An optimized bicistronic chimeric antigen receptor against GPC2 or CD276 overcomes heterogeneous expression in neuroblastoma

• Text
• PDF

Abstract

Chimeric antigen receptor (CAR) T cell therapies targeting single antigens have performed poorly in clinical trials for solid tumors due to heterogenous expression of tumor-associated antigens (TAAs), limited T cell persistence, and T cell exhaustion. Here, we aimed to identify optimal CARs against glypican 2 (GPC2) or CD276 (B7-H3), which were highly but heterogeneously expressed in neuroblastoma (NB), a lethal extracranial solid tumor of childhood. First, we examined CAR T cell expansion in the presence of targets by digital droplet PCR. Next, using pooled competitive optimization of CAR by cellular indexing of transcriptomes and epitopes by sequencing (CITE-Seq), termed P-COCC, we simultaneously analyzed protein and transcriptome expression of CAR T cells to identify high-activity CARs. Finally, we performed cytotoxicity assays to identify the most effective CAR against each target and combined the CARs into a bicistronic “OR” CAR (BiCisCAR). BiCisCAR T cells effectively eliminated tumor cells expressing GPC2 or CD276. Furthermore, the BiCisCAR T cells demonstrated prolonged persistence and resistance to exhaustion when compared with CARs targeting a single antigen. This study illustrated that targeting multiple TAAs with BiCisCAR may overcome heterogenous expression of target antigens in solid tumors and identified a potent, clinically relevant CAR against NB. Moreover, our multimodal approach integrating competitive expansion, P-COCC, and cytotoxicity assays is an effective strategy to identify potent CARs among a pool of candidates.

Authors

Meijie Tian, Adam T. Cheuk, Jun S. Wei, Abdalla Abdelmaksoud, Hsien-Chao Chou, David Milewski, Michael C. Kelly, Young K. Song, Christopher M. Dower, Nan Li, Haiying Qin, Yong Yean Kim, Jerry T. Wu, Xinyu Wen, Mehdi Benzaoui, Katherine E. Masih, Xiaolin Wu, Zhongmei Zhang, Sherif Badr, Naomi Taylor, Brad St. Croix, Mitchell Ho, Javed Khan