An in-vitro model for tumor immunotherapy with antibody heteroconjugates

CD18 Antibody Application Blocks Unwanted Off-Target T Cell Activation Caused by Bispecific Antibodies

Simple Summary

Bispecific antibodies are a very effective immunotherapy against different types of cancer since they activate T cells in the presence of tumor cells. However, they can cause severe side effects, such as a systemic inflammation called cytokine release syndrome. We aimed to clarify an important mechanism that causes cytokine release syndrome. In cocultures of T cells with endothelial cells or lymphoid cells, application of bispecific antibodies can induce T cell activation and cytokine release in the absence of tumor cells. By blocking the adhesion molecule CD18, this interaction is interrupted and the unwanted T cell activation is diminished. CD18 blockade, however, does not interfere with T cell activation when tumor cells are present. Therefore, CD18 blockade could prevent side effects of bispecific antibodies without decreasing the anti-tumor effect.

Abstract

T cell-recruiting bispecific antibodies (bsAbs) are successfully used for the treatment of cancer. However, effective treatment with bsAbs is so far hampered by severe side effects, i.e., potentially life-threatening cytokine release syndrome. Off-target T cell activation due to binding of bispecific CD3 antibodies to T cells in the absence of target cells may contribute to excessive cytokine release. We report here, in an in vitro setting, that off-target T cell activation is induced by bsAbs with high CD3 binding affinity and increased by endothelial- or lymphoid cells that act as stimulating bystander cells. Blocking antibodies directed against the adhesion molecules CD18/CD54 or CD2/CD58 markedly reduced this type of off-target T cell activation. CD18 blockade—in contrast to CD2—did not affect the therapeutic activity of various bsAbs. Since CD18 antibodies have been shown to be safely applicable in patients, blockade of this integrin holds promise as a potential target for the prevention of unwanted off-target T cell activation and allows the application of truly effective bsAb doses.

Hybrid Antibodies in Cancer Diagnosis and Therapy

Monoclonal Antibodies (Mabs) represent a promising tool for cancer diagnosis and theraphy.

Administration of MAbs alone or conjugated to cytotoxic agents has been attempted but has significant limitations. Another potentially effective approach is the use of bispecific or bifunctional antibodies where the capacity to recognize the tumor cell and the toxic agent or lymphocyte activation molecule are united in one MAb. The hybrid molecule can be produced by chemical linkage between the two parentalantibodies, or alternatively by a biological approach that consists in the fusion of the two selected hybridomas. In the resulting quadroma cell the hybridoma immunoglobulin chains recombine randomly to form the bifunctional MAb.

In different in vitro and in vivo models, bifunctional MAbs against tumor and CDS at nanomolar concentration has been shown to promote tumor cell killing by cytotoxic T cells.

Specific localization of chemotherapeutic drugs in xenografted tumors has been demonstrated in mice pretreated with hybrid MAbs.

The advantages of the hybrid MAb approach are that it should reduce the MAb biodistribution problem and that it involves no chemical manipulation between the functional agent and the MAb molecules.

Characterization of a bispecific FLT3 X CD3 antibody in an improved, recombinant format for the treatment of leukemia

FLT3 is a receptor-tyrosine-kinase that is expressed on leukemic cells of the myeloid and lymphoid lineage rather specifically. We here report on the construction and selection of bispecific FLT3 X CD3 antibodies in a new recombinant format, termed Fabsc, that resembles the normal antibody structure more closely than the well-established bispecific single chain (bssc)-format. Our preferred antibody, which emerged from an initial selection procedure utilizing different FLT3- and CD3-antibodies, contains the FLT3-antibody 4G8 and the CD3-antibody UCHT1. The 4G8 X UCHT1 Fabsc-antibody was found to be superior to a bssc-antibody with identical specificities with respect to (i) affinity to the target antigen FLT3, (ii) production yield by transfected cells, and (iii) the diminished formation of aggregates. T-cell activation in the presence and absence of cultured leukemic cells and killing of these cells was comparable for both molecules. In addition, the 4G8 X UCHT1 Fabsc-antibody was found to induce T-cell activation and efficient killing of leukemic blasts in primary peripheral blood mononuclear cell (PBMC) cultures of acute myeloid leukemia (AML) patients. In these experiments, the bispecific molecule was clearly superior to an Fc-optimized monospecific FLT3-antibody described previously, indicating that within PBMC of AML patients the recruitment of T cells is more effective than that of natural killer cells.

Specific and effective targeting cancer immunotherapy with a combination of three bispecific antibodies

For the purpose of establishing a new adoptive immunotherapy for bile duct carcinoma (BDC), we previously constructed two kinds of bispecific antibodies (bsAbs), anti-MUC1 x anti-CD3 (M x 3) and anti-MUC1 x anti-CD28 (M x 28), which activate T cells and form bridges between them and MUC1-expressing tumor cells. In our previous studies [Cancer Res. 56 (1996) 4205] specific targeting therapy (STT) consisting of i.v. administration of lymphokine activated killer cells with a T cell phenotype (T-LAK) sensitized with two kinds of bsAbs to human BDC-grafted severe combined immunodeficient (SCID) mice demonstrated remarkable inhibition of tumor growth. However, complete cures could not be obtained. In order to improve antitumor efficacy, we have paid attention to anti-CD2 monoclonal antibodies (mAbs), thought to play an important roles in signal transduction in T cell activation or control of T cell receptor (TCR)-driven activation. Therefore, we developed another bsAb, anti-MUC1 x anti-CD2 (M x 2), in order to examine if this would show synergism with the two previously described bsAbs. The combination of the three bsAbs (M x 3, M x 28 and M x 2 bsAbs) showed highest cytotoxicity against MUC1-expressing BDC cells when given simultaneously with peripheral blood mononuclear cells (PBMCs) or T-LAK cells in vitro. When 2 x 10(7) T-LAK cells sensitized with different combinations of bsAbs were administered four times i.v. to BDC-grafted SCID mice, the best therapeutic result was obtained with a combination of all three bsAbs. These results indicate usefulness of combination of three bsAbs for targeting cancer immunotherapy.

Role of target antigen in bispecific-antibody-mediated killing of human glioblastoma cells: a pre-clinical study

Bispecific antibodies (bsAbs) directed to tumor-associated antigens and to receptors mediating T-cell activation, such as the TCR/CD3 complex and the co-stimulatory CD28 molecule, are capable of activating T cells at the surface of tumor cells, resulting in tumor-cell killing. Here we report the pre-clinical characterization of bispecific-antibody fragments (bsFab2) directed to 2 different glioblastoma-associated antigens: the EGF receptor (EGFR) and a chondroitin-sulfate proteoglycan (CSPG). Using cultured glioblastoma cells expressing both target antigens, we found that the ability of anti-tumor x anti-CD28 bsFab2 to mediate "targeted T-cell co-stimulation" is superior for constructs targeting the CSPG molecule, correlating with an approximately 6-fold higher expression level of this antigen on the cell surface. In contrast, bsFab2 triggering CD3 are more effective if they contain EGFR-target specificity. This indicates that the activity of anti-tumor x anti-CD3 constructs critically depends on properties of the antigen other than its expression level on the cell surface, e.g., its mobility in the membrane. These findings prompted us to use EGFR-targeting bsFab2 in an ongoing clinical trial with glioma patients.

Tumor-growth inhibition with bispecific antibody fragments in a syngeneic mouse melanoma model: the role of targeted T-cell co-stimulation via CD28

The ability of bispecific antibodies with anti-tumor x anti-CD3 specificity to mediate the killing of tumor cells by activated T cells has been demonstrated in many in vitro experiments. Moreover, long-term survival of lymphoma-bearing mice has been observed after treatment with such reagents. The therapeutic effect of bispecific antibodies in solid-tumor models has been less impressive, in particular if fragmented antibodies were used to avoid systemic T-cell activation by bispecific constructs binding to Fc-receptor-positive cells. Here we report that bispecific anti-tumor x anti-CD3-fragments markedly inhibit intraperitoneal as well as pulmonary tumor growth in mice inoculated with B16 melanoma cells, resulting in the long-term survival of animals. Therapeutic success critically depends on the number of recruitable effector cells at the site of tumor growth. A second bispecific construct triggering the co-stimulatory CD28-molecule on the T-cell surface increased tumor-cell killing in vitro and in vivo, despite rather low avidity of this reagent to mouse T cells. Finally, long-term-surviving animals showed improved survival after i.v. rechallenge with tumor cells, indicating that bispecific antibodies are capable of inducing long-lasting protective immunity.

Targeting of drugs to cell surface receptors

The new approach to the treatment of cancer or to immunomodulation is drug targeting. The effort to achieve either an absolute or a relative amplification of the tumoricidal effect of anticancer drugs through increased generation or acquisition of reactive molecules at the tumor site or a reduction of the toxic molecules available to the periphery has led to a number of strategies. Among them are (1) targeting using antibodies to their fragments, hormones, carbohydrates, and growth factors; (2) retargeting using bispecific antibodies; (3) construction of chimeric genes; (4) streptavidin-biotin based immunotherapy; (5) prodrug activation strategies (ADEPT); (6) antibody-targeted superantigens; and (7) gene delivery for the purpose of gene therapy.

Chemospecificity and cross-reactivity of target cell recognition by human CD56+ NK and LAK cells

Inhibition of specific cytotoxicity of highly purified (> 95%) human CD56+ NK and LAK cells against K562 tumour cells was studied with various sugar acetates. Maximum inhibitory specificity was obtained with 60%-deacetylated penta-acetates of mannose, galactose, glucose, or 80%-deacetylated penta-O-acetate of N-acetyl neuraminic acid. The inhibition was strictly dosedependent and 100% inhibition was achieved in the concentration range of 500-1000 nmoles/ml with all four sugar acetate samples. Enhancement of specific cytotoxicity in the presence of rhamnogalacturonan (RG; 500 ng/ml), acting as a bridging molecule, was also inhibited in a dose-dependent manner with the same inhibitory specificity and within the same concentration range indicating involvement of the same number of sugar acetate-specific receptors. Moreover, formation of lytic CD56+ effector cell/tumour cell (E/T) conjugates was equally well inhibited whereas formation of total E/T conjugates was only partially inhibited (NK: 44-73%; LAK: 46-50%). E/T conjugate formation in the presence of RG was enhanced. Inhibition of the enhancement of formation of lytic E/T conjugates in the presence of RG was again completely accomplished with the same inhibitory specificity and within the same concentration ranges as recorded for E/T conjugate formation in the absence of RG. However, inhibition of total E/T conjugate formation was again only partially achieved at the given concentrations. The data support the assumption of an NK cell receptor with specificity for acetylated carbohydrate moieties on target cells or on bridging molecules such as RG.

Cytotoxicity of cytokine-induced killer cells coated with bispecific antibody against acute myeloid leukemia cells

Various types of cytokines have been used in in vitro experiments to generate cytokine-induced killer (CIK) cells that are reactive to patient acute myeloid leukemia (AML) cells. Of these CIK cells, interleukin-2 (IL-2)-activated peripheral blood mononuclear cells, i.e., lymphokine-activated killer (LAK) cells, with the initial addition of the anti-CD3 monoclonal antibody (T3 LAK cells), are the most potent cytotoxic lymphocytes, and have marked proliferative capacity. The cytotoxicity of such T3 LAK cells against CD13+ AML cells is further enhanced by the addition of anti-CD3 x anti-CD13 bispecific antibody (BsAb) during the cytotoxicity assay. The combined use of T3 LAK cells and the BsAb can be used for ex vivo purging of CD13+ AML cells in autologous bone marrow transplantation. Other cytokines, such as IL-7 or IL-7 in combination with IL-2, or newly identified cytokines, will also be tested in attempts to obtain more specific and more potent effector cells. Studies of methods to increase the susceptibility of AML cells to CIK are also required.

Induction of anergy in resting human T lymphocytes by immobilized anti-CD3 antibodies

How the T cell receptor (TcR)/CD3 complex mediates not only the induction of T cell activation but also suppressive effects like T cell anergy or apoptosis is not well understood. Here we describe a series of preincubation and restimulation experiments which demonstrate that primary stimulation of resting, unseparated human T cells with mitogenic doses of immobilized anti-CD3 antibodies induces hyporesponsiveness upon restimulation of the cells. Various costimuli can prevent this type of anergy to a variable degree if present during the preincubation period, phorbol 12-myristate 13-acetate (PMA) being the most and anti-CD4 antibody the least effective. If employed together with anti-CD3 antibody during the restimulation phase of the assay, interleukin (IL)-2, IL-4 and anti-CD28 antibody break anergy almost completely. Proliferation induced by a submitogenic dose of anti-CD3 antibody supplemented by costimulatory signals (anti-CD2, anti-CD4, anti-CD28, IL-2, IL-4 or PMA) does not result in hyporesponsiveness. Taken together, these results support a modified view of the two-signal model for T cell activation according to which anergy induction in resting T cells occurs if primary proliferation is induced by high density triggering of the TcR/CD3 complex in the absence of accessory signals. We discuss possible implications of these findings for the induction of peripheral tolerance.

Superantigen-based tumor therapy: in vivo activation of cytotoxic T cells

We have recently demonstrated that the superantigen staphylococcal enterotoxin A (SEA) targets in vitro activated cytotoxic T lymphocytes against tumor cells expressing major histocompatibility complex (MHC) class II antigens. In this report we analyze the use of SEA as an immunoactivator in vivo. Treatment of mice with SEA activated a fraction of CD3+ T cells apparently as a function of their T cell receptor V beta expression. SEA induced interleukin-2 receptor expression and proliferation in both CD4+ and CD8+ T cells. This proliferative response was dose-dependent (0.1-100 micrograms/mouse), peaked during day 1 after treatment and declined to background levels within 4 days. The cytotoxic response, measured as cytotoxicity to SEA-coated MHC class II+ target cells (staphylococcal-enterotoxin-dependent cell-mediated cytotoxicity, SDCC), was maximal at a dosage of 1 microgram SEA/mouse. The SDCC was confined to the CD8+ T cell compartment, peaked 2 days after treatment and declined to background levels within 4 days. A second injection of SEA on day 5 after the first SEA treatment resulted in SDCC function with kinetics and magnitude identical to that seen after one injection. These results pave the way for the use of SEA in the treatment of MHC class II+ tumors.

Use of bispecific antibodies in the therapy of tumors

Progress toward an understanding of the construction and use of BsAb in therapy has been considerable. The importance of accessory (adhesion) molecules as well as the requirements for killing and the mechanisms by which cytotoxicity is mediated are being clarified. New approaches to simultaneous activation and targeting of effector cells have been developed. Most important, limited clinical trials have demonstrated little toxicity and in several instances promising responses and long-term survivals, if not cures. It seems likely, therefore, that BsAb will be very useful tools for therapy of tumors that may be most efficacious as an adjunct tumor therapy after surgery, chemotherapy, and/or irradiation in order to further reduce, and to potentially eliminate, tumor cells in the patient. Clearly, much remains to be done before BsAb are used routinely for therapy, but the results thus far demonstrate the considerable potential of BsAb to redirect and focus natural immune mechanisms in the treatment of tumors.

Killing of human leukaemia/lymphoma B cells by activated cytotoxic T lymphocytes in the presence of a bispecific monoclonal antibody (alpha CD3/alpha CD19)

Bispecific antibodies (BsAb) can be used to retarget T cells irrespective of their specificity to certain target cells inducing target cell lysis. We have tested the efficacy of the BsAb SHR-1, directed against the T cell antigen CD3 and the B cell antigen CD19 to induce (malignant) B cell kill by T cells as measured in a 51Cr-release assay. Two cytotoxic T cell clones (CTL), expressing TCR alpha beta or TCR gamma delta, were effective in killing CD19 expressing B cell lines at different stages of differentiation in the presence, but not in the absence, of the BsAb. CD19- target cells were not killed. Fresh CD19+ leukaemia/lymphoma cells were also efficiently killed by SHR-1 preincubated CTL clones. In addition, phytohaemagglutinin (PHA) or CD3-activated IL-2 expanded peripheral blood mononuclear cells (PBMC) of normal donors did so after 2 weeks of stimulation. A concentration of 100 ng/ml of the BsAb was sufficient to obtain optimal lysis of all target cells tested. These results show that fresh human leukaemia/lymphoma cells, freshly derived from active lymphoblastic leukaemia (ALL) as well as non-Hodgkin's lymphoma (NHL) patients, can be effectively killed in the presence of this BsAb by activated T cells.

Cell retargeting by bispecific monoclonal antibodies. Evidence of bypass of intratumor susceptibility to cell lysis in human melanoma

Intratumor heterogeneity for susceptibility to cytotoxic T lymphocytes (CTL)-mediated lysis represents a major obstacle to cancer adoptive immunotherapy. To overcome the heterogeneity observed in terms of susceptibility of target cells to cell-mediated lysis, in this study we used two purified bispecific monoclonal antibodies (bsmAbs) that recognize molecules expressed by cytotoxic effector cells (CD3 and IgG Fc receptorial molecules), as well as one high molecular weight melanoma-associated antigen (HMW-MAA). The ability of these reagents to enhance or induce a relevant in vitro cytotoxic activity by a CTL clone (CTL 49) isolated from PBL of a melanoma patient was tested on a large panel of autologous and allogeneic melanoma cell lines and clones. Functional studies revealed that the CTL 49 clone lysed all the HMW-MAA+ tumor lines in the presence of bsmAbs and that these reagents affected the target lysis in a cooperative fashion. The effectiveness of bsmAbs in overcoming the heterogeneous susceptibility of human melanoma cells to cell-mediated lysis may find practical implications in cancer adoptive immunotherapy.

Miniantibodies: use of amphipathic helices to produce functional, flexibly linked dimeric FV fragments with high avidity in Escherichia coli

We have designed dimeric antibody fragments that assemble in Escherichia coli. They are based on single-chain FV fragments, with a flexible hinge region from mouse IgG3 and an amphiphilic helix fused to the C-terminus of the antibody fragment. The sequence of the helix was taken either from that of a previously reported four-helix bundle design or from a leucine zipper, optionally extended with a short cysteine-containing peptide. The bivalent fragments associate in vivo, either with covalent linkage or with a monomer-dimer equilibrium, and results from ultracentrifugation sedimentation studies and SDS-PAGE are consistent with dimers. All constructs are able to bind to surface-bound antigen under conditions in which only bivalent but not monovalent antibody fragments bind. The covalent bundle helix construct shows binding characteristics nearly identical to those of the much larger whole mouse antibody, resulting in substantially more stable immunoglobulin-antigen complexes than in the case of monovalent fragments. This modular design of natural and engineered protein domains directly leads to a boost of avidity, and it allows the construction of bispecific antibody fragments in functional form in E. coli.

Monoclonal antibody-targeted superantigens: a different class of anti-tumor agents

The bacterial superantigen staphylococcal enterotoxin (SE) A (SEA) directs cytotoxic T lymphocytes (CTLs) expressing particular sequences of the T-cell receptor (TCR) beta chain to lyse tumor cells expressing major histocompatibility complex (MHC) class II molecules, which serve as receptors for SEs. We now report that chemical conjugates of SEA and the colon carcinoma-reactive monoclonal antibodies (mAbs) C215 or C242 mediate T cell-dependent destruction of colon carcinoma cells lacking MHC class II molecules. SEA was covalently linked to the mAbs C215 and C242 via a PEG-based hydrophilic spacer. The C215-SEA conjugate targeted CD4+ as well as CD8+ CTLs to lyse a panel of colon carcinoma cells lacking MHC class II molecules. T-cell recognition of mAb-SEA conjugates was SEA specific, since SEB-selective T-cell lines with potent cytotoxic activity towards Raji cells coated with SEB did not respond to the C215-SEA conjugate. Unconjugated SEA did not induce T-cell lysis of MHC class II- colon carcinoma cells but efficiently directed CTLs against MHC class II+ Raji cells and certain interferon-treated MHC class II+ colon carcinoma cells. These results suggest that SEA-mAb conjugates retain the SEA-related selectivity for certain TCR beta-chain variable region (V beta) sequences but, in contrast to unconjugated SEA, mediate the TCR interaction in a MHC class II-independent manner. The cytotoxic activity mediated by C215-SEA and C242-SEA conjugates was blocked by excess of C215 mAb and C242 mAb, respectively, showing that the specificity in the targeting of mAb-SEA conjugates is defined by the antigen reactivity of the mAb. These results demonstrate that bacterial superantigens may be successfully conjugated to mAb with preserved T cell-activating capacity. The circumvention of MHC class II binding of SEs by conjugation to mAb suggests that such conjugates may find general application as antitumor agents, taking advantage of the extreme T cell-activating potency of superantigens.

Target cell-induced T cell activation with bi- and trispecific antibody fragments

Previously we proposed a concept for tumor immunotherapy in which two different bispecific antibody conjugates, an anti-target x anti-CD3 and an anti-target x anti-CD28 conjugate, induce the activation of resting human T cells upon binding to the respective tumor target cells. After in vivo application of these reagents, this model of a "target cell-induced T cell activation" envisages the destruction of target cells by in situ activated T cells. Obviously however, for in vivo application, the use of Fc-free antibody fragments is mandatory to prevent binding of the conjugates to Fc receptor-positive cells which would lead to Fc-mediated T cell activation. Here we report a simplification of published procedures for the generation of bispecific Fab-hybrid fragments, univalent for each antigen. We demonstrate that an anti-target x anti-CD3/anti-target x anti-CD28 combination of such hybrids, as well as an identical combination of covalently coupled F(ab')2 fragments, mediate "target cell-induced T cell activation" in an in vitro test system. Thus, these reagents may be capable of inducing an in situ activation of human T cells upon systemic in vivo application according to the concept outlined above. A trispecific conjugate with anti-target, anti-CD3- and anti-CD28 specificity appears to be unsuitable for this purpose because it activates resting T cells in soluble form without requiring immobilization through binding via its anti-target portion.

Augmentation of interleukin-2-induced activation of human melanoma tumor-infiltrating lymphocytes by heteroconjugate antibody

Heteroconjugate (HC) antibody (anti-CD3 mAb x anti-p97 melanoma mAb) or monomeric anti-CD3 mAb by itself did not induce proliferation of uncultured melanoma tumor-infiltrating lymphocytes (TILs). They also failed to induce IL-2 production in uncultured TILs, although anti-CD3 mAb, but not HC antibody, stimulated IL-2 production in peripheral blood mononuclear cells (PBMCs). Sequential treatment of uncultured TILs from p97-antigen-positive (p97+) melanomas with HC antibody, followed by washing and incubation with interleukin-2 (IL-2), induced significantly higher proliferation than incubation with IL-2 alone. HC antibody pretreatment led to significantly greater results than with anti-CD3 mAb at a 1 ng/ml level in IL-2-induced proliferation of TILs from p97+ melanomas, similar to those with anti-CD3 mAb at a level of 100 ng/ml. HC antibody (1 ng/ml) pretreatment did not enhance IL-2-induced proliferation of either TILs from p97- melanomas or PBMCs, while anti-CD3 mAb enhanced the proliferation of TILs from some p97- melanomas and PBMCs. Regardless of the pretreatment of uncultured TILs with HC antibody or anti-CD3 mAb, IL-2-activated TILs were cytotoxic primarily only to autologous tumor cells, and their phenotypes remained the same. Thus, HC antibody can augment IL-2-induced activation of TILs only from p97+ melanomas, without altering their pattern of cytotoxicity or phenotype. The findings were consistent with observations at the clonal level. In contrast to anti-CD3 mAb, HC pretreatment of uncultured TILs from only p97+ melanoma prior to limiting-dilution analysis increased the number of proliferating TIL clones, including autologous tumor-specific cytotoxic T lymphocyte clones. These results suggest that use of HC antibody in vivo would be more advantageous than anti-CD3 mAb, with regard to augmentation of IL-2-induced TIL activation.

Gene transfer by retrovirus-derived shuttle vectors in the generation of murine bispecific monoclonal antibodies

The present study reports on the use of gene transfer by retrovirus-derived shuttle vectors in the generation of hybrid hybridomas secreting bispecific monoclonal antibodies. neo- and dhfr- genes were infected into distinct murine hybridomas, thus conferring a dominant resistance trait to geneticin (G418) and to methotrexate. The vectors employed were replication-deficient and dependent on complementation by a helper virus provided by the irradiated packaging lines. After cocultivation with the relevant packaging cell lines, stable hybridoma lines expressing the selectable markers were easily obtained and were then suitable for conventional somatic fusion. This high-efficiency method was used to generate two bispecific monoclonal antibodies simultaneously targeting molecules expressed on cytotoxic cells (i.e., T lymphocytes and natural killer cells) against a human melanoma-associated antigen.

Preliminary trial of specific targeting therapy against malignant glioma

The results of specific targeting therapy by use of lymphokine-activated killer (LAK) cells treated with bispecific antibody in 10 patients with malignant glioma were compared with the results of therapy with untreated LAK cells in 10 other patients. Both treated and untreated cells were given locally. The bispecific antibody was an anti-CD3 monoclonal antibody (mAb) chemically conjugated to anti-glioma mAb. Of the 10 patients given specific targeting therapy, 4 showed regression of tumour, and in another 4 patients computed tomography and histology suggested eradication of the glioma cells left behind after surgery. No recurrence was detected in the 10 to 18 months of follow-up. Patients receiving untreated LAK cells had recurrences within 1 year except in 1 case, and 8 patients died within 4 years. So far specific targeting therapy seems to be a new useful form of adoptive immunotherapy.

Lectins and antibodies as tools for studying cellular interactions

Specific interactions between multiple cell types are critical for a variety of processes central to the development, homeostasis and immune defense of multicellular organisms. Studies designed to elucidate how cells communicate through physical encounters have exploited exogenously supplied factors to bypass intrinsic recognition mechanisms and facilitate cellular conjugation. In this review, we compare the relatively nonspecific agglutinating properties of lectins and the selective cell targeting capabilities of antibodies and bispecific antibody constructs for studying cell-cell interactions in immunobiology. In addition, we discuss a novel system for inducing cellular interactions which closely resembles native receptor-mediated conjugation. In this system, surrogate receptors promote specific cell-cell interactions without hindering endogenous receptor-ligand interactions at the cell-cell interface which may be important in mediating physiologic cellular responses.

Lymphokine activated killer cells

Various subpopulations of human leukocytes may be induced by lymphokines to exert cytotoxic activity. In man major histocompatibility complex non-restricted tumor cell lysis by interleukin-2 (IL-2) induced peripheral blood lymphocytes is attributed mainly to natural killer cells. These T cell receptor negative large granular lymphocytes are called lymphokine activated killer (LAK) cells. In order to explore the potential of LAK cells in tumor therapy, several clinical studies have been conducted, using IL-2 alone or in combination with ex vivo IL-2-activated peripheral blood lymphocytes. Objective responses have reproducibly been achieved only in renal cell carcinoma and malignant melanoma and were associated with considerable toxicity. In view of restricted efficacy and increasing doubts as to whether LAK cells indeed account for the in vivo observed responses, more recent strategies focus on tumor antigen specific cytotoxic T cells or tumor infiltrating lymphocytes. Successful translation of this approach into clinical practice, however, may be dependent on some basic problems of tumor immunology to be solved which were thought to be by-passed by the LAK cell approach.

Biochemical characterization of a component in extracts of Viscum album enhancing human NK cytotoxicity

Enhancement of human NK cytotoxicity in the presence of fresh Viscum album extract and some commercial V. album extracts Iscador correlated strictly with an increased formation of lytic effector cell/K562 tumor cell conjugates in the single-cell assay. Both activities were completely destroyed by pretreatment of V. album extracts with pectinase, hemicellulase, amyloglucosidase and alpha-glucosidase, but not with proteases and RNase, i.e., the activities are linked to a polysaccharide. The active component in V. album extract was non-dialysable at a molecular weight cutoff of 10,000. Inhibition of both activities was observed with D-galacturonic acid, poly-galacturonic acid and pectins. The site of galacturonic acid-specific interaction could be identified on the effector cells. The rate of effector cell/tumor cell conjugate formation in the presence of V. album extracts, as well as the abrogation of both activities by pretreatment of V. album extracts with exoglycosidases specific for sugars other than galacturonic acid indicated an action of the NK cytotoxicity-enhancing component on the basis of a bridging mechanism. However, no conclusive results could be obtained for the structural specificity of the site interacting with the target cells.