G Protein-Coupled Receptor Class C Group 5 Member D (GPRC5D) and B-Cell Maturation Antigen (BCMA) Bi-Specific Dual Chimeric Antigen Receptors (CARs) Effectively Address Antigen Escape and Tumor Heterogeneity Challenge in Multiple Myeloma (MM)

Chia-Wei Chang, Hui Ding, Lukasz Wlodarek, Kevin Nguyen, Yonghua Luo, Mandana Karimi, Shailaja P. Rao, Arpita S. Pal, Yi-shin Lai, Sihui Guan, Weimin Li, Weijie Lan

Overland Pharmaceuticals, South San Francisco, CA

BCMA-directed CAR-T therapy shows remarkable efficacy in multiple myeloma (MM) patients, however, challenges like heterogeneous tumor BCMA expression, antigen escape, and limited treatment options upon resistance still post significant hurdles in relapsed patients. Dual targeting of MM-associated antigens could address these challenges and offer durable efficacy in BCMA-resistant MM patients. In this study, we present the development of a bi-specific CAR targeting GPRC5D and BCMA simultaneously, demonstrating its efficacy, both in vitro and in vivo.

Altogether, 41 GPRC5D and 36 BCMA VHH binders were identified from synthetic library screening and alpaca immunization. These binders were incorporated into lentiviral CAR constructs and tested for NFAT mediated signals in a Jurkat reporter platform. CARs with high antigen specific response and low non-specific and tonic signaling (Table 1) against antigen positive and negative target cell lines were prioritized for downstream functional assessments. Multiple GPRC5D and BCMA CARs were shown with better in vitro efficacy than benchmark (BM) GPRC5D and BCMA CARs (Table 1). Long-term and repetitive tumor challenge assays were performed using MM tumor lines OPM-2, NCI-H929 and RPMI-8226 to assess in vitro anti-tumor activities of the selected CARs. Several CARs were proved to have superior durable tumor growth inhibition comparable to BMs (Table 1). Next, bi-specific CARs (bsCAR) targeting both GPRC5D and BCMA were created using validated monovalent binders. During in vitro tumor restimulation assay against NCI-H929 tumor line, multiple bsCAR designs demonstrated superior long-term tumor growth control versus BM (Table 1). In this setting, our lead bsCAR T cells showed comparable antigen specific proliferation to BCMA BM and GPRC5D BM after 2 rounds of NCI-H929 challenging in vitro (Table 1). For cytokine production, the lead bsCAR T cells also generated high levels of GM-CSF, Granzyme-B, IFN-γ and TNF-α upon antigen stimulation (Table 1). Furthermore, in antigen escaping models, GPRC5D and BCMA bsCAR T cells efficiently suppressed growth of cells expressing only GPRC5D or BCMA antigens but spared double negative cells. Finally, in a disseminated NSG mouse model bearing MM cell line OPM-2, our top bsCAR T cells demonstrated highly efficient tumor growth inhibition, exhibiting competitive profiles to BCMA and GPRC5D BMs (Table 1) with an improved survival rate observed at Day 34 post-infusion.

In conclusion, we have developed a novel VHH-based GPRC5D and BCMA bi-specific CAR product that efficiently suppresses tumor growth in vitro and in vivo. The dual- antigen targeting bi-specific CAR-T cells maintained robust anti-tumor efficacy even in models of antigen-escape. Together, these findings suggest the promising potential for treating MM patients resistant to BCMA-directed therapies due to antigen-escape or heterogenous BCMA expression.

Table 1. Characterization of tonic AND antigen-specific signaling and anti-tumor activities of GPRC5D CAR, BCMA CAR, and bi-specific CARs