Bone Marrow Transplantation (2000) 25, Suppl. 2, S50–S53  2000 Macmillan Publishers Ltd All rights reserved 0268–3369/00 $15.00 www.nature.com/bmt Treatment of patients with malignant lymphomas with monoclonal

H Tesch, A Engert, O Manzke, V Diehl and H Bohlen

Klinik I fuer Innere Medizin, Universitaet zu Koeln, Koeln, Germany

Summary: Results and discussion

Malignant lymphomas represent a heterogenous group of B and -derived malignancies. Most lymphomas Native monoclonal antibodies are sensitive to chemo- and radiotherapy, however many patients will eventually relapse. Immunothera- Since the first description of therapy using monoclonal anti- peutic approaches including monoclonal antibodies, bodies in 1979, several phase I and II trials have been or vaccination approaches may offer an alter- initiated to evaluate both safety and antitumoral activity of native treatment of chemotherapy-resistant residual this approach. Native MoAbs can kill a tumor cell through cells especially in cases with low tumor burden or various mechanisms including complement activation, anti- residual disease following chemo- or radiotherapy. body-dependent cellular cytotoxicity (ADCC), phago- Monoclonal antibodies have been successfully applied in cytosis of -coated tumor cells, inhibition of cell their native form, or coupled with radioisotopes or tox- cycle progression, and induction of apoptosis.2,3 Alterna- ins to selectively destroy lymphoma cells and promising tively, MoAbs can eliminate a tumor cell by inhibiting results in early clinical trials have been obtained. Alter- growth factor receptors or molecules involved in signal natively, bispecific antibodies and idiotypic vaccination transduction and cell proliferation. strategies are used to target autologous T cells to elimin- The group of R Levy at Stanford reported on promising ate lymphoma cells. A humanized anti-CD20 antibody results in B-lymphocytic malignancies using anti-idiotypic showed excellent results in chemotherapy refractory monoclonal antibodies (anti-Id MoAbs).4,5 Side-effects lymphomas and has recently been approved for clinical were moderate and included chills, fever, dyspnea, nausea, application in CD20 positive lymphomas. Bone diarrhea, myalgia, and the development of human anti- Marrow Transplantation (2000) 25, Suppl. 2, S50–S53. mouse antibodies (HAMA). Complete and partial Keywords: malignant lymphoma; monoclonal anti- remissions had been observed together with a strong corre- bodies; immunotherapy lation of anti-Id MoAb-mediated signal transduction and tumor regression as measured by the induction of cellular protein tyrosine phosphorylation.6 Several trials evaluated Malignant lymphomas represent a heterogeneous group of the effect of the Campath-1H MoAb directed against the B and T cell-derived tumors. Most entities respond to CDw52 , expressed on B and T cell lymphomas. In chemo- and/or radiotherapy and, depending on the histo- addition to common side-effects such as fever, rigor and logical subtype, up to 80% complete remissions (CR) can neutrocytopenia, severe opportunistic infections due to the be achieved. The main reason for tumor recurrence is the depletion of CD4- and CD8-positive T were development of lymphoma cell clones resistant to the initial reported.7 High response rates up to 70% were reported therapy. The use of monoclonal antibodies (MoAbs) or when Campath-1H was administered in CLL patients.8 other immunotherapeutic strategies may be useful to eradi- Exciting results have been demonstrated in clinical phase cate such chemotherapy-resistant tumor cells. Clinical trials I/II trials with the chimeric mouse/human anti-CD20 MoAb in lymphoma patients showed a strong correlation between IDEC-C2B8 (IgG1), which binds complement and induces poor outcome and the presence of minimal residual disease, ADCC as well as programmed cell death (apoptosis).2 A which can be detected by polymerase chain reaction (PCR)1 large phase II trial revealed a response rate of 50% in low at very high sensitivity. Also, autologous stem cell prep- and intermediate grade lymphomas with significantly better arations from patients with lymphomas frequently contain responses in patients with low tumor masses. At our depart- residual lymphoma cells. Since residual lymphoma cells ment more than 80 patients with relapsed or refractory lym- may contribute to an overt clinical relapse, the elimination phomas including mantle cell lymphomas were treated with of these residual cells may have a great clinical impact and IDEC-C2B8 so far and toxicities were generally low. How- is followed in purging strategies. ever, patients with a higher tumor load in the peripheral blood developed signs of severe release.9 Current trials investigate the role of anti-CD20 antibody treatment Correspondence: Dr V Diehl, Klinik I fuer Innere Medizin, Universitaet in high grade lymphomas and in combination with chemo- zu Koeln, Joseph-Stelzmann-Str 9, 50924 Koeln, Germany therapy. Antibody treatment of lymphomas H Tesch et al S51 Radioimmunoconjugates Grossbard and colleagues39 performed similar phase I stud- ies in patients with B-NHL using the anti-CD19 MoAb B4 Radioimmunoconjugates may offer a number of advan- coupled to blocked ricin (bR). In this study nine out of 25 tages: (1) endocytosis is not required for cytotoxicity; (2) patients developed HAMA and HARA, and one CR and the use of an isotope enables the eradication of antigen- two PR were reported. In a phase II trial, 50 patients were negative ‘bystander’ tumor cells; (3) radioimmunoconju- treated with ITs after high-dose chemotherapy and stem cell gates allow biodistribution experiments to detect the local- transplantation at a reduced dose of 30 ␮g/kg/day over 7 ization of the conjugate at the tumor site. Antibodies can days. In 34 evaluable patients no relapse occurred between either be labeled directly using 131iodine, 99mtechnetium or 99m 90 111 1 and 3 years after transplantation. The Cancer and Acute indirectly ( Tc, Y, In) using bifunctional chelating Leukemia Group B (CALGB) initiated a randomized multi- agents. center phase III study in which patients in CR after high- Several clinical trials using non-myeloablative and mye- dose chemotherapy receive anti-B4-bR over a 7-day loablative doses of radiolabeled MoAbs (R-MoAbs) were period).18 conducted in lymphoma patients. Excellent response rates were achieved with the 131I-labeled B1 MoAb against the CD20 antigen using a single non-myeloablative dose of 30– Bispecific monoclonal antibodies 10 160 mCi in patients with refractory lymphomas. Of 28 Bispecific monoclonal antibodies (BiMAbs) recognize two treated patients, 14 achieved a CR and eight PR. Non- different via their two binding sites.19 BiMAbs can hematologic toxicities were generally mild but hematologic be used to crosslink lymphoma cells with immunological toxicities proved dose-limiting. effector cells like , T cells or NK cells which High-dose radioimmunotherapy seems to be even more can lyse tumor cells. An alternative approach is the design superior compared to low-dose therapy. Out of 43 patients of BiMAbs directed against a tumor cell antigen and a pro- with relapsed NHL, which were initially trace-labeled, 24 11 drug-converting enzyme such as alkaline phosphatase to demonstrated satisfactory tumor imaging. Subsequently, convert relatively non-toxic cytostatic prodrugs like mito- these patients received a single infusion of the R-MoAb up mycin phosphate or etoposide phosphate into the active to 777 mCi. The dose limiting myelosuppression required cytostatic drugs mitomycin and etoposide, respectively. In reinfusion of autologous bone marrow in 15/19 patients. addition, toxins like saporin or radioisotopes conjugated to Two PR and 16 CR were observed among 24 patients, and can be targeted to tumor cells via BiMAbs. nine patients remained in CR over a period of 3 to 53 BiMAbs can be generated by chemical crosslinking of months. In a recently published phase II trial, 21 patients two monoclonal antibodies or F(abЈ) fragments,20 by were treated with therapeutic doses (345–785 mCi) of 131I- 12 somatic fusion of two distinct hybridoma cell lines (hybrid- B1, with subsequent autologous stem cell transplantation. hybridomas or tetradoma),19 or by recombinant antibody A summary of these phase I/II studies showed an overall technology.21 Most BiMAbs aim to activate T cells by survival of 93% with a median follow-up of 2 years. 22 ␥ ␤ crosslinking of the TCR/CD3 complex. After binding of In contrast to -emitters, -emitters deliver sufficient the BiMAb to the TCR/CD3 complex, the T-lymphocytes doses over a distance of 100 to 1000 cell diameters without 13 become pre-activated requiring further stimulation to fully the need for hospitalization. In a recently published trial, 23 90 exert their cytotoxic activity. The additional costimulating Y-labeled anti-Id MoAbs were used for the treatment of molecule can be provided by cytokines such as IL-2, IL- nine patients with refractory NHL. Toxicities were mild and 12, and TNF-␣, or by CD28-mediated costimulation. two patients achieved a CR and one patient a PR. Haagen et al24 tested several CD3xCD19 BiMAbs against B cell lines and primary lymphoma cells. These Immunotoxins BiMAbs induced remarkable cytotoxic activity in the pres- ence of activated T cells, as well as proliferation and cyto- Immunotoxins (ITs) can selectively bind and destroy their toxicity in resting T cells. We used bispecific CD3xCD19 specific target cell upon internalization.14 They consist of antibodies together with monospecific antibodies recogniz- a binding chain (usually a ) and a toxin ing the CD28-molecule on T lymphocytes. Our results molecule, either covalently bound via a chemical show that CD3xCD19 bispecific antibodies in combination crosslinker or generated genetically by recombinant tech- with CD28 monospecific antibodies can activate T cells to nology. The toxin is of plant or bacterial origin and includes lyse autologous lymphoma cells in vitro and in vivo.Ina in most instances ribosome-inactivating proteins to arrest clinical trial with 10 patients with relapsed B cell lym- protein synthesis.15 Ricin is a plant toxin most frequently phoma who were treated with a locoregional application of used for the construction of chemically linked ITs. ITs were CD3xCD19 bispecific antibodies together with CD28 anti- also generated using recombinant DNA technology by the bodies only moderate toxicities were found. The analysis fusion of antibody VH genes to bacterial toxins such as of immunological responses revealed temporary increases diphtheria toxin (DT)16 or Pseudomonas exotoxin-A in the T/B-ratio of the peripheral blood, as well as enhanced (ETA).17 activity. Vitetta and colleagues38 used anti-CD22 and anti-CD19 A phase I trial with three patients with low- and inter- antibodies coupled to deglycosylated ricin A-chain (dgA) mediate-grade NHL receiving CD3xCD19 BiMAb (SHR- for the treatment of B-NHL with dose-limiting toxicities 1) i.v. over 0.5 h in escalating doses ranging from 10 ␮g (DLT) including vascular leak syndrome, fever and myal- on day 1 to 5 mg on day 9,25 showed moderate toxicities gia. In 5/14 patients a PR of short duration was observed. inclusing chills, fever, and thrombocytopenia. Transient

Bone Marrow Transplantation Antibody treatment of lymphomas H Tesch et al S52 decreases in splenomegaly and lymphoma occured in two nized four NHL patients with autologous ex vivo generated of three patients. There were no detectable human anti- dendritic cells pulsed with idiotypic protein. None of the mouse or anti-rat antibodies. Another clinical investigation patients developed anti-idiotypic antibodies, however, in all in four patients suffering from end-stage low-grade B-NHL patients a substantial increase in the lymphoma-specific used a cocktail of two different BiMAbs directed against cellular response was demonstrated. The complete different on the CD22 antigen and saporin.26 All remission observed in one patient was most likely due to patients showed rapid but transient reduction of tumor a specific effect. Another patient with mini- volume. mal residual disease that was only detectable by PCR con- BiMAb-mediated phagocytosis of malignant cells can be verted to be PCR-negative. We currently investigate this achieved through targeting of macrophages.27 F(abЈ) frag- approach in patients with multiple myeloma. In our study ments recognizing the B cell differentiation markers CD19 myeloma patients are vaccinated with autologous dendritic and CD37 were chemically linked to the F(abЈ) portion of cells pulsed with the patient’s paraprotein. Twelve patients a MoAb which reacts with Fc␥RI/CD64 expressed on mon- were vaccinated by this approach. Four patients developed ocytes, macrophages and dendritic cells. Both BiMAbs idiotype-specific antibodies and in three patients an mediated tumor cell lysis by -derived macro- idiotype-specific T cell response could be detected. phages.

Vaccination with monoclonal antibodies Conclusions Vaccination against tumor-associated antigens is limited by the problem of obtaining sufficient amounts of purified anti- gen, as well as the poor immune response due to tolerance The major goal in future studies is the treatment of residual against autologous tumor antigens.28 However, active spe- disease in patients who have undergone first- or second- cific immunotherapy circumventing tumor tolerance is line treatment. The combination of different therapeutic implied by the network theory.29 This hypothesis suggests strategies, ie chemotherapy, radiotherapy, and immunother- that tumor antigen-like structures within the antigenic apy, will probably increase therapeutic efficacy and thereby environment of anti-idiotypic determinants are presented on the clinical outcome of the patients. variable regions of immunoglobulins. According to the net- Despite the promising results using unconjugated work theory by Jerne,29 idiotopes can induce a cascade of MoAbs, several obstacles remain. The development of neu- idiotypic antibodies termed Ab1, Ab2, Ab3, etc. The Ab2 tralizing human antibodies can be minimized by con- is an anti-idiotypic antibody directed against the variable structing chimeric humanized MoAbs or recombinant sin- region of the anti-tumor antibody (Ab1), thereby mimicking gle chain antibodies. New technologies to design high- the conformation of the original antigen like an internal affinity human MoAbs include the phage display tech- image. Thus, Ab2 itself can induce specific anti-tumor nique35 or transgenic ‘humanized’ mice.36 Recombinant response in the absence of the original tumor antigen and DNA technology can be used for site-directed modifi- can therefore serve as a vaccine. cations in order to select high-affinity antibodies. Immune responses can also be elicited against the sur- The selection of antigen-negative tumor cell variants face-Ig idiotypes expressed on malignant B cells.30 Active among the heterogeneous tumor cell population can, at least immunization against idiotypic determinants on murine B in part, be circumvented by the application of MoAb cock- cell lymphoma was demonstrated to inhibit tumor growth.31 tails, ie mixtures of MoAbs against different antigens on An improved immune response requires conjugation of the the same target cell, or by active tumor-specific vaccination Ig to a immunogenic carrier protein like keyhole-limpet approaches. Such cocktails of MoAbs have successfully hemocyanin (KLH).32 In a clinical trial conducted by Levy been used to purge bone marrow from residual lymphoma et al,9 NHL patients in clinical remission after chemo- cells,37 with a combination of three antibodies depleting the therapy were immunized with idiotype derived from their malignant cells in a magnitude of up to 6 logs. B cell lymphoma. Lymphoma cells obtained from tumor The ultimate goal to use monoclonal antibodies to cure biopsies of these patients were fused with a heterohybri- patients with relapsed or refractory lymphomas has not yet doma. The Ig-idiotype was isolated from the cell culture been reached. However, we have learned to use monoclonal supernatant, conjugated to KLH, and subsequently used as antibodies and their conjugates. Preliminary results are vaccine. Two patients showed humoral responses to the Ig- promising and large clinical trials in patients with residual idiotype, four patients had a cellular response, and one disease are required to analyze the role of monoclonal patient both. Side-effects were mild and two patients with antibodies in more detail. measurable disease had CR up to 24 months, six patients remained disease-free up to 10 months and one patient had an early relapse. A recent vaccination approach employs dendritic cells, which have been demonstrated to be the most effective anti- Acknowledgements gen-presenting cells.33 These cells express high levels of adhesion molecules like ICAM-1 and LFA-3, as well as This work was supported by the Deutsche Forschungsgemein- MHC-I, MHC-II and immunostimulatory proteins such as schaft, the Deutsche Krebshilfe, and the Bundesministerium fuer CD80 (B7-1), CD86 (B7-2), and CD40.31 Hsu et al34 immu- Bildung und Forschung.

Bone Marrow Transplantation Antibody treatment of lymphomas H Tesch et al S53 References mediated by human Fc receptors for IgG. Immunol Today 1989; 10: 92–97. 1 Drexler HG, Borkhardt A, Janssen JWG. Detection of 21 Schmidt M, Hynes NE, Groner B, Wels W. A bivalent single- chromosomal translocations in leukemia–lymphoma cells by chain antibody-toxin specific for ErbB-2 and the EGF recep- polymerase chain reaction. Leuk Lymph 1995; 19: 359–380. tor. Int J Cancer 1996; 65: 538–546. 2 Tedder TF, Engel P. CD20: a regulator of cell-cycle pro- 22 Perez P, Hoffman RW, Shaw S. Specific targeting of cytotoxic gression and B lymphocytes. Immunol Today 1994; 15: T-cells by anti-T3 conjugates. Nature 1985; 316: 354–356. 450–454. 23 Segal DM, Garrido MA, Perez P et al. Targeted cytotoxic cells 3 Yao XR, Scott DW. Expression of protein tyrosine kinases in as a novel form of cancer immunotherapy. Mol Immunol 1988; the Ig complex of anti-mu-sensitive and anti-mu-resistant B- 25: 1099–1103. cell lymphomas: role of the p55blk kinase in signaling growth 24 Haagen IA, Geerars AJ, de Lau WBM et al. The efficacy of arrest and apoptosis. Immunol Rev 1993; 132: 163–186. CD3 × CD19 bispecific monoclonal antibody (BsAb) in a 4 Meeker TC, Lowder J, Maloney DG et al. A clinical trial of clonogenic assay: the effect of repeated addition of BsAb and anti-idiotype therapy for B-cell malignancy. Blood 1995; 65: interleukin-2. Blood 1995; 85: 3208–3212. 1349–1363. 25 De Gast GC, Van Houten AA, Haagen IA et al. Clinical 5 Brown SL, Miller RA, Horning SJ et al. Treatment of B-cell experience with CD3xCD19 bispecific antibodies in patients lymphomas with anti-idiotype antibodies alone and in combi- with B-cell malignancies. J Hematother 1995; 4: 433–437. nation with alpha interferon. Blood 1989; 73: 651–661. 26 French RR, Hamblin TJ, Bell AJ et al. Treatment of B-cell 6 Vuist WMJ, Levy R, Maloney DG. Lymphoma regression lymphomas with combination of bispecific antibodies and induced by monoclonal anti-idiotypic antibodies correlates saporin. Lancet 1995; 346: 223–224. with their ability to induce Ig signal transduction and is not 27 Ely P, Wallace PK, Givan AL et al. Bispecific-armed, inter- prevented by their tumor expression of high levels of Bcl-2 feron gamma-primed -mediated phagocytosis of protein. Blood 1994; 83: 899–906. malignant non-Hodgkin’s lymphoma. Blood 1996: 87: 3813– 7 Rai KR, Hoffman M, Janson D et al. Immunosuppression and 3821. opportunistic infections (OI) in patients with chronic lympho- 28 Greene MI. Cellular and genetic basis of immune reactivity cytic leukemia (CLL) following Campath 1-H therapy. Blood to tumor cells. Contemp Top Mol Immunol 1980; 11: 81–87. 1995; 10 (Suppl.): 1381. 29 Jerne N. Towards a network theory of the . 8 Anagnostopoulos A, Fassas A, Papaminas N et al. Immuno- Ann Immunol 1974; 125: 373–389. therapy with Campath-1H for patients with lymphoprolifer- 30 Kwak LW, Campbell MJ, Czerwinski DK et al. Induction of ative disease. Blood 1995; 10 (Suppl.): 3297. immune responses in patients with B-cell lymphoma against 9 Winkler U, Jensen M, Manzke O et al. Cytokine-release syn- the surface-immunglobulin idiotype expressed by their tumors. drome in patients with B-cell chronic lymphocytic leukemia New Engl J Med 1992; 327: 1209–1215. and high counts after treatment with an anti-CD20 31 Campbell MJ, Carroll W, Kon S et al. Idiotype vaccination monoclonal antibody (rituximab, IDEC-C2B8). Blood 1999; against murine B-cell lymphoma. Humoral and cellular 94: 2217–2224. responses elicited by tumor derived and 10 Kaminski MS, Zasadny KR, Francis IR et al. Iodine-131-anti- its molecular subunits. J Immunol 1987; 139: 2825–2833. B1 radioimmunotherapy for B-cell lymphoma. J Clin Oncol 32 Campbell MJ, Esserman L, Byars NE et al. Development of 1996; 14: 1974–1981. a new therapeutic approach to B-cell malignancy. The induc- 11 Press OW, Eary JF, Appelbaum FR et al. Radiolabeled-anti- tion of by the host against cell surface receptor on body therapy of B-cell lymphoma with autologous bone mar- the tumor. Int Rev Immunol 1989; 4: 251–270. row support. New Engl J Med 1993; 329: 1219–1224. 33 Sallusto F, Lanzavecchia A. Efficient presentation of soluble 12 Press OW, Eary JF, Appelbaum FR et al. Phase II trial of antigen by cultured human dendritic cells is maintained by 131I-Bi (anti-CD20) antibody therapy with autologous stem granocyte/macrophage colony stimulating factor plus interleu- cell transplantation for relapsed B-cell lymphomas. Lancet kin-4 and downregulated by tumor necrosis factoralpha. J Exp 1995; 346: 336–340. Med 1994; 179: 1109–1118. 13 White CA, Halpern SE, Parker BA et al. Radioimmunotherapy 34 Hsu FJ, Benike C, Fagnoni F et al. Vaccination of patients of relapsed B-cell lymphoma with yttrium-90 anti-idiotype with B-cell lymphoma using autologous antigen-pulsed den- monoclonal antibodies. Blood 1996; 87: 3640–3649. dritic cells. Nature Med 1996; 2: 52–58. 14 Eiklid K, Olsnes S, Pihl A. Entry of lethal doses of abrin, 35 Griffiths AD, Williams SC, Hartley O et al. Isolation of high ricin and modeccin into the cytosol of HeLa cells. Exp Cell affinity human antibodies directly from large synthetic reper- Res 1980; 126: 321–326. toires. EMBO J 1994; 13: 3245–3260. 15 Endo Y, Tsurugi K. RNA N-glycosidase activity of ricin A- 36 Lonberg N, Taylor LD, Harding FA et al. Antigen-specific chain. Mechanism of action of the toxic lectin ricin on eukary- human antibodies from mice comprising four distinct genetic otic ribosomes. J Biol Chem 1987; 262: 8128–8130. modifications. Nature 1994; 368: 856–859. 16 Greenfield L, Bjorn MJ, Horn G et al. Nucleotide sequence 37 Gribben JG, Freedman AS, Neuberg D et al. Immunologic of the structural gene for diphtheria toxin carried by coryneb- purging of marrow assessed by PCR before autologous bone acteriophage beta. Proc Natl Acad Sci USA 1983; 80: 6853– marrow transplantation for B-cell lymphoma. New Engl J Med 6857. 1991; 325: 1525–1533. 17 FitzGerald D, Pastan I. Targeted toxin therapy for the treat- ment of cancer. J Natl Cancer Inst 1989; 81: 1455–1463. References added in proof 18 Engert A, Diehl V, Schnell R et al. A phase-I study of an anti-CD25 ricin A-chain immunotoxin (RFT5-SMPT-dgA) in 38 Vitetta ES, Stone M, Amlot P et al. Phase 1 immunotoxin patients with refractory Hodgkin’s lymphoma. Blood 1997; trial in patients with B-cell lymphoma. Cancer Res 1991; 51: 89: 403–410. 4052–4058. 19 Milstein C, Cuello AC. Hybrid hybridomas and their use in 39 Grossbard ML, Lambert JM, Goldmacher VS et al. Anti-B4- immunohistochemistry. Nature 1983; 305: 537–541. blocked ricin: a phase I trial of 7-day continuous infusion in 20 Fanger MW, Shen L, Graziano RF et al. Cytotoxicity patients with B-cell neoplasms. J Clin Oncol 1993; 11: 726.

Bone Marrow Transplantation