Edmund C. Lattime, PhD, professor of surgery and molecular genetics, microbiology, and immunology, UMDNJ-Robert Wood Johnson Medical School (RWJMS) and associate director for education and training, the Cancer Institute of New Jersey; and (right to left) Robert E. Weiss, MD, associate professor, surgery, RWJMS; Roderich Schwarz, MD, PhD, associate professor, surgery, RWJMS; Scott R. Schell, MD, PhD, associate professor, surgery, RWJMS. Not pictured; Mark Stein, MD, assistant professor, medicine, rwjms.

New immune-based gene therapy approaches to cancer by Edmund C. Lattime

One of the greatest advances in healthcare has been the development of vaccines for infectious disease. The immune system efficiently recognizes an unlimited array of disease causing agents and has developed highly effective mechanisms for their destruction. Despite having this broad repertoire, the immune system has learned to distinguish self from non-self and usually it does not attack its own tissues. The last feature makes difficult the harnessing of the immune system as a means of treating cancer, although it remains an attractive strategy and has been a goal of scientists for decades. While there have been some encouraging results, this goal has for the most part gone unrealized. Despite this lack of success, there is currently renewed enthusiasm for the potential of cancer vaccines. The reasons are twofold: First, we have become more knowledgeable about the complexity of the immune system and how it is regulated. Second, we are just now realizing that we have underestimated the ability of the tumor to directly compromise the immune response and thus escape immune destruction. Combining advances in both areas has led to a new generation of vaccine strategies for the treatment of cancer. Studies being carried out by the Immunotherapy Group at The Cancer Institute of New Jersey are focused on developing next-generation approaches to immune-based gene therapy by combining expertise in basic immunology, medical oncology, surgical oncology, and urology.

The development of an immune response requires the highly regulated interaction of a number of different types of white blood cells (WBC) (See Fig. 1). When exposed to a potential target (antigen), cells called antigen-presenting cells or dendritic cells (DC) take up antigenic material, are activated, and then travel to the lymph nodes. There they interact with T and B lymphocytes, resulting in the generation of antibodies and lymphocyte populations that can kill cells bearing the antigen. In addition to effector populations, regulatory cells that enhance or inhibit the end stage effector response are activated.

Our studies have focused on analyzing the tumor microenvironment in patients with a variety of tumors to determine the presence or absence of a tumor-specific immune response, and of immune regulatory molecules that would provide targets for manipulation. This information, coupled with the growing knowledge of the regulatory pathways involved in the development of an effective immune response, allowed us and others to identify a series of immune regulatory cytokines overexpressed in tumors that have the potential to suppress the development of antitumor immunity. We have primarily focused our studies on interleukin 10 (IL10), a cytokine normally elicited in the process of the development of an immune response as a means of dampening the response and preventing runaway inflammation. Using pre-clinical tumor models in mice, we found that the tumor-associated IL10 completely inhibited the ability of mice to develop effective antitumor immunity. Our experimental tumors uniformly grew in IL10 wild type mice without development of an anti-tumor response. In addition, the tumor-bearing mice were completely non responsive (anergic) to immunization with traditional vaccines. When IL10 was inhibited using either neutralizing antibody or IL10 knockout mice, a significant number of mice developed effective immunity and rejected the tumor challenge. We have gone on to show that IL10 was exerting its suppressive activity by inhibiting the maturation of antigen-presenting dendritic cells. This observation led us to develop (a combination vaccine by engineering) a recombinant vaccinia virus that encoded both the gene for granulocyte-macrophage colony stimulating factor (GMCSF), a factor that stimulates DC function, and tumor antigen. Use of this vaccine allowed us to break the anergy and effectively immunize tumor-bearing mice.

Having identified the association between tumor induced IL10 and its downstream effects on antigen presentation, we have developed a novel strategy of using recombinant poxvirus, administered intra-tumorally in the case of accessible solid tumors or intra-vesically (instilled into the bladder cavity) in bladder cancer, to transfer the GMCSF gene, as well as genes encoding a number of immune co-stimulatory molecules, to tumors. We hypothesize that this approach will overcome the tumor-associated inhibition of DC activity leading to a productive antitumor response. Initial clinical trials of intra-tumoral vaccinia-GMCSF produced encouraging responses in patients with metastatic melanoma. We are currently accruing to a trial of intra-vesical fowlpox-GMCSF vs. intra-vesical fowlpox-TRICOM (immune co-stimulatory molecules) in patients with advanced bladder cancer.

Figure 1. Tumor-associated cytokines inhibit anti-tumor immunity. The generation of cytokines by tumor cells directly or by non-tumor cells under the influence of tumor has a profound effect on the development of anti-tumor immunity and the responsiveness to tumor vaccines. Factors such as interleukin 10 (IL10), transforming growth factor beta (TGFß), and vascular endothelial growth factor (VEGF), can inhibit the maturation and activation of antigen-presenting dendritic cells (DC) as well as regulate the generation of effector and/or regulatory T cells, with the end effect of allowing the tumor to escape immune recognition and surveillance.

As an extension of our pre-clinical studies that demonstrated the activity of recombinant virus encoding both GMCSF and antigen, we have recently shown that the level of immunity induced is significantly increased when the vaccine is given at the tumor site. We hypothesize that the combination of tumor antigen with the positive cytokine overcomes anergy by expanding tumor-specific T cell populations already present in small numbers in the draining lymph node. We are in the process of translating these findings to clinical trials where we will administer these combination vectors intra-tumorally in patients with breast and pancreatic cancer. In addition to determining direct anti-tumor effects of such therapies, we have designed correlative laboratory studies evaluating how these therapies are influencing the development of anti-tumor immunity. We hope that the information gained from the early trials and their correlative studies will be critical to the design of the next generation of clinical trials.

Edmund Lattime, PhD, is a professor of surgery and molecular genetics, microbiology, and immunology at RWJMS. He obtained his PhD from Rutgers University, followed by postdoctoral training at the Memorial Sloan Kettering Cancer Center, where he was a member of the faculty for 10 years. Dr. Lattime has served on a number of NCI and ACS study sections. He is associate director for education and training and the director of surgical oncology research at The Cancer Institute of New Jersey. He currently serves on the editorial boards of Cancer Research and The Journal of Clinical Oncology.§