Clinical regressions and broad immune activation following combination therapy targeting human NKT cells in myeloma

Abstract

Natural killer T (iNKT) cells can help mediate immune surveillance against tumors in mice. Prior studies targeting human iNKT cells were limited to therapy of advanced cancer and led to only modest activation of innate immunity. Clinical myeloma is preceded by an asymptomatic precursor phase. Lenalidomide was shown to mediate antigen-specific costimulation of human iNKT cells. We treated 6 patients with asymptomatic myeloma with 3 cycles of combination of α-galactosylceramide-loaded monocyte-derived dendritic cells and low-dose lenalidomide. Therapy was well tolerated and led to reduction in tumor-associated monoclonal immunoglobulin in 3 of 4 patients with measurable disease. Combination therapy led to activation-induced decline in measurable iNKT cells and activation of NK cells with an increase in NKG2D and CD56 expression. Treatment also led to activation of monocytes with an increase in CD16 expression. Each cycle of therapy was associated with induction of eosinophilia as well as an increase in serum soluble IL2 receptor. Clinical responses correlated with pre-existing or treatment-induced antitumor T-cell immunity. These data demonstrate synergistic activation of several innate immune cells by this combination and the capacity to mediate tumor regression. Combination therapies targeting iNKT cells may be of benefit toward prevention of cancer in humans.

Trial registration: ClinicalTrials.gov NCT00698776.

Figures

Figure 1

Study schema. All patients underwent leukapheresis > 2 weeks before initiating therapy to harvest DC progenitors. Treatment protocol consisted of 3 cycles of lenalidomide (10 mg/d) for 3 weeks followed by 1 week of rest. KRN7000 DCs were infused on day 7 of each cycle.

Figure 2

Changes in NKT cells. The presence of NKT cells was monitored by flow cytometry. PBMCs obtained from patients before starting therapy (PreC1), days 7 and 14 of cycle 1, before starting cycle 2 (PreC2), and at the completion of therapy (Post) were analyzed for the presence of iNKT cells using flow cytometry. The cells were stained with anti-CD3, Vα24, and Vβ11 antibodies. PBMCs were also analyzed using CD1d dimer either unloaded (Dimer control) or loaded with α-Galcer (Dimer Galcer). (A) Data from a representative patient during first cycle of therapy. (B) Summary of pooled data. Only patients with frequency of NKT cells above 0.01% at baseline were used to reliably estimate posttreatment decline in NKT cells. *P < .05; **P < .01.

Figure 3

Changes in NK cells. PBMCs obtained from patients before staring therapy (PreC1), before cycle 2 (PreC2), and before cycle 3 (PreC3) as well as day 7 and 14 of each cycle were analyzed for the presence of CD3−, CD56+ NK cells using flow cytometry. Changes in surface expression of CD56, NCRp46, as well as NKG2D on the surface of the CD3−, CD56+ NK cells were also monitored by flow cytometry. (A) Summary of pooled data for frequency of circulating NK cells during therapy, reflected as fold change on day 7 and 14 compared with baseline; *P < .05. (B) Representative data from a single patient. (C) Changes in expression of CD56 on NK cells during cycle 1. (Left panel) Data from a representative patient. (Right panel) Summary of pooled data from all patients; P < .05. (D) Changes in expression of NKG2D on NK cells during cycle 1. (Left panel) Data from a representative patient. (Right panel) Summary of pooled data from all patients; ** P = .07. (E) Changes in expression of NCRp46 on NK cells during cycle 1. (Left panel) Data from a representative patient. (Right panel) Summary of pooled data from all patients.

Figure 4

Changes in other innate cells and cytokines. (A) Changes in expression of CD16 on monocytes during cycle 1. PBMCs obtained from patients before starting therapy (PreC1) as well as from cycle 2 (PreC2) were analyzed for the presence of CD16 on CD14+ monocytes using flow cytometry. (Left panel) Data from a representative patient. (Right panel) Summary of pooled data from all patients. (B) Changes in eosinophils. A complete blood count was obtained weekly as part of routine care of the patients. Shown are the changes in eosinophil count as a percentage of white blood cells during therapy. (C) Changes in plasma levels of soluble IL2 receptor during therapy. Plasma obtained from blood collected on patients before starting therapy (PreC1), before starting cycle 2 (PreC2), before starting cycle 3 (PreC3), as well as days 7 and 14 of each cycle was analyzed for the presence of soluble IL2-receptor (sIL-2R) using a multiplex Luminex assay. Shown are the changes in sIL2R expressed as fold change from baseline (before starting cycle 1).

Figure 5

Clinical effects of therapy. (A) Comparison of tumor-associated monoclonal Ig (M spike) at baseline versus the value at best response. (B) Differential kinetics of M spike versus serum-free light chains in a patient (P5) with progressive increase in M spike after completion of therapy.

Figure 6

Changes in T-cell immunity. (A) Changes in cytokine profile of T cells. PBMCs from pre- and posttherapy time points were thawed and stimulated with PMA and ionomycin. The percentage of CD3+ T cells expressing individual cytokines (IFNγ, IL2, IL17, and IL4) was evaluated using intracellular cytokine staining and flow cytometry. Shown is the fold change in cytokine production on days 7 and 14 of cycle one as well as before starting cycle 2 (PreC2) compared with cytokine production by T cells before starting therapy (preC1). (B-C) PBMCs from baseline (pretreatment) were stimulated overnight with an overlapping peptide library derived from SOX2 (B) or a pool of peptides from viral antigens (CEF) as a control (C). Reactivity to the peptide pools was assayed with the detection of IP10 in the supernatant. Stimulation index refers to fold increase in IP10 over control wells. (D) Changes in SOX2 reactivity during therapy in a patient P2 with clinical response to therapy. PBMCs were stimulated overnight with peptide pools derived from SOX2 as in panel B. After overnight culture, the presence of IP10 in supernatant was assayed using Luminex.