Blockade of beta-adrenergic receptors reduces cancer growth and enhances the response to anti-CTLA4 therapy by modulating the tumor microenvironment

Abstract

The development of immune checkpoint inhibitors (ICI) marks an important breakthrough of cancer therapies in the past years. However, only a limited fraction of patients benefit from such treatments, prompting the search for immune modulating agents that can improve the therapeutic efficacy. The nonselective beta blocker, propranolol, which for decades has been prescribed for the treatment of cardiovascular conditions, has recently been used successfully to treat metastatic angiosarcoma. These results have led to an orphan drug designation by the European Medicines Agency for the treatment of soft tissue sarcomas. The anti-tumor effects of propranolol are suggested to involve the reduction of cancer cell proliferation as well as angiogenesis. Here, we show that oral administration of propranolol delays tumor progression of MCA205 fibrosarcoma model and MC38 colon cancer model and increases the survival rate of tumor bearing mice. Propranolol works by reducing tumor angiogenesis and facilitating an anti-tumoral microenvironment with increased T cell infiltration and reduced infiltration of myeloid-derived suppressor cells (MDSCs). Using T cell deficient mice, we demonstrate that the full anti-tumor effect of propranolol requires the presence of T cells. Flow cytometry-based analysis and RNA sequencing of FACS-sorted cells show that propranolol treatment leads to an upregulation of PD-L1 on tumor associated macrophages (TAMs) and changes in their chemokine expression profile. Lastly, we observe that the co-administration of propranolol significantly enhances the efficacy of anti-CTLA4 therapy. Our results identify propranolol as an immune modulating agent, which can improve immune checkpoint inhibitor therapies in soft tissue sarcoma patients and potentially in other cancers.

Fig. 1. Pharmacological ADRB blockade by propranolol delays tumor growth of MCA205 fibrosarcoma.

A, B Tumor growth curves (A) and survival rates (B) of control, and propranolol (PRO) treated mice (n = 10). C Tumor weight on day 16 (n = 8–9). D Representative Ki67 (brown) IHC staining on MCA205 tumor tissues from control and propranolol treated mice; hematoxylin counterstain; scale bar 50 µm. E Dot plot showing quantification of Ki67 staining in nine control mice and 11 propranolol treated mice. **p < 0.01, n.s. not significant, according to multiple t test with Bonferroni correction. Tumor growth and survival data were analyzed using TumGrowth software. Mean ± SEM are depicted.

Fig. 2. Propranolol treatment reduces tumor angiogenesis.

A–C Quantification of angiogenic marker Vegfa gene expression in MCA205 (A), MC38 (B), and SVR (C) tumors from control mice and propranolol (PRO) treated mice by qRT-PCR (n = 5). D–F Quantification of Kdr gene expression in MCA205 (D), MC38 (E), and SVR (F) tumors from control mice and propranolol (PRO) treated mice by qRT-PCR (n = 5). G Representative CD34 (brown) IHC staining of MCA205 tumor tissue from control mice or propranolol treated mice; hematoxylin counterstain; scale bar 100 µm (left panels), and 50 µm (right panels). H Dot plot showing quantification of CD34 staining by percentages of DAB + area in 9 control mice and 11 propranolol treated mice. *p < 0.05, **p < 0.01, n.s. Not significant, according to multiple t test with Bonferroni correction. Mean ± SEM are depicted.

Fig. 3. Propranolol treatment increases CD4 + T cell infiltration in MCA205 tumors.

A–F Single cell suspensions were made from excised MCA205 tumors at the experimental endpoint and analyzed by flow cytometry. A Representative flow cytometry dot plot of CD4 + and CD8 + T cells in the TME. Quantification of CD4 + T cells (B), regulatory T cells (C), and CD8 + T cells (D) (n = 4). Quantification of CD137 + , PD1 + CD137 + PD1 + CD4 + T cells (E), and CD8 + T cells (F) among all live cells in the TME (n = 4). G, H Tumor growth kinetics (E) and survival rates (F) of control, and propranolol treated (PRO) nude mice (n = 13). For tumor growth and Kaplan–Meier curves, statistical analyses were performed using TumGrowth software. For other comparisons, multiple t tests with Bonferroni correction for multiple comparison were used. ***p < 0.005, n.s. Not significant. Mean ± SEM are depicted.

Fig. 4. ADRB blockade by propranolol reduces intratumoral MDSCs and upregulates PDL1 expression on tumor associated macrophages (TAMs).

Single cell suspensions were made from excised MCA205 tumors in immune competent mice and T cell deficient nude mice at the experimental endpoint and analyzed by flow cytometry (n = 3–5). A, B Representative flow cytometry dot plot (A) and quantification of MDSCs (CD11b + F4/80- GR1 + ) (B) in the TME of tumors from control and propranolol (PRO) groups in immune competent mice. C Quantification of TAMs (CD11b + F4/80+) in samples from control and propranolol treated immune competent mice. D, E, Representative histograms and quantifications of mean fluorescence intensity (MFI) of PDL1 on TAMs from immune competent mice (D), and T cell deficient nude mice (E). MFI quantification of MHC II (F) and CD206 (G) expression on TAMs in immune competent mice. Multiple t tests with Bonferroni correction for multiple comparison were used for statistical testing. *p < 0.05, **p < 0.01. Mean ± SEM are depicted.

Fig. 5. Macrophages are sensitive to ADRB stimulation and propranolol treatment leads to TAMs with distinct gene expression profiles.

A Quantification of PDL1 MFI on non-polarized macrophages upon isoprenaline (ISO) or propranolol (PRO) treatment. isoprenaline and/or propranolol’s effect on the surface expression of MHC II (B), CD86 (C), or CD206 (D) under M1 polarizing condition with IFNγ and LPS or under M2 polarizing condition with IL4. E, F, RNA-seq analysis showing heatmap (E) and volcano plot (F) of genes differentially regulated in TAMs from control or propranolol treated mice. Data show two biological replicates. G Gene ontology analysis showing the biological processes most significantly enriched within genes that are differentially expressed between TAMs isolated from control mice or propranolol treated mice. *p < 0.05, **p < 0.01, ***p < 0.005, according to multiple t test with Bonferroni correction for multiple comparison. Mean ± SD are depicted.

Fig. 6. ADRB blockade by propranolol (PRO) improves the efficacy of anti-CTLA4 in MCA205 and MC38 tumor model.

A Treatment regimen and experimental setup of immune checkpoint inhibitor (ICI) tumor studies. B, C Tumor growth kinetics (B) and Kaplan–Meier survival curves (C) of C57BL/6 mice inoculated with MCA205 cancer cells, treated with anti-CTLA4 and PRO. D, E Tumor growth kinetics (D) and Kaplan–Meier survival curves (E) of C57BL/6 mice inoculated with MC38 cancer cells, treated with anti-CTLA4 and PRO. Each line represents one animal. n = 10–15 per group. Pie charts indicate the response rate of treated mice. Therapeutic response is defined as having stable tumor volume for more than 7 days (four measurements). Shaded areas, included for easy comparison of the different treatment groups, represent the terminal tumor growth time frame of the PBS + water group. Statistical analyses were performed using TumGrowth software. *p < 0.05, **p < 0.01, ***p < 0.005.

Fig. 7. Propranolol combined with anti-CTLA4 increases the number of intratumoral CD8 + T cells, reduces tumor angiogenesis, and provides long lasting immune memory against MCA205 cancer cells.

A–C Representative CD8 (A) and CD34 (C) IHC staining of MCA205 tumor tissues; hematoxylin counterstain; scale bar 100 µm (upper panels), and 50 µm (lower panels). B, D Dot plot showing quantification of CD8 (B), and CD34 staining (D). Splenocytes from cured mice were re-stimulated ex vivo with MCA205 cancer cells for 24 h, and intracellular cytokine expressions were quantified by flow cytometry. E Representative flow cytometry dot plots showing the expression of TNFα and IFNγ on CD4 + T cells (upper panels) or CD8 + T cells (lower panels) with memory phenotype from naïve mice (left panels) or anti-CTLA4 + PRO treated mice that had shown complete tumor regression(right panels). F Percentages of tumor reactive T cells (IFNy + or TNFa + ) with memory phenotype (CD44hi) among CD8 + or CD4 + T cells after 24 h ex vivo re-stimulation. *p < 0.05 according to multiple t test with Bonferroni correction. Mean ± SEM are depicted.