Abstract One of the most promising candidates for Chimeric Antigen Receptor (CAR) T cell therapy beyond CAR19 is treatment of multiple myeloma with CAR T cells targeting B cell maturation antigen (BCMA). However, current reports of BCMA CAR in the clinic have a median progression free survival of 11.8 months, suggesting that targeting BCMA alone may not be sufficient. Evidence suggests patients treated with BCMA-targeted therapies may be thwarted by BCMA negative relapse. We have developed a novel second generation CAR T cell against another multiple myeloma target: transmembrane activator and CAML interactor (TACI). Mice were immunized against TACI and we designed a new scFv targeting TACI based on the resulting antibodies. Anti-TACI CAR T cells are cytotoxic in vitro and in vivo against multiple myeloma. To overcome single antigen loss, we designed bispecific tandem scFv CAR T cells targeting BCMA and TACI. These two antigens share a natural ligand, A Proliferation-Inducing Ligand (APRIL), which has also been used to design a dual-targeting CAR called TriPRIL. We show that these dual targeting CARs, based on tandem scFv or natural ligand design, have similar efficacy against wild type multiple myeloma models. However, this changes in the context of single antigen loss. We have characterized the proliferative (population doublings) and activation capability (CD69) of these CARs, as well as their memory (CCR7, CD45RA) and exhaustion phenotype (PD-1, TIM-3, LAG-3), with long term exposure to single antigen. Our studies show that sensitivity to antigen density differs between the tandem bispecifics and the natural ligand CAR. These data provide insight into how structural differences between dual-targeting CAR T cells affects their function.