In the past year, this project has consisted of testing multiple factors that could affect the function of chimeric antigen receptors (CARs). Chimeric antigen receptors consist of several components, the antigen-recognition moiety that is usually derived from a monoclonal antibody, a extracellular region that connects the antigen-recognition moiety to the transmembrane portion, costimulatory domains such as 4-1BB and CD28, and T cell activation domains such as CD3-zeta. We were among the first investigators to test anti-CD19 CAR T cells. In the past year, we published important results in the Journal of Clinical Oncology reporting a 73% overall response rate when patients with advanced lymphoma were treated with anti-CD19 CAR T cells. This same work also reported the important association between serum interleukin-15 levels and anti-lymphoma responses. We also reported the longest continuous complete remissions of diffuse large B-cell lymphoma that was treated with anti-CD19 CAR T cells. Four patients with complete remissions of 38 to 56 months were reported in a paper that is in press at Molecular Therapy. We have constructed multiple new CARs over the past 2 years to test various components of CARs. T cells are transduced with the various CARs by using a gammaretroviral vector, and in vitro assays are carried out.
The aim i s to find CARs that impart T cells with the ability to kill cancer cells and proliferate without producing large amounts of potentially toxic inflamatory cytokines. We have found that changing the hinge and transmembrane regions, costimulatory domains, or T cell activation domains all cause profound differences in CAR function. Following extensive in vitro testing, we found that CARs with hinge and transmembrane regions from CD8 led to lower levels of cytokine production and activation-induced cell death than CARs with hinge and transmembrane regions from CD28. CARs with either CD8 or CD28 hinge and transmembrane regions could both eliminate tumors from mice. This work has resulted a paper in press at Molecular Therapy. This work has also lead to a new fully-human anti-CD19 CAR that is currently being tested in a clinical trial. We have enrolled 16 patients on a clinical trial of the CAR resulting from this work. The CAR is called Hu19-CD828Z. It has hinge and transmembrane regions from the CD8 molecule. Of 16 treated patients, 15 patients are evaluable for anti-lymphoma response so far. The overall response rate is 73% and the complete remission rate is 47%. One notable preliminary finding is a decreased rate of severe neurologic toxicity compared to previous anti-CD19 CAR clinical trials. This work resulted in an oral presentation at the American Society of Hematology meeting in December 2016. We have completed a formal comparison of the new Hu19-CD828Z CAR to our previously-used FMC63-28Z CAR. We found a 5% rate of Grade 3 or 4 clinical neurologic toxicity with teh new Hu19-CD828Z CAR versus a 50% rate of Grade 3 or 4 clinical neurologic toxicity with the older FMC63-28Z CAR. We have found lower levels of cytokine production with T cells expressing Hu19-CD828Z versus T cells expressing FMC63-28Z. Another aspect of this project is developing CARs against B-cell antigens other than CD19. We have developed 3 functional CARs that target the B-cell antigen CD20. We plan to utilize this knowledge of anti-CD20 CARs to either conduct a clinica trial of an anti-CD20 CAR or to design a bispecific CAR including CD20 binding and a domain targeting another B-cell antigen. Targeting more than 1 antigen simultaneously is important because many cases of lymphoma do not express CD19. We have selected an optimal bicistronic CAR construct that targets both CD19 and CD20. The bicistronic construct encodes 2 CARs, one of the CARs is the previously -described Hu19-CD828Z CAR. The second CAR is an anti-CD20 CAR designated Hu20-CD8BBZ. The entire bicistronic construct is designated Hu19-CD828-Hu20BB, and the construct is encoded by a gammaretroviral vector. We have submitted an Investigational New Drug Application (IND) to the FDA in order to test T cells transduced with this bicistronic construct in humans. We plan to initiate a clinical trial treating lymphoma patients with these T cells in the fall of 2019.
