|
Advertisement |
|
|
||
 |
|
|||||
|
|||||
|
 |
||||||
|
||||||
|
||||||
|
© 1999 American Society for Clinical Oncology
Karnofsky Lecture: Immunotherapy of LymphomaFrom the Division of Oncology, Stanford University School of Medicine, Stanford, CA. Address reprint requests to Ronald Levy, MD, Division of Oncology,M207, Stanford University School of Medicine, Stanford, CA 94395-5306;email levy{at}leland.stanford.edu MONOCLONAL ANTIBODIES are the first example of the payoff for cancer treatment that comes from our knowledge of the immune system. Monoclonal antibodies were the product of a fundamental discovery, and they are now changing the paradigm of how diseases are diagnosed and treated. Harold Varmus1 has pointed out that most of the revolutionary changes that have occurred in biology and medicine are rooted in new methods. Those new methods are in turn usually rooted in fundamental discoveries. The discovery of monoclonal antibodies began with the work of Klinman et al2 (Fig 1). They transferred cells from an immunized mouse to a syngeneic, identical twin that had been lethally irradiated and then cultured small pieces of the recipient's spleen. Some of the spleen fragments contained only one lymphocyte from the donor. Those cultures were then stimulated in vitro with the antigen, and they produced monoclonal antibodies. With this technology, Klinman et al were able to show that monoclonal antibodies could be produced against any antigen.
In 1975, Kohler and Milstein3 then discovered that the antibody-producing cells from a mouse spleen could be immortalized by being fused with myeloma cells. These fused cells (hybridomas) were able to live forever and produce antibodies one at a time. This discovery provided researchers with a set of reagents that could be used indefinitely. Kohler and Milstein were interested in basic scientific questions: how the immune system works, how genes from two cells work when mixed together, and how proteins can assemble when two cells combine. However, this research resulted in a new and powerful tool for generating diagnostic reagents and even new drugs. After these groundbreaking studies, researchers involved in cancer and immune-system research were suddenly able to tap the mouse immune system to search for differences between cancer cells and normal cells. Researchers were then able to make panels of monoclonal antibodies and test them for the ability to recognize antigens expressed by tumor cells that would not be expressed by normal cells. In one of our early experiments, Levy et al4 analyzed a panel of hybridoma antibodies for their ability to recognize antigens on the surface of leukemic cells and on normal cells taken from the same person (Fig 2). In this manner, we discovered molecules on the surface of leukemia cells that would help us in diagnosing the disease and that would serve as candidates for therapeutic attack. The engine of discovery was the use of monoclonal antibodies. Once the antibodies were identified, the candidates found, and the hybridomas grown, unlimited quantities of these antibodies were available. The stage was set for researchers to use these monoclonal antibodies to treat patients.
The first attempt at using monoclonal antibodies in cancer treatment was reported by Nadler et al5 in 1980. They treated a lymphoma patient by use of a monoclonal antibody. However, their experiment did not work because the target that they chose was exported by the tumor cells into the serum of the patient, thereby blocking the antibody's ability to reach the tumor. However, the feasibility and safety of the maneuver was established. In 1982, we reported the first successful use of a monoclonal antibody in the therapy of a patient with B-cell lymphoma.6 The patient experienced a complete remission of his disease in response to the monoclonal antibody, without any side effects. This remission lasted for 7 years. He is currently alive and healthy, now 17 years later. The secret to the success of this therapy was to choose a good target on the cancer cell. Every B cell makes an immunoglobulin. Each B cell's immunoglobulin is unique and different. When a particular B cell develops into a lymphoma, that lymphoma is a clone of cells, all of which have the same immunoglobulin on their surface. The unique portion of the immunoglobulin, the idiotype, is an ideal target for therapy. It is present on all the cancer cells and is not present on any of the normal cells in the body. The difficulty in using this target, however, is that every tumor has a unique target, and, therefore, a different monoclonal antibody must be made for each patient. We developed such custom-made monoclonal antibodies for a series of patients and tested them for therapeutic effects. Table 1 summarizes the outcome and current status of each of the 52 patients we treated. The majority of the patients had clinically significant, sometimes complete, regressions of their tumors. Several of the complete remissions lasted 10 years or longer. This was quite a remarkable effect, considering that the treatment had virtually no toxicity.
We believe that this treatment was successful because the target was special—that is, a target that needs to be on the surface of a lymphocyte for it to survive. In addition, the target transduces a death signal when it is cross-linked by a monoclonal antibody. Indeed, Renschler et al7 showed that this death-signal transduction can be accomplished in a different manner. They screened a library of random peptides and found peptides that could bind to the target for each individual patient. When these peptides were linked together so that they cross-linked the idiotype molecules, a signal was transduced into the cell, ultimately leading to the death of the cell. Later, Vuist et al8 showed that the response of our patients to the antibody therapy correlated with the ability of the antibody to transduce this signal. Later, as researchers attempted to generalize this maneuver and choose targets that were more generic, that could be used for more than one person's tumor, molecules on the surface of B cells that are signal transducers or growth-factor receptors became the candidates. When Nabil Hanna was given the opportunity to choose a target against which to design a generic antibody, he had a variety of possibilities for targets then known to exist on the surface of B lymphocytes. We had become aware of a dozen or more of these targets, including growth-factor receptors and functional coreceptors. Of these, the least understood target, even today, is the CD20 molecule, originally discovered by Stashenko et al.9 It is known that the CD20 molecule is present only on cells in the B-lymphocyte lineage. It is not expressed by pro-B cells or by plasma cells but is expressed by mature lymphocytes, both normal and malignant. A chimeric anti-CD20 monoclonal antibody was created by Reff et al.10 When this antibody was infused into monkeys, it depleted the B lymphocytes, not only from the blood but also from the tissues.11 Clinical trials using this monoclonal antibody against the CD20 molecule were then designed. The first trial was initiated by Maloney et al12 in 1993. In that trial, a single-dose escalation of the antibody was used in 15 patients. A subsequent trial involving a fixed dose of the antibody in 47 patients was performed in a multicenter setting.12 In both of these trials, half of the patients had clinical regressions of their tumors. An additional trial of 166 patients was conducted between 1995 and 1996 which proved that the agent was safe and confirmed that the combined complete and partial remission rate was approximately 50%.13 Antonio Grillo-Lopez analyzed the data from this last trial (Fig 3) and showed most of the patients actually experienced some tumor shrinkage. The tumor regressions lasted for a median duration of 13 months. These results led to the approval of this drug by the Food and Drug Administration. In the first year since the drug was approved, over 14,000 patients were treated with it in the United States alone. We now know that this drug can be used repeatedly: when the first remission wears off and the tumors progress, the patient can be treated and respond again (Davis et al, manuscript in preparation). Remarkably, some of these subsequent remissions have lasted longer than the first remission. The magnitude of the impact that this treatment is going to have on our management of patients with lymphoma has not yet been realized. Certainly this treatment method can be combined with other modes of therapy because it has so little toxicity.
Many new clinical trials using this antibody are now being conducted. The original studies of low-grade lymphoma have now been extended to include aggressive lymphoma, mantle-cell lymphoma, posttransplantation lymphoproliferative disorder, myeloma, and autoimmune diseases. Untreated patients, relapsed patients, maintenance therapy, treatment before chemotherapy, treatment combined with chemotherapy, bone marrow transplantation, and cytokines are all being investigated. An example of such a trial is one that we performed in conjunction with other centers, in which this antibody was combined with a short course of interferon treatment (manuscript submitted for publication). Thirty-four patients were entered onto the study. The response rate was approximately the same as those in earlier studies. However, the duration of these responses was on the order of 23 months, approximately double that observed previously with antibody alone. This was a small number of patients, and larger prospective trials are needed to establish this point; however, I think the principle is clear—monoclonal antibodies can be combined with other agents to achieve enhanced effects. Currently, there are eight monoclonal antibodies approved by the Food and Drug Administration for therapy of a variety of diseases and conditions, including renal transplantation, organ transplantation, heart disease, infectious diseases, autoimmune diseases, and two different malignancies: rituximab (CD20) for B-cell lymphoma and trastuzumab (HER2) for breast cancer (Table 2). There are now at least 75 monoclonal antibodies in late-stage clinical trials for a variety of disease indications, including cancer; we can anticipate that at least some of those will become approved drugs in the near future.
Initially, there was no reason to believe that antibodies alone would be so effective. The original intention was to use the antibodies as delivery devices—"homing" devices—to direct other agents of therapy, such as chemotherapeutic drugs, toxins, or radioactivity, to the tumor cells. The obvious advantage of this would be that an antitumor effect would be achieved not only from the antibody but also from the agent that was attached to it. For example, a radioactive nuclide could be delivered to some tumor cells and to neighboring tumor cells that were not necessarily targeted by the antibody. However, this same cross-fire effect would also reach normal cells in the vicinity of the tumor cell, such as bone marrow cells, and cause undesirable side effects. The clinical studies of Kaminsky et al14 and Liu et al15 demonstrated the dramatic antitumor effects of a radiolabeled version of an anti-CD20 antibody. In a pivotal trial conducted by Kamisky et al,16 involving 60 patients, an impressive result was obtained. Because of concerns that the radiation might cause increased toxicity, this trial was performed in patients whose disease was refractory to chemotherapy and progressive. Only approximately 25% of the patients in this trial had responded to their last course of chemotherapy and even those responses had lasted only an average of less than 6 months. These same patients were subsequently treated with the radiolabeled version of the CD20 antibody and achieved significantly more frequent and longer durations of response. I turn now from the topic of monoclonal antibodies to that of tumor vaccines. In doing so, I move from the experiment that has clearly worked to the experiment that is still in progress. The proven use for monoclonal antibodies has so far been for the treatment of patients in relapse. By contrast, the most likely use for tumor vaccines will be for patients in remission. The advantages of this shift are obvious. Active vaccination induces immunologic memory. It induces both antibodies and T lymphocytes. The response is heterogeneous and polyclonal and does not depend on one specific target. In addition, it is not as subject to the potential limitation caused by mutation of the target in the tumor cells. Therefore, it would be highly desirable to induce an immune response by patients against their own tumors. The question again becomes what the target should be. One might think that any target which worked for passive antibodies should potentially be a good target for a vaccine. However, the problem is immunologic tolerance. The normal B-cell molecules, such as CD20, are difficult to induce an immune response against because humans are immunologically tolerant. Therefore, we returned to our original target: the immunoglobulin idiotype. The question then became, Can the patients make an immune response against the idiotype of their own tumor? To do this, a biopsy must first be performed to obtain a sample of a patient's tumor. A hybridoma is then made by fusion of the tumor cells to a myeloma cell. These hybridomas secrete the immunoglobulin-bearing idiotype made by the tumor cell. The idiotype is then purified and chemically coupled to a foreign protein to make it more recognizable by the immune system. The patients selected for these studies were in remission after having received standard chemotherapy, and, hopefully, their immune systems had recovered. These patients were immunized with their own tumor-derived idiotype protein. The question again was, Can they do it? Do patients have the ability to respond to this vaccine and make immune responses against their tumors? The answer is yes.17 In the example shown in Fig 4, the serum of the patient contained antibodies after immunization that reacted with her tumor cells and not with normal cells in the biopsy specimen. Approximately half of the vaccinated patients made either an antibody response, a T-cell response, or both. The next question was, Does the immune response make any difference?
Figure 5 shows the updated follow-up of our vaccinated patients.18 The results show a superior outcome for the patients who made the immune response, compared with those who did not. When the survival was examined from the time of the patients' original diagnoses, with a median follow-up time of 8 years, the entire vaccinated group had a superior outcome, compared with that of a historical control group from our own database at Stanford. The entire benefit that occurred in this vaccinated group is attributable to the subgroup that made the immune response, because there have been no mortality events in that subgroup, as opposed to the historical control group and the subgroup that did not make an immune response (Fig 6). Thus the data from this study show a strong correlation between the ability of the patient to make an immune response and a better clinical outcome. This correlation, of course, is not a proof of efficacy. Prospective, controlled, randomized, clinical trials will be needed, and such trials are now in the planning stages.
In the meantime, new methods of immunization are being developed. For example, technologies exist that can potentially make the customized tumor vaccine approach practical so that many patients can receive such treatment. The first approach takes advantage of new methods for the production of proteins in plants. For instance, one can produce genetically engineered tobacco plants to express proteins from mouse cells. Investigators at the Biosource Corporation have developed a form of the tobacco mosaic virus into which foreign genes can be inserted.20 This virus can then infect a plant and take over its protein synthesis machinery. The foreign protein—in this case, the tumor-derived idiotype—can then be extracted from the leaves of the plant and used as a vaccine. McCormick et al21 have tested this method in a preclinical model and it does appear to work. Animals vaccinated with a product produced in the leaves of a tobacco plant were protected from the growth of the lymphoma. This technology could make it possible to turn tobacco, a cause of cancer, into a treatment for cancer! A second technology, which may be successful, is the use of naked DNA as a vaccine. Stevenson et al22 suggested that there is no need to make the idiotype protein for use as a vaccine. Instead, the gene coding for the protein can be cloned into a mammalian expression vector, which can be injected directly into a muscle. The protein, the gene, or both find their way to the draining lymph nodes and to the antigen-presenting cells (dendritic cells) that process the protein and "talk" to the immune system. This approach of naked DNA vaccination would be the easiest of all. One could amplify the genes from the tumor, place them in a vector, inject them into a host, and achieve an immune response. This also works in an animal model.23 The naked DNA vaccination method is currently undergoing clinical trial in patients with lymphoma. What will the future targets be? The idiotype is a special case that we know about because of our understanding of the immune system and our knowledge that every B cell has a unique and different immunoglobulin on its surface. Obviously, it would be desirable to apply this technology that we are developing from this special situation to other types of cancer, for which we do yet have targets that are completely tumor-specific. In the past, using hybridoma technology, we searched for antigens that are differentially expressed on tumor cells. In the future, we will be looking for genes that are differentially expressed by tumor cells. All the genes expressed by humans will soon be identified. DeRisi et al24 developed a strategy for searching for genes that are over- or underexpressed in any cell. The genes are isolated and placed onto an array of tens of thousands of genes at a time. Genes expressed from various sources (eg, tumor cells) could be compared with those from other sources (eg, normal cells). In an example of this method, Alizadeh et al25 collected a series of approximately 10,000 different genes that are expressed only in lymphocytes and made an array of these genes. The data already obtained using this "lymphochip" has disclosed genes uniquely expressed in large-cell lymphoma. In the next decade, genomics will be the engine of discovery, the way that researchers will discover new targets, just as hybridomas were used in the last two decades. The discovery process has come full circle, and researchers are now working in a new era in which new targets will be discovered. It will be a daunting problem to deal with the massive amount of data that will be generated by these new genetic methods. Advances in informatics will certainly be necessary. However, the big challenge will then be to translate that information to benefit patients with cancer. We need and depend on a continuing supply of clinical investigators who know the diseases and the current therapies and their limitations. As Harold Varmus1 put it, "Clinical advances require clinical research—an embattled enterprise that needs help to recruit, train and support talented people." We can only hope that we will have enough talented people who can be trained and supported in carrying out this important task. ACKNOWLEDGMENTS I thank the following people for their contributions to the work that I have presented here while they were at the Stanford University School of Medicine: Richard Miller, MD, and David Maloney, MD, PhD, who performed the first clinical trial on a patient that used an anti-idiotype antibody; John Timmerman, MD, who is running our current clinical trials; Shinichiro Kon, MD, Michael Campbell, PhD, Stanford Stewart, MD, Kris Thielemans, MD, PhD, Timothy Meeker, MD, Mark Kaminski, MD, Laura Esserman, James Lowder, Neil Berinstein, Mi Hua Tao, Sherri Brown, Charles Starnes, PhD, Thomas Chen, MD, PhD, Larry Kwak, MD, David Bahler, Itzak Hakim, Frank Hsu, Andrew Zelenetz, Athanasea Syrengelas, Clemens Caspar, MD, Wim Vuist, PhD, Carmen Wong, PhD, Thomas Davis, Markus Renschler, MD, Craig Okada, MD, PhD, Volker Reichardt, PhD, Arcangelo Liso, MD, Alison McCormick, PhD, and Edgar Engleman, MD, who worked on the biology of the lymphoma cells and who developed strategies for vaccination; my wife, Shoshana Levy, PhD, who has been a critical contributor to all of my work; and Henry Kaplan, MD, and Saul Rosenberg, MD, my teachers and mentors, who through their pioneering work on the treatment of Hodgkin's disease established the paradigm of clinical research that we all follow today. REFERENCES 1. Varmus H: Common science. Worldlink Mag XI:34-36, 1998 2. Klinman NR, Segal GP, Gerhard W, et al: Obtaining homogeneous antibody of desired specificity from fragment cultures, in Haber E, Krause RM (eds): Antibodies in Human Diagnosis and Therapy. New York, NY, Raven Press, 1977, pp 225-236 3. Kohler G, Milstein C: Continuous cultures of fused cells secreting antibody of predefined specificity. Nature256:495-497, 1975[Medline]
4.
Levy R, Dilley J, Fox RI, et al: A human thymus-leukemia antigen defined by hybridoma monoclonal antibodies. Proc Natl Acad Sci U S A76:6552-6556, 1979
5.
Nadler LM, Stashenko P, Hardy R, et al: Serotherapy of a patient with a monoclonal antibody directed against a human lymphoma-associated antigen. Cancer Res40:3147-3154, 1980 6. Miller RA, Maloney DG, Warnke R, et al: Treatment of B cell lymphoma with monoclonal anti-idiotype antibody. N Engl J Med306:517-522, 1982[Medline]
7.
Renschler WF, Bhatt RR, Dower WJ, et al: Synthetic peptide ligands of the antigen binding receptor induce programmed cell death in a human B-cell lymphoma. Proc Natl Acad Sci U S A91:3623-3627, 1994
8.
Vuist W, Maloney D, Levy R: Lymphoma regression induced by monoclonal anti-idiotypic antibodies correlates with their ability to induce Ig signal transduction and is not prevented by tumor expression of high levels of bcl-2 protein. Blood83:899-906, 1994 9. Stashenko P, Nadler LM, Hardy R, et al: Characterization of a human B lymphocyte-specific antigen. J Immunol125:1678-1685, 1980[Abstract]
10.
Reff ME, Carner K, Chambers KS, et al: Depletion of B cells in vivo by a chimeric mouse human monoclonal antibody to CD20. Blood83:435-445, 1994 11. Maloney D, White C, Bodkin D, et al: IDEC-C2B8: Results of a Phase I multiple-dose trial in patients with relapsed non-Hodgkin's lymphoma. J Clin Oncol15:3266-3274, 1997[Abstract]
12.
Maloney D, Liles T, Czerwinski D, et al: Phase I clinical trial using escalating single dose infusion of chimeric anti-CD20 monoclo-nal antibody (IDEC-C2B8) in patients with recurrent B-cell lymphoma. Blood84:2457-2466, 1994 13. McLaughlin P, Grillo-Lopez A, Link B, et al: Rituximab chimeric anti-CD20 monoclonal antibody therapy for relapsed indolentlymphoma: Half of patients respond to a four-dose treatment program. J Clin Oncol16:2825-2833, 1998[Abstract]
14.
Kaminski MS, Zasadny KR, Francis IR, et al: Iodine-131-anti-B1 radioimmunotherapy for B-cell lymphoma. J Clin Oncol14:1974-1981, 1996 15. Liu SY, Eary JF, Petersdorf SH, et al: Follow-up of relapsed B-cell lymphoma patients treated with iodine-131-labeled anti-CD20 antibody and autologous stem-cell rescue. J Clin Oncol16:3270-3278, 1998[Abstract] 16. Kaminski MS, Zelenetz AD, Press O, et al: Multicenter, phase III study of iodine-131 tositumomab (anti-B1 antibody) for chemotherapy-refractory low-grade or transformed low-grade non-Hodgkin's lymphoma (NHL). Blood 92:316a, 1998 (abstr 1296) 17. Kwak LW, Campbell MJ, Czerwinski DK, et al: Induction of immune responses in patients with B cell lymphoma against the surface immunoglobulin idiotype expressed by their tumors. N Engl J Med327:1209-1215, 1992[Abstract]
18.
Hsu FJ, Caspar CB, Czerwinski D, et al: Tumor-specific idiotype vaccines in the treatment of patients with B cell lymphoma: Long-term results of a clinical trial. Blood89:3129-3135, 1997 19. Kaplan EL, Meier P: Non-parametric estimation from incomplete observations. J Am Stat Assoc53:457-481, 1958
20.
Kumagai MH, Turpen TH, Weinzettl N, et al: Rapid, high-level expression of biologically active alpha-trichosanthin in transfected plants by an RNA viral vector. Proc Natl Acad Sci U S A90:427-430, 1993
21.
McCormick A, Kumagai M, Hanley K, et al: Rapid production of specific vaccines for lymphoma by expression of the tumor derived single chain Fv epitopes in tobacco plants. Proc Natl Acad Sci U S A96:703-708, 1999 22. Stevenson FK, Zhu D, King CA, et al: Idiotypic DNA vaccines against B-cell lymphoma. Immunol Rev145:211-228, 1995[Medline] 23. Syrengelas A, Chen T, Levy R: DNA immunization induces protective immunity against B cell lymphoma. Nat Med2:1038-1041, 1996[Medline]
24.
DeRisi JL, Iyer VR, Brown PO: Exploring the metabolic and genetic control of gene expression on a genomic scale. Science278:680-686, 1997 25. Alizadeh AA, Eisen MB, Davis RE, et al: The lymphochip: A specialized cDNA microarray for the genomic-scale analysis of gene expression in normal and malignant lymphocytes. Cold Spring Harb Symp Quant Biol (in press) This article has been cited by other articles:
|
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
|
|||||||||||
|
|||||||||||
|
|
Copyright © 1999 by the American Society of Clinical Oncology, Online ISSN: 1527-7755. Print ISSN: 0732-183X
|
