Investigating macrophage function as a mechanism of resistance to daratumumab in relapsed refractory multiple myeloma patients.

Abstract

e20547 Background: Daratumumab (Dara), an anti-CD38 antibody, has a variety of Fc-dependent immune effector mechanisms of action including macrophage-mediated antibody-dependent cellular phagocytosis (ADCP). However, it remains unclear which mechanism, when disrupted, drives in vivo resistance. Given widespread use of this agent for treatment of multiple myeloma, identifying how to overcome Dara resistance has become an urgent clinical need. We hypothesized that inhibiting macrophage activity, via increased plasma cell surface expression of the CD47 “don’t eat me” signal, could be a mechanism of acquired Dara resistance. Methods: We retrospectively identified 11 patients with relapsed refractory multiple myeloma who progressed on Dara and had sequential bone marrow core biopsies before and after Dara with both samples containing > 20% plasma cells. Nine patients had sufficient material for analysis. Core biopsy slides were stained and analyzed for expression of CD47 (H-scoring for total intensity on plasma cells, as well as annotation of membrane vs. cytosolic localization), CD68 (cell counts of macrophages), and CD4/CD8/FOXP3 triple stain (cell counts of CD4 and CD8 effector and regulatory T-cells). Results: Median age was 57 years. Median R-ISS stage was 2. Two patients had high-risk cytogenetics at diagnosis. Median time from diagnosis to progression on Dara was 56.3 months (95% CI 37.43 – 88.92). Median overall survival was 69.5 months (95% CI: 46.5 – 97.1). Plasma cells in all samples expressed CD47 (H-score median = 96, range 43 - 290). However, in contrast to our hypothesis, there was no significant change in CD47 H-score at progression on Dara (H-score median = 100, range 29 – 140). Furthermore, we observed no significant change in CD68+ macrophage counts nor CD68+ localization at Dara progression. We did note a trend toward CD47 expression shifting from membrane to cytosol at Dara resistance ( p = 0.14). Consistent with prior studies, in our cohort we observed no influx in Tregs but a markedly increased CD8/CD4 ratio at resistance ( p = 0.01). Conclusions: In vitro and murine studies have proposed ADCP as an important mechanism of Dara efficacy. However, we did not observe signatures of resistance in patients being driven by this pathway. Our work suggests that anti-CD47 therapy, leading to activation of macrophages to eliminate tumor cells, will potentially be efficacious for both the Dara-naïve and Dara-refractory settings.