AUTHORS
Julie O'Neal, Julie Ritchey2, Michael P. Rettig, PhD3, Susan Gladney4, Gabriel J. Haas, MS5, Audrey H. Heathcote6, Jingyu Xiang, MD, MSCI7, Ramzi Abboud, MD8, Linda Eissenberg9, Feng Gao, MD, PhD, MPH, MS10, Armin Ghobadi, MD11, John F. DiPersio, MD12
AFFILIATIONS
1 Washington University School of Medicine, St. Louis, Missouri, United States
2 Washington University School of Medicine, St. Louis, Missouri, United States
3 Washington University School of Medicine, St. Louis, Missouri, United States
4 Washington University School of Medicine, St. Louis, Missouri, United States
5 Washington University School of Medicine, St. Louis, Missouri, United States
6 Washington University School of Medicine, St. Louis, Missouri, United States
7 Washington University School of Medicine, St. Louis, Missouri, United States
8 Washington University School of Medicine, St. Louis, Missouri, United States
9 Washington University School of Medicine, St. Louis, Missouri, United States
10 Washington University School of Medicine, St. Louis, Missouri, United States
11 Washington University School of Medicine, St. Louis, Missouri, United States
12 Washington University School of Medicine, St. Louis, Missouri, United States
ABSTRACT
Background: Multiple myeloma (MM), a malignancy of mature plasma cells remains mostly incurable despite improvement in treatment options, including BCMA targeted immunotherapies. Targeting alternative cell surface proteins with chimeric antigen receptor expressing T-cells (CAR-T) cells could contribute to cure. FCRL5 (FcRH5/IRTA2/CD307), is expressed on most MM cells with limited expression on healthy tissues, making it an ideal CAR target with a positive safety profile. FCRL5 encodes a type I transmembrane protein with nine IgG-like extracellular domains (D1-9) with D1 being most distal to the membrane and D9 the most membrane proximal.
Methods: To develop FCRL5 CAR-T, we generated anti-FCRL5 human antibodies using immunization strategies designed to generate antibodies targeting either specifically D9 or more distal domains (D1-8), and obtained the single chain variable fragment (scFv) sequences for each clone. We engineered OPM2 cells to express full length extracellular FCRL5 (D1-9) or a series of extracellular truncated FCRL5 proteins (D1-6, D4-6, D1-7, D1-8, D8, or D9) which via flow cytometry and in in vitro CAR-T cytotoxicity assays, enabled us to map epitope antibody binding sites. To test in vivo efficacy of FCRL5 CAR-T cells, NSG mice were injected i.v. with 1x106 OPM2-FCRL5 (luciferase expressing) cells and were treated two weeks later with 2x106 CAR-T cells. We also developed and tested CAR-T efficacy in a MOLP2 (luciferase expressing) systemic model of MM by injecting 3x106 MOLP2 i.v. into NSG mice. For both models, survival was assessed via Kaplan Meier and longitudinal tumor burden was measured with bioluminescent imaging (BLI).
Results: Antibodies from mice immunized with D9 bound to cells expressing D9 alone and D1-9; but surprisingly, the epitope binding site(s) extended to either D8 (because they bound to/killed OPM2-D1-8 cells) or to D7 (because they also bound to/killed OPM2-D1-7 cells). D1-8 targeted antibodies bound to and killed OPM2 cells expressing full length D1-9 as expected, and to OPM2 cells expressing D4-6, D1-6, or D8. These studies suggest we can target distinct regions of FCRL5 with CAR-T cells. We generated and tested a series of FCRL5 CAR-T cells targeting each of the epitope regions our CAR-T cells target in an NSG/OPM2-FCRL5 (luciferase expressing; full length D1-9) xenograft mouse model. Compared to mice left untreated or treated with non-transduced T-cell controls, mice treated with FCRL5 CAR-T targeting D1-6, D4-6, D7-9, D8, or D8-9 showed significant reduction of tumor burden (assessed by longitudinal bioluminescent imaging; BLI) and survival (Kaplan Meier).
Because the MOLP-2 cell line endogenously expresses low levels of FCRL5, similar to the levels we observed on primary MM cells, we developed an NSG/MOLP-2 xenograft model. NSG/MOLP-2 mice showed multiple features of human MM: bone marrow localized disease, serum monoclonal antibody levels that increased with disease burden, and bone destruction. A dose titration informed our study design to compare efficacy of FCRL5 CAR-T cells. MOLP-2 bearing mice were treated with either 500,000 or 2x106 FCRL5 CAR-Ts targeting D7-9 (clones 947, 948) or D8-9 (clones 951, 980) which led to significant extension of survival, reduced tumor burden, and protection of bone health. Across multiple studies, 948 FCRL5 CAR-T cells demonstrated superior efficacy compared to the other clones in this model. 951-FCRL5 CAR-T cells showed next- best efficacy. Of note, mice treated with CAR-Ts expressing either 948 or 951 FCRL5 binders also showed efficacy in tumor re-challenge experiments (948>951) suggesting that both promote significant CAR-T persistence in vivo.
Conclusions: 948 and 951 FCRL5 CAR-T cells target membrane proximal FCRL5 domains and showed efficacy in two MM xenograft models; one of these two candidates may serve as our clinical candidate. Alternatively, the best performing BCMA CAR-T product, ciltacabtagene autoleucel (Carvykti) targets two separate BCMA epitopes, and our data demonstrates possibility of rational FCRL5 bi-targeting. We are working to test our hypothesis that a CAR-T bi-targeting the FCRL5 protein (tandem or hairpin format) with one scFv from a membrane proximal targeting scFv (948 or 951) and an scFv targeting a more distal epitope (D4-6; clone 973), will prove superior to single targeting. The goal of this program is to develop a highly effective and safe FCRL5 CAR-T product for the treatment of multiple myeloma.
DISCLOSURES
Julie O'Neal: NeoImmuneTech - Patents & Royalties; Wugen/Allotera Therapeutics - Patents & Royalties