In some patients, adoptive immunotherapy with tumor-reactive T-cells can mediate the regression of large amounts of disease. However, many patients do not respond, requiring increased understanding of the factors required for successful immunotherapy. For a T-cell to be functional in vivo it must be able to 1) recognize antigens on tumors, 2) become activated and proliferate in vivo, and 3) migrate efficiently to the tumor site. Substantial progress has been made in the 1st of these steps. For melanoma, many antigens have been cloned which are specifically recognized by tumor-reactive T-cells. This has allowed active immunization with specific melanoma peptides, resulting in the ability to generate T-cells with high reactivity against melanoma. Furthermore, gene transfer technology has allowed the introduction of receptor genes into T-cells in order to redirect their specificity. Natural T-cell receptors against melanoma have been cloned, and transduction of nonspecific T-cells with these genes confers the ability to recognize melanoma cells. In addition, with Zelig Eshhar from the Weizmann Institute in Israel, we have pioneered the use of chimeric antibody/T-cell receptor fusion genes in order to redirect specificity against common cancers. This technique, which joins antibody variable regions with T-cell signaling chains, allows the generation of large numbers of T-cells highly specific against common cancers and has the potential to greatly enhance the generalizeability of T-cell based immunotherapy. We are currently utilizing this technology in a clinical trial for ovarian cancer, and have developed receptors against colon cancer and tumor vasculature, while other labs have now developed receptors against breast cancer and HIV. These advances have allowed the treatment of patients with large numbers of highly specific T-cells capable of recognizing and lysing tumor cells and producing large amounts of cytokines upon encounter with tumor antigens. Despite this high level of specific recognition, though, many patients still do not respond, and other factors besides tumor recognition must be involved, such as lack of proliferation and poor migration of adoptively transferred T-cells. Little work has been performed to study these aspects of adoptive T-cell transfer. Our hypothesis is that the presence of circulating antigen-reactive T-cells is not sufficient to induce tumor regresssion, but enhanced activation and proliferation of T-cells in vivo and improved migration to the tumor site will be required for these treatments to succeed. My program focuses on understanding the factors that govern the in vivo activation, expansion and trafficking of tumor-reactive T-cells. Delineation of the elements which can drive T-cell expansion and migration to tumor may allow the design of improved adoptive immunotherapies for cancer. To study whether T-cell expansion was limited by the weak immunogenicity of tumor antigens, we evaluated the in vivo proliferative capacity of T-cells reactive against potent antigens, such as those found on allogeneic antigen presenting cells (APC). We found that adoptively transferred alloreactive, cultured T-cells could proliferate highly in vivo following the administration of allogeneic APC. In order to confer tumor specificity to these T-cells, they were gene modified with a chimeric receptor composed of an antibody variable region joined to a T-cell signaling chain, which we have previously demonstrated to redirect T-cell specificity. These """"""""dual specific T-cells"""""""" proliferated to high levels and exhibited strong antitumor activity in vivo following immunization with allogeneic antigen presenting cells. Thus, in vivo expansion of cultured, adoptively transferred T-cells can be achieved in murine models with strong stimulation of the T-cell receptor using an appropriately potent antigen. To test this hypothesis in the human, we constructed a chimeric receptor recognizing folate binding protein (FBP), which is overexpressed in the majority of epithelial ovarian cancers. Dual specific T-cells were generated by transducing this chimeric receptor gene into autologous alloreactive T-cells from patients. In an ongoing clinical trial for ovarian cancer, we are administering these dual specific T-cells followed by immunization with allogeneic APC. This strategy of generating dual specific T-cells may be widely applicable to the treatment of a large number of common cancers. Besides proliferation, T-cell activation is critical for a successful immune response and is central to effector function and migration. Numerous studies have demonstrated that T-cells need to be activated in order to proliferate and to develop effector function including the ability to specifically lyse target cells or produce cytokines such as interferon-g in response to antigen. Additionally, it is clear that activation status is intrinsically linked to the ability to migrate to sites of inflammation. For example, naive T-cells circulate in blood and secondary lymphoid tissues in search of antigens presented on the surface of antigen presenting cells. Once activated, adhesion molecules are altered on the surface of effector T-cells to allow them to migrate to sites of inflammation, via tethering on E- and P-selectin and the activation of integrins by chemokine receptors. Because T-cell activation is such a critical factor to mediate T-cell proliferation, effector function, and migration, we investigated methods to optimally activate T-cells with dendritic cells. We developed novel methods to retrovirally transduce DCs with tumor antigen genes in order to stimulate both CD4 and CD8 positive T-cells. In addition, we found that multiple inflammatory signals were required to fully activate DCs in order to maximally stimulate tumor-reactive T-cells. We are conducting a clinical trial in order to translate these insights. In addition to in vivo activation and proliferation of lymphocytes, an important factor in the development of effective immunotherapy is the ability of T-cells to migrate to tumor deposits. Therefore, we are studying the factors which direct the in vivo migration of T-cells to tumor by examining T-cell interactions with adhesion molecules and chemokines. Many tumors produce chemokines, such as GRO-a and IL-8, which may function as angiogenic or autocrine growth factors. However, we have found that cultured, tumor-reactive T-cells do not express the receptors necessary to respond to many of these factors. Therefore, we are modifying T-cells with appropriate chemokine receptor genes to determine if this alters their in vivo migration to tumor. As an initial model, we have transduced T-cells with the gene encoding CXCR2, which confers responsiveness to GRO-a and IL-8. We found that CXCR2 transduced T-cells were capable of signaling in response to GRO-a as evidenced by calcium ion mobilization and cytokine production. In addition, CXCR2 transduced T-cells specifically migrated in vitro towards GRO-a as well as melanoma supernatants. These studies clearly demonstrate that we can alter the migratory phenotype of T-cells by transducing them to express chemokine receptor genes. We are currently performing in vivo studies to evaluate whether this strategy effects their migration to inflammatory sites as well as tumor. These studies may provide insights into the development of improved T-cell therapies with enhanced T-cell survival, proliferation and migration to tumor in vivo.
