Multi-Omic Single-Cell Characterization of Paired and Serial Samples in Responding and Non-Responding Patients Receiving BCMA Bispecific Antibody Therapy for Multiple Myeloma

Background: Bispecific T-cell antibodies targeting B-cell maturation antigen (BCMA-BsAb) have demonstrated significant clinical efficacy in relapsed refractory multiple myeloma with deep and durable responses. However, there are limited studies thus far characterizing the impact of treatment on plasma cells (PCs) and the tumor microenvironment (TME) using high dimensional single-cell multi-omics sequencing on paired and/or serial samples from patients receiving BCMA-BsAb. Methods: We performed single-cell RNA sequencing (scRNA-seq), single-cell T/B cell receptor sequencing (scTCR-seq and scBCR-seq) on paired baseline and on/post-treatment fresh bone marrow (BM) aspirate samples from 14 RRMM patients that received BCMA-BsAb therapy, including standard-of-care teclistamab (N=8) and BCMA-BsAb on clinical trials (N=6). For each sample, the BM mononuclear cells (BMMNCs) were isolated, and PCs were enriched by CD138+ selection. cDNA libraries from BMMNCs and PCs were then used for scRNA-seq, scBCR-seq and/or scTCR-seq. Single-cell data was processed as previously described (Dang et al., Cancer Cell, 2023). Results: Among the 14 patients, 12 were responders (partial response (PR), (N=1); very good partial response (VGPR), N=1; complete response (CR), N=10) while 2 were non-responders (stable disease (SD), N=1); progressive disease (PD), N=1). We profiled 171,713 high-quality cells including 95,488 TME cells (among which 2,384 have paired scBCR-seq and 21,516 have paired scTCR-seq data) and 76,225 PCs (among which 70,460 have paired scBCR-seq data). Unsupervised clustering analysis revealed 24 different TME cell types. Malignant cells were identified based on BCR clonotypes as well as inferred copy number alterations. We profiled the dynamics of TME and myeloma cells in both responders and non-responders. In responders, we observed a consistent decrease in pDCs and mature B cells in 9 out of 9 patients (with more than 100 TME cells), an increase in Lymphoid Primed Multipotent Progenitors in 8 out of 9 patients, and an increase in NKT cells in 7 out of 9 patients in responding samples compared with baselines. For patients with paired baseline and progression samples (N=5 with more than 100 TME cells), we observed a consistently increased proportion of macrophages, megakaryocyte progenitors, and stroma cells compared with baselines. CD8+ T cells in progressed samples showed increased exhausted and stressed phenotypes. Additionally, we observed a decrease in TCR diversity in responding samples. Specifically, we found the expansion of effector CD8+ T cell and NKT cell clones in 5 out of 6 responders with available paired TCR data. Comparatively, 2 out of 5 progressed samples showed expansion of effector CD8+ T cell or NKT cell clones, while 1 had no expanded T cell clones, and 2 had expanded MAIT or interferon-stimulated T cells. For myeloma cells, the patient with PD as best response had negative BCMA expression at both baseline and post-treatment, while the SD patient has a very low BCMA transcriptional level at baseline, which increased after treatment. There were two samples from initial responders with evidence of early measurable residual disease (MRD) relapse that were BCMA negative, while there were two samples at the time of disease progression from initial responders which showed recovered BCMA expression. The comparison of myeloma cells between baseline and on-treatment samples showed an overall downregulation of BCMA, as well as MHC class I molecules, including HLA-A/B/C/E, in relapsed/refractory samples, with very few upregulated genes observed. Gene set enrichment analysis reveals the downregulation of antigen processing/presentation and adaptive immune response in relapsed/refractory samples. Conclusions: We comprehensively characterized the cellular and molecular properties of TME and myeloma cells in 14 BCMA-BsAb treated patients with paired samples collected and identified different factors determining patient responsiveness. We observed the expansion of cytotoxic lymphocytes in responding samples, whereas progressed samples showed an accumulation of exhausted T cells in TME and downregulated BCMA and MHCI molecules in myeloma cells. These findings highlight potential targets for enhancing therapeutic strategies and overcoming resistance in RRMM treatment.