Lenalidomide Enhances Immune Checkpoint Blockade-Induced Immune Response in Multiple Myeloma

Abstract

Purpose: PD-1/PD-L1 signaling promotes tumor growth while inhibiting effector cell-mediated antitumor immune responses. Here, we assessed the impact of single and dual blockade of PD-1/PD-L1, alone or in combination with lenalidomide, on accessory and immune cell function as well as multiple myeloma cell growth in the bone marrow (BM) milieu.

Experimental design: Surface expression of PD-1 on immune effector cells, and PD-L1 expression on CD138(+) multiple myeloma cells and myeloid-derived suppressor cells (MDSC) were determined in BM from newly diagnosed (ND) multiple myeloma and relapsed/refractory (RR) multiple myeloma versus healthy donor (HD). We defined the impact of single and dual blockade of PD-1/PD-L1, alone and with lenalidomide, on autologous anti-multiple myeloma immune response and tumor cell growth.

Results: Both ND and RR patient multiple myeloma cells have increased PD-L1 mRNA and surface expression compared with HD. There is also a significant increase in PD-1 expression on effector cells in multiple myeloma. Importantly, PD-1/PD-L1 blockade abrogates BM stromal cell (BMSC)-induced multiple myeloma growth, and combined blockade of PD-1/PD-L1 with lenalidomide further inhibits BMSC-induced tumor growth. These effects are associated with induction of intracellular expression of IFNγ and granzyme B in effector cells. Importantly, PD-L1 expression in multiple myeloma is higher on MDSC than on antigen-presenting cells, and PD-1/PD-L1 blockade inhibits MDSC-mediated multiple myeloma growth. Finally, lenalidomide with PD-1/PD-L1 blockade inhibits MDSC-mediated immune suppression.

Conclusions: Our data therefore demonstrate that checkpoint signaling plays an important role in providing the tumor-promoting, immune-suppressive microenvironment in multiple myeloma, and that PD-1/PD-L1 blockade induces anti-multiple myeloma immune response that can be enhanced by lenalidomide, providing the framework for clinical evaluation of combination therapy.

Figure 1. Increased frequency of PD1 and PD-L1 expression in MM bone marrow microenvironment

Cell surface expression of PD1 (CD279) and PD-L1 (CD274) is shown on CD138⁺MM cells, immune effector cells, and immune-suppressive MDSC in ND-MM and RR-MM. A. Cell surface expression of PD-L1 is quantitated on patient MM cells obtained from patients with ND-MM (n=6), RR-MM (n=10), and compared to healthy donor plasma cells (HD, n=3). Data represent percentage of PD-L1 expressing CD138⁺MM cells. Representative histogram plots of PD-L1 expression (red) relative to control (gray) is shown on a gated population of CD138⁺ plasma cells. Top panel demonstrates PD-L1 expression on MM cell lines (MM1.S, OPM2, H929), and lower panel represents PD-L1 expression on BM CD138⁺ plasma cells from patients with ND-MM and RR-MM, as well as HD-BM. B. Cell surface expression of PD1 is quantitated on immune effector cells (CD4T cells, CD8T cells, NK cells and NKT cells) from patients with ND-MM (n=6) and RR-MM (n=10) compared to healthy donors (HD, n=10). Data represent percentage of PD1 coexpressing CD4T cells (top left graph), CD8T cells (top right graph), NK cells (lower left graph), and NKT cells (lower right graph) in BM of patients with ND-MM and RR-MM compared to HD-PBMC. Representative histogram plots of PD1 expression (blue) versus control (grey) on BM immune effector cells: CD4T cells, CD8T cells, NK cells, and NKT cells with gating strategy is shown by multi-parameter dot plots (right panel). C. The frequency of PD-L1 cell surface expression is shown in monocytic MDSC (mMDSC) and neutrophilic MDSC (nMDSC) compared to antigen presenting cells (APC) from BM of patients with ND-MM (left graph) and RR-MM (right graph). Representative flow cytometry histogram of PD-L1 expression (red) versus control (grey) on MDSC in healthy donor and RR-MM bone marrow (right panel). With gating strategy for mMDSC and nMDSc is shown by multi-parameter dot plots (right panel).

Figure 2. Checkpoint blockade overcomes bone marrow stroma-mediated MM growth

PD1/PD-L1 signaling role in the bidirectional interaction between MM cells and bone marrow stroma cells (BMSC) is shown in coculture of BMSC and MM cells. A. BMSC effect on PD-L1 expression in MM cells is demonstrated by multi-parameter flow cytometry analysis of MM cell-BMSC cocultures. Representative histogram plot for PD-L1 expression in CD138⁺MM cell population (red) versus control (grey) is demonstrated in MM cell line (H929) alone and cultured with BMSC. B. Checkpoint blockade effect on BMSC-mediated MM growth is demonstrated by CFSE-flow analysis in cocultures of CD138⁺MM cells from patient with RR-MM with BMSC. Shown are multi-parameter-dot plots for MM cells alone and cultured with BMSC with or without single and dual blockade of PD1 and PD-L1. CD138⁺CFSE^low cell population represents live/growth MM cells (large gated box) and CD138⁺CFSE^high cell population represents non-dividing/dead MM cells (small gated box).

Figure 3. Checkpoint blockade enhances immune effector cell-mediated anti-MM responses in MM bone marrow

A. Impact of checkpoint blockade on immune effector cell-mediated anti-MM response is demonstrated by a multi-parameter CFSE/PI flow cytometry analysis. CFSE labeled target CD138⁺MM cells and autologous immune effector cells (CD3T cells and NK cells) were cultured with or without anti-PD1 and anti-PD-L1, alone or in combination. Effector cell mediated-cytotoxicity was determined by CD138⁺CFSE⁺PI⁺ apoptotic/dead MM cells. Fold change is relative to control (effector cell-mediated MM cytotoxicity without checkpoint blockade). Impact of checkpoint blockade on effector cytokines IFNγ (B) and Gzm-B (C) mediating cytotoxicity against MM is shown in effector cells in cocultures of CD138+MM cells and autologous effector cells.

Figure 4. Checkpoint blockade partially reverses MDSC-mediated MM growth and immune suppression in MM bone marrow

A. Impact of checkpoint blockade on MDSC-mediated tumor growth is demonstrated in the bone marrow of patients with RR-MM by CFSE-flow cytometry analysis. CFSE labeled CD138⁺MM cells and autologous MDSC were cultured in the absence or presence of anti-PD1 and anti-PD-L1, alone or in combination. Viability/growth of CD138⁺MM cells (CD138⁺CFSE^low) is shown by representative histogram plots. B. Effect of checkpoint blockade on MDSC-mediated immune suppression is shown by intracellular effector cytokine analysis in RR-MM BM. Autologous T cells cultured either alone or with mMDSC and nMDSC in the absence or presence of anti-PD1 and anti-PD-L1, alone or in combination. Gated boxes demonstrate percent intracellular IFNγ expression in T cells cultured alone (top panel), with mMDSC (middle panel), and with nMDSC (lower panel) with or without dual PD1 and PD-L1 blockade. C. Impact of checkpoint blockade with lenalidomide on MDSC-mediated immune suppression is shown in RR-MM bone marrow by flow cytometry intracellular cytokine analysis. A representative bar graph shows percent intracellular Gzm-B expression in all effector cells cultured alone or with autologous mMDSC and nMDSC (top graph). A representative bar graph of intracellular Gzm-B expression is shown in each effector cell population from the coculture of effector cells with autologous mMDSC (middle graph) and with autologous nMDSC (lower graph).

Figure 5. Lenalidomide reduces expression of PD1 and PD-L1 and enhances checkpoint blockade-induced MM cytotoxicity in MM bone marrow

Impact of lenalidomide on cell surface expression of PD1 on effector cells (A) and PD-L1 on CD138+MM cells (B), as well as CD14+myeloid cells and MDSCs (C) in RR-MM bone marrow is shown by representative histogram plots. Percent PD1 expression on the effector cells and PD-L1 on the CD14+myeloid cells and MDSCs of untreated BM cells (blue) and lenalidomide treated BM cells (red) is shown relative to control (grey). D. Impact of lenalidomide on checkpoint blockade-induced MM cytotoxicity is shown in a representative graph of RR-MM bone marrow. Mononuclear cells from patient with RR-MM bone marrow were labeled with CFSE and cultured in the absence or presence of anti-PD1 and anti-PD-L1, alone or in combination, and with or without addition of lenalidomide. Shown is a percent Apoptotic/Dead (CD138⁺CFSE⁺PI⁺) MM cells in BMMC (top graph). Impact of lenalidomide on checkpoint blockade-induced effector cell activity is shown in RR-MM bone marrow by intracellular effector cytokine analysis. RR-MM bone marrow cells were cultured with anti-PD1 and anti-PD-L1 alone, or in combination, or with the addition of lenalidomide. Representative bar graph shows intracellular expression of IFNγ in each effector cell populations in RR-MM bone marrow (lower graph).