AFFILIATIONS
1 Sylvester Comprehensive Cancer Center, University of Miami, Miller School of Medicine, Miami, Florida, United States
2 Division of Myeloma, Sylvester Comprehensive Cancer Center, University of Miami, Miami, Florida, United States
3 Division of Transplantation and Cellular Therapy, Sylvester Comprehensive Cancer Center, University of Miami, Miller School of Medicine, Miami, Florida, United States
4 Division of Transplantation and Cellular Therapy, Sylvester Comprehensive Cancer Center, University of Miami, Miller School of Medicine, Miami, Florida, United States
5 Sylvester Comprehensive Cancer Center, University of Miami, Miller School of Medicine, Miami, Florida, United States
6 Sylvester Comprehensive Cancer Center, University of Miami, Miller School of Medicine, Miami, Florida, United States
7 Division of Myeloma, Sylvester Comprehensive Cancer Center, University of Miami, Miami, Florida, United States
8 Division of Myeloma, Sylvester Comprehensive Cancer Center, University of Miami, Miami, Florida, United States
9 Division of Myeloma, Sylvester Comprehensive Cancer Center, University of Miami, Miami, Florida, United States
10 Division of Myeloma, Sylvester Comprehensive Cancer Center, University of Miami, Miami, Florida, United States
11 Division of Myeloma, Sylvester Comprehensive Cancer Center, University of Miami, Miami, Florida, United States
12 Division of Myeloma, Sylvester Comprehensive Cancer Center, University of Miami, Miami, Florida, United States
13 Division of Myeloma, Sylvester Comprehensive Cancer Center, University of Miami, Miami, Florida, United States
14 Division of Myeloma, Sylvester Comprehensive Cancer Center, University of Miami, Miami, Florida, United States
15 Division of Myeloma, Sylvester Comprehensive Cancer Center, University of Miami, Miami, Florida, United States
16 Division of Myeloma, Sylvester Comprehensive Cancer Center, University of Miami, Miami, Florida, United States
17 Division of Myeloma, Sylvester Comprehensive Cancer Center, University of Miami, Miami, Florida, United States
18 Division of Transplantation and Cellular Therapy, Sylvester Comprehensive Cancer Center, University of Miami, Miller School of Medicine, Miami, Florida, United States
19 Division of Myeloma, Sylvester Comprehensive Cancer Center, University of Miami, Miami, Florida, United States
ABSTRACT
Background: BCMA- and GPRC5D-directed bispecific antibodies (BsAbs) have transformed relapsed/refractory multiple myeloma (RRMM) treatment, but resistance limits durable clinical benefit. Although BsAbs directly engage T cells against target-expressing plasma cells, sustained anti-myeloma efficacy may also depend on additional immune interactions. Antigen-presenting monocytes can support CD4? T cells, which in turn promote CD8? effector function, whereas monocytic myeloid-derived suppressor cells (M-MDSCs) suppress T-cell activity. How the balance between antigen-presenting and suppressive monocyte programs evolves during BsAb therapy and relates to response and resistance remains undefined.
Methods: Thirty patients with RRMM receiving anti-BCMA (n=22) or anti-GPRC5D (n=8) BsAbs were classified by progression-free survival (PFS) as responders (n=16; median PFS not reached after a median follow-up of 17.4 months) or non-responders (n=14; median PFS, 2.5 months). A total of 85 longitudinal peripheral blood samples were collected at pre-treatment, best response, and at progression in non-responders or latest on-treatment follow-up in responders. Paired single-cell RNA (scRNA-seq) and T-cell receptor sequencing of peripheral blood mononuclear cells (PBMCs) profiled immune composition and phenotype, tracked T-cell clonotypes, and inferred ligand-receptor interactions. Whole-genome sequencing (WGS) was performed on bone marrow plasma cells from available cases.
Results: scRNA-seq profiled 621,647 PBMCs, including 18,788 circulating plasma cells (CPCs). WGS identified biallelic TNFRSF17 loss at progression in one non-responder after BCMA BsAb therapy, with no biallelic target loss in remaining patients. In patients without target loss, baseline and paired longitudinal profiles were evaluated to identify pre-existing and treatment-emergent features of resistance.
At baseline, cytotoxic CD8? T cells from non-responders had higher exhaustion-signature scores (PDCD1, LAG3, HAVCR2, TIGIT, and TOX; P=0.03). Classical monocytes in both groups exhibited a comparable immunosuppressive M-MDSC-like program, characterized by high S100A8/S100A9 and low HLA-DR expression. Although baseline CPC abundance was similar, non-responder CPCs were enriched for inflammatory and pro-survival pathways, including non-canonical NF-κB signaling.
Longitudinally, classical monocytes in responders reduced the M-MDSC-like suppressive signature (P<0.001) and increased MHC-II antigen presentation (P=0.004), whereas non-responder monocytes retained the suppressive program and failed to increase antigen presentation. This impaired monocyte activation in non-responders was accompanied by disrupted reciprocal monocyte-CD4? T-cell cross-talk during treatment, including reduced ICAM1-LFA-1-mediated adhesion between monocytes and Th1 and cytotoxic CD4? T cells, together with reduced Th1-derived IFNG signaling to monocytes relative to responders (P=0.01 and P=0.03). Non-responders also showed a decline in cytotoxic-effector programs in ZEB2? CD8? T cells, whereas responders maintained these programs (P=0.03). Despite similar expansion of baseline-detected clonotypes in both groups, only those from responders gained cytotoxic-effector programs, whereas non-responder clonotypes retained elevated exhaustion-associated programs and showed declining cytotoxic-effector signatures (P=0.02).
CPCs persisted longitudinally in non-responders but declined in responders from a median of 0.3% to 0.0% (P=0.002). Paired differential expression of residual responder CPCs (n=3) showed enhanced pro-apoptotic signaling (DAPK1, TRAF3 up; MCL1, BCL2 down). Persistent CPCs in non-responders showed reduced DAPK1, elevated CCND1, and attenuated immune-interaction programs, consistent with diminished immune responsiveness and greater proliferative potential.
Conclusion: Baseline CD8? T-cell exhaustion and CPC pro-survival signaling represent candidate blood-based biomarkers of poor response to BCMA- or GPRC5D-directed BsAbs. Longitudinally, resistance in patients without target-antigen loss was characterized by failure of monocytes to transition from a suppressive to an antigen-presenting program and disruption of monocyte-CD4? T-cell cross-talk. Despite clonal expansion, T-cell clonotypes remained exhausted rather than gaining cytotoxic-effector programs, while CPCs persisted. Together, these findings support a coordinated cellular mechanism of BsAb resistance without target-antigen loss.