Reprogramming tumor immune microenvironment by milbemycin oxime results in pancreatic tumor growth suppression and enhanced anti-PD-1 efficacy

Abstract

Pancreatic ductal adenocarcinoma (PDAC) has a survival rate of 12%, and multiple clinical trials testing anti-PD-1 therapies against PDAC have failed, suggesting a need for a novel therapeutic strategy. In this study, we evaluated the potential of milbemycin oxime (MBO), an antiparasitic compound, as an immunomodulatory agent in PDAC. Our results show that MBO inhibited the growth of multiple PDAC cell lines by inducing apoptosis. In vivo studies showed that the oral administration of 5 mg/kg MBO inhibited PDAC tumor growth in both subcutaneous and orthotopic models by 49% and 56%, respectively. Additionally, MBO treatment significantly increased the survival of tumor-bearing mice by 27 days as compared to the control group. Interestingly, tumors from MBO-treated mice had increased infiltration of CD8⁺ T cells. Notably, depletion of CD8⁺ T cells significantly reduced the anti-tumor efficacy of MBO in mice. Furthermore, MBO significantly augmented the efficacy of anti-PD-1 therapy, and the combination treatment resulted in a greater proportion of active cytotoxic T cells within the tumor microenvironment. MBO was safe and well tolerated in all our preclinical toxicological studies. Overall, our study provides a new direction for the use of MBO against PDAC and highlights the potential of repurposing MBO for enhancing anti-PD-1 immunotherapy.

Keywords: CD8+ T-cells; ICD; ICI; STING; anti-PD-1; anti-helminthic drug; chemokines; drug repurposing; immune checkpoint inhibitors; immunogenic cell death; immunomodulation; pancreatic tumor; tumor suppression.

Graphical abstract

Figure 1

MBO suppresses the growth of PDAC cell lines and suppresses their colony-formation ability (A–H) Growth-suppressive effects of MBO in (A) MiaPaCa-2, (B) AsPC-1, (C) BXPC3, (D) Panc-1, (E) SUIT-2, (F) PO2-Luc cells, (G) MiaPaCa-2 GR, and (H) Capan-2. The cells were treated with varying concentrations of MBO for 24, 48, and 72 h. Cell viability was analyzed using the SRB assay and the data were plotted using GraphPad Prism 8. (I–L) Effects of MBO on colony-formation ability of (I) AsPC1, (J) SUIT-2, (K) MiaPaCa-2, and (L) PO2. The cells were treated with various concentrations of MBO, and the colonies were stained using SRB dye. Each experiment was repeated three times with at least three replicates in each experiment. Statistically significant when compared with control. ∗p < 0.05, ∗∗p < 0.01.

Figure 2

MBO induces apoptotic cell death in PDAC cells (A) MiaPaCa-2, (B) AsPC1, (C) BXPC3, and (D) SUIT-2 cells were treated with various concentrations of MBO for 48 h, and the cells were processed for annexin-V/APC apoptosis assay. Results were analyzed using FlowJo software and plotted using GraphPad Prism 8. Each experiment was repeated three times with at least three replicates in each experiment. Statistically significant when compared with control. ∗p < 0.05, ∗∗p < 0.01.

Figure 3

MBO suppresses growth of PDAC tumors in subcutaneous and orthotopic models by inducing apoptosis (A) Tumor curve representing the growth of subcutaneously implanted MiaPaCa-2 tumors in control, 2.5 mg/kg MBO, 5 mg/kg MBO, and 10 mg/kg MBO. (B) Average tumor weight recorded on the final day of the experiment. (C) Average concentration of MBO in tumors analyzed by LC-MS/MS method. (D) Average luminescence curve representing the growth of orthotopically implanted PO2-Luc tumors in control versus 5 mg/kg MBO. (E) Average tumor weight recorded on the final day of the experiment. (F) Representative images of orthotopically implanted tumors from control versus MBO groups. (G) Representative tumor sections depicting apoptosis in orthotopically implanted tumors from control and MBO-treated groups. (H) Quantitative analysis of apoptosis induction in tumor sections (n = 3). Statistically significant differences between groups are reported as p values in the panels. Statistically significant when compared with control. ∗p < 0.05, ∗∗p < 0.01.

Figure 4

MBO treatment increases CD8⁺ T cell infiltration in the TME Frequencies of CD8⁺ T cells and CD4⁺ T cells within tumors (A) and spleens (B) as a percentage of viable cells. FACS analysis was performed as described in materials and methods. Immune cell frequencies were analyzed using FlowJo software and plotted using GraphPad Prism 8. Statistically significant differences between groups are reported as p values in the panels. Statistically significant when compared with control. ∗p < 0.05, ∗∗p < 0.01.

Figure 5

Tumor-suppressive effect of MBO is CD8 T cell dependent, and MBO increases the survival of tumor-bearing mice (A) Representative tumor sections depicting tumor-infiltrating CD8⁺ T cells after MBO treatment as confirmed by IHC. (B) Quantitation of CD8⁺ T infiltrates by IHC. (C) Tumor curve representing growth of subcutaneously implanted PO2-Luc tumors following treatment with control, MBO, anti-CD8α alone, and anti-CD8α + MBO groups. (D) Average tumor weight recorded on the final day of the experiment. (E) Survival proportions of control and MBO-treated groups. Statistically significant differences between groups are reported as p values in the panels. Statistically significant when compared with control. ∗p < 0.05, ∗∗p < 0.01.

Figure 6

MBO synergizes with anti-PD-1 immunotherapy (A) Tumor curve representing growth of orthotopically implanted PO2-Luc tumors following treatment with control, MBO, anti-PD-1 alone, and anti-PD-1 + MBO. (B) Average tumor weight recorded on the final day of the experiment. (C) Quantification of various CD8⁺ T cell populations. (D) Quantification of various CD4⁺ T cell populations. (E) Quantification of IFN-γ, CXCL10, and CXCL9 in tumor lysates. Statistically significant differences between groups are reported as p values in the panels. Statistically significant when compared with control. ∗p < 0.05, ∗∗p < 0.01.

Figure 7

MBO induces immunogenic cell death marker expression (A) Expression of HSP90 in AsPC1, PO2, and SUIT-2 cells. (B) Quantification of HSP90 in respective cell lines. (C) Expression of CRT in AsPC1, PO2, and SUIT-2 cells. (D) Quantification of CRT in respective cell lines. (E) Expression of MHC-1 in PO2 and SUIT-2 cells. Statistically significant differences between groups are reported as p values in the figures. Each experiment was repeated three times with at least three replicates in each experiment. Statistically significant when compared with control. ∗p < 0.05, ∗∗p < 0.01.

Figure 8

MBO downregulates PI3K/Akt signaling while activating STING signaling in PDAC cells (A) Heatmaps depicting the up- and downregulated proteins in MiaPaCa-2 cells following treatment with MBO. The red color signifies upregulated proteins, while the green color denotes downregulation. The values generated from RPPA were analyzed further by IPA. PI3K/Akt was the most downregulated pathway, while STING signaling was the most upregulated signaling pathway. (B) Graphical summary depicting the mechanism of action of MBO in PDAC. Statistically significant when compared with control. ∗p < 0.05, ∗∗p < 0.01.