Potentiation of Tumor Lysis by a Bispecific Antibody That Binds to CAI9-9 Antigen and the Fctreceptor Expressed by Human Large Granular Lymphocytes1

Home , Lysis

[CANCER RESEARCH 50. 7123-7128, November 15. 1990] Potentiation of Tumor Lysis by a Bispecific Antibody that Binds to CAI9-9 Antigen and the FCTReceptor Expressed by Human Large Granular Lymphocytes1

Irma GarcÃade Palazzo, Cicek Gercel-Taylor, Joanne Kitson, and Louis M. Weiner2

Department of Medical Oncology, Fox Chase Cancer Center, Philadelphia, Pennsylvania 19111

ABSTRACT tumor have been rarely observed. Attempts to enhance antibody efficacy by concomitant systemic activation of defined effector Murine monoclonal antibody therapy of human cancer rarely induces populations using 7-interferon (6-9) and interleukin-2 (10) clinical responses. Antibody-induced cellular infiltrates rarely accumulate have not led to notable enhancement of clinical effects, even at sites of tumor, even in clinically responding lesions. Thus, the ability of these antibodies to promote host effector cell-mediated lysis of tumor though these cytokines are potent enhancers of in vitro ADCC. via antibody-dependent cellular cytotoxicity (ADCC) has not been har ADCC requires interaction of antibody Fc domains with Fc receptors on effector cells (2-4). Since human and murine nessed by existing treatment approaches. One potential explanation is that ADCC requires binding of antibody Fc domains to cellular I-'o antibody Fc domains have similar affinities for human Fc receptors, and therapeutically administered murine antibodies must com receptors, infused murine antibodies must compete with vast pete with vast excesses of human IgG for FCT receptor occupancy. excesses of human immunoglobulin for Fc receptor occupancy. Chemically linked antibody heteroconjugates that bind selected target This competition may inhibit promotion of ADCC at sites of and effector cell structures via distinct Fab portions can mediate lysis of murine antibody-bound tumor. Thus, the clinical relevance of malignant cells in vitro in the presence of human serum. This approach ADCC remains to be investigated. To address this problem, addresses a potentially major obstacle to antibody therapy. Production chemically linked AHCs have been designed to bind, via their of bispecific monoclonal antibodies with similar specificities and superior Fab domains, to tumor antigens and to epitopes on Fc7 recep in vivo biodistribution characteristics would thus have potential clinical tors that are outside these receptors' immunoglobulin-binding applications. We have prepared and purified a bispecific, monovalent monoclonal antibody and evaluated its in vitro effects. The IgG 1-secreting epitopes (11). Fc7R I, expressed by human macrophages and hybridoma line 3G8 (Â«-humanFc-yR III) was fused with the hybridoma polymorphonuclear leukocytes, and the membrane-anchored line CA19-9, which produces an IgGl antibody that binds to a glycopro- form of Fc7R III, expressed by human large granular lympho tein shed by gastrointestinal cancers. Multiple clones with bispecific cytes, have been found to be efficient effector cell cytotoxicity binding properties were identified. CA19-9 x 3G8 clonal supernatants triggers (12) and AHC components (13-15). These results are and purified antibody, but not the parent antibodies, efficiently mediated analogous to those obtained by others using AHCs that focus specific in vitro lysis of cells of the SW948 line by human large granular lysis of T-lymphocytes via the CD3/T-cell receptor signal trans- lymphocytes (LGLs). Human serum-resistant target cell lysis augmen duction complex (16-18). Since the promotion of lysis by these tation at low effectontarget ratios was seen using picogram amounts of antibody. In contrast, the IgG2a variant of CA19-9, which also promotes antibodies is resistant to excess human immunoglobulin, bis ADCC by LGLs, was unable to augment lysis of SW948 cells when pecific antibodies may be efficient in promoting in vivo ADCC. effectors were preincubated with human serum. This bispecific, mono Monovalent bispecific antibodies prepared by somatic cell valent monoclonal antibody is an efficient promoter of the anti-tumor fusion of two hybridoma lines have been constructed. These effects of LGLs in physiological concentrations of human serum. In vivo antibodies retain the properties of their corresponding chemi models that evaluate treatment efficacy and promotion of inflammatory cally linked AHCs ( 19-23). Such antibodies, which retain native tumor infiltrates by bispecific monoclonal antibodies are required to immunoglobulin configurations, are likely to be more suitable assess the therapeutic potential of these novel constructs. for clinical use than AHCs since the purity of AHCs and stability of chemical linkages may be suboptimal. We have INTRODUCTION prepared a bispecific monoclonal antibody derived by somatic cell fusion of hybridomas secreting antibodies to the tumor- The ability of selected murine monoclonal antibodies to associated antigen CAI9-9 (24) and Fc7R III, the Fc7 receptor promote in vitro tumor lysis by human monocytes (1) and large for aggregated immunoglobulin expressed by large granular granular lymphocytes (2-4) via ADCC3 has prompted extensive lymphocytes (25, 26). This antibody is a potent promoter of clinical evaluations of these agents in patients with diverse lysis of an allogeneic CA19-9-expressing human tumor cell malignancies. Clinically significant responses to antibody ther line by interleukin-2-activated large granular lymphocytes, even apy have been unusual (5), and cellular infiltrates at sites of in the presence of excess human immunoglobulin.

Received 3/6/90; accepted 8/13/90. The costs of publication of this article were defrayed in part by the payment MATERIALS AND METHODS of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. 1This work was supported by National Cancer Institute Grants CA06926. Cell Lines and Antibodies CAO1130, andCA50633. 2To whom requests for reprints should be addressed, at Department of Medical Hybridoma lines 3G8 and 19-9 were utilized for fusion. 3G8, a gift Oncology, Fox Chase Cancer Center. 7701 Burholme Avenue. Philadelphia. PA 19111. of Dr. J. Unkeless, secretes a murine IgGl antibody that binds to Fc7R 3The abbreviations used are: ADCC. antibody-dependent cellular cytotoxicity; III (27). This antibody has been used to construct heteroconjugates AHC, antibody heteroconjugate; IL-2, interleukin 2: rIL-2, recombinant IL-2; capable of enhancing lysis of malignant targets by peripheral blood HAT. hypoxanthine. aminopterin and thymidine; MEM. minimal essential me mononuclear cells activated by rIL-2 (13-15). The CA19-9 hybridoma dium: LGL, large granular lymphocyte; E:T, effectortarget ratio; PBS. phos phate-buffered saline; SDS-PAGE, sodium dodecyl sulfate-polyacrylamide gel line (American Type Culture Collection, Rockville, MD) secretes an electrophoresis; ELISA, enzyme-linked immunosorbent assay: LAK, lymphokine- IgGl murine antibody that binds to a cell surface glycoprotein shed by activated killer cell; Fc^R, Fc> receptor. malignant adenocarcinoma cells derived from the gastrointestinal tract 7123

Downloaded from cancerres.aacrjournals.org on September 28, 2021. © 1990 American Association for Cancer Research. BISPECIFIC ANTIBODY TARGETING CAI9-9 AND Fc-fR III (24). 3G8 was grown in Dulbecco's MEM containing 20% fetal calf normal donors were incubated for 4 h with 5lCr-labeled SW948 cells at serum, in 5% CO2 atmosphere at 37Â°C,keeping the concentration of various E:T ratios in the presence of varying concentrations of super cells between 5 x 10s and 1 x JO'/ml. CA19-9 cells were grown in natants from different clones with dual binding specificity. CA 19-9 Iscove Dulbecco's modified media containing 10% fetal calf serum at and 3G8 supernatants were used alone in this system as negative concentrations of 1 x 106/ml in 5% CO2 atmosphere at 37Â°C.The controls. The IgG2Â«CAI9-9 antibody was used as a positive control. IgG2Â«variant of CA19-9 was a kind gift of Dr. Z. Steplewski. The Lytic units were calculated by the method of Pross (34). fluorescein isothiocyanate-labeled antibodies Leu 19 (a-CD-56), Leul le (a-CD16) and Leu4 (o-CD3) were purchased from Becton-Dickenson Purification of Bispecific Antibody (Mountain View, CA). The SW948 and SW1116 cell lines were ob Supernatants from hybrid hybridoma clone 158 were clarified by tained from the American Type Culture Collection (3). centrifugation (1000 x g), precipitated with saturated (50%) ammo Cell Selection Procedures nium sulfate, and dialyzed against PBS for 48 h with frequent changes of the buffer. This crude antibody preparation was applied to an G418 Resistance. The methodology for producing bispecific mono immunoaffinity column (Affi-Gel 10-anti-mouse IgGl). Nonspecific clonal antibodies was adapted from Lanzavecchia et al. ( 19) and Suresh material was removed by washing with 10 m\i Tris-0.5 M NaCl, pH et al. (23). 3G8 cells were transfected with the plasmid pko neo, which 7.5, and eluted with 100 HIMglycine buffer at pH 2.5. This purified carries the neomycin-like resistance gene (28). The procedure for trans- antibody, which contained both of the parental antibodies and the fection was performed using DEAE dextran as described previously bispecific antibody, was further fractionated on the cation-exchange (29, 30). Transfected cells were able to grow in 2 mg/ml G418 (Sigma column, S-Sepharose FF (Pharmacia, Uppsala, Sweden). Antibody was Chemical Co., St. Louis, MO). At this concentration, the wild-type eluted with a linear gradient of 10 m\i sodium phosphate from 0-0.5 3G8 cells were uniformly killed. M NaCl, pH 7.0, and fractions were monitored by absorbance at 280 HAT Sensitivity. To select 3G8 cells sensitive to HAT, HAT-sensi- nm. Samples from the fractions collected by chromatography were tive, G418-resistant cells were grown in the presence of increasing subjected to SDS-PAGE on a 12.5% acrylamide gel under reducing concentrations of 8-azaguanine until cells were adapted to grow in 30 conditions (35). jig/ml of this agent. The 8-azaguanine-resistant 3G8 cells were subcloned by limiting dilution. Clones that possessed excellent growth Quantitation of Mouse IgGl by ELISA characteristics, antibody secretion, G418 resistance, and HAT sensitiv The concentration of antibody in culture supernatants was deter ity were expanded and used for fusions. mined by ELISA (36). Ninety-six-well polystyrene plates were coated with anti-mouse IgGl (Sigma), diluted 1:250, for 24 h at 4Â°C(4 /jg/ Fusion Procedure ml). PBS with 2% bovine serum albumin was used to block nonspecific The doubly adapted 3G8 cells were fused with HAT-resistant, G418- binding. After washing, mouse IgGl (0.013-10 ^g) was added to the sensitive CAI9-9 cells. Cells (1 x 10") of each line were pelleted at wells and incubated at 37Â°Cfor 1 h. The plates were washed twice with room temperature and the cells were combined in 30 ml of cold serum- PBS/1% bovine serum albumin and incubated with peroxidase conju free opti-MEM. Cells were centrifuged for 5 min at 200 x g at 4Â°C, gated anti-mouse IgG (1:1000) for 1 h. Antibody reactivity was esti resuspended gently in 30 ml of serum-free opti-MEM, and washed mated by developing the color reaction using the substrate O-phenyle- twice in the same fashion. After the second wash the supernatant was nediamine. The test wells were incubated with different dilutions (un aspirated completely and 1 ml of 30% polyethylene glycol was added diluted 1:10, 1:100) of clone 158 supernatants and ascites, while known dropwise over 3 min at room temperature. Following gentle dissociation concentrations of IgGl were added to control wells to develop a of the cell pellet and centrifugation at 200 x g for 5 min, the cells were standard curve for each assay. The concentration of mouse protein in washed and resuspended in opti-MEM containing HAT and 2 mg/ml the clonal supernatants was estimated from the standard curve. G418. Cells were added to 96-well plates at concentrations of 200, 100, 20, and 5 cells/well. Media were not changed until there was evidence RESULTS of growth in the wells. Screening. Cloned supernatants were screened for binding by Screening: Immunohistochemistry immunohistology to SW1116 cells and purified polymorpho nuclear leukocytes. Of 96 clonal supernatants screened for The supernatants of wells exhibiting good clonal growth were screened for binding to SW1116 and/or SW948 colorectal carcinoma binding to both cell preparations, 22 exhibited bispecific bind cell lines (both of these cell lines express CAI9-9 antigen) and to ing. The cells from one well with excellent bispecific binding polymorphonuclear leukocytes, which express FcR III. The binding of were selected for further study. A clone from these cells, des supernatants to the different cytospin preparations was assessed by ignated cl.158, was repeatedly subcloned and clonal progeny immunohistology using the avidin-biotin conjugate procedure as de that secreted antibody with dual specificity were identified. scribed by Hsu et ai. (31). All clones that expressed double binding Characterization of Clonal Supernatant. To demonstrate that specificity were subcloned twice and rescreened. Supernatants from cl.158 supernatant contained active bispecific antibody it was these clones were screened by cytotoxicity assay. necessary to show that this supernatant augmented the lysis of an appropriate tumor target line by interleukin-2-activated pe Cytotoxicity Assays ripheral blood mononuclear cells. Varied concentrations of Peripheral blood mononuclear cells were activated in rIL-2 as de cl. 158 supernatants were screened for their ability to effect lysis scribed before (2). LGLs were purified on a discontinuous Percoli of SW948 colorectal carcinoma cells by human peripheral blood gradient as described by Timonen et al. (32). Following Percoli density mononuclear cells that had been activated by 1000 units/ml of gradient centrifugation, no additional purification of LGLs by negative rIL-2 (Fig. 1) in a standard Mchromium release assay. Undiluted selection procedures was attempted. Flow cytometry was performed as cl.158 supernatant, containing approximately 6 units/ml mu described to characterize the phenotype of the Percoli gradient-derived rine antibody by quantitative ELISA and yielding an assay cell populations (33). The ability of clonal supernatants to promote lysis of SW948 colo concentration of 1.5 ng/ml, was unable to promote tumor lysis. rectal carcinoma cells by human peripheral blood mononuclear cells However, lysis potentiation was evident at murine antibody activated in 1000 units/ml of rIL-2 was screened using a standard assay concentrations ranging from 750 ng/ml-75pg/ml. Since "chromium release assay (3). In these experiments, IL-2-activated about 33% of the antibody of this clone was later found to be unfractionated peripheral blood lymphocytes or purified LGLs from bispecific, bispecific antibody at concentrations as low as 25 7124

100- confined to the LGL-containing fractions 1 and 2. In this representative experiment the LGLs exhibited characteristic granules on Giemsa staining, while >80% of the cells in the T- cell fraction expressed CD3 when analyzed by flow cytometry. Because 69-77% of the cells in the LGL fraction expressed CD3 as well, the precise phenotype of the cells responsible for O bispecific antibody-mediated lysis is unclear; experiments that use cell sorter-purified CD3+ CD16+, CD3+ CD16", and other O) 50- a. defined populations are in progress to address this issue. Purification and SDS-PAGE Analysis of the Bispecific Anti bodies. Tissue culture supernatants of cl.158 were fractionated by successive steps involving ammonium sulfate precipitation, l affinity, and ion-exchange chromatographies. The peak frac tions with mobility on S-Sepharose chromatography between the 19-9 and 3G8 parental peaks augmentated killing of 51Cr- labeled SW948 cells by IL-2-activated LGLs. SDS-PAGE of 0.5 the peak activity fractions (Fig. 2) demonstrated that the bis Effector to Target Ratio pecific product (lane B) contained the light chains of both 19- Fig. 1. Human peripheral blood lymphocytes were activated in 1000 units/ml 9 and 3G8 antibodies, while lanes A and C each contained the rlL-2 for 72-96 h. These cells were incubated with SW948 colon carcinoma cells light chain of one of the parent antibodies. The unequal densi at various effectontarget ratios in the presence of various supernatant dilutions that yielded final antibody concentrations of 1 Mg/ml-100 pg/ml and bispecific ties of the dual light chain bands in lane B are consistent with antibody concentrations ranging (approximately) from 250 ng/ml-25 pg/ml. partial purification of the bispecific antibody from the parental Cytotoxicity was evaluated with standard 5lCr-release assays. The results of two products (lane D). The antibody in lane B was used for further representative experiments were averaged: points, actual mean percentages of studies. Heavy chains from the parent antibodies were not specific lysis. A, no bispecific antibody; A. 25 pg/ml bispecific antibody; D, 250 pg/ml bispecific antibody; â€¢2.5 ng/ml bispecific antibody; O. 25 ng/ml bispecific distinguishable by the methods of analyses described above. antibody; â€¢.250 ng/ml bispecific antibody. Biological Characterization of Purified Antibody. In contrast to its IgGl isotype the IgG2Â«variant of CAI9-9 promotes pg/ml were able to promote tumor cell lysis (see below). Max ADCC by LGLs. Because this antibody has the same tumor imal induction was seen at concentrations exceeding 25 ng/ml. specificity as our bispecific antibody, it was possible to directly Supernatants of the IgGl parental 19-9 and 3G8 hybridomas compare the efficiency of classic ADCC with bispecific anti were unable to augment lysis in these assays. Fig. 1 also shows body-promoted cytotoxicity. The IgG2Â«variant of CAI 9-9 was that maximal cytotoxicity requires higher effectontarget ratios a potent mediator of ADCC in the absence of human serum at low antibody concentrations. At 250 pg/ml concentrations but not when effector cells were preincubated in 20% pooled of bispecific antibody, 74% specific lysis was attained; however, human serum. In contrast, the lysis potentiation of partially lysis fell off rapidly at lower effectontarget ratios and did not purified CAI9-9 x 3G8 was not affected by human serum (Fig. differ from LAK lysis at 5:1 (not shown), 1:1, or 0.5:1 ratios. 3). The resistance of these biological effects to inhibition by In contrast, identical 74% lysis was seen at 1:1 effectontarget human serum was a major advantage of the bispecific antibody ratios when 250 ng/ml concentrations of bispecific antibody when compared to the IgG2Â«variant of CAI9-9. The data in were used; this level of lysis was maintained even at 0.5:1 Fig. 4 suggest additional advantages in that lysis augmentation effector:target ratios. At 25:1 effectontarget ratios the concen occurred at picogram concentrations of the bispecific antibody, tration of bispecific antibody required to achieve half-maximal while microgram/milliliter concentrations of the IgG2Â«variant augmentation of lysis was about 250 pg/ml. At 5:1 and 1:1 were required to see equivalent effects. It is noteworthy that ratios the corresponding concentrations were approximately these low concentrations of bispecific antibody promoted lysis 500 pg/ml and 25 ng/ml, respectively. The efficiency of this of antigen-shedding tumor cells. Since these effects were lysis is even more impressive, considering that only 15-20% of achieved at 5:1 E:T ratios, it is evident that bispecific antibody- IL-2-activated lymphocytes express Fc-yR III. Preincubation of mediated tumor cell lysis is efficient with regard to the require effectors with 3G8 antibody to saturate FcyR III eliminates ments for antibody and cellular delivery to tumor when com bispecific antibody promotion of tumor lysis (data not shown). pared with an antibody of identical tumor specificity that me The effects of the bispecific antibody required CA 19-9 diates its effects via a classic ADCC mechanism. antigen expression, since clonal supernatants were unable to augment lysis of cells of the Daudi line, which do not express DISCUSSION this antigen (data not shown). Although 15% of mature mononuclear phagocytes express This is the first report of a bispecific monoclonal antibody- Fc7R III, cl. 158 did not augment lysis of SW948 cells by short- prepared by somatic cell fusion of hybridoma clones secreting term cultured human monocytes that do not express Fc7R III. antibodies that bind to a tumor antigen and an Fc7 receptor, To demonstrate that LGLs are the cell population that mediates respectively. The resulting antibody possesses in vitro biological lysis augmentation by this antibody, LGL- and T-cell-enriched characteristics that are at least equivalent to chemically linked fractions were purified by Percoli gradient centrifugation and AHCs that bind to other tumor antigens and FcyR III (13-15). assayed for lysis of SW948 and Daudi cells, in the presence and Furthermore, this antibody has physical properties and func absence of cl.158. Table 1 shows that the LGL-enriched frac tional characteristics that are similar to those of bispecific tions were responsible for the bulk of non-major histocompat- monoclonal antibodies that bind tumor antigens and the TcR/ ibility complex-restricted killing of SW948 and Daudi cells and CD3 complex on T-cells (19-23). that antibody-induced lysis augmentation of SW948 targets was Of the three cell populations used thus far in bispecific 7125

Table l Medialion of clone 158 biological effects by large granular lymphocytes "Cr-labeled targets were incubated with cells purified from a modified Percoli gradient. Effectortarget ratios were 1:1 for SW948 and 5:1 for Daudi. Statistical comparisons of paired values were done by / test. lysis of5t(-Ã•15856.3"34.7*'f11.6' lysisof<-)15883.8 of cells positiveforCD1632.1 Cell type (Per colifraction)LGLs2

63.5"12.67.5% 73.624.6!Cr-Daudi(+)15877.174.429.9CD369.0 77.182.9 21.24.13.6CD5641.819.33.2 T-cells5

5.6Â° 6% 4.6O--SW948(+)15879.1Â° 84.4% P = 0.01 2. *P = 0.02. ' P = 0.021.

ABCD 80

60 T -66,000 z 40 LU O ce -45,000 LU n.

-36,000 20 -29,000

NO CLONE 158 19.9x3G8 19-9(IgG2a) ANTIBODY (1:100) (PURIFIED) Fig. 3. IL-2-activated peripheral blood mononuclear cells were incubated with "Cr-labeled SW948 cells at 5:1 E:T ratios plus the indicated antibodies in the absence (open columns) and presence (hatchedcolumns) of 20% autologous human serum. Clone 158 supcrnatants were used at 1:100 dilution, containing approxi mately 5 ng/ml bispecific antibody. The purified bispecific antibody was used at 1 (Jg/ml, while the IgG2n variant of CA19-9 was used at 3.5 /Jg/ml. Columns, mean % lysis for four experiments; bars, Â±SEM. Fig. 2. SDS-PAGE (12.5%) of S-Sepharose-fractioned bispecific antibody under reducing conditions. A, 3G8 peak; B, cl.158 peak; C, 19-9 peak; D, unfractionated cl.158 after affinity purification; ordinate, molecular weight. presence of human serum, may improve the therapeutic activity ofIL-2. antibody therapy research, T-cells have received the greatest A few technical points require mention. We found that hybrid attention, with the antibodies identifying tumor cells and sup hybridoma supernatants containing high concentrations of an plying the stimulus for receptor-mediated activation through tibody were not active in these cytotoxicity assays. Since 8.5- the CD3/TcR. We chose to develop antibodies that redirect the 50% of hybrid hybridoma clonal supernatants contain bispecific toxicity of cells that express membrane-anchored Fc^R III antibodies (23), the intact monospecific bivalent parental anti because such cells (i.e., LGLs) are presumably designed to bodies are also secreted and presumably compete for binding mediate antibody-directed functions and are potent effectors of to antigen at higher concentrations, thus inhibiting antibody- in vitro ADCC (2-4). The ability of bispecific antibodies to mediated effector-target conjugation. More important, purifi focus the cytotoxic activity of LGLs may have important ther cation of the bispecific antibody posed various problems due to apeutic consequences. These cells are the major mediators of the heterogeneous nature of the antibodies produced from the in vitro lysis of natural killer-resistant and fresh, autologous or clone. Fusion of the identical isotype antibodies 19-9 and 3G8 allogeneic malignant cells (LAK activity) (37). However, accu made separation of the bispecific from its parental antibodies mulation of LGLs at sites of tumor does not occur in patients very difficult. With this clone, the strong cation exchanger S- treated with IL-2 at various doses and schedules (38); thus, Sepharose FF was most useful; however, several other ion- LAK activity has not been clinically exploited with current IL- exchange chromatography resins did not yield adequate sepa 2 therapy approaches. Bispecific antibodies that promote cyto ration of the various antibody species. For example, hydroxyl- toxic interactions of LGLs and malignant cells, even in the apatite HPLC, which has been used to resolve other bispecific 7126

significant advantage in the therapy in vivo of established ma lignancies, in which competing human immunoglobulin, low antibody concentrations, and low effector:target ratios may be anticipated.

en ACKNOWLEDGMENTS in The authors gratefully acknowledge the expert secretarial assistance of Isabella Day.

REFERENCES

1. Steplewski. Z.. LÃ¼beck.M. D., and Koprowski. H. Human macrophages armed with murine immunoglobulin G2a antibodies to tumors destroy human cancer cells. Science (Washington. DC). 221: 865-867. 1983. 2. Ortaldo. J. R., Woodhouse. C., Morgan. A. C., Herberman. R. B.. Cheresh. D. A., and Reisfeld. R. Analysis of effector cells in human antibody-dependent cellular3566-3571. cytotoxicity 1987." with murine monoclonal antibodies. J. Immunol.. 138: 3. Weiner. L. M.. Zarou, C. M.. O'Brien, J.. and Ring, D. Effector character Concentration of Murine IgG(/ml) istics of the IgGj murine monoclonal antibody 113F1. J. Biol. Response Modif., 8: 227-237, 1989. Fig. 4. 5lCr-labeled SW948 cells were incubated with 5:1 cffectortarget ratios 4. Munn. D. H., and Cheung. N. V. lnterleukin-2 enhancement of monoclonal of IL-2-activated lymphocytes in the presence of various concentration of anti antibody mediated cellular cytotoxicity against human melanoma. Cancer bodies in this representative example of four separate experiments. â€¢.purified Res., 47: 6600-6605, 1987. 19-9 x 3G8 bispecific antibody; O, IgG2u variant of CAI 9-9 antibody. Lysis of 5. Houghton, A. N., and Scheinberg. D. A. Monoclonal antibodies: potential targets was determined in a standard 4-h 5'Cr-release assay. applications to the treatment of cancer. Semin. Oncol.. 13: 165-179. 1986. 6. Weiner. L. M.. Steplewski,eple Z.. Koprowski, H.. Sears. H. F., Litwin. S., and Comis. R. L. Biologic effects of 7-interferon pre-treatment followed by antibodies, was not useful with our antibody. Selection of monoclonal antibody 17-1A administration in patients with gastrointestinal parental antibody pairs for production of hybrid hybridomas carcinoma. Hybridoma. 5 (Suppl. 1):S65-S77. 1986. 7. Weiner, L. M., Steplewski. Z.. Koprowski, H.. Litwin. S., and Comis, R. L. should involve screening of the purified parental antibodies Divergent dose-related effects of 7-imerferon therapy on in vitro antibody- with a variety of purification strategies to ensure that the dependent cellular and nonspecific cytotoxicity by human peripheral blood monocytes. Cancer Res., 48: 1042-1046, 1988. parental antibodies can be adequately resolved. 8. Weiner. L. M., Moldofsky, P. J.. Gatenby, R. A., O'Dwyer, J., O'Brien, J., It was not expected that this antibody would be more efficient Litwin. S., and Comis. R. L. Antibody delivery and effector cell activation in than its bivalent, monospecific antitumor parent in mediating a phase II trial of recombinant 7-interferon and murine monoclonal antibody in vitro tumor cell lysis, except in the presence of human serum. CO 17-1A in advanced colorectal carcinoma. Cancer Res., 48: 2568-2573, 1988. However, this proved to be the case. As with other bispecific 9. Tempero. M. A., Pour, P. A., Debida, E., Herlyn. D., and Steplewski, Z. monoclonal antibodies that monovalently bind their target an Monoclonal antibody CO 17-1A and leukapheresis immunotherapy of pan creatic cancer. Hybridoma. 5 (Suppl. 1):S133-S138. 1986. tigens, very low concentrations of this antibody were required 10. Bajorin. D.. Chapman, P., Kuricka. J.. Cordon-Cardo. C., Weite, K.. Mer- to promote relevant target cell lysis. With this bispecific anti telsmann. R.. Melamed. M.. Oettgen, H. F.. Old, L. J.. and Houghton, A. body, promotion of LGL cytotoxicity was observed at antibody N. Phase I trial of a combination of R24 mouse monoclonal antibody and recombinant interleukin-2 in patients with melanoma. Proc. Am. Soc. Clin. concentrations as low as 25 pg. By contrast, the concentrations Oncol., 6:210, 1987. of the bivalent, monospecific CAI 9-9 IgG2Â«antibody required 11. Segal, D. M., and Wunderlich. J. R. Targeting of cytotoxic cells with to see similar effects were in the microgram range. Monovalent heterocrosslinked antibodies. Cancer Invest., 6: 83-92. 1988. 12. Fanger, M. W., Shen. L., Graziano, R. F.. and Guyre, P. M. Cytotoxicity binding may confer an advantage by targeting less accessible mediated by human Fc receptors for IgG. Immunol. Today. 112: 230-235, antigen on cell membranes or by permitting closer apposition 1989. of effector and target cells. Alternatively, cross-linking of Fc7R 13. Shen. L.. Guyre. P. M.. Anderson, C. L.. and Fanger, M. W. Heteroantibody- mediated cytotoxicity: antibody to the high affinity Fc receptor for IgG III on LGLs may be more efficient when antibody Fab, rather mediates cytotoxicity by human monocytes that is enhanced by interferon-7 than Fc domains, interact with the receptor. and is not blocked by human IgG. J. Immunol., 137: 3378-3382, 1986. 14. Karpovsky, B.. Titus, J. A., Stephany, D. A., and Segal, D. M. Production Irrespective of the mechanism, the efficiency of those anti of target-specific effector cells using hetero-cross-linked aggregates contain bodies in promoting relevant target cell lysis is noteworthy, as ing anti-target cell and anti-Fc-y receptor antibodies. J. Exp. Med.. 760: are the low effector:target ratios required to promote tumor 1686-1701. 1984. 15. Titus. J.A., Perez, P.. Kaubisch. A.. Garrido. M. A., and Segal. D. M. Human lysis. In the setting of low effectortarget ratios, relatively higher K/natural killer cells targeted with hetero-cross-linked antibodies specificially concentrations of bispecific antibodies are required, but these lyse tumor cells in vitro and prevent tumor growth in vivo. J. Immunol., 139: concentrations are at least 1000-fold lower than those of con 3153-3158. 1987. 16. Jung, G., Honsik, C. J., Reisfeld, R. A., and Muller-Eberhard. H. J. Activa ventional monoclonal antibodies. We found that equivalent tion of human peripheral blood mononuclear cells by anti T3: killing of lysis augmentation may be achieved with combinations of either tumor target cells coated with anti-target-anti-T3 conjugates. Proc. Nati. Acad. Sci. USA, 83: 4479-4483. 1986. high concentrations of antibody and low effectortarget ratios 17. Perez. P., Hoffman. R. W., Shaw. S., Bluestone, J. A., and Segal, D. M. or low antibody concentrations and higher effectontarget ratios. Specific targeting of cytotoxic T cells by anti-T3 linked to anti-target cell This suggests the possibility that efficient antibody promotion antibody. Nature (Lond.). 316: 354-356, 1985. 18. Perez, P., Hoffman. R. W.. Titus. J. A., and Segal, D. M. Specific targeting of tumor lysis via Fc7R 111requires a critical number of inter of human peripheral blood T cells by heteroaggregates containing anti-T3 actions between this Fc7R and the targeted tumor antigens. cross-linked to anti-target cell antibodies. J. Exp. Med., Â¡63:166-178, 1986. 19. Lanzavecchia, A., and Scheidegger, D. The use of hybrid hybridomas to Further exploration into the roles of antibody concentration target human cytotoxic T lymphocytes. Eur. J. Immunol., 17:105-111. 1987. and saturation of LGL-associated FcyR III in promoting tumor 20. van Dijk, J.. Warnaar. S. O.. Van Eendenburg, J. D. H., Thienpont. M., lysis by this bispecific antibody are under way. The results of Braakman, E., Boot, J. H. A., Fleuren, G. J., and Bolhuis. R. L. H. Induction of tumor-cell lysis by bi-specific monoclonal antibodies recognizing renal- the current study suggest that the efficiency of bispecific anti cell carcinoma and CD3 antigen. Int. J. Cancer, 43: 344-349. 1989. body-redirected in vitro tumor cell lysis may prove to be a 21. Staerz. U. D.. Kanagawa. O.. and Bevan, M. J. Hybrid antibodies can target 7127

sites for attack by T cells. Nature (Lond.). 314: 628-631. 1985. with cloned Epstein-Barr virus (EBV) DNA. Virology. 157: 58-66. 1987. 22. Staerz, U. D., and Bevan, M. J. Hybrid hybridoma producing a bispecific 31. Hsu. S. M.. Raine. L., and FÃ¤nger.H. The use of antiavidin antibody and monoclonal antibody that can focus effector T-cell activity. Proc. Nati. Acad. avidin-biotin-peroxidase complex in immunoperoxidase technics. Am. J. Sci. USA, 83: 1453-1457, 1986. Clin. Pathol., 75:816-821, 1981. 23. Suresh, M. R.. Cuello, A. C.. and Milstein. C. Bispecific monoclonal anti 32. Timonen, T.. Ortaldo. J. R., and llcberman. R. B. Characteristics of human bodies from hybrid hybridomas. Methods Enzymol., 121: 210-228. 1986. large granular lymphocytes and relationship to natural killer and K cells. J. Exp. Med., 663: 569-582. 1981. 24. Koprowski. H.. Steplewski, Z.. Mitchell, K.. Herlyn. M.. Herlyn. D., and 33. Weiner, L. M., O'Dwyer. J.. Kitson. J.. Comis. R. L.. Frankel. A. E., Bauer. Fuhrer. J. P. ColorÃ©ela!carcinoma antigens detected by hybridoma antibod ies. Somat. Cell Genet.. 5: 957-972. 1979. R. J., Ki.ni.ul M. S.. and Groves. E. S. Phase I evaluation of an anti-breast 25. Unkeless, J. C., Scigliano, E., and Freedman. V. H. Structure and function carcinoma monoclonal antibody 260F9-recombinant ricin A chain immuno- of human and murine receptors for IgG. Annu. Rev. Immunol.. 6: 251-281. conjugate. Cancer Res.. 49: 4062-4067, 1989. 1988. 34. Pross, H. F.. Baines. M. T.. Rubin. P.. Shragge. P., and Patterson, M. S. 26. Perussia, B., Starr, S., Abraham. S.. Fanning. V.. and Trinchieri. G. Human Spontaneous human lymphocyte mediated cytotoxicity against tumor target cell. IX. The quantitation of natural killer cell activity. J. Clin. Immunol.. /: natural killer cells analyzed by B73.1. a monoclonal antibody blocking Fc 51-63. 1981. receptor functions. J. Immunol., AJO:2I33-2141, 1983. 35. Laemmli. U. K. Cleavage of structural proteins during the assembly of the 27. Fleit, H. B., Wright, S. D., and Unkeless, J. C. Human neutrophil Fc> head of bacleriophage T4. Nature (Lond.). 27: 680-685, 1970. receptor distribution and structure. Proc. Nati. Acad. Sci. USA, 79: 3275- 36. Del.au. W. B. M.. Van Loon. A. E.. Heije. K., Valerio. D.. and Bast. B. J. 3279, 1982. E. G. Production of hybrid hybridomas based on HAT"-neomycin' double 28. van Doren. K.. Hanahan. D., and Gluzman. V. Infection of eucaryotic cells mutants. J. Immunol. Methods, /17: 1-8, 1989. by helper-independent recombinant adenoviruses: early region 1 is not oblig 37. Damle, N. K.. Doyle. L. V.. and Bradley. E. C'. lnterleukin-2 activated human atory for integration of viral DNA. J. Virol.. 50: 606-614. 1984. killer cells are derived from phenotypically heterogeneous precursors. J. 29. Banerji. J.. Olson. L., and Schaffner. W. A lymphocyte-specific cellular Immunol.. 137: 2814-2822. 1986. enhancer is located downstream of the joining region in immunoglobulin 38. Rubin. J. T.. Elwood. L. J.. Rosenberg. S. A., and Lotze, M. T. Immunohis- heavy chain genes. Cell, 33: 729-740. 1983. IM,lirniir.il correlates of response to recombinant in(erlcukin-2 based im- 30. Bartsch. D.. Brichacek. B.. and Hirsch, I. Transfection of human lymphocytes munotherapy in humans. Cancer Res., 49: 7086-7092, 1989.

7128

Downloaded from cancerres.aacrjournals.org on September 28, 2021. © 1990 American Association for Cancer Research. Potentiation of Tumor Lysis by a Bispecific Antibody that Binds to CA19-9 Antigen and the Fc γ Receptor Expressed by Human Large Granular Lymphocytes

Irma García de Palazzo, Cicek Gercel-Taylor, Joanne Kitson, et al.

Cancer Res 1990;50:7123-7128.

Updated version Access the most recent version of this article at: http://cancerres.aacrjournals.org/content/50/22/7123

E-mail alerts Sign up to receive free email-alerts related to this article or journal.

Reprints and To order reprints of this article or to subscribe to the journal, contact the AACR Publications Subscriptions Department at [email protected].

Permissions To request permission to re-use all or part of this article, use this link http://cancerres.aacrjournals.org/content/50/22/7123. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC) Rightslink site.