Our laboratory and clinical research efforts have been devoted to studying the cellular immune reactivity against established cancers in experimental animals and in humans and the translation of these findings to the development of effective immunotherapies for patients with cancer. Our early studies evaluated the immunologic properties and anti-tumor effects of the administration of interleukin-2 (IL-2) to experimental animals and later in human clinical trials. These studies in the Surgery Branch, NCI demonstrated that the systemic administration of IL-2 could lead to durable objective cancer regressions in patients with metastatic melanoma or metastatic kidney cancer and ultimately led to the approval of IL-2 by the Food and Drug Administration as a treatment for patients with these metastatic cancers. These clinical studies of IL-2 administration led to studies to identify the immune cells capable of recognizing cancer. Attempts to generate lymphoid cells with specific anti-tumor activity led to the description of tumor infiltrating lymphocytes (TIL) in both mice and humans that could be grown from the stroma of solid tumors. TIL exhibited MHC restricted recognition of tumor-associated antigens present on mouse and human tumors. We next cloned the genes encoding antigens recognized by TIL and have described over 50 different human cancer antigens restricted by a wide variety of MHC antigens. Many of these antigens, including the differentiation antigens, MART-1 and gp100 were then used in cancer immunization trials in patients using vaccines based on peptides, recombinant poxviruses, adenoviruses, DNA, and peptide-pulsed dendritic cells. Although it was possible to raise up to 30% of all circulating CD8+ T cells with anti-tumor reactivity, it did not appear that these high levels could mediate tumor regression in patients with established cancer. Studies are continuing to attempt to identify factors that limit the anti-tumor response to cancer vaccines and to explore possible manipulations to improve the efficacy of these vaccines. The ability to identify specific anti-tumor T cells in patients led to new generations of adoptive cell transfer studies. Adoptive cell transfer has several theoretical advantages compared to other immunotherapy approaches. In cell transfer therapies highly selected cells with high avidity for recognition of tumor antigens can be activated ex vivo to exhibit anti-tumor effector function, expanded to large numbers, and tested in vitro to identify the exact subpopulations and effector functions that are required for cancer regression in vivo. Perhaps most importantly it is possible to manipulate the host prior to the cell transfer to provide an altered environment for the transferred cells. We developed an improved method for the preparation of TIL selected to have high avidity for tumor recognition, often based on the recognition of several tumor antigens, and containing both CD4 and CD8 subpopulations. Of 93 patients who received these new TIL preparations following a lymphodepleting chemotherapy, 52 patients (50%) experienced an objective cancer regression assessed by standard RECIST criteria. Most patients had progressed through prior IL-2 administration and most had received prior chemotherapy. We are now attempting to improve the results of cell therapies and are conducting detailed analyses of the factors involved in mediating successful tumor regression. Manipulation of host immune factors that might increase the effectiveness of adoptive cell therapy are also being studied.
