S100A8/A9 drives T cell exhaustion and compromises the therapeutic efficacy of bispecific antibodies in multiple myeloma.

Abstract Background: S100A8/A9 is a calcium-binding alarmin released during cellular stress and inflammation, modulating immune responses. We previously identified S100A8/A9 as a mediator of resistance to anti-BCMA CAR T-cell therapy through suppression of cytotoxic function. Here, we evaluate the role of S100A8/A9 in patients with multiple myeloma (MM) treated with bispecific antibodies (BsAb) targeting BCMA or GPRC5D. We perform in vitro and in-vivo validation to elucidate the underlying mechanisms contributing to suppression of T cell-directed therapy efficacy and test salvage strategies. Methods: Serum S100A8/A9 levels were correlated with depth of response (DOR) and progression-free survival (PFS) in patients receiving bispecific antibody therapy (n=19). We conducted extensive phenotyping of peripheral blood mononuclear cells (PBMCs) and quantified antigen-specific cytokine production and cytotoxicity by spectral flow cytometry. Post BsAb relapse EMD samples were interrogated with the Visium spatial transcriptomic platform. To study mechanistic effects, healthy donor T cells were incubated with recombinant S100A8/A9, and dynamic changes in phenotype, antigen-specific cytokine production, and cytotoxic activity were assessed. In vitro and in vivo salvage experiments were performed using blocking antibodies against S100A8/A9 in combination with bispecific therapeutics. Results: Patients with short PFS (n=11,<5 months, median 63 days) had significantly higher serum S100A8/A9 levels both at baseline and during therapy compared to those with longer PFS (n=8, median 503 days; p<0.05). S100A8/A9 levels during treatment inversely correlated with PFS (r=-0.52, p< 0.05). T cells from patients with short PFS exhibited features of exhaustion, including increased TOX+ CD3+ T cells (p=0.004) and higher CD8+ PD-1+ T cells. Upon stimulation with MM cell lines, PBMCs from these patients produced significantly lower levels of TNF-α, GM-CSF, IFN-γ, and IL-2 in both CD4+ and CD8+ subsets (p<0.05). CD4+ and CD8+ cytokine production inversely correlated with TOX expression (CD4+: r² = –0.6, p<0.05; CD8+: r² = –0.77, p<0.01). Antigen-specific cytotoxicity was significantly reduced in patients with short PFS compared to those with long PFS (p<0.001). Spatial transcriptomics from extramedullary collected samples showed colocalization of S100A8/A9 expressing CD14 monocytes with exhausted T cells in the tumor microenvironment of patients undergoing BsAb therapy. To measure the direct effect of S100A8/A9 on T cell functionality contributing to the impairment of T cell-directed therapy, we exposed healthy donor T cells to S100A8/A9. Incubation with S100A8/A9 induced T cell exhaustion in a dose-dependent manner, as evidenced by increased TOX+ CD4+ and TOX+ CD8+ populations, and impaired BsAb-mediated cytotoxicity. Importantly, blockade of anti- S100A8/A9 with monoclonal antibodies (mAbs) restored BsAb-mediated cytotoxicity (p=0.032) and prevented induction of TOX in healthy donor and patient samples, supporting a causal role of S100A8/A9 in T cell dysfunction. Direct incubation of T cells with S100A8/A9 caused a dose-dependent decrease in pERK signaling that was reversed with mAb salvage. In vivo treatment with anti-S100A8/A9 mAb activity showed a trend towards improved survival in combination with bispecific therapy as compared to bispecific therapy alone in transgenic tumor bearing mice. Conclusions: S100A8/A9 directly promotes T cell exhaustion and impairs T cell effector function, thereby reducing the efficacy of BsAb therapy in MM. Elevated serum S100A8/A9 levels are associated with inferior clinical responses and shorter PFS. Mechanistically, S100A8/A9 induces TOX expression and suppresses cytokine production and cytotoxicity in T cells. Salvage with monoclonal antibodies not only restores function but also prevents T cell exhaustion mediated by S100A8/A9 exposure. These findings position S100A8/A9 as a novel biomarker of resistance and a potential therapeutic target to enhance the efficacy of T cell-engaging immunotherapies in multiple myeloma.