Enhancing the Efficacy of CAR T Cells

The impressive response rates of chimeric antigen receptor (CAR) modified T cells in B-ALL and high grade B cell lymphomas illustrate their remarkable capacity to kill chemoresistant cancer. However, much work remains to be done if CAR T cells are to cure a high fraction of patients with B cell malignancies and if this potency is to be extended to other chemoresistant cancers. Experience in B cell malignancies has identified primary mechanisms of resistance to CAR T cells as antigen escape and T cell failure, which are generally mutually exclusive, such that antigen escape occurs in cases where CAR T cells persist whereas antigen positive recurrence is associated with disappearance of CAR T cells. A significant factor driving antigen escape is a requirement for high antigen density for CAR T cell function. Unlike natural T cells which are activated by low antigen density (e.g. ~10 peptides/cell), CAR T cells require high antigen density for activation, estimated to be ~10,000 antigens/cell. This is most well illustrated by experience with the CD22-CARs for B-ALL. CD22-CARs induce high response rates in both CD19+ and CD19neg/lo B-ALL, however this is followed by a significant rate of relapse associated with expression of CD22lo B-ALL. The basis for the CAR T cell requirement for high antigen density remains incompletely understood and efforts are underway to lower the antigen density required for CAR T cell activation. But this property could be leveraged to provide a therapeutic window for targeting antigens with low level expression on normal tissues, which may be useful for extending the efficacy of CAR T cells to solid tumors. Furthermore, acknowledgement of the CAR requirement for high antigen density as well as the general property of tumor heterogeneity leads to the prediction that monospecific CAR T cells are unlikely to eradicate most cancers. We have therefore focused efforts on optimizing approaches to develop bispecific CAR T cells and are currently testing bivalent, bispecific CD19/22-CAR T cells capable of activating in response to CD19 and/or CD22 expression. We hypothesize that simultaneous targeting of two antigens will lower the risk of antigen loss escape in B cell malignancies. The second major cause of resistance following CAR T cell therapy is T cell failure, most commonly as a result of T cell exhaustion. Current understanding of the biology of human T cell exhaustion remains incomplete. My laboratory has developed a model of "T cell exhaustion in a dish", which uses a synthetic biology model of tonically signaling CAR T cells to induce excessive TCR zeta phosphorylation and drive T cell exhaustion. We have observed that temporary cessation of TCR zeta phosphorylation using a small molecule based system to regulate CAR protein expression levels leads to remarkable rejuvenation of exhausted T cells. The rejuvenation induced via temporary cessation of zeta phosphorylation is significantly more potent than that induced by PD-1 blockade. These results suggest that regulatable CAR expression systems could prevent and/or reverse T cell exhaustion by attacking the "root cause", namely excessive or prolonged zeta phosphorylation.