Impact of prior BCMA-targeted therapy on CAR-T expansion and host T-cell phenotypes in patients with Relapsed/Refractory multiple myeloma receiving BCMA CAR-T therapy

Abstract Background: B-cell maturation antigen (BCMA)-targeted therapies such as chimeric antigen receptor T-cell (CAR-T) therapy and bispecific antibodies have demonstrated therapeutic efficacy with robust response rates and improved survival in patients with relapsed/refractory multiple myeloma (RRMM). Prior exposure to BCMA-targeted therapies has been shown to negatively impact the efficacy of subsequent BCMA-directed CAR-T cell therapy and is associated with an altered immune landscape. However, the effect of prior exposure to BCMA-targeted therapy on CAR-T expansion, self-renewing host T cell phenotypes, and their association with therapeutic efficacy in patients with RRMM receiving BCMA CAR-T have yet to be characterized. Methods From cryopreserved peripheral blood mononuclear cells (PBMCs), we performed spectral flow cytometry with a ~40-plex panel on serial timepoints ranging from day -5 before BCMA-directed CAR-T infusion, up to month 12 post CAR-T infusion. BCMA-directed CAR were gated by staining with fluorophore-conjugated soluble BCMA antigen or a rabbit anti-camelid VHH antibody. Data was acquired on a Cytek Aurora. The BCMA CAR-Ts for RRMM were either ciltacabtagene autoleucel (ciltacel) given as standard of care, or through an investigator-initiated trial of a UCSF-manufactured BCMA-directed CAR-T for triple-class refractory MM (NCT05577000). Group 1 of collections included 10 RRMM patients receiving BCMA-directed CAR-T without prior BCMA-targeted therapy (anti-BCMA naïve group). Group 2 of collections were 3 RRMM patients receiving BCMA-directed CAR-T with prior BCMA-targeted therapy (anti-BCMA exposed group), which included idecabtagene vicleucel, belantamab mafadotin, teclistamab, and an investigational BCMA-directed bispecific T cell engager (TNB-383B, TeneoBio). Statistical comparisons between anti-BCMA naïve and anti-BCMA exposed groups were performed using the Mann-Whitney U test to evaluate differences in the frequency of CAR-T cells from total CD45+ cells and the frequency of non-CAR T cell lymphocytes expressing specific phenotypic markers. Results BCMA-directed CAR-T expansion, which was observed across patients up to day 30 in the anti-BCMA naïve group (median % CAR-T from total CD45+ events = 13.62%), was significantly diminished in the anti-BCMA exposed group (median % CAR-T from CD45+ = 0.34%, p=0.04). Diminished expansion in the anti-BCMA exposed group was seen both with ciltacabtagene autoleucel and the UCSF-manufactured CAR-T. At baseline prior to BCMA CAR-T infusion, the anti-BCMA exposed group had a significantly lower % of non-CAR CD8+ T cells compared to the anti-BCMA naïve group that were T-cell factor 1 (TCF1) positive (median 6.9% vs 36.3%, P 0.03) or C-X-C chemokine receptor type 4 (CXCR-4) positive (median 8.1% vs 44.9%, P 0.03). After CAR-T therapy infusion, the anti-BCMA exposed group also demonstrated a higher % of CD4+ T cells expressing granzymes B and K (median 7.3% vs 2.0%, P 0.02). Together, these findings suggest that prior exposure to BCMA-targeted therapy is associated with a depletion of host T cells with putative self-renewing capacity before CAR-T treatment, and skewing towards a more effector host T cell phenotype after CAR-T infusion. With regards to efficacy, we also observed higher baseline levels of non-CAR CXCR4+ CD8+ with best clinical response of CR/sCR compared to SD/VGPR after CAR-T therapy. Conclusions Prior BCMA-targeted therapy exposure may impact the host immune setpoint and downstream responses to subsequent BCMA-directed CAR-T therapy, including a shift from self-renewing non-CAR host T cell populations towards effector phenotypes, and impaired CAR-T expansion after CAR-T infusion. Further clinical follow up and correlative analysis to be presented will elucidate whether these immune features of prior BCMA-targeted therapy are related to clinical response and relapse.