Emerging Immunotherapies for Non-Hodgkin Lymphomas: The Tortoise Approaches the Finish Line
O. Press
DOI: https://doi.org/10.7326/0003-4819-132-11-200006060-00012
IF: 39.2
2000-06-06
Annals of Internal Medicine
Abstract:Disseminated malignancies continue to pose vexing problems for both patients and health care providers. Although localized malignant lesions can often be extirpated with surgery or radiation therapy, most metastatic neoplasms are difficult to cure despite systemic chemotherapy. Furthermore, the alopecia, nausea, myelosuppression, neurotoxicity, and nephrotoxicity that often accompany conventional chemotherapy are potent deterrents for many patients. Consequently, investigators and patients with cancer have long sought alternative therapies that might be effective while avoiding the dreaded side effects of chemotherapy. Immunologic interventions have been particularly appealing because they use natural components of the immune system to selectively eradicate malignant cells while leaving normal cells unharmed. Immunotherapeutic attempts have included active approaches, such as development of tumor vaccines, and passive or adoptive immunotherapy, in which components of the immune system (antibodies, interferons, interleukins, activated T lymphocytes) are produced exogenously before they are administered to patients with cancer. Of these strategies, monoclonal antibodies have achieved the most recent success and publicity. Paul Ehrlich is widely credited as the visionary who initially promulgated the concept of therapeutic anticancer antibodies in the late 1800s and early 1900s. Little progress was made toward realization of his vision until 1975, when Kohler and Milstein published their seminal paper describing the mouse hybridoma method (1). This technique permitted large-scale production of high-quality, homogeneous antibodies of defined specificity. The imaginations of researchers, journalists, and the lay public were soon captured by the potential of monoclonal antibodies, and unrealistic predictions concerning the rapidity with which monoclonal antibodies would replace conventional cancer therapies were widely published in the lay and scientific literature. Unfortunately, early preclinical and clinical trials with monoclonal antibodies encountered a variety of obstacles: a paucity of well-defined tumor-specific antigens, interference from circulating free antigens shed from tumor cells into the bloodstream, the immunogenicity of murine antibodies, slow and heterogeneous penetration into tumor sites, and inadequate potency. Exaggerated predictions of early success with anticancer antibodies gradually degenerated into disillusionment by funding agencies, investors, the lay public, and scientists themselves when two decades transpired without U.S. Food and Drug Administration approval or marketing of a single therapeutic antibody for cancer. Attendance at antibody-oriented scientific meetings dwindled dramatically in the 1980s, biotechnology companies that focused on antitumor antibodies went bankrupt, National Institutes of Health grants in the area became difficult to obtain, and investigators in the field were often derisively regarded as stubborn proponents of a failed technology. Despite the pessimism that enveloped the field for a decade, gradual progress was slowly being made in the laboratory and in the clinic. Technological advances permitted the ex vivo production of monoclonal antibodies in bioreactors or hollow fiber systems rather than in mouse ascitic fluid. Basic scientists exploited developments in molecular biology to develop chimeric and humanized antibodies that were less immunogenic than murine antibodies and interacted more effectively with human effector cells and complement (2). Scientists discovered phage display methods that permitted rapid selection and development of high-affinity antibodies of any defined specificity. Molecular engineering permitted the large-scale production and purification of single-chain antibodies, minibodies, and other antigen-combining proteins with sizes ranging from 25 to 150 kD. Progress was also being made in the clinic, where investigators soon recognized that hematologic malignancies, particularly B-cell lymphomas, were ideal for testing the potential of anticancer antibodies. Lee Nadler (3) demonstrated the safety and negligible toxicity of monoclonal antibody infusions in patients with lymphoma. Ron Levy and Richard Miller conducted brilliant studies at Stanford University with anti-idiotypic antibodies that recognize the tumor-specific portion of the surface immunoglobulin molecule present on malignant B lymphocytes. The first patient treated with anti-idiotypic antibodies achieved a complete clinical remission that lasted more than 7 years, and 33 of 52 patients treated with this sophisticated method achieved partial or complete remissions (4, 5). Despite the undeniable efficacy of this approach, anti-idiotypic antibodies have not been widely adopted or commercially developed because of the logistic, financial, and temporal constraints inherent in producing and purifying patient-specific antibodies. Nevertheless, the proof of principle provided by anti-idiotypic antibodies encouraged other investigators, including our group in Seattle, to focus on antibodies directed against lineage-specific pan B antigens, particularly the CD20 molecule. This molecule is reliably expressed on both malignant and normal B lymphocytes, but not on other cell types. In 1985, we reported the first clinical trial using monoclonal anti-CD20 antibodies and documented the safety and promise of this approach (6). Two of four patients treated with murine anti-CD20 antibodies demonstrated tumor regressions, including one patient with adenopathy in whom 90% of the tumor disappeared. Unfortunately, the responses observed in this pilot study were transient and partial. Maloney subsequently obtained more durable responses using a molecularly engineered, chimeric anti-CD20 antibody now known as rituximab (7). The human Fc portion of this antibody fixes complement, recruits host effector cells, and mediates apoptosis much more effectively than its murine predecessors. These advantages translated into improved clinical efficacy as well; 50% of treated patients with indolent lymphomas and 30% to 35% of patients with aggressive lymphomas achieved objective remissions with minimal toxicity in multicenter trials (8, 9). Remissions lasted approximately 1 year in the typical patient, and most patients were delighted with the lack of alopecia, myelosuppression, or nausea associated with rituximab therapy. These compelling studies led to Food and Drug Administration approval of rituximab for treating relapsed lymphomas in November 1997, making it the first monoclonal antibody approved for cancer therapy. Despite the unequivocal success and popularity of rituximab, it is not a panacea. Half of the patients with relapsed low-grade lymphomas and 60% to 70% of patients with relapsed aggressive lymphomas do not respond to rituximab. Responses, when they do occur, last only a median of 10 to 12 months, and no compelling evidence yet demonstrates that overall survival is prolonged with rituximab (or any other therapy for relapsed indolent lymphomas). Several strategies are being used to improve current results, including rituximab administration to patients with newly diagnosed lymphoma, combination therapy with standard chemotherapy or other biological agents, and integration of rituximab with high-dose chemoradiotherapy and stem-cell transplantation. Whether these protocols will improve patient outcomes remains to be seen. However, a preliminary trial by Czuczman and colleagues that involves administering rituximab with standard CHOP chemotherapy (cyclophosphamide, doxorubicin, vincristine, and prednisone) to 40 patients with newly diagnosed follicular lymphoma yielded a response rate of 95% and a progression-free survival rate of 74% after a median follow-up of 29 months (10). An alternative approach to augment the efficacy of anti-CD20 antibodies involves conjugating them to radionuclides, an approach that permits tumor-specific targeting of cytocidal ionizing radioactivity to neoplastic cells. This strategy has three major advantages. First, radiolabeled antibodies are not heavily reliant on the host immune system, which is often woefully deficient in patients with lymphoma. Second, the particles emitted by radioisotopes such as iodine-131 and yttrium-90 are effective over many cell diameters, permitting eradication of antigen-negative tumor mutant clones and cells embedded deep in tumor masses, which would be unaffected by nonradioactive antibodies. Finally, the cytocidal effects of the ionizing particles are very potent, permitting higher overall response rates, higher complete response rates, and longer remission durations than have been reported with unconjugated antibodies. Studies using minimally toxic, nonmyeloablative doses of iodine-131tositumomab (anti-CD20) (11) and yttrium-90Y2B8 (12, 13) have demonstrated response rates of 67% to 80%, including complete remission rates of 27% to 40% in patients with relapsed lymphomas. More impressively, 100% of patients with newly diagnosed indolent lymphomas responded to nonmyeloablative doses of iodine-131tositumomab in a recent trial, and more than 50% had complete remission (14). Although these trials of low-dose radioimmunotherapy are extremely encouraging, it appears that few patients will be permanently cured with nonmyeloablative doses of radiolabeled antibodies. Our group in Seattle has explored the use of myeloablative doses of iodine-131tositumomab in conjunction with autologous stem-cell transplantation in patients with multiply-relapsed B-cell lymphomas. This high-dose approach yielded objective remissions in 86% of treated patients with relapsed B-cell lymphomas, including a complete response rate of 79% (15-17). Many of these responses have been very durable, with 11 of 29 patients remaining in continuous complete remission without any other therapy for 5 to 10 years. Recent pilot studies at our institution suggest that myeloablati