No Evidence of BCMA Expression Loss or Systemic Immune Impairment after Treatment with the BCMA-Targeted Antibody-Drug Conjugate (ADC) Belantamab Mafodotin (Belamaf) in the DREAMM-1 and DREAMM-2 Trials of Patients with Relapsed/Refractory Multiple Myeloma (RRMM)

Introduction: Belamaf, a first-in-class humanized ADC, binds BCMA expressed on differentiated B cells and myeloma (MM) cells. Belamaf monotherapy demonstrated deep and durable responses in the DREAMM-2 trial of patients (pts) with RRMM. Belamaf eliminates MM cells by a multimodal mechanism of action (MoA): direct cell killing and anti-MM immune response. We explored whether complete target loss and/or immune impairment were observed in pts who received belamaf in the DREAMM-1 and -2 trials. Methods: DREAMM-1 was a phase 1 dose-escalation study to investigate the safety, PK/PD, immunogenicity, and clinical activity of belamaf. DREAMM-2 was an open-label, two-arm, phase 2 study of belamaf 2.5 mg/kg or 3.4 mg/kg IV Q3W. Both trials enrolled adult pts with RRMM with ≥3 prior lines of therapy including an immunomodulatory drug, proteasome inhibitor, and an anti-CD38 monoclonal antibody (DREAMM-2 only). Free sBCMA, a shed form of BCMA that circulates in blood after cleavage from the membrane, was measured using an electrochemiluminescence assay on the MesoScale Discovery platform (Alliance Pharma, Malvern, PA; detection range 1.95-1000 ng/mL). Absolute concentrations of sBCMA were examined at baseline (BL), with absolute concentration and fold-change from BL assessed at best achieved response and latest progression. After the first post-infusion timepoint, samples were taken >4 days post-infusion to minimize belamaf interference. In this post hoc analysis, association of sBCMA with time-on-study was modeled using a linear mixed model to adjust for longitudinal categorical response, with time modeled as a quadratic function of square-root-time-on-study. A generalized additive mixed model was used to adjust the time-association for M-protein among pts with apparent secretory disease. Immune cell populations in peripheral blood were analyzed using flow cytometry and clinical hematology tests. Neutrophil-to-lymphocyte ratio and total lymphocyte counts were modeled by categorical pt visit adjusted for continuous or count-based biomarkers; T cell subpopulation counts were modeled using negative binomial mixed effects models. Results: At progression, sBCMA levels were detectable in 98% (50/51) of eligible pts in DREAMM-1 and 98.9% (181/183) of eligible pts in DREAMM-2, regardless of response status. Decreased sBCMA vs predose levels was observed immediately post-infusion in all response groups. In nonresponders, sBCMA returned to predose levels within 1 cycle. Responders (≥partial response) had a quantitatively lower but measurable sBCMA level, suggesting BCMA loss is not commonly observed in pts receiving belamaf. In 97% (64/66) of pts who responded but later progressed in DREAMM-2, sBCMA levels showed a pronounced drop during response but returned to near BL upon progression (Fig A). This suggests membrane BCMA expression is not lost following belamaf treatment and complete target loss is not the primary mechanism driving tumor escape in these pts. Previously, sBCMA levels have been shown to correlate with disease burden markers (e.g., M-protein) and ISS stage; we modeled sBCMA correcting for these associations to evaluate on-treatment dynamics, predicting a decrease in sBCMA with increasing time-on-treatment. After correcting for lower average BL and on-treatment sBCMA levels in responders, sBCMA levels tended to decrease over time. Longitudinal monitoring of pts' systemic immune cell populations showed no change in immune cell ratios (Fig B) or major cell populations regardless of response status. Immune cell profiles of responders and nonresponders were similar and consistent over time suggesting belamaf does not detrimentally affect total lymphocyte numbers while mediating anti-MM activity. Conclusions: Belamaf resistance and immune escape mechanisms are not well understood. We found no evidence that BCMA expression is completely lost after belamaf treatment. When sBCMA decreased after belamaf treatment, levels returned to BL upon progression. Regardless of response status, belamaf does not appear to negatively impact total lymphocyte numbers. These data may help inform clinical sequencing of belamaf with other BCMA-targeted therapies. Funding: GSK (205678/NCT03525678; 117159/NCT02064387). Drug linker technology licensed from Seagen, Inc; monoclonal antibody produced using POTELLIGENT Technology licensed from BioWa. Figure 1View largeDownload PPTFigure 1View largeDownload PPT Close modal