Exploring T cell subsets as predictors of response to BCMA targeting bispecific antibody therapy in multiple myeloma.

7567 Background: BCMA targeting bispecific Antibody (bsAb) therapy has shown unprecedented response and survival rates in patients with relapsed/refractory multiple myeloma (MM). The clinical activity of bsAb therapy has been shown to depend on T-cell function, yet clinical parameters, such as absolute lymphocyte count (ALC) have not been found to be associated with response to bsAbs. The aim of the present study was to elucidate whether distinct T cell subsets, such as CD4 and CD8 counts as well as their proportion within the ALC are associated with outcome of BCMA targeting bsAb therapy. Methods: We retrospectively collected data on 79 patients who had been treated with at least one full dose of a BCMA targeting bsAb. T cell subsets, including ALC, total CD3, CD4 and CD8, were measured from peripheral blood within 7 days prior to bsAb therapy initiation. Statistical analysis was performed with SAS version 9.4 using univariate and multivariate logistic regression models. Results: Median age of the patient cohort was 72 (31-84) years with 40/79 (51%) being male and 15/79 (19%) being African American. Median lines of therapy was 5 (2-12) with 78/79 (99%) patients being triple class refractory and 50/79 (63%) being penta-drug refractory. 59/79 (75%) of patients had at least one prior stem cell transplant (SCT) with 35/79 (44%) having had at least two SCTs. 26/79 (33%) patients had prior exposure to BCMA targeting therapy, with 18/79 (23%) and 8/79 (10%) previously having received belantamab mafodotin or BCMA targeting Car-T cell therapy respectively. Cytokine release syndrome (CRS) occurred in 38/79 (48%) of patients with the majority (92%) being grade 1 with the remainder being grade 2. Median ALC was 1.02 (0.25-5) x10 3 /mL, median CD4 count was 0.27 (0.04-1.13) x10 3 /mL while median CD8 count was 0.5 (0.07-2.97) x10 3 /mL. Overall response (OR) rate of the whole cohort was 82% (69/79) with best responders (³very good partial response) comprising 51% (40/79). While total ALC, CD4 and CD8 counts did not appear to impact response to BCMA targeting bsAb therapy, the ratio of CD4 to ALC proved to be significantly associated with OR (univariate p=0.04) and best response (univariate p= 0.004, multivariate p=0.01) . The CD8 to ALC ratio had no significant impact in this patient population. The only other factor to show a significant association with response was the number of prior lines of therapy with higher numbers being associated with worse response (p=0.046). Conclusions: Our study suggests that an increased proportion of CD4 cells within the ALC is significantly associated with better response to BCMA targeting bsAb. While future studies are needed to elaborate on CD4 function in bsAB therapy, our findings imply that therapeutic strategies to increase the CD4 proportion could lead to improved bsAb therapy effectiveness.