Chimeric antigen receptor (CAR)-T therapy has recently shown great promise against multiple myeloma (MM) by targeting the B cell maturation antigen (BCMA). However, the production of autologous CAR-T cells on an individualized basis has significant challenges such as manufacture time, reproducibility, consistency and cost. In addition, viral-vector-produced CAR-T cells tend to be comprised of differentiated T cell subsets, which correlate with poor in vivo durability. Using the high-fidelity NextGEN™ (NG) CRISPR gene editing system in combination with the non-viral piggyBac™ (PB) DNA transposition technology, which produces CAR-T cells predominantly in a highly desirable stem cell memory T cell subset, we have developed P-BCMA-ALLO1, an off-the-shelf allogeneic CAR-T cell product that overcomes these limitations and can be given to any MM patient.

The NG system was used to disrupt both TCR and MHCI expression in P-BCMA-ALLO1 to reduce possible alloreactivities. The DNA cleavage specificity is guaranteed by a truly dimeric system requiring the simultaneous binding of two obligate homodimers. High-throughput DNA sequencing analysis was performed on human T cells with gene disruption of TCRα, β2M or PD-1. Indel frequencies of amplified genomic on-target and off-target sites were evaluated from T cells that had been edited with either single or dual guide RNAs (gRNA)s. Herein, an average of 100,000 sequences were analyzed per site and non-edited T cells were included as a background reference. For these targets, the NG system demonstrated high-efficiency gene editing (>84% for TCRα, >64% for β2M, and >50% for PD-1) that was performed in resting T cells. No indel events were detected at any off-target site, proving the exquisite specificity of this system.

P-BCMA-ALLO1 was incapable of mediating allo-responses such as graft-vs-host disease (GvHD) and graft rejection. Using a clinically-scalable purification method, P-BCMA-ALLO1 cells were enriched to be >99.5% TCR/MHCI double-negative and alloreactivities were evaluated by standard mixed lymphocyte reaction for either cell proliferation or IFNγ production in two separate assays: a 7-12 day CFSE-based cell proliferation assay and a 20-hour IFNγ-ELISPOT assay. Compared to non-edited CAR-T cells, P-BCMA-ALLO1 did not exhibit any detectable response to allogeneic PBMC stimulation, nor did it trigger any allogeneic T cell proliferative response above background level of cell proliferation or IFNγ production.

Pre-clinical evaluation of P-BCMA-ALLO1 demonstrated optimal T cell phenotype and potent in vitro and in vivo efficacy. More than 95% of the P-BCMA-ALLO1 cells were CAR+ and greater than 50% of the product were CD45RA+CD62L+ stem cell memory T cells (TSCM), a highly desirable early memory T cell subset that exhibits the greatest capacity for self-renewal and the multipotent capacity to the entire spectrum of memory and effector T cell subsets. In addition, the CAR-T cells did not express the exhaustion markers PD-1, TIM-3 and LAG-3, nor did they exhibit any evidence of tonic signaling as IFNγ production in the absence of target antigen. In a short-term 24-hour in vitro co-culture assay with BCMA+ MM cell lines, P-BCMA-ALLO1 secreted high levels of IFNγ and mediated robust (70-90%) killing of cell targets. In addition, more than 80% of the cells proliferated in a longer 72-hour CFSE-based proliferation assay. P-BCMA-ALLO1 also showed strong potency in a NSG mouse model engrafted with MM.1S cells, an aggressive human MM-derived tumor model. While all control animals showed rapid tumor growth by bioluminescent imaging (BLI) and succumbed to disease within five weeks, all treated mice survived for the duration of the study. After P-BCMA-ALLO1 administration, tumor burden was rapidly reduced to the limit of detection by BLI within 7-10 days. Lastly, several occurrences of tumor relapse were observed in some of these animals, but were subsequently controlled without re-administration of the product. Altogether, these data demonstrate that P-BCMA-ALLO1 exhibited significantly reduced alloreactivity yet retained potent effector function, suggesting its safe and effective use for treating any MM patient.


Wang: Poseida Therapuetics: Employment. Barnett: Poseida Therapeutics, Inc.: Employment. Martin: Poseida Therapeutics, Inc.: Employment. Hermanson: Poseida Therapeutics: Employment. Li: Poseida Therapuetics: Employment. Smith: Poseida Therapeutics, Inc.: Employment. Wu: Poseida Therapuetics: Employment. Cranert: Poseida Therapuetics: Employment. Rengarajan: Poseida Therapeutics, Inc.: Employment. Tan: Poseida Therapeutics, Inc.: Employment. Tong: Poseida Therapuetics: Employment. Codde: Poseida Therapeutics, Inc.: Employment. Down: Poseida Therapeutics, Inc.: Employment. Ostertag: Poseida Therapeutics, Inc.: Employment, Equity Ownership. Shedlock: Poseida Therapeutics, Inc.: Employment.

Author notes


Asterisk with author names denotes non-ASH members.

Sign in via your Institution