Inhibition of PCSK9 enhances the antitumor effect of PD-1 inhibitor in colorectal cancer by promoting the infiltration of CD8+ T cells and the exclusion of Treg cells

Abstract

Immunotherapy especially immune checkpoint inhibitors (ICIs) has brought favorable clinical results for numerous cancer patients. However, the efficacy of ICIs in colorectal cancer (CRC) is still unsatisfactory due to the poor median progression-free survival and overall survival. Here, based on the CRC models, we tried to elucidate novel relapse mechanisms during anti-PD-1 therapy. We found that PD-1 blockade elicited a mild antitumor effect in these tumor models with both increased CD8⁺ T cells and Treg cells. Gene mapping analysis indicated that proprotein convertase subtilisin/kexin type 9 (PCSK9), low-density lipoprotein receptor, transforming growth factor-β (TGF-β), and CD36 were unexpectedly upregulated during PD-1 blockade. To investigate the critical role of these proteins especially PCSK9 in tumor growth, anti-PCSK9 antibody in combination with anti-PD-1 antibody was employed to block PCSK9 and PD-1 simultaneously in CRC. Data showed that neutralizing PCSK9 during anti-PD-1 therapy elicited a synergetic antitumor effect with increased CD8⁺ T-cell infiltration and inflammatory cytokine releases. Moreover, the proportion of Treg cells was significantly reduced by co-inhibiting PCSK9 and PD-1. Overall, inhibiting PCSK9 can further enhance the antitumor effect of anti-PD-1 therapy in CRC, indicating that targeting PCSK9 could be a promising approach to potentiate ICI efficacy.

Keywords: CD8+ T cells; PCSK9; PD-1; regulatory T cells; tumor microenvironment.

Figure 1

In vivo antitumor effect of anti-PD-1 antibody. (A) Tumor volume and tumor weight in the MC38 tumor model. Anti-PD-1 mAb administration was 5 mg/kg, n = 4 mice/group. (B) Tumor volume and tumor weight of CT26 tumor model. Anti-PD-1 mAb administration was 5 mg/kg, n = 3 mice/group. (C, D) Flow cytometry analysis of tumor-infiltrating T cells for mice treated with PBS or anti-PD-1 antibody in MC38 tumor model (C) and in CT26 tumor model (D). (E, F) IHC staining of CD8a and Foxp3 in CT26 tumors. "*" means p-value < 0.05 and "**" means p-value < 0.01.

Figure 2

Quantitative real-time PCR (qRT-PCR) analysis of IFN-γ, granzyme B, and TNF-α gene expression in MC38 tumors (A) and CT26 tumors (B). RT-qPCR analysis of PCSK9, LDLR, TGF-β, and CD36 gene expression in tumor of the MC38 tumor model (C) and the CT26 tumor model (D). "*" means p-value < 0.05 and "**" means p-value < 0.01 while "ns" means not statistically significant.

Figure 3

Tumor volume and tumor weight in MC38 tumor model (A) and CT26 tumor model (B) treated with anti-PD-1 or anti-PCSK9 antibody. Analysis of IFN-γ, granzyme B and PCSK9 protein level in tumor of MC38 tumor model (C) and CT26 tumor model (E) by ELISA. Quantitative analysis of LDLR expression on cell membrane in MC38 tumors (D) and CT26 tumors (F) by immunofluorescence. Quantitative RT-PCR analysis of CD36 and TGF-β gene expression in MC38 tumors (G) and in CT26 tumors (H). "*" means p-value < 0.05 and "**" means p-value < 0.01 while "ns" means not statistically significant.

Figure 4

(A) IHC staining of CD8a in tumor of the MC38 tumor model treated with anti-PD-1 or anti-PCSK9 antibody and quantitative analysis of positive particles and (B) for the CT26 tumor model. (C) Flow cytometry analysis of CD45⁺, CD3⁺, and CD8⁺ T-cell infiltration in MC38 tumors. "*" means p-value < 0.05 and "**" means p-value < 0.01.

Figure 5

(A) IHC staining of Foxp3 in MC38 tumors treated with anti-PD-1 or anti-PCSK9 antibody and quantitative analysis of positive particles and (B) for CT26 tumors. (C) Flow cytometry analysis of CD4⁺ and Foxp3⁺ T-cell infiltration in MC38 tumors. (D, E) Tumor volume in the mice under the treatment of anti-PD-1 and anti-PCSK9 antibody with/without anti-CD8, anti-CD4, anti-NK1.1, and anti-CSF1R antibody. "*" means p-value < 0.05 and "**" means p-value < 0.01 while "ns" means not statistically significant.