Chimeric Antigen Receptor T Cells Targeting LILRB4, an Immunoreceptor Mediating T-Cell Suppression, Are Potently Effective in Multiple Myeloma

Abstract

Significant progress has been made in the treatment of multiple myeloma (MM) however, no strategy to date has proven curative and relapse invariably occurs. We recently developed an integrated pipeline to identify biologically and therapeutically relevant targets for MM immunotherapy by using Mass-Spectrometry analyses from seven MM cell lines and RNA-seq data from 900+ patients. Starting from 4,000+ candidates, ILT3 aka LILRB4, an immunoreceptor tyrosine-based inhibition motif-containing receptor emerged as a promising candidate target: it has absent/low expression in most normal tissues and high and frequent expression in many primary patient samples we used for validation, similarly to BCMA (Di Meo et al., Cell Reports Medicine 2023). We sought to examine its functional role in MM and develop Chimeric Antigen Receptor (CAR) T-cells targeting LILRB4. Interrogating RNA-seq data from MM patients, LILRB4 expression results higher in patients with high-risk cytogenetics such as t:(14;16), t:(14;20), and t:(8;14) translocations compared to patients with no genomic abnormalities. Investigating function, we find no difference between cell proliferation of U266/KMS11/OPM2 genetically depleted of LILRB4 (KO) compared to controls by measuring cell growth in vitro for 120 hours and MTS assays. Similarly, when KO and control cells are seeded in plastic chambers containing a porous membrane and serum-free medium, both KO and control cells are attracted to the lower chambers containing FBS, suggesting that LILRB4 is also not required for MM cell migration. Despite these initial observations, RNAseq data from LILRB4 KO and overexpressing (OE) U266 MM cells identify ~3,000 differentially expressed genes compared to controls. Thus, we next investigate potential extrinsic effects on T-cell proliferation by co-culturing T-cells from healthy donors with LILRB4 KO or OE U266 MM cells (T:E ratio 1:2). T cells are previously activated with anti-CD3/CD28-coated beads and IL-2. 5 days after activation, T cells and LILRB4 KO U266 MM are seeded in the same well plate. After 48h of co-culture, T-cell proliferation is decreased (-30%) in the presence of LILRB4 U266 OE MM cells, whereas it is increased (+50%) when the co-culture includes LILRB4 U266 KO cells. CD25 expression is also reduced by 20% on T cells co-cultured with LILRB4 OE cells and increased by the same extent when T cells are co-cultured with LILRB4 KO cells, suggesting a role as a negative immune receptor. This effect seems dependent on the direct cell-cell interaction given that when T and MM cells are separated by a transwell in co-culture, no effect on the T cell proliferation is observed. Finally, we develop multiple Chimeric Antigen Receptor (CAR) constructs targeting LILRB4, by using different single-chain variable fragments (scFVs) and rearranging the variable regions of the heavy (VH) and light (VL) regions. We used lentiviral LILRB4 scFV-41BB-CD3zeta vectors. LILRB4 CAR T-cells specifically kill U266, H929 and KMS11 MM cell lines in vitro with cytotoxicity ranging from 30 to 80% depending on the construct (T:E ratio 1:10). Equal efficacy is observed in U266 cells resistant to Bortezomib and BCMA KO U266cells. No effect on LILRB4 null cells is noted. These data suggest that LILRB4 is a negative immune receptor that is active on malignant plasma cells mediating T-cell suppression, similar to previously described data in leukemia (Deng et al., Nature 2018). Our work supports the rationale development of ILT3 CAR-T cell therapy as a viable therapeutic approach for MM patients especially high-risk patients. Ongoing studies are investigating the functional role of LILRB4 in mediating the cross-talk between MM ad immune cells.