Correlation between enhancement of graft-versus-leukemia effects following allogeneic bone marrow transplantation by rIL-2 and increased frequency of cytotoxic T-lymphocyte precursors in murine myeloid leukemia.

A model of mouse acute myeloid leukemia (mAML) was used to study the effector mechanism mediating the graft-versus-leukemia (GVL) effects in recipients of allogeneic bone marrow cells (BMC). mAML-bearing SJL/J (H-2s) mice were lethally irradiated and then transplanted with a mixture of BMC and spleen cells (SC) derived from normal syngeneic or allogeneic mice. To augment the GVL effect, recipients were injected intraperitoneally with recombinant human interleukin-2 (rIL-2) (1.2 x 10(5) IU) for 3 consecutive days, starting one day post BMC + SC transplantation. Spleen cells from treated recipients were adoptively transferred to untreated secondary SJL/J mice to test for the existence of residual tumor cells. All the secondary recipients of SC from mAML-bearing SJL/J mice rescued with syngeneic (SJL/J) or allogeneic (B10.S) BMC+SC (H-2s) differing at minor antigens of the histocompatibility complex (MiHC) developed leukemia and died. In sharp contrast, none of the secondary recipients of SC obtained from identical mAML-bearing mice rescued with B10.S BMC + SC but activated in vivo with IL-2 developed leukemia. Adoptive recipients of SC obtained from mAML-bearing recipients of major histocompatibility complex (MHC)-disparate (C57BL/6, H-2b) cells remained free of leukemia regardless of the use of rIL-2. In parallel with the in vivo findings, a 4-day in vitro exposure of splenocytes to 6 x 10(3) IU/ml rIL-2 resulted in a 5- to 20-fold increase in the frequency of alloreactive cytotoxic T-lymphocyte (CTL) precursors (CTLp) across MiHC and MHC barriers and a 2- to 6-fold increase in their cytotoxic activity. Our data suggest that augmentation of GVL effects by rIL-2 may be due to CTL activation by rIL-2, not excluding the potential beneficial role of rIL-2-activated allogeneic natural killer cells and MHC non-restricted killer cells. Cumulatively, our results suggest potentially beneficial effects of rIL-2, when used jointly with bone marrow transplantation or allogeneic cell therapy, on eradication of leukemia.

Acute/relapsing experimental autoimmune encephalomyelitis: induction of long lasting, antigen-specific tolerance by syngeneic bone marrow transplantation.

Experimental autoimmune encephalomyelitis (EAE) is an inducible autoimmune disease widely used as a model of the acute/relapsing stage of multiple sclerosis. We have previously shown that treatment of EAE-mice with high doses of cyclophosphamide (CY) (350 mg kg), followed by syngeneic bone marrow transplantation (SBMT), completely abrogates the clinical paralytic signs and even prevents the appearance of new relapses in the chronic-relapsing model of the disease. In the present study we examined whether this treatment protocol induces long term tolerance and whether this tolerance is antigen-specific. EAE was induced by immunization with spinal cord homogenate (MSCH) in complete Freund's adjuvant (CFA). The treatment with CY and SBMT was performed on day 6 post immunization. Treated and untreated mice were rechallenged with MSCH, or a non-relevant antigen (OVA) in CFA at various stages after the first paralytic attack. In contrast to previous data showing that animals recovering from acute EAE are usually refractory to re-induction of the disease, repeated injections of MSCH at different sites from the initial immunization, followed by i.v. injection of inactivated Bordetella bacteria, 2, 4 and 6 months after the initial EAE-induction, caused a severe and usually lethal relapse in all the untreated, control animals. Mice treated with CY and SBMT were resistant to all rechallenges with the same encephalitogenic inoculum. Following the second rechallenge, peripheral lymph node cells were examined in vitro for their proliferative responses to myelin antigens or to OVA. Lymphocytes obtained from CY+SBMT treated mice did not proliferate in vitro in response to myelin basic protein (MBP), but proliferated against OVA, when immunized with this antigen, after SBMT. Adoptive transfer of lymphocytes from tolerant mice to naive recipients did not transfer resistance to EAE-induction. Our results indicate that high doses of CY, followed by SBMT, induce long term antigen-specific tolerance presumably by a mechanism of clonal deletion or anergy.

Allogeneic cell-mediated immunotherapy for eradication of minimal residual disease: comparison of T-cell and IL-2 activated killer (LAK) cell-mediated adoptive immunotherapy in murine models.

In the course of allogeneic bone marrow transplantation (BMT), avoiding graft-versus-host disease (GVHD) while retaining the antileukemic effects of the T cells remains a major challenge. T-cell depletion (TCD) reduces the incidence of GVHD but increases the relapse rate after allogeneic BMT. We attempted to develop a regimen that would retain or increase the graft-versus-leukemia effect induced by donor T cells while preventing GVHD. Immunosuppressed mice were given immunocompetent donor cells, i.e., fresh lymphocytes or lymphokine-activated killer (LAK) cells differing from the host in major (MHC) or minor (MiHC) histocompatibility antigens. Engraftment of donor cells was documented by polymerase chain reaction analysis. Administration of MHC- and MiHC-incompatible allogeneic LAK cells, especially in conjunction with recombinant interleukin-2 (rIL-2), increased disease manifestations and mortality associated with GVHD. On the other hand, irradiated LAK cells or TCD-LAK cells prevented GVHD in both mice models studied. Phenotypic analysis of LAK cells demonstrated that CD8(+)-equivalent (Lyt-2) T cells are of significance in aggravation of GVHD. The in vitro cytotoxic capacity of LAK cells against MHC-nonrestricted target cells was not reduced by either irradiation or TCD. These results provide the background for designing improved protocols for immunotherapy of residual disease after BMT. In addition, the data imply that antitumor effects may be retained by irradiated rIL-2-activated allogeneic cells without causing GVHD. Whereas unmodified allogeneic LAK cells can induce more effective graft-versus-leukemia effects at the cost of GVHD, irradiated allogeneic donor LAK cells might play some role in eradication of minimal residual disease following autologous or allogeneic BMT without causing GVHD.

Failure of cyclosporine to induce graft-vs-host disease or graft-vs-leukemia after syngeneic bone marrow transplantation in mice.

Cyclosporine administration can result in graft-vs-host disease (GVHD) after syngeneic or autologous bone marrow transplantation (BMT). However, data on its anti-tumor effects are limited. We have tried to produce cyclosporine-induced GVHD or graft-vs-leukemia (GVL) against two Ia-bearing murine leukemias. BALB/c mice undergoing syngeneic BMT after total-body irradiation received 1 mg/kg or 10 mg/kg cyclosporine or dextrose intraperitoneally for 30 days post-transplant, followed by 5 x 10(3) BCL1 murine leukemia cells 1 or 3 weeks after stopping cyclosporine. Similarly, SJL/J mice undergoing syngeneic BMT received 10 mg/kg cyclosporine or dextrose intraperitoneally for 30 days post-transplant, and were inoculated with 5 x 10(3) or 10(5) murine acute myeloid leukemia (mAML) cells 3 weeks after stopping cyclosporine. No clinical or histological evidence of GVHD was seen, and leukemia-free and overall survival was similar with cyclosporine and dextrose. We conclude that under the experimental conditions used, it is not possible to induce syngeneic GVHD with cyclosporine in BALB/c or SJL/J mice. In the absence of GVHD, this approach is also not associated with any GVL effect against murine leukemias BCL1 and mAML.

Immunomodulation of autoimmunity in MRL/lpr mice with syngeneic bone marrow transplantation (SBMT).

MRL-lpr/lpr mice spontaneously develop a severe autoimmune syndrome, characterized by massive generalized lymphadenopathy, arthritis, arteritis, dermatitis and immune complex-mediated glomerulonephritis. Bone marrow transplantation (BMT) from MHC-matched systemic lupus erythematosus (SLE)-resistant donors to susceptible recipients has proved effective in correcting autoimmune manifestations in autoimmune-prone mice. We investigated the effect of syngeneic BMT from MRL/lpr (donor) to immunocompromised MRL/lpr (recipient), after purging the bone marrow inoculum with MoAbs against mature T cells (anti-Thy 1.2). All the untreated mice developed lymphadenopathy and by the age of 36 weeks five of the eight were dead; in contrast, all the mice which underwent syngeneic BMT following acute immunosuppression with total body irradiation (900 cGy) (TBI) remained disease-free. In an additional experiment, it was found that conditioning with cyclophosphamide (CY) before BMT was more effective than TBI in inhibiting delayed-onset autoimmune manifestations (mean survival 350 days in the CY group and 305 days in the TBI group, versus 197 days in untreated controls). Under both immunosuppressive regimens T cell-depleted bone marrow grafts produced far better results than did unmanipulated BMT. Following syngeneic BMT the incidence of proteinuria and the level of serum anti-DNA (dd) antibodies were significantly reduced, compared with that of the age-matched untreated controls. CY was more effective than TBI in reducing the anti-DNA titres. Likewise, T depletion of bone marrow inocula before BMT induced a more drastic drop in autoantibodies, following both CY and TBI conditioning protocols. After syngeneic BMT (either CY or TBI) no signs of lymphadenopathy were observed even at an advanced age. Upon histopathological examination, the BMT-treated mice displayed normal glomeruli with occasional minimal signs of glomerulonephritis. Syngeneic T cell-depleted BMT following acute cytoreduction of anti-self immune lymphocytes may represent a new therapeutic approach for drug-resistant autoimmune diseases.

Enhancement of GVL effect with rhIL-2 following BMT in a murine model for acute myeloid leukemia in SJL/J mice.

A murine model for acute myeloid leukemia (mAML) was used to study graft-vs.-leukemia (GVL) effects on residual leukemic cells across both major (MHC) and minor histocompatibility antigens (mHA) barriers. In addition, the therapeutic effect of recombinant human interleukin-2 (rhIL-2)-administered postsyngeneic and allogeneic bone marrow transplantation (BMT) was examined. SJL/J mice inoculated with mAML cells were exposed later to total body irradiation (TBI) and transplanted with bone marrow cells (BMC) mixed with spleen cells derived from normal syngeneic (SJL/J), congenic (B10.S), or allogeneic (C57BL/6) donor mice. One-half of the mice in each group received low dose rhIL-2 for 3 days starting 1 day post-BMT. Spleen cells from treated recipients were transferred to secondary naive SJL/J mice for in vivo detection of residual tumor cells. At a tumor load of 10(5) cells per animal, none of the mice rescued with SJL/J or B10.S cells was cured since 100% of secondary recipients developed leukemia. Concomitant treatment of recipients of B10.S cells with rhIL-2 induced GVL effects since none of the secondary recipients developed leukemia after 2 years. All adoptive recipients of mice rescued with C57BL/6 cells remained free of leukemia after 2 years whether or not rhIL-2 was injected. The potency of the GVL effects observed across mHA and MHC were tumor-cell dose dependent since fewer animals inoculated with 10(6) mAML cells were cured. Only marginal GVL effects were noticed following syngeneic BMT and rhIL-2. Our results sustain the importance of the GVL effects in the treatment of myeloid leukemia and demonstrate that immunotherapy with rhIL-2 following BMT can enhance the therapeutic effect induced by the allograft.

Inhibition of acute, experimental autoimmune encephalomyelitis by the synthetic immunomodulator linomide.

Linomide (LS-2616, quinoline-3-carboxamide) is a synthetic immunomodulator that stimulates natural killer cell activity and activates several lymphocytic subpopulations in experimental animals and humans. In this study we determined the effect of oral treatment with linomide on the development of experimental autoimmune encephalomyelitis, an animal model for immune-mediated human demyelinating disorders. Experimental autoimmune encephalomyelitis was induced in SJL/J mice and in an outbred strain of rats (Sabra) by subcutaneous injection of spinal cord homogenate in adjuvant followed by inoculation with Bordetella pertussis. Linomide was administered in drinking water, at an estimated dose of 50 to 100 mg/kg/day. None of the linomide-treated mice (0/41) and Sabra rats (0/15) developed any clinical or pathological signs of experimental autoimmune encephalomyelitis, whereas almost all control animals (48/53 and 18/19, respectively) were severely paralyzed and 64.5% died from the disease. Lymphocytes obtained from linomide-treated animals had reduced in vitro proliferative responses to guinea pig myelin basic protein, proteolipid protein of the myelin, and tuberculin-purified protein derivative, unlike antigen-independent proliferation which was rather unaffected. Natural killer cell activity (tested by a cytotoxic assay on radiolabeled YAC-1 target cells) was significantly enhanced in mice treated with linomide. Our results indicate that modulation of the immune system with linomide leads to complete inhibition of experimental autoimmune encephalomyelitis in the absence of systemic immunosuppression. Linomide could therefore be of use in future clinical trials for the treatment of human autoimmune demyelinating disorders.

Spontaneous and IL-2-induced anti-leukemic and anti-host effects against tumor- and host-specific alloantigens.

The graft-vs-leukemia (GVL) effect and its connection to graft-vs-host disease (GVHD) were studied in F1 hybrid mice after parental bone marrow transplantation. (BALB/c x C57BL/6) F1 mice (F1) were sublethally and lethally irradiated and inoculated with 10(4) BCL1 cells, a murine leukemia of BALB/c (BALB) origin. Subsequently mice were transplanted with F1 (syngeneic), BALB (allogeneic to the host but syngeneic to the tumor), C57BL/6 (B6) (allogeneic to both host and tumor) spleen cells, or spleen cells mixed with syngeneic (F1) bone marrow cells, respectively. Fresh BALB (H-2d) and B6 (H-2b) spleen cells, which induce equally strong GVHD in the F1 recipient against H-2b and H-2d alloantigens, respectively, were assayed for their ability to induce GVL effects on BCL1 (H-2d) cells with and without in vivo administration of human rIL-2. In order to increase the GVL effects, spleen cells were incubated for 4 days with rIL-2 to generate IL-2-activated killer cells. The results of adoptive transfer experiments showed that only immunocompetent B6 cells led to consistent eradication of leukemia. Neither BALB nor syngeneic F1 cells induced curative GVL effects regardless of whether spleen cells were further stimulated with rIL-2 in vitro, in vivo, or both. Our data support the theory that GVL effects are caused by T cell-dependent immune responses of allogeneic T cells and not by MHC-nonrestricted NK or rIL-2-activated NK cells, independently of the GVHD process itself. Hence, recognition and killing of tumor cells is most probably a MHC-restricted T cell-dependent process.

Prevention and reversal of adoptively transferred, chronic relapsing experimental autoimmune encephalomyelitis with a single high dose cytoreductive treatment followed by syngeneic bone marrow transplantation.

A chronic relapsing form of experimental autoimmune encephalomyelitis (CR-EAE) was induced in SJL/J mice by adoptive transfer of lymph node cells (LNC) sensitized to guinea pig myelin basic protein (GMBP). We examined the efficacy of high dose immunosuppressive regimens (cyclophosphamide [CY] 300 mg/kg or total body irradiation [TBI] 900 cGy) followed by syngeneic bone marrow transplantation (SBMT) in prevention and treatment of already established CR-EAE. Treatment with TBI and SBMT on day 5 after the induction of CR-EAE, just before the onset of clinical signs, completely inhibited the appearance of the paralytic signs. The same treatment, applied 4 d after the clinical onset of the disease, led to a significant regression of the paralytic signs and to a total inhibition of spontaneous relapses during a follow-up period of 2 mo. Challenge of mice with GMBP+CFA 78 d after the passive induction of CR-EAE induced a relapse of the disease 7 d later in almost all of the untreated mice; in contrast, the same challenge given to TBI+SBMT-treated mice caused a delayed relapse (30 d later) in only a minority (3/7) of the challenged mice. In vitro lymphocytic proliferative responses to GMBP and purified protein derivative were significantly lower in TBI/SBMT-treated mice before and after the GMBP challenge, although these mice were fully immunocompetent, as evidenced by their normal lymphocytic proliferation to concanavalin A (ConA) and the FACS analysis of their lymphocytic subpopulations. A similar beneficial therapeutic effect was observed in mice treated with CY followed by SBMT, after the onset of CR-EAE. Our results could support possible clinical applications of similar therapeutic strategies, involving acute immunosuppression followed by stem cell transplantation and retolerization of the reconstituting immune cells in life-threatening neurological and multisystemic autoimmune diseases.

Treatment of chronic-relapsing experimental autoimmune encephalomyelitis with the synthetic immunomodulator linomide (quinoline-3-carboxamide).

Linomide is a synthetic immunomodulator that enhances natural killer cell activity and significantly activates several lymphocytic cell subpopulations in both experimental animals and humans. In this study we examined the effect of linomide (80 mg per kg per day in drinking water) on mice with chronic-relapsing experimental autoimmune encephalomyelitis (CR-EAE), a T-cell-mediated organ-specific autoimmune disease that resembles human multiple sclerosis. None of the mice (n = 17) that were treated with linomide from day 7 after disease induction developed any clinical or histopathological signs of CR-EAE, as compared to 19 of 20 untreated controls that were severely paralyzed and had extensive demyelinating lesions in the central nervous system. Linomide-treated animals were also resistant to an induced attack by a booster injection with a murine spinal cord homogenate. When administered to mice exhibiting severe clinical signs of paralysis, linomide inhibited both spontaneous and induced relapses. Linomide treatment protected mice from passively induced CR-EAE as well, when given from the day of injection with myelin-basic-protein-specific lymphocytes. Lymphocytes obtained from linomide-treated mice had a reduced in vitro proliferative response to the myelin basic protein and to the tuberculin purified protein derivative, whereas the mitogenic response to concanavalin A was not affected. Natural killer cell and lymphokine-activated killer cell activities were enhanced. These results suggest that linomide regulates autoimmunity in the absence of systemic immunosuppression. Since linomide is very well tolerated in experimental animals and humans, it might be used in the treatment of multiple sclerosis.

Investigations on the role of inflammatory and anti-inflammatory agents on the treatment of murine B cell leukemia by recombinant human interleukin-2.

Human recombinant interleukin-2 (rIL-2) was used in an effort to find effective biological therapy against the murine B-cell leukemia (BCL1), a spontaneous, nonimmunogenic, highly lethal leukemia of Balb/c origin. High dose rIL-2 (10(5) Cetus units 3 times a day for 5 days) was proven to be curative for mice inoculated with up to 10(4) BCL1 cells. Considering the fact that mice which showed more clinical signs of rIL-2 related toxicity had higher leukemia-free survival, and in view of the fact that BCL1 is a totally nonimmunogenic tumor, we have investigated the possible role of nonspecific effects of inflammatory and anti-inflammatory agents in order to speculate on the possible therapeutic role of inflammation on the overall effect of rIL-2 in the treatment of BCL1. According to the results presented in this work, it appears that although mediators of inflammation may play some role against leukemia, their overall effect in comparison with high-dose rIL-2 is relatively insignificant.

Prevention and treatment of relapse by bone marrow transplantation.

High, myeloablative doses of chemoradiotherapy represent the treatment of choice for a large number of malignant hematological diseases that cannot be successfully treated with conventional chemotherapy. Residual tumor cells following high dose chemotherapy represent the most common treatment failure, resulting in frequent relapse following autologous bone marrow transplantation (ABMT) and even allogeneic bone marrow transplantation (BMT). Graft vs leukemia (GVL) effects mediated by immunocompetent donor lymphocytes represent a major therapeutic potential of allogeneic BMT which results in reduced rate of relapse, especially when immune interactions between immunocompetent donor's T lymphocytes and host allogantigens is apparent, a reaction which might result in graft vs host disease (GVHD). Recent experiments in animal models of murine leukemias suggest that post-transplant immunotherapy may be successfully accomplished by lymphokine-mediated immunotherapy (LMI) and cell-mediated immunotherapy (CMI). Following allogeneic BMT, provided GVHD can be prevented by T-cell depletion, CMI may be amplified by repeated administration of immunocompetent donor's lymphocytes in graded increments following successful induction of chimerism and sustained hematopoiesis. GVL effects induced by CMI may be further potentiated by in-vivo administration of a short course of recombinant human interleukin-2 (rhIL2). Taken together, our data suggest that post-transplant immunotherapy by cytokines and adoptive cell therapy may successfully prevent relapse in patients at high-risk and even result in complete elimination of tumor cells following overt relapse. Thus, immunotherapy may represent an optimal approach for prevention and treatment of minimal residual disease.

Immunotherapy of minimal residual disease in conjunction with autologous and allogeneic bone marrow transplantation (BMT).

Recent investigations in animal models of human lymphoid and myeloid leukemia suggest that induction of immune-mediated antitumor effects is feasible at the stage of minimal residual disease (MRD) using allogeneic immunocompetent lymphocytes following initial reconstitution with T cell depletion and/or activation of reconstituting syngeneic or allogeneic immune cells by recombinant interleukin-2 (rIL2). Pilot clinical trials in patients with leukemias and lymphomas at high risk to relapse following autologous bone marrow transplantation (BMT) suggest that beneficial antitumor effects may be achieved at the stage of MRD by home immunotherapy as soon as hematopoietic reconstitution occurs using rIL2 (Cetus) and alpha interferon (Roferon A) (Hoffmann LaRoche). Although results obtained from our open trial seem encouraging, prospective randomized trials and longer observation periods are needed in order to confirm immune-mediated antitumor effects in conjunction with autologous BMT in patients with malignant hematological disorders at high risk to relapse. Likewise, it seems that amplification of anti-leukemia effects following allogeneic BMT is feasible by post-transplant infusion of donor's peripheral blood lymphocytes for prevention and/or treatment of relapse.

Chronic-relapsing experimental autoimmune encephalomyelitis (CR-EAE): treatment and induction of tolerance, with high dose cyclophosphamide followed by syngeneic bone marrow transplantation.

We examined the effect of acute immunosuppression with high dose cyclophosphamide (CY), followed by syngeneic T-cell-depleted bone marrow transplantation (SBMT) on chronic-relapsing autoimmune encephalomyelitis (CR-EAE) induced in SJL/J mice by immunization with mouse spinal cord homogenate (MSCH) in adjuvant. Treatment of mice on day 9 post immunization, before the appearance of clinical signs of the disease, delayed the onset of paralysis, but did not affect its clinical course. Treatment on day 2-3 after the first clinical signs led to complete regression of the disease. During a period of 3 months, only one of the 15 mice treated after the the onset of CR-EAE relapsed, as compared to a total of 21 relapses in the 15 untreated animals. A rechallenge with MSCH in adjuvant on day 78 after immunization induced a severe relapse in all untreated mice, with 78% mortality; in contrast, only 25% of mice treated with CY and SBMT relapsed when similarly rechallenged. Lymphocytes from mice treated with CY and SBMT showed reduced in vitro proliferative responses to myelin basic protein (GMBP) and PPD, even after the rechallenge with MSCH. Our results show that high dose CY for elimination of immunocompetent lymphocytes, followed by SBMT rescue, suppresses CR-EAE and induces tolerance to the immunizing antigens. These results may encourage attempts to apply a similar therapeutic principle in life-threatening human neurological autoimmune diseases.

Control of relapse due to minimal residual disease (MRD) by cell-mediated cytokine-activated immunotherapy in conjunction with bone marrow transplantation.

In conclusion, lessons from the animal model of lymphoid leukaemia suggest that in the setting of allogeneic BMT, under certain conditions GVL effects may be separable from GVHD; more specifically, GVL effects may be induced despite development of tolerance of donor cells against allogeneic host alloantigens. The latter phenomenon suggests that either curative GVL effects may be inducible despite subclinical GVHD or alternatively that effector cells of GVL may recognize different tumour-associated targets different from cell surface determinants of 'normal' alloantigens. Alternatively, effector cells of GVL may be distinguished from effector cells of GVHD. It is tempting to suggest that NK and IL2-aspirated NK cells may play a major role as effector cells of GVL in an MHC non-restricted fashion, different from classical CD8+ cytotoxic cells that certainly play a major role in GVHD and GVL. Once proven, the latter hypothesis may help develop new and safer therapeutic approaches since NK cells and products of the NK cell family are unlikely to play a major role, if any, in GVHD. The feasibility of induction of GVL-like effects by MHC non-restricted effector cells, such as that observed by CMI, most likely through cytokine-activated NK cells, seems promising because such effector mechanisms may be utilized clinically through either adoptive transfer of in vitro-activated lymphocytes or activation of lymphocytes in vivo by administration of cytokines such as IL2 and alpha IFN. Similarly, induction of CCI following ABMT may permit establishment of GVL-like effects with no major risk of GVHD. Our animal models suggest that both approaches may be beneficial and perhaps even combined. From a practical standpoint, activation of antitumour effector cells in vivo is much more feasible, in comparison with the cumbersome and expensive technologies for large-scale in vitro manipulation of IL2-activated 'LAK' cells or tumour-infiltrating lymphocytes ('TIL') at dose ranges required for obtaining clinically meaningful responses. No less important is the fact that more potent immunotherapy may be inducible by cytokine combinations (such as IL2 and alpha IFN). We are currently investigating additional cytokine combinations in order to attempt to optimize antitumour effects inducible by allogeneic and syngeneic lymphocytes since it appears logical that amplifying in vivo antitumour responses by multiple cytokine combinations may yield better antitumour effects.