CDK4/6 Inhibition Augments Antitumor Immunity by Enhancing T-cell Activation

Abstract

Immune checkpoint blockade, exemplified by antibodies targeting the PD-1 receptor, can induce durable tumor regressions in some patients. To enhance the efficacy of existing immunotherapies, we screened for small molecules capable of increasing the activity of T cells suppressed by PD-1. Here, we show that short-term exposure to small-molecule inhibitors of cyclin-dependent kinases 4 and 6 (CDK4/6) significantly enhances T-cell activation, contributing to antitumor effects in vivo, due in part to the derepression of NFAT family proteins and their target genes, critical regulators of T-cell function. Although CDK4/6 inhibitors decrease T-cell proliferation, they increase tumor infiltration and activation of effector T cells. Moreover, CDK4/6 inhibition augments the response to PD-1 blockade in a novel ex vivo organotypic tumor spheroid culture system and in multiple in vivo murine syngeneic models, thereby providing a rationale for combining CDK4/6 inhibitors and immunotherapies.Significance: Our results define previously unrecognized immunomodulatory functions of CDK4/6 and suggest that combining CDK4/6 inhibitors with immune checkpoint blockade may increase treatment efficacy in patients. Furthermore, our study highlights the critical importance of identifying complementary strategies to improve the efficacy of immunotherapy for patients with cancer. Cancer Discov; 8(2); 216-33. ©2017 AACR.See related commentary by Balko and Sosman, p. 143See related article by Jenkins et al., p. 196This article is highlighted in the In This Issue feature, p. 127.

Figure 1. Small molecule CDK4/6 inhibitors enhance IL-2 secretion from T cells

A, Plot of replicate Z scores from screening of small molecule compounds capable of enhancing IL-2 secretion from stimulated PD-1-Jurkat cells with compounds annotated as GSK3α/β or CDK4/6 inhibitors labeled. B, Quantification of IL-2 levels by ELISA from PD-1-Jurkat cells treated with PMA/ionomycin or 1 µM CDK4/6 inhibitors and stimulated as indicated for 18h. Results shown as mean ± SD (UT, n=2; other conditions, n=5) (*p<0.05). C, Quantification of IL-2 levels by ELISA from primary human CD4+ T cells treated with 100 nM palbociclib or trilaciclib and stimulated as indicated. Results shown as mean ± SD (UT, n=2; other conditions, n=4) (*p<0.05). D, Immunoblot for CDK4 and CDK6 from PD-1-Jurkat cells transiently transfected with the indicated siRNA. E, Quantification of IL-2 levels from PD-1-Jurkat cells after transient transfection with siRNA against Cdk4 or Cdk6 and stimulated as indicated for 18h. Results shown as mean ± SD (n=4) (*p<0.05). Cytokine profiling analysis from human patients using patient-derived organotypic tumor spheroids (PDOTS) cultured in 3-dimensional culturing system at day 1 (F) and day 3 (G). Freshly-obtained patient samples were digested into spheroids and treated with indicated drugs in the 3-D microfluidic system. Cytokine secretion was analyzed by Luminex and expressed as log2-fold change relative to untreated control.

Figure 2. CDK4/6 inhibition de-represses NFAT activity

A, Quantification of IL-2 levels from PD-1-Jurkat cells treated with 1 µM palbociclib and/or 1 µM cyclosporine A (CsA) and stimulated as indicated for 18h. Results shown as mean ± SD (UT, n=2; other conditions, n=4) (*p<0.05). B, Immunoblot for levels of phospho-S172 and total NFAT2 after treatment of PD-1-Jurkat cells with 1 µM Palbociclib and stimulated as indicated for 18h. C, Immunoblot for NFAT4 from nuclear and cytoplasmic fractions of PD-1-Jurkat cells treated with 1 µM Palbociclib and stimulated as indicated for 18h. D, Normalized luminescence of PD-1-Jurkat cells transiently transfected with NFAT-FLuc and RLuc-SV40 reporters after treatment with 1 µM Palbociclib and stimulated as indicated for 18h. Results shown as mean ± SD (n=3) (*p<0.05). E, Relative levels of Il-2, Il-3, and GM-CSF mRNA as measured by qPCR from PD-1-Jurkat cells treated with 1 µM Palbociclib and stimulated as indicated for 8h. Results shown as mean ± SD (n=3). *p<0.05 by two-way ANOVA with Bonferonni correction for multiple comparisons.

Figure 3. Analysis of immune infiltrates of lung tumor after CDK4/6 inhibition

Genetically engineered mouse model (GEMM) harboring the Kras^LSL-G12DTrp53^fl/fl mutation was induced by Ad-CRE recombinase for lung tumors. After verification of tumor formation by MRI scan, mice were then treated with either trilaciclib (trila) or palbociclib (palb) every day for 7 days, after which lung tissues were collected for FACS analysis. Results shown are pooled from three independent experiments. Lung infiltrating T cells percentage among total CD45⁺ leukocytes (A) or absolute cell number (B) after treatment with trilaciclib (n=8) or palbociclib (ctrl, n=4, Palb., n=5). (*p<0.05, ***p<0.001) c, BrdU incorporation by T cells shows proliferation affected by CDK4/6 inhibitors trilaciclib or palbociclib. Mice without (naïve, upper panel) or with (TMB, lower panel) Kras^LSL-G12DTrp53^fl/fl (KP) allograft tumor were treated with trilaciclib or palbociclib, followed by systemic BrdU injection (I.P.). BrdU incorporation within different T cell subpopulations Treg (CD4⁺Foxp3⁺) and Tconv (CD4⁺Foxp3⁻) was determined by flow cytometry (n=6) (*p<0.05, **p<0.01). D, Expression levels of PD-1 and CTLA-4 in CD4⁺ or CD8⁺ T cells infiltrated at tumor site after treatment (ctrl, n=4, Palb., n=5) (*p<0.05). E, Changes in levels of CD11b⁺ and CD11c⁺ myeloid subpopulations after trilaciclib (n=8) or palbociclib (ctrl, n=4, Palb., n=5) treatment (*p<0.05, **p<0.01).

Figure 4. Tumor antigen-experienced T cells exhibit greater sensitivity to CDK4/6 inhibition

A, IL-2 production from Tconv cells after trilaciclib treatment. CD4⁺CD25⁻ Tconv cells were isolated from either naïve or tumor bearing (TMB) mice and treated with trilaciclib at indicated concentrations, in the presence of CD3 and CD28 stimulation. IL-2 production was determined 3 days after the treatment and normalized with untreated control. (n=3) (*p<0.05) B, Increased IFNγ production in CD8⁺ T cells by trilaciclib treatment, in the presence of Treg. CD8⁺ T cells from naïve or TMB mice were isolated and co-cultured with CD4⁺CD25⁺ Treg cells (5:1 ratio), in the presence of different concentrations of trilaciclib as indicated. IFNγ production was determined 3 days after the treatment and normalized with untreated control. (n=3) (*p<0.05, **p<0.01) C, Violin plot of expression levels of NFAT-regulated genes determined by single-cell RNA-seq of tumor-infiltrating CD3⁺ T cells from KP GEMM mice 7 days after trilaciclib treatment. D, t-Distributed Stochastic Neighbor Embedding (t-SNE) plot showing distinct homogenous groups of T cells identified with density based clustering (dbscan). E, Heat map showing transcriptional levels of genes from each cell that are important for T cell activation and suppression, and IL-2 and TCR signaling. The status of each cell including treatment status and group identification is shown below the heat map as bar graphs. Each column represents one cell.

Figure 5. CDK4/6 inhibitor elicits anti-tumor immunity and enhance cell death induced by anti-PD-1 antibody ex vivo

A, Quantification of tumor volume changes by MRI scan after treatment with trilaciclib. Left panel, waterfall plot shows tumor volume response to the treatment. Each column represents one mouse. Right panel, representative MRI scan images (one out of 24 scanned images of each mouse) show mice lung tumors before and after the treatment. Circled areas, heart. B. Live (AO=green) /Dead (PI=red) analysis of murine-derived organotypic tumor spheroid (MDOTS) cultured in 3-D microfluidic culture at day 0, day 3 and day 6 following treatment of CDK4/6 inhibitors trilaciclib or palbociclib (100nM) alone or in combination with PD-1 antibody (10 µg/ml) as indicated. Upper panel, quantification results of live/dead analysis; lower panel, representative images of deconvolution fluorescence microscopy shows live/dead cells at day 6 after indicated treatment. Statistical analysis is calculated by comparing the indicated treatment group with DMSO+IgG group at day 6. (*p<0.05, **p<0.01, ***p<0.001) Scale bar, 50 µm C. Cytokine secretion from MC38 MDOTS were expressed as log2-fold change (L2FC) relative to untreated control after indicated treatment.

Figure 6. Combination treatment of CDK4/6 inhibitor synergize anti-PD-1 antibody induced anti-tumor immunity through T cell

A, Tumor growth curves of MC38 (left) or CT26 (right) treated with CDK4/6 inhibitor or PD-1 antibody alone or in combination. MC38 murine cancer cells were injected subcutaneously into C57BL/6 mice. The mice were treated with either CDK4/6 inhibitor (trilaciclib or palbociclib, 100 mg/kg) intermittently (3 days on, 4 days off) with or without PD-1 antibody (200 µg/mouse, 3 times a week) as indicated starting from day 3 (MC38) or day 7 (CT26). Tumor volumes were monitored every 2~3 days. Each graph shows representative results from two independent experiments. (left panel, n=8; right panel, n=10) (*p<0.05, **p<0.01, ***p<0.001) B, Individual traces of tumor volume of CT26 tumors over time after treatment with palbociclib and anti-PD-1, either alone or in combination (n=8). C, Quantification of cytokine production produced by MC38 tumor infiltrating T lymphocytes. At the end of the treatment (day 17), mice were sacrificed and TILs were isolated from the tumor for cytokine analysis for IL-2 from CD4⁺ T cells (left panel) and IFNγ from CD8+ T cells (right panel). (*p<0.05, **p<0.01, ***p<0.001) D, Cytokine production of IFNγ from CD8⁺ T cells from inguinal lymph nodes of mice with MC38 tumors treated with trilaciclib at the end of treatment (day 17). E, Tumor growth curves of CT26 treated with palbociclib (100mg/kg) and PD-1 antibody (200 µg/mouse) with or without anti-CD4 (400 µg/mouse) or anti-CD8 (400 µg/mouse) depletion antibodies. The depletion antibody treatment started at day −3 before tumor implantation was continued twice a week. Palbociclib and PD-1 were dosed at the same schedule shown in panel d starting from day 7. The graph shows representative result of two independent experiments, and the dosing and tumor measurement were performed by different people. (n=10) (***p<0.001, ****p<0.0001)