Response to Mosunetuzumab in Patients with B-Cell Lymphomas Relapsed or Refractory after CAR-T Cell Therapy Is Associated with Changes in Lymphocyte Counts

Introduction: Mosunetuzumab is a CD20/CD3 bispecific antibody that redirects endogenous T cells to kill malignant B cells by concomitantly binding to CD3 on T cells and CD20 on B cells. A multicenter phase I/II study evaluating mosunetuzumab in relapsed/refractory (r/r) non-Hodgkin lymphomas, including patients r/r after chimeric antigen receptor-modified T-cell therapies (CAR-T), was the basis for FDA approval for mosunetuzumab in follicular lymphoma (NCT02500407; Budde et al. Lancet Oncol 2022). We retrospectively assessed the impact of mosunetuzumab on lymphocyte counts and CAR T cells in responding and non-responding patients.

Methods: Samples were collected prospectively from participants enrolled on NCT02500407 and exploratory post hoc analyses were performed on a subset of patients who had received prior CAR-T. P-values unadjusted for multiple comparisons are reported. We examined changes in absolute lymphocyte counts (ALC) and CAR-T transgene levels, as well as CD4 counts and CD8 counts, at baseline (C1D1) and prior to cycle 2 day 1 (C2D1) of mosunetuzumab. We also examined length of time from CAR-T to mosunetuzumab. CAR transgene levels were assessed by quantitative PCR.

Results: 30 patients had cycle 1 day 1 (C1D1) and C2D1 paired samples available for analysis. 27 patients had large B-cell lymphoma (90%) and 3 patients had low-grade follicular lymphoma (10%). The doses of mosunetuzumab received ranged from 6mg – 60mg IV or 1.6mg – 20mg SC. The median time from CAR-T infusion to mosunetuzumab was 161 days (range: 35-1071); 40% of patients (12/30) received mosunetuzumab within 100 days of CAR-T, 43% of patients (13/30) received mosunetuzumab > 100-days to 1 year after CAR-T, and 17% (5/30) of patients received mosunetuzumab > 1 year after CAR-T infusion. We found that patients who responded to mosunetuzumab had a longer time from CAR-T infusion than non-responders (mean 342 days [range: 98-587] vs 110 days [range: 70-150], respectively, p = 0.02). Although there was no difference in baseline ALC, we found that responding patients had higher ALCs on C2D1 (p=0.005). For 15 patients with available T-cell subsets, CD4 counts were higher on C2D1 for responders (p=0.05). We also investigated changes in CAR transgene levels. 14/20 patients who received commercially available standard of care CAR-T products had detectable CAR transgene levels vs only 1/10 patients who received an investigational CAR-T due to assay limitations. Thus, we examined changes in CAR transgene levels in those patients who had received commercially available standard of care CAR-T (n=20). No difference was observed between responders and non-responders in CAR transgene copy numbers at baseline; however, on C2D1 after one cycle of mosunetuzumab, non-responding patients had a trend toward decrease in CAR transgene levels while responding patients had persistence or increase in CAR transgene levels (p=0.07).

Conclusions: We observed higher peripheral blood lymphocytes after one cycle of mosunetuzumab in responding patients. Of interest, an increase in lymphocytes in responding patients had previously been described after blinatumomab (Klinger et al. Blood 2012). These findings may indicate a class effect of bispecific antibodies in general. We also found that patients who were r/r to CAR-T and responding to mosunetuzumab had a trend towards an increase in CAR transgene levels compared with patients not responding to mosunetuzumab. This suggests an interaction between CAR-T cells and bispecific antibodies. Finally, it is notable that post-CAR patients responding to mosunetuzumab had a longer time from CAR-T infusion, perhaps reflecting recovery from lymphodepletion. Further investigations are needed to understand the mechanism(s) behind these observations.