaccinated with a polyclonal IL-7-transduced population as well. Using the same vaccination approach with a mixture of hIL2-producing NIH-3T3 ‘bystanding’ fibroblasts with parental WEHI3 leukemic cells, we saw a protective immune response against WEHI3 leukemia. The observation that the protective vaccination effects vanished with a 10-fold increase of local IL-2 production at the vaccination site is in accord with previous reports of an ‘optimal dose’ of local IL-2 production with no protective effect either below or above this ‘window’.39 The mGM-CSF has been reported to be the most effective cytokine in some cancer vaccine models.26 However, in our model, vaccination with mGM-CSF-producing WEHI3 cells resulted in only 20% survival of i.v. challenged mice, and the combination of mGM-CSF-, IL-2- and IL-7-producing vaccines did not lead to either additive or synergistic effects. The level of mGM-CSF production in WEHI cells is less than what has been seen in other cell lines using MFG-based vectors. This is most likely due to an expression problem in WEHI3 cells. Therefore, it is conceivable that increases in the expression of mGM-CSF in this system still might result in more effective immunization. No differences in development of leukemia and survival times could be detected when non-irradiated, viable, IL-7and mGM-CSF-producing WEHI3 cells were injected i.v. as compared to the unaltered parental leukemia cells, demonstrating the absence of any immediate strong immune response. However, within 4 weeks, the mice could be successfully pre-immunized with irradiated cytokine-producing leukemic cells, indicating that low-immunogenic murine leukemia cells can be rendered immunogenic to some extent by this transduction/vaccination approach. This vaccination strategy, used as treatment for pre-established leukemia was unsuccessful. No significant difference in development of leukemia survival was detected between treated and control groups. Two explanations for this outcome are possible: (1) the levels of either the T cell growth factors (IL-2, IL-7) and/or potential enhancement of antigen presentation (GMCSF) may not have been great enough to amplify any inherent immune response or break possible immuno-tolerance; and (2) the leukemia developed too rapidly to give an emerging immune response a period of time to prevent progression of the leukemia. Other models of immune modulation of leukemia showed effectiveness only in the state of ‘minimal disease’.40 Taken together, we identified IL-7 and IL-2 as effective cytokines in our non-immunogenic WEHI3 leukemia/vaccination model, with only marginal immunostimulation by GM-CSF. Immunizing mice with a pre-established leukemia burden did not cure or result in a prolongation of survival, indicating that improved methods of immunization are needed.

References 1 Marx J. Cancer vaccines show promise at last news. Science 1989; 244: 813–815. 2 Mathe G. Active immunotherapy. Adv Cancer Res 1971; 14: 1–36.

3 Wittaker JA. Annotation: Immunotherapy in the treatment of acute leukemia. Br J Haematol 1980; 45: 859–865. 4 Reizenstein P. Adjuvant immunotherapy with BCG of acute myeloid leukemia: a 15-year follow up. Br J Haematol 1990; 75: 288–299. 5 La Cour Petersen E, Hokland P, Ellegaard J. Adjuvant stimulation with Corynebacterium parvum during maintenance chemotherapy of acute leukemia. Cancer Immunol Immunother 1983; 16: 88–92. 6 Mathe G. Passive, adoptive, and active immunotherapy: a review of clinical trials in cancer. Cancer Detect Prev 1987; 1 (Suppl.): 279–290. 7 Butturini A, Bortin MM, Gale RP. Graft versus leukemia following bone marrow transplantation. Bone Marrow Transplant 1987; 2: 233–242. 8 Brenner MK, Heslop HE. Immunotherapy of leukemia. Leukemia 1992; 6: 76–79. 9 Adler A, Chervenick PA, Witeside TL, Lotzova E, Herberman RB. Interleukin 2 induction of lymphokine-activated killer (LAK) activity in the peripheral blood and bone marrow of acute leukemia patients. II. Feasibility of LAK generation in children with active disease and in remission. Blood 1989; 74: 1690–1697. 10 Archimbaud E, Bailly M, Dore JF. Inducibility of lymphokine activated killer (LAK) cells in patients with acute myelogenous leukaemia in complete remission and its clinical relevance. Br J Haematol 1991; 77: 328–334. 11 Lotzova E, Savary CA, Herberman RB. Induction of NK cell activity against fresh human leukemia in culture with interleukin 2. J Immunol 1987; 138: 2718–2727. 12 Dawson MM, Johnston D, Taylor GM, Moore M. Lymphokine activated killing of fresh human leukemias. Leukemia Res 1986; 10: 683–688. 13 Oshimi K, Oshimi Y, Akutsu M, Takei Y, Saito H, Okada M, Mozoguchi H. Cytotoxicity of interleukin-2 activated lymphocytes for leukemia and lymphoma cells. Blood 1986; 68: 938–948. 14 Montagna D, Arico M, Montiti E, De Benedetti F, Maccario R. Identification of HLA-unrestricted CD8+/CD28− cytotoxic T-cell clones specific for leukemic blasts in children with acute leukemia. Cancer Res 1995; 55: 3835–3839. 15 Fisch P, Weil-Hillmann G, Uppenkamp M, Hank JA, Chen BP, Sosman JA, Bridges A, Colamonici OR, Sondel PM. Antigen-specific recognition of autologous leukemia cells and allogeneic class-I MHC antigens by IL-2 activated cytotoxic T cells from a patient with acute T-cell leukemia. Blood 1989; 74: 343–353. 16 Sosman JA, Oettel KR, Smith SD, Hank JA, Fisch P, Sondel PM. Specific recognition of human leukemic cells by allogeneic T cells. II. Evidence for HLA-D restricted determinants on leukemic cells that are crossreactive with determinants present on unrelated nonleukemic cells. Blood 1990; 75: 2005–2016. 17 Oettel KR, Wesley OH, Albertini MR, Hank JA, Iliopolis O, Sosman JA, Voelkerding K, Wu SQ, Clark SS, Sondel PM. Allogeneic T-cell clones able to selectively destroy Philadelphia chromosome-bearing (Ph1+) human leukemia lines can also recognize Ph1− cells from the same patient. Blood 1994; 83: 3390–3402. 18 Townsend ARM, Rothbard J, Gotch FM, Bahadur G, Wraith D, McMichael AD. The epitopes of influenza nucleoprotein recognized by cytotoxic T lymphocytes can be defined with short synthetic peptides. Cell 1986; 44: 959–968. 19 Lurquin C, Van Pel A, Mariame B, De Plaen E, Szikora JP, Janssens C, Reddehase MJ, Lejeune J, Boon T. Structure of the gene of tumtransplantation antigen P91A: the mutated exon encodes a peptide recognized with Ld by cytolytic T cells. Cell 1989; 58: 293–303. 20 Fearon ER, Pardoll DM, Itaya T, Golumbek P, Levitsky JW, Simons H, Karasuyama H, Vogelstein B, Frost P. Interleukin-2 production by tumor cell bypasses T helper function in the generation on an antitumor response. Cell 1990; 60: 397–403. 21 Gansbacher B, Zier K, Daniels B, Cronin K, Gilboa E. Interleukin 2 gene transfer into tumor cells abrogates tumorigenicity and induces protective immunity. J Exp Med 1990; 172: 1217–1224. 22 Tepper RI, Pattengale PK, Leder P. Murine IL-4 displays potent anti-tumor activity in vivo. Cell 1989; 57: 503–512. 23 Golumbek PT, Lazenby AJ, Levitzky HI, Jaffee LM, Karasuyama H, Baker M, Pardoll DM. Treatment of established renal cancer by tumor cells engineered to secrete interleukin-4. Science 1991; 25: 713–716. 24 Dougherty GJ, Thacker JD, Comora SH, McBride WH. Inhibitory


25

26

27

28

29 30

31

32

effect of locally produced IL-6 on tumor growth in vivo. Cancer Immunol Immunother 1994; 38: 339–345. McBridge WH, Thacker JD, Comora S, Economou JS, Kelley D, Hogge D, Dubinett SM, Dougherty GJ. Genetic modification of a murine fibrosarcoma to produce interleukin 7 stimulates host cell infiltration and tumor immunity. Cancer Res 1992; 52: 3931– 3937. Dranoff G, Jaffee E, Lazenby A, Golumbek P, Levitsky H, Brose K, Jackson V, Hamada H, Pardoll D, Mulligan RC. Vaccination with irradiated tumor cells engineered to secrete murine granulocyte–macrophage colony-stimulating factor stimulates potent, specific, and long-lasting anti-tumor immunity. Proc Natl Acad Sci USA 1993; 90: 3539–3543. Asher AL, Mule JJ, Kasid A, Restifo NP, Salo JC, Reichert CM, Jaffe G, Fendy B, Kriegler M, Rosenberg SA. Evidence for a paracrine immune effect of tumor necrosis factor against tumors. J Immunol 1991; 146: 3227–3234. Gansbacher B, Bannererij R, Daniels B, Zier K, Cronin K, Gilboa E. Retroviral vector-mediated interferon-gamma gene-transfer into tumor cells generates potent and long lasting antitumor immunity. J Exp Cancer Res Med 1990; 50: 7820–7825. Townsend SE, Allison JP. Tumor rejection after direct constimulation of CD8+ T cells by B7-transfected melanoma cells. Science 1993; 259: 368–370. Porgador A, Tzehoval E, Katz A. IL-6 gene transfection into LL carcinoma tumor cells suppresses the malignant phenotype and confers immunotherapeutic competence against parental metastatic cells. Cancer Res 1992; 52: 3679–3686. Muul LM, Spiess PJ, Director EP, Rosenberg SA. Identification of specific cytolytic immune responses against autologous tumor in humans bearing malignant lymphoma. J Immunol 1987; 138: 989–995. Ackerstein A, Kedar E, Slavin S. Use of recombinant human interleukin-2 in conjunction with syngeneic bone marrow transplantation in mice as a model for control of minimal residual dis-

33

34

35

36 37

38 39

40

ease in malignant hematologic disorders. Blood 1991; 78: 1212–1215. Ciolli V, Gabriele L, Sestili P, Varano F et al. Combined interleukin-1/interleukin-2 therapy of mice injected with highly metastatic Friend leukemia cells: host antitumor mechanisms and marked effects on established metastases. J Exp Med 1991; 173: 313–322. Vallera DA, Taylor PA, Aukerman SL, Blazar BR. Antitumor protection from the murine T-cell leukemia/lymphoma EL4 by the continuous subcutaneous coadministration of recombinant macrophage colony-stimulating factor and interleukin-2. Cancer Res 1993; 53: 4280. Ferrantini M, Proietti E, Santodonato L, Gabriele L et al. Alphainterferon gene transfer into metastatic Friend leukemia cells abrogated tumorigenicity in immunocompetent mice: antitumor therapy by means of interferon-producing cells. Cancer Res 1993; 53: 1107–1112. Ralph P, Nakoinz I. Direct toxic effects of immunopotentiators on monocytic, myelomonocytic, and histiocytic or macrophage tumor cells in culture. Cancer Res 1977; 37: 546–550. McBride WH, Thacker JD, Comora S, Economou JS, Kelley D, Hogge D, Dubinett SM, Dougherty GJ. Genetic modification of a murine fibrosarcoma to produce interleukin-7 stimulates host cell infiltration and tumor immunity. Cancer Res 1992; 52: 3931– 3937. Markowitz D, Goff S, Bank A. Construction and use of a safe and efficient amphotropic packaging cell line. Virology 1988; 167: 400–406. Miller AR, McBride WH, Moen RC, Schuck BC, Glasby SE, Economou JS. Interleukin-2 producing fibroblast cell line abrogates tumorigenicity in a murine tumor model: colon. An approach to genetically engineered vaccines. Surgical Forum 1993; 49: 512. Hagenbeek A, de Grot CJ, Martens ACM. Cytokine gene mediated immunotherapy of leukemia. Studies in the Brown Norway rat acute myelomonocytic leukemia employing the interleukin-2 gene. Blood 1993; 82 (Suppl.): 254a.

405