Monocytes and Granulocytes Reduce CD38 Expression Levels on Myeloma Cells in Patients Treated with Daratumumab

Abstract

Purpose: Daratumumab treatment results in a marked reduction of CD38 expression on multiple myeloma cells. The aim of this study was to investigate the clinical implications and the underlying mechanisms of daratumumab-mediated CD38 reduction.Experimental Design: We evaluated the effect of daratumumab alone or in combination with lenalidomide-dexamethasone, on CD38 levels of multiple myeloma cells and nontumor immune cells in the GEN501 study (daratumumab monotherapy) and the GEN503 study (daratumumab combined with lenalidomide-dexamethasone). In vitro assays were also performed.Results: In both trials, daratumumab reduced CD38 expression on multiple myeloma cells within hours after starting the first infusion, regardless of depth and duration of the response. In addition, CD38 expression on nontumor immune cells, including natural killer cells, T cells, B cells, and monocytes, was also reduced irrespective of alterations in their absolute numbers during therapy. In-depth analyses revealed that CD38 levels of multiple myeloma cells were only reduced in the presence of complement or effector cells, suggesting that the rapid elimination of CD38high multiple myeloma cells can contribute to CD38 reduction. In addition, we discovered that daratumumab-CD38 complexes and accompanying cell membrane were actively transferred from multiple myeloma cells to monocytes and granulocytes. This process of trogocytosis was also associated with reduced surface levels of some other membrane proteins, including CD49d, CD56, and CD138.Conclusions: Daratumumab rapidly reduced CD38 expression levels, at least in part, through trogocytosis. Importantly, all these effects also occurred in patients with deep and durable responses, thus excluding CD38 reduction alone as a mechanism of daratumumab resistance.The trials were registered at www.clinicaltrials.gov as NCT00574288 (GEN501) and NCT1615029 (GEN503). Clin Cancer Res; 23(24); 7498-511. ©2017 AACR.

Trial registration: ClinicalTrials.gov NCT00574288 NCT01615029.

©2017 American Association for Cancer Research.

Figures

Figure 1. Treatment with daratumumab plus lenalidomide-dexamethasone is associated with reduced CD38 expression on MM cells

(A) CD38 expression was measured on BM-localized MM cells in a subset of GEN503 patients before start of treatment (baseline) (n=6), 16 weeks after the initiation of daratumumab plus lenalidomide-dexamethasone treatment (n=6) and also at the time of progression (PD) during daratumumab plus lenalidomide-dexamethasone therapy (n=3) (B) CD38 expression was also determined in patients with circulating MM cells. Before start of treatment 5 out of 9 patients had circulating tumor cells, while 1 week after the first infusion these cells were only detected in 2 patients. In one patient we were able to determine CD38 levels on circulating tumor cells at the time of progression (PD). (C) Only in one patient, circulating MM cells were detectable during the whole treatment period allowing the longitudinal measurement of CD38 expression and absolute numbers. CD38 expression was determined by using HuMax-003-FITC, which binds to a different epitope compared to daratumumab, thereby excluding the possibility that binding of daratumumab masked the detection of CD38. Data are presented as mean ± s.e.m. P-values between the indicated groups were calculated using a paired Student’s t-test; *, P<0.05; ns, not significant. Abbreviations: MFI, median fluorescence intensity; BM, bone marrow; PB, peripheral blood.

Figure 2. Daratumumab alone or combined with lenalidomide-dexamethasone reduces CD38 expression levels on normal white blood cell subsets irrespective of changes in their absolute numbers

(A) Changes in CD38 expression levels and absolute numbers (abs) of various white blood cell populations during daratumumab monotherapy (GEN501 study; n=17) or daratumumab combined with lenalidomide and dexamethasone (GEN503 study; n=9). (B) Representative flow cytometry histogram overlays depicting cell surface expression of CD38 on various white blood cell populations from a representative patient treated with daratumumab in the GEN501 study and a representative patient treated with daratumumab plus lenalidomide-dexamethasone in the GEN503 trial at different time points: before the first daratumumab infusion (red histogram) and during daratumumab treatment (blue histogram). (C) CD38 expression levels of various white blood cell subsets were measured both before (pre) and directly after the end of the first infusion (post) in a different group of patients treated with daratumumab in our institution (n=8). In 3 of these 8 patients circulating MM cells were also detected and analyzed for CD38 expression levels. Median duration of the first infusion was 6 hours and 55 minutes. (D) Flow cytometry dot plots from the 3 patients who had circulating MM cells showing CD38 and CD138 expression on the different PBMC subsets and the circulating MM cells before and directly after the first daratumumab infusion. CD38 expression was determined by using HuMax-003-FITC, which binds to a different epitope compared to daratumumab, thereby excluding the possibility that binding of daratumumab masked the detection of CD38. Data are presented as mean ± s.e.m. P-values between the indicated groups were calculated using a paired Student’s t-test; *, P<0.05; ** P<0.005; ***, P<0.0005; ****, P<0.00005, ns not significant. Abbreviations: MFI, median fluorescence intensity.

Figure 3. Daratumumab-mediated CD38 reduction on MM cell lines is only observed in the presence of effector cells or complement

UM9, UM9-CD38, and Daudi cells were treated with daratumumab (10 μg/ml) or IgG-b12 control antibody in the absence of complement or effector cells (A), in the presence of pooled unheated human serum as a source of complement (B), or in the presence of freshly isolated PBMCs from healthy donors as effector cells (C). The left panel shows for each of these conditions the median fluorescence intensity (MFI) of CD38 on surviving MM cells treated with daratumumab or control IgG-b12 antibody. The right panel shows for each condition the percentage lysis, which was determined and calculated as described in the materials and methods. ADCC- and CDC-assays were performed as described in materials and methods. (D) Ex vivo ADCC and CDC assays were performed with BM aspirates from 7 and 10 MM patients, respectively. CDC- and ADCC-assays were performed as described in materials and methods. The black bars show median fluorescence intensity (MFI) of CD38 on surviving MM cells as determined by flow cytometry, and white bars show the percentage of MM cell lysis mediated by daratumumab in ADCC (left) or CDC assays (right). Data are presented as mean ± s.e.m. P-values between the indicated groups were calculated using a paired Student’s t-test; *, P<0.05; **, P<0.005; ns, not significant. Abbreviations: DARA, daratumumab; MFI, median fluorescence intensity; PBMC, peripheral blood mononuclear cells.

Figure 4. Monocytes and granulocytes reduce CD38 expression of MM cells

(A) UM9 cells were co-cultured with PBMCs or with purified white blood cell populations, isolated from healthy donor buffy coats as described in the materials and methods. The left panel shows the median fluorescence intensity (MFI) of CD38 on surviving UM9 cells treated with 10 μg/ml daratumumab or IgG-b12 control antibody in the presence of the different effector cells. The right panel shows the lysis of the UM9 cells mediated by the different white blood cell populations. (B) UM9 cells were stained with CellTrace Violet and then opsonized with daratumumab or IgG-b12 control antibody which were both labeled with AF488. These UM9 cells were subsequently co-cultured with white blood cells. Uptake of AF488 signal by white blood cell subsets was analyzed by flow cytometry. The left panel shows the median fluorescence intensity (MFI) values for the AF488 signal of the different white blood cell subsets. Gating on cells with a negative signal for CellTrace Violet ensured that the increase in AF488 signal of PBMC was not caused by phagocytosis of complete UM9 cells (right panel). Data are presented as mean ± s.e.m. P-values between the indicated groups were calculated using a paired Student’s t-test; *, P<0.05; **, P<0.005; ns, not significant. Abbreviations: MFI, median fluorescence intensity; PBMC, peripheral blood mononuclear cells; PMN, polymorphonuclear leukocytes (granulocytes); WBC, white blood cells.

Figure 5. Transfer of membrane fragments from UM9 cells to monocytes

(A) UM9 cells were stained with CellTrace Violet and then opsonized with AF488-labeled daratumumab. These UM9 cells were subsequently co-cultured for 2 hours with monocytes which were pre-treated with or without polyclonal human IgG to block Fcγ receptors. The signal for AF488-daratumumab on monocytes and UM9 cells was analyzed by flow cytometry. Shown are the median fluorescence intensity (MFI) values for the AF488 signal. Pre-treatment of monocytes with polyclonal IgG partially prevented the loss of AF488-labeled daratumumab on UM9 cells. At the same time the increase in AF488-labeled daratumumab on monocytes was significantly lower in monocytes pre-treated with polyclonal IgG compared to untreated monocytes. (B) Monocytes were incubated with or without daratumumab-AF488 or IgG-b12-AF488-treated UM9 cells. The cell membrane of these UM9 cells was also stained with PKH-26. After a 2-hour co-incubation, monocytes experienced a significant increase of both PKH-26 and AF488-labeled daratumumab. The uptake of PKH-26 by monocytes was only observed when UM9 cells were opsonized with daratumumab, but not if they were opsonized with the IgG-b12 control antibody. The left panel shows representative flow cytometry dot plots (left dot plot: monocytes only; right dot plot: membrane-stained (PKH-26) and daratumumab-AF488-opsonized UM9 cells in the presence of monocytes; monocytes are depicted in blue; UM9 cells are depicted in red). The right panel shows the median fluorescence intensity (MFI) of the cell membrane dye PKH-26 on monocytes treated under different conditions. Data are presented as mean ± s.e.m. P-values between the indicated groups were calculated using a paired Student’s t-test; *, P<0.05; **, P<0.005; ns, not significant (C) Membrane-stained (PKH-26) and daratumumab-AF488-opsonized UM9 cells were incubated with CellTrace Violet-labeled monocytes. Confocal scanning laser microscopy shows the transfer of both daratumumab-AF488 (green) and cell membrane (red) from UM9 cells to CellTrace Violet-labeled-monocytes (blue) in two representative cytospins. Monocytes (yellow arrows) which are in close proximity to UM9 cells generally take up both AF488-labeled daratumumab and membrane fragments from UM9 cells, while monocytes (white arrows), which are not attached to UM9 cells are generally negative for both AF488-daratumumab and membrane staining. Abbreviations: DARA, daratumumab; MFI, median fluorescence intensity.

Figure 6. Several cell membrane proteins also decrease on MM cells exposed to daratumumab in the presence of effector cells

(A) Cell lines were incubated with daratumumab (10 μg/ml) or IgG-b12 control antibody (10 μg/ml) in the presence of pooled unheated human serum as source of complement or freshly isolated PBMCs as effector cells, as described in materials and methods. Cells were then harvested to determine CD138 (UM9, and UM9-CD38 cells) or CD19 (Daudi) expression levels on surviving cells by flow cytometry. (B) The effect of daratumumab on several other membrane proteins known to be expressed on MM and lymphoma cells (CD44, CD49d, CD54, and CD56) was tested in the presence of monocytes (upper panel) or pooled unheated human serum as a source of complement (lower panel), as described in materials and methods. After a 2-hour incubation, cells were harvested to determine expression levels of CD38, CD44, CD49d, CD54, and CD56 on surviving cells by flow cytometry. P-values between the indicated groups were calculated using a paired Student’s t-test. (C) Expression of CD138 on circulating MM cells was determined before (pre) and directly after the first infusion of daratumumab (post) in 3 patients. The circulating tumor cells were CD56 positive in 1 of these 3 patients, and expression levels of this protein were also analyzed before and directly after the first daratumumab infusion. (D) Longitudinal data representation of CD56 expression levels on BM-localized MM cells from 14 patients treated in GEN501 and GEN503 studies with CD56-positive MM cells from whom sequential BM samples were available. There was marked reduction of CD56 expression levels during therapy but also at the time of progression (PD). The reduction was most prominent in patients with high CD56 expression. The left panel shows the absolute median fluorescence intensity (MFI) values; the middle panel shows the percent change of MFI for each individual patient; and the right panel shows a representative flow cytometry histogram overlay depicting cell surface expression of CD56 on MM cells at different time points: before start of treatment (green histogram), during daratumumab treatment (blue histogram), and at the time of progressive disease (PD; red histogram). (E) Paired BM samples (patient 1; treated in GEN501) and paired peripheral blood samples (patient 2; treated in GEN501) were used to analyze expression levels of CD38, CD49d, CD44, CD54, CD138, and CD56 on MM cells by flow cytometry. The first sample was obtained before start of daratumumab treatment and the second sample was obtained 4 weeks after the first daratumumab infusion. Shown is the percent change in MFI of the different membrane proteins, as compared to baseline values. Flow cytometry histogram overlays depict surface expression of CD38, CD56, CD49d, and CD138 on BM-localized MM cells from patient 1 (before the first daratumumab infusion (red histogram) and 4 weeks after the first daratumumab infusion (blue histogram)). Data are presented as mean ± s.e.m. P-values between the indicated groups were calculated using a paired Student’s t-test; *, P<0.05, **, P<0.005, ***, P<0.0005, ****, P<0.00005; ns, not significant. Abbreviations: DARA, daratumumab; MFI, median fluorescence intensity; PBMC, peripheral blood mononuclear cells; BM, bone marrow; PB, peripheral blood.