| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
EDITORIAL |
From the Departments of Surgery (HKL, TMC), Pathology and Immunology (HKL), and Medicine (MAM), Duke University Medical Center, Durham, North Carolina.
Correspondence: Address correspondence to: H. Kim Lyerly, MD, Department of Surgery, Duke University Medical Center, Durham, NC 27710; Fax: 919-681-7970: E-mail: k.lyerly{at}cgct.duke.edu
Clinically effective cancer immunotherapy has been a long-term objective of many investigators. The identification and cloning of a number of tumor associated and tumor specific antigens directly address the fact that human cancer cells do express antigens. Recent clinical developments, such as the demonstrated anti-tumor activity of specific monoclonal antibodies (anti-CD20 and anti-Her2/neu), have contributed to renewed enthusiasm in immunotherapy. In addition, cellular therapies such as donor lymphocyte infusion in chronic myelogenous leukemia and non-myeloablative bone marrow allograft for renal cell carcinoma have shown remarkable clinical promise. Remarkable advances in our basic understanding of immunity, and the requirements for the induction of specific immune responses, has led to a variety of exciting specific immunotherapy approaches. Elucidating the requirements of antigen presentation with co-stimulatory molecules to stimulate immune responses has highlighted the importance of antigen presenting cells, such as dendritic cells, in triggering primary immune responses.
The development of effective therapeutic cancer vaccines has also progressed, both in the development of new approaches and in the evaluation of established vaccine candidates in clinical trials. Promising new strategies have included the use of gene modified tumor vaccines and dendritic cell based vaccines.1 Early stage clinical trials that target specific tumor antigen targets, or use dendritic cells as adjuvants, have demonstrated some promising results. In addition, results are being reported from large scale clinical trials of more established vaccine candidates. For example, a prospective, randomized phase III clinical trial of autologous tumor cell vaccines for colorectal cancer performed in the Netherlands demonstrated an improvement in disease-free survival, and a trend toward improved overall survival, which lead to the approval of this treatment for colon cancer.2
The oncology community anxiously awaits further trials of promising immunotherapy strategies. Unfortunately, the results of clinical trials that will prove clinical benefit for any of the vaccine strategies often require many years of follow-up. For example, the prospective randomized phase III trial for colon cancer vaccines cited above took close to 10 years to complete. Based on the enormous time and cost of performing randomized, phase III clinical trials of therapeutic vaccines, it would be unrealistic to assume that every promising tumor vaccine will be thoroughly evaluated in phase III testing. Therefore, it is extremely important to develop intermediate markers to prioritize which agents to test in prospective randomized phase III trials. Clearly, traditional surrogate markers such as tumor response will be used to prioritize these studies, and tumor vaccine candidates that have significant activity in reducing or eradicating measurable tumor burden should, and will be, evaluated first.
An important point to consider, however, is that most cancer vaccines are designed to reduce and eliminate micrometastasis, and prevent recurrence or metastasis. Therefore, some potentially effective vaccines may have little effect on established metastatic disease, but may be useful in preventing clinical tumor progression or metastasis in patients with minimal tumor burden. Due to the heterogeneity of tumor progression, clinical studies that document significant improvements in disease-free survival due to a novel agent will be required to enroll large numbers of patients. In fact, due to patient selection bias, many biostatisticians suggest that these be executed as prospective randomized trials. Clinical investigators are left with a dilemma. How do they prioritize promising strategies in which phase I data may not show toxicity but traditional phase II data demonstrate that tumor shrinkage does not exist?
One proposed intermediate endpoint to prioritize promising vaccines for further evaluation has been the induction of an immune response by the vaccine of choice. This seems obvious, because one may assume that a vaccine must elicit some sort of immune response to be effective. Unfortunately, a number of issues cloud this strategy. First, it is not clear what type of immune response to detect and measure. Some have suggested that antibody responses are predictive of effective antitumor immunity while others have suggested that cytotoxic T cells are most predictive. In addition, murine adoptive transfer experiments suggest that cytokine secreting cells are most likely to be associated with an anti-tumor effect. Next, it is not clear what part of the body should be analyzed. Although peripheral blood is the most common site to sample, others have suggested that effective immune responses will necessarily lead to effector cells migrating to sites of tumor and have directed their efforts to these sites rather than peripheral blood. Finally, it is not clear if the magnitude and durability of the immune response is important. New assays such as flow based quantitation of antigen specific T cells that use peptide MHC tetramers, enzyme linked immunosorbent assays (ELISPOT), and quantitative intracellular cytokine analysis allow for the direct detection of immune response in samples, and these suggest significant levels of immunity are generated in response to natural pathogens.3 These levels may not be reached by vaccines required to detect tumor antigens that may be closely related to self proteins.
The long-term objective will be the establishment of true surrogates of effective anti-tumor immunity. These studies will first require that clinical benefit is proven. Nonetheless, intermediate markers of biologic activity may help to prioritize the many vaccine strategies that are currently under development. The report by Tsioulias and colleagues4 in this volume of Annals of Surgical Oncology proposes one possible way of monitoring the response to a melanoma vaccine, i.e. the detection of circulating immune complexes. The authors evaluated whether the induction of immune complexes that consist of the melanoma tumor antigen TA90 and immunoglobulin (TA90-IC) during postoperative adjuvant immunotherapy with the allogeneic melanoma vaccine CancerVax could serve as a surrogate marker of vaccine efficacy. They observed that TA90-IC seroconversion was associated with improved disease-free and overall survival compared with persistent positivity and concluded that TA90-IC can serve as a surrogate marker for the clinical efficacy of CancerVax. The authors should be congratulated for their successful efforts to collect laboratory and clinical data over a several-year interval on the 219 immunized patients. It is not currently clear how the TA90-IC assay compares with other methods of monitoring responses to immunization used by these investigators such as delayed-type hypersensitivity skin testing (DTH).5 Statistically significant correlation of various assays with each other and clinical outcome would increase the likelihood that the phenomena detected in the assays is a reliable intermediate marker.
In summary, biologic markers that may be helpful in demonstrating intermediate endpoints of a therapeutic intervention, and may be ultimately useful in predicting clinical efficacy, have become important to the development of immunotherapy strategies, as they have for many new therapies for cancer. Whether different assays or markers will need to be used for each strategy or whether one set of assays will suffice for most active immunotherapy approaches is not clear. What is clear is that standardization of reagents, conditions, and interpretation of assay results in appropriately powered prospective studies will be necessary in order for the evaluation of the numerous approaches to be undertaken. In the interim, the in vivo activity of agents with putative specific biologic activity may be used to demonstrate that the target activity has been achieved.
Acknowledgments
Supported by NIH grants U01CA72162–03, 5-P01CA47741–08, and 1-P01-CA78673–01A1. M.A.M. is a recipient of an American Society of Clinical Oncology Career Development Award and supported by NIH grant M01RR00030.
Received for publication December 5, 2000. Accepted for publication December 8, 2000.
REFERENCES
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
