Egress of CD19(+)CD5(+) cells into peripheral blood following treatment with the Bruton tyrosine kinase inhibitor ibrutinib in mantle cell lymphoma patients

Abstract

Ibrutinib (PCI-32765) is a highly potent oral Bruton tyrosine kinase (BTK) inhibitor in clinical development for treating B-cell lymphoproliferative diseases. Patients with chronic lymphocytic leukemia (CLL) often show marked, transient increases of circulating CLL cells following ibrutinib treatments, as seen with other inhibitors of the B-cell receptor (BCR) pathway. In a phase 1 study of ibrutinib, we noted similar effects in patients with mantle cell lymphoma (MCL). Here, we characterize the patterns and phenotypes of cells mobilized among patients with MCL and further investigate the mechanism of this effect. Peripheral blood CD19(+)CD5(+) cells from MCL patients were found to have significant reduction in the expression of CXCR4, CD38, and Ki67 after 7 days of treatment. In addition, plasma chemokines such as CCL22, CCL4, and CXCL13 were reduced 40% to 60% after treatment. Mechanistically, ibrutinib inhibited BCR- and chemokine-mediated adhesion and chemotaxis of MCL cell lines and dose-dependently inhibited BCR, stromal cell, and CXCL12/CXCL13 stimulations of pBTK, pPLCγ2, pERK, or pAKT. Importantly, ibrutinib inhibited migration of MCL cells beneath stromal cells in coculture. We propose that BTK is essential for the homing of MCL cells into lymphoid tissues, and its inhibition results in an egress of malignant cells into peripheral blood. This trial was registered at www.clinicaltrials.gov as #NCT00114738.

Figures

Figure 1

Transient mobilization of lymphocytes in MCL patients treated with ibrutinib. The mean percentage change of ALC (over baseline) is graphed against treatment time. (A) Ibrutinib treatment was in 35-day cycles of 28 days on and 7 days off. Mean percentage ALC change and percentage of the sum of perpendicular diameters (SPD) change are plotted to treatment cycles (n = 9). (B) Ibrutinib treatment was continuous with no gap in treatment. Mean percentage ALC change compared with baseline (plotted as mean + standard error [SE]; n = 17) is plotted against treatment time. Time points plotted are D0 (day zero), D1, D8, D15, D22, and end of cycles 1, 2, 3, 4, and 5. (C) The absolute count (ABS) and percentage of CD19+CD5+ vs CD19+CD5– cells following 1 week of ibrutinib treatment. Note the statistically significant change in the absolute count of CD19+CD5+ cells of MCL patients following treatment. *P < .05 (paired t test; n = 16). (D) Flow plot of gated lymphocytes of PBMC samples from a representative MCL patient before and after ibrutinib treatment (560 mg per day) for 7 days. PBMCs were stained with CD3, CD19, and CD5. Note increase of CD19+CD3– and CD19+CD5+ population after 7 days of drug treatment. (E) Fluorescent probe occupancy assay of BTK in PBMCs isolated from a representative MCL patient before treatment (predose), and after 4 hours, 24 hours after first dose, before treatment on the eighth day (Day 8, predose), and 4 hours after the eighth day dose (Day 8, 4HR). The arrows point to the 75 kDa BTK band on a scanned fluorescent gel (top) and a western blot (bottom). (F) An average of >90% occupancy of BTK by drug is achieved in MCL patients who were administered ibrutinib in the first week, determined by fluorescent probe assays (n = 6).

Figure 2

CD19+CD5+ cells have decreased CXCR4, CD38, and Ki67 expression following ibrutinib treatment. (A) Reduction of CD38 expression (mean fluorescence intensity ratio [MFIR]) in CD19+CD5+ cells but not CD19+CD5– cells during 4 weeks of treatment in 4 patients treated with ibrutinib. (B) Surface CD38 expression (left panel; *P < .05) and intracellular Ki67 (right panel; **P < .01) is significantly reduced following 1 week of treatment. MFIR of intracellular phospho-ERK (pT202/Y204/ERK1/2) of CD20+CD5+ cells from healthy participants or MCL patients treated with ibrutinib before treatment (day 1 [D1]) and after 1 week of treatment (D8) (lower panel; *P < .05). (C) Significant reduction of surface CXCR4 expression (MFIR) in CD19+CD5+ cells following 1 week of ibrutinib treatment (n = 14; *P < .05). The line between the dot plots shows the mean. (D) CXCR4 and CD38 expression from LN biopsies and PBMCs (PB) of three MCL patients (patients A, B, and C) not treated with drug. (E) Plasma chemokine and cytokine concentrations on day 8 (D8; left) or day 29 (D29; right) of ibrutinib-treated MCL patients compared with pretreatment times 100% (n = 9).

Figure 3

Ibrutinib inhibits migration of MCL cells beneath stromal cells (pseudo-emperipoiesis) and the formation of CXCL12-stimulated cortical actin. (A) Phase contrast (left panel) and 4,6 diamidino-2-phenylindole (DAPI) staining (right panel) of Mino and BM stromal cell M2-10B4 coculture 24 hours after Mino cells were pretreated with vehicle (dimethylsulfoxide [DMSO]) (panels a and b) or ibrutinib (1000 nM) (panels c and d). Highlighted yellow outlines show typical cobblestone appearance of migrated Mino cells beneath stromal cells. Black arrow points to a cell that is adhered on top of stromal cells but not migrated underneath. Yellow arrow points between two migrated Mino cells. (B) Mino cell and stromal cell coculture of Mino cells loaded with 5-(and-6)-(((4-chloromethyl)benzoyl)amino)tetramethylrhodamine (red), and M2 cells loaded with 5-chloromethylfluorescein diacetate (green). (A) Stromal cells alone and (B) in coculture with Mino cells. (C) Mino cells pretreated with G protein coupled receptor inhibitor pertussis toxin at 200 ng/mL or (D) ibrutinib at 1000 nM. (C) Mino cells were stimulated with CXCL12 at 100 ng/mL and stained with rhodamine-phalloidin to determine actin polymerization and were counterstained with DAPI to identify nuclei of cells. Phalloidin staining of Mino cells (A) before and (B) after CXCL12 stimulation and after treatment with (C) pertussis toxin at 200 ng/mL or (D) ibrutinib at 100 nM. Magnification, ×200. (D) Mino cells were pretreated with ibrutinib, pertussis toxin, or vehicle for 30 minutes and then placed on a stromal cell–populated plate. After 4 hours, coculture was washed several times and migrated, and adhered Mino cells were counted in a flow cytometer with calibrated beads after staining with hCD19. Both pertussis toxin and ibrutinib dose-dependently inhibited migration and adhesion of Mino cells (left panel). Mino cells stimulated with CXCL12 and treated with vehicle or drug were stained with phalloidin, and its intensity was determined by using flow cytometry (right panel). (E) Ibrutinib (100 nM) inhibited pseudo-emperipoiesis of primary MCL (hCD19+ cells) in coculture with M2-10B4 stromal cells (left panel). Actin polymerization as assessed by phalloidin staining was significantly reduced by ibrutinib treatment in primary MCL cells. **P < .01; ***P < .001. One-way analysis of variance (ANOVA) compared with vehicle control.

Figure 4

Ibrutinib suppresses BCR- and coculture-stimulated signaling and cytokine and chemokine production of MCL cells. (A) Mino cells pretreated with vehicle or ibrutinib (10, 100, or 1000 nM) were cultured alone (with anti-IgM stimulation) or in coculture with murine M2-10B4 stromal cells. M2-10B4 cells alone were pretreated with either vehicle or 100 nM ibrutinib (M2 only). Mino cells in coculture, Mino cells alone, or M2-10B4 cells alone were collected after 48 hours, lysed, and subjected to sodium dodecyl sulfate polyacrylamide gel electrophoresis and western blot analysis of pY223 BTK, pY1217 PLCγ2, pT202/Y204 ERK, pS473 AKT, and pT183/Y185 JNK. The phosphoblot was scanned, and signals were quantified (right panel); pBTK, pPLCγ2, pERK, pAKT, and pJNK signals were normalized to BTK, PLCγ2, ERK, AKT, and JNK total protein, respectively. Signals from each treatment were compared with vehicle-treated Mino cells. (B) Conditioned media from Mino cells only without stimulation, Mino cells stimulated with anti-IgM, or Mino cells in coculture with M2-10B4 or M2-10B4 alone were collected after 48 hours and analyzed for human cytokines and chemokines (IL-10, CCL22, CCL3, CCL4, TNF-α, CCL17, and CCL21). Note media from murine M2 cells alone do not react with human cytokines or chemokines.

Figure 5

Ibrutinib inhibits BCR-activated adhesion of MCL cells. (A-B) Jeko1 or HBL2 cells pretreated with 100 nM ibrutinib or DMSO vehicle were stimulated with anti-IgM for 0, 0.5, 1, 2, 5, or 10 minutes and then immunoblotted for pBTK (Y223), pERK, and pAKT. (C-D) Jeko1 and HBL2 cells pretreated with drug were subjected to adhesion assays on plates precoated with anti-IgM and either (C) fibronectin or (D) VCAM-1. PMA stimulations were used as a positive control for the assay and to show specificity of drug response. The n values represent the number of independent experiments performed in triplicate; the average of these experiments is displayed. **P < .01; ***P < .001. One-way ANOVA compared with vehicle control. PMA, phorbal 12-myristate 13-acetate.

Figure 6

Ibrutinib inhibits CXCL12-/CXCL13-activated adhesion and migration of MCL cells. (A) Mino (left panel) or Jeko1 (right panel) cells pretreated with vehicle or ibrutinib 10, 100, or 1000 nM were either stimulated with anti-IgM, CXCL12, or CXCL13 or treated with medium (Med) for 15 minutes and then immunoblotted for pBTK, pPLCγ2, pAKT, pERK, and pJNK. (B-C) Mino or Jeko1 cells were treated with 100 nM ibrutinib and subjected to adhesion assays on plates precoated with (B) CXCL12 or CXCL13 and fibronectin or (C) VCAM-1. The n values represent the number of independent experiments performed in triplicate; the average of these experiments are displayed. (D) Cells from MCL cell lines Mino, Jeko1, and JVM-1 were treated with increasing concentration of ibrutinib and subjected to a chemotaxis migration assay in transwell plates with filters coated with VCAM-1, and CXCL12 or CXCL13 was added into the lower chamber as a chemoattractant. Ibrutinib dose-dependently inhibited CXCL12- and CXCL13-mediated migration of Mino cells, and CXCL12-mediated migration of Jeko1 and JVM-1 cells. *P < .05; **P < .01; ***P < .001. One-way ANOVA compared with vehicle control.

Figure 7

Ibrutinib inhibits BCR- and chemokine-induced adhesion of primary MCL cells. (A) CD19+ cells isolated from PBMCs of healthy volunteers or MCL patients were treated with ibrutinib at 100 nM for 10 minutes prior to cell harvest, lysate generation, and western blotting. Primary MCL cells have increased BTK activity compared with B lymphocytes from healthy volunteers, which is further inhibited by ibrutinib in a dose-dependent manner. Tyrosine phosphorylation sites of PLC-γ are also inhibited dose-dependently by drug. (B) Ibrutinib inhibited adhesion of primary MCL cells to fibronectin- or VCAM-1–coated plates co-coated with anti-IgM, CXCL12, or CXCL13 (*P < .05; **P < .01). The n values represent the number of independent experiments performed in triplicate; the averages of these experiments are displayed. (C) Schematic of mechanism of action of ibrutinib. Since BTK is downstream to both BCR and CXCR4 signaling, ibrutinib inhibits chemokine and BCR-mediated cell adhesion/migration of malignant cells thereby disrupting the microenvironment in the tissues, LNs, and BM, which results in the malignant cells egressing and eventually entering the peripheral circulation where they are presumably cleared.