266759 Engineering Chimeric Antigen Receptors Targeting an Endogenous Murine Tumor Associated Antigen

Wednesday, October 31, 2012: 3:36 PM
Pennsylvania East (Westin )
Cary F. Opel1, Matthias T. Stephan2 and K. Dane Wittrup1,3, (1)Chemical Engineering, Massachusetts Institute of Technology, Cambridge, MA, (2)Clinical Research Division, Fred Hutchinson Cancer Research Center, Seattle, WA, (3)Biological Engineering, Massachusetts Institute of Technology, Cambridge, MA

Adoptive cell transfer (ACT) therapy for the treatment of cancer is the act of administering lymphocytes to a patient with the goal of achieving an active immune response against tumor cells. A variety of methods have emerged to modify the lymphocytes in order to enhance their anti-tumor potency. One increasingly effective method is the use of chimeric antigen receptors (CAR) to target the lymphocytes specifically to tumor associated antigens. CARs are surface bound proteins consisting of an intracellular activation domain fused to an extracellular targeting domain. These receptors enable the lymphocytes to bypass many of the restrictions on the endogenous activation pathway, potentially allowing for a more effective immune response to the target tumor cells.

Despite many recent successes using lymphocytes expressing CARs in a clinical setting, some unforeseen side effects have been reported, and the design criteria for the most effective CAR domain structure have yet to be clearly established. Of critical importance to the design of a CAR is the selection of internal signaling domains. Weakly activating domains may not elicit a strong enough response to eradicate the cancer, while potent activating domains may cause an autoimmune response with severe side effects. Additionally, some activating domains may enhance other beneficial properties such as the survival, persistence, or proliferation of the transferred lymphocytes. First generations CARs used only the activating domain of CD3, while more recent second and third generation CARs combine two or three internal activating domains to enhance the effectiveness of the transferred lymphocytes. Unfortunately, the exact benefits and side effects of the various domain combinations remain unclear. This work aims to elucidate some of the design parameters that enable the most effective targeting of tumor cells by CAR expressing lymphocytes.

To study the effects of different activating domains, a model system was developed using the poorly immunogenic murine melanoma cell line, B16F10, and an antibody, TA99, targeting gp75, an endogenous tumor associated antigen expressed on the surface of B16F10 cells. The targeting domain of a CAR was created using an scFv constructed from the variable regions of the TA99 antibody. Preliminary cytotoxic assays using a first generation CAR showed promising results, so several CARs were constructed using various combinations of internal signaling domains and their effectiveness was tested in vitro.

Unlike many other model systems for adoptive cell transfer (ACT), this model will be expanded to use immune competent C57BL/6 mice. Typically, the effectiveness of CARs is tested on immunodeficient mice using human cell lines and targeting human antigens. In these models, off-target effects as well as interactions with the immune system are not evaluated, creating potentially misleading results. By using a fully endogenous system, a more accurate measurement of the effectiveness of various activating domains may be achieved, providing a more accurate prediction of how treatments will perform in a clinical setting.

Extended Abstract: File Not Uploaded