Nanoparticle delivery of innate immune agonists combined with senescence-inducing agents promotes T cell control of pancreatic cancer

Abstract

Pancreatic ductal adenocarcinoma (PDAC) has quickly risen to become the third leading cause of cancer-related death in the United States. This is in part because of its fibrotic tumor microenvironment (TME) that contributes to poor vascularization and immune infiltration and subsequent chemo- and immunotherapy failure. Here, we investigated an immunotherapy approach combining delivery of stimulator of interferon genes (STING) and Toll-like receptor 4 (TLR4) innate immune agonists by lipid-based nanoparticle (NP) coencapsulation with senescence-inducing RAS-targeted therapies, which can remodel the immune suppressive PDAC TME through the senescence-associated secretory phenotype. Treatment of transplanted and autochthonous PDAC mouse models with these regimens led to enhanced uptake of NPs by multiple cell types in the PDAC TME, induction of type I interferon and other proinflammatory signaling pathways, increased antigen presentation by tumor cells and antigen-presenting cells, and subsequent activation of both innate and adaptive immune responses. This two-pronged approach produced potent T cell-driven and type I interferon-mediated tumor regression and long-term survival in preclinical PDAC models dependent on both tumor and host STING activation. STING and TLR4-mediated type I interferon signaling was also associated with enhanced natural killer and CD8⁺ T cell immunity in human PDAC samples. Thus, combining localized immune agonist delivery with systemic tumor-targeted therapy can orchestrate a coordinated type I interferon-driven innate and adaptive immune response with durable antitumor efficacy against PDAC.

Fig. 1.. Systemic administration of NPs can deliver cargo to tumor and immune cells in PDAC TME with minimal toxicity.

(A) Schematic representation of immuno-NP design. (B) NP hydrodynamic size as assessed by dynamic light scattering (DLS). (C) Measurement of NP surface charge as assessed by zeta potential. (D) KPC1 PDAC tumor cells expressing luciferase-GFP were injected orthotopically into the pancreas of 8–12 week old C57BL/6 female mice. Following tumor formation, mice received a single dose of fluorescently labeled immuno-NPs by intravenous (i.v.) injection. Flow cytometry analysis of DiI⁺ NP uptake in indicated cell types 48hr later is shown (n = 7 mice per group). Tumor cells were defined as GFP⁺, and stromal cells as CD45^-GFP^-. (E) PDAC-bearing KPC GEMM mice were i.v. injected with a single dose of fluorescently labeled immuno-NPs. Flow cytometry analysis of DiD-labeled NP uptake in different cell types 48hr later is shown (n = 3 mice per group). (F) Representative immunofluorescence (IF) staining of KPC1 orthotopic transplant PDAC tumors for DiI-labeled immuno-NPs. Scale bars, 100 μm. (G) Representative IF staining of KPC GEMM PDAC tumors for DiD-labeled immuno-NPs. Scale bars, 100 μm. (H) Plasma ALT and AST concentrations in naïve wild-type (WT) C57BL/6 mice treated with empty- or immuno-NPs weekly and vehicle (V) or trametinib (1 mg/kg) and palbociclib (100 mg/kg) (T/P) 4 times per week for 21 days (n = 3 to 5 mice per group). Dotted lines indicate established range for normal AST and ALT concentrations. (I) Representative hematoxylin and eosin (H&E) staining of livers from WT C57BL/6 mice treated as in (H). Scale bars, 100 μm. (J) Change in weight of naïve WT C57BL/6 mice treated as indicated by arrows (n = 3 to 5 mice per group). Sac, sacrifice. Error bars, mean ± SEM. P values were calculated using Kruskal-Wallis with Dunn’s multiple comparisons tests. ns, not significant.

Fig. 2. T/P pre-treatment enhances immuno-NP uptake, IFN and cytokine production, and antigen presentation in tumor cells and APCs.

(A) Schematic of KPC orthotopic transplant model and 2-week treatment schedule. (B) Flow cytometry analysis of DiI-labeled NP uptake in indicated cellular compartments in KPC1 transplant PDAC tumors from mice treated with vehicle (V) or trametinib (1 mg/kg) and palbociclib (100 mg/kg) (T/P) for 2 weeks and empty- or immuno-NPs for 48 hr (n = 4 to 8 mice per group). Tumor cells were defined as GFP⁺, and stromal cells as CD45^-GFP^-. (C) RT-qPCR analysis of IFN pathway and SASP gene expression in KPC1 transplant PDAC tumors from mice treated as in (B) (n = 3 mice per group). A.U., arbitrary units. (D) Representative IF staining of KPC GEMM PDAC tumors from mice treated as in (B) for IFN-β in DCs (CD11c⁺), macrophages (F4/80⁺), and tumor cells (CK19⁺). Quantification of mean fluorescent intensity (MFI) of total IFN-β expression in tissues is shown on the right (n = 5 to 11 technical replicates per group). Scale bars, 100 μm. (E) Schematic of KPC1 cell line treatment schedule. (F) RT-qPCR analysis of IFN pathway and SASP gene expression in KPC1 PDAC cells treated with vehicle (V) or trametinib (25 nM) and palbociclib (500 nM) (T/P) for 1 week and empty- or immuno-NPs for 48 hr (n = 3 biological replicates per group). (G) Representative histograms (left) and quantification of MHC-I (H-2k^b) MFI (right) on KPC1 PDAC cells treated as in (F) (n = 6 biological replicates per group). (H) RT-qPCR analysis of antigen presentation/processing gene expression in KPC1 transplant PDAC tumors from mice treated as in (B) (n = 3 mice per group). Error bars, mean ± SEM. P values were calculated using one-way ANOVA with Tukey’s post-tests (B,C,D,F,H) or Kruskal-Wallis with Dunn’s multiple comparisons tests (B,C,F,G,H) depending on the normality of the distribution. **** P <0.0001, *** P <0.001, ** P <0.01, * P <0.05. ns, not significant.

Fig. 3.. Combinatorial immuno-NP and T/P treatment activates NK and CD8⁺ T cell immunity in PDAC.

(A to C) Flow cytometry analysis of total CD45⁺ immune cells (A), T cell frequencies and activation markers (B), and NK cell frequencies and activation markers (C) in KPC1 orthotopic transplant PDAC tumors from mice treated with vehicle (V) or trametinib (1 mg/kg) and palbociclib (100 mg/kg) (T/P) for 2 weeks and empty- or immuno-NPs for 48 hr (n = 6 to 8 mice per group). (D) Representative immunohistochemical (IHC) staining of KPC1 orthotopic transplant PDAC tumors from mice treated as in (A). Quantification of the number of Granzyme B (GZMB)⁺ cells per field is shown in the inset (n = 3 to 6 mice per group). Scale bar, 50μm. (E) Representative IF staining of PDAC tumors from KPC GEMM mice treated as in (A) (left). Quantification of NK1.1⁺ NK cell and CD8⁺ T cell MFI is shown on right (n = 2 to 6 technical replicates per group ). Scale bars, 100 μm. (F) Representative IF staining for TNF-α in CD11c⁺ DCs (left) and F4/80⁺ macrophages (right) in PDAC tumors from KPC1 transplant mice treated as in (A). Scale bars, 100 μm. (G) Quantification of combined TNF-α MFI in macrophages and DCs from IF staining in (F) (n = 4 to 5 technical replicates per group). (H) Flow cytometry analysis of DCs in KPC1 transplant PDAC tumors from mice treated as in (A) (n = 6 to 8 mice per group). Error bars, mean ± SEM. P values were calculated using one-way ANOVA with Tukey’s post-tests (A,B,C,E,G) or Kruskal-Wallis with Dunn’s multiple comparisons tests (B,C,H) depending on the normality of the distribution. **** P <0.0001, *** P <0.001, ** P <0.01, * P <0.05. ns, not significant.

Fig. 4.. Immuno-NP and T/P regimens produce anti-tumor efficacy and long-term survival in preclinical PDAC models.

(A) Waterfall plot of the response of KPC1 transplant PDAC tumors after treatment with vehicle (V) or trametinib (1 mg/kg) and palbociclib (100 mg/kg) (T/P) 4 times per week and empty- or immuno-NPs weekly for 2 weeks (n = 12 to 13 mice per group). (B) Representative H&E staining of KPC1 transplant PDAC tumors from mice treated with vehicle (V) or trametinib (1 mg/kg) and palbociclib (100 mg/kg) (T/P) for 2 weeks and empty- or immuno-NPs for 48 hr. Quantification of percent of tumor area covered in necrosis is shown in the inset (n = 5 to 6 mice per group). Scale bar, 500μm. (C) Kaplan-Meier survival curve of mice harboring KPC1 transplant PDAC tumors treated with vehicle (V) or trametinib (1 mg/kg) and palbociclib (100 mg/kg) (T/P) 4 times per week and empty- or immuno-NPs weekly (n = 7 to 8 mice per group). (D) Waterfall plot of the response of KPC GEMM PDAC tumors to treatment as in (A) (n = 5 to 9 mice per group). (E) Representative H&E staining of KPC GEMM PDAC tumors from mice treated as in (A). Quantification of percent of tumor area covered in necrosis is shown in the inset (n = 4 to 7 mice per group). Scale bar, 500μm. (F) Kaplan-Meier survival curve of PDAC-bearing KPC GEMM animals treated as in (A) (n = 8 to 10 mice per group). Arrows indicate when an individual mouse was taken off both treatment regimens for the duration of the study. Error bars, mean ± SEM. P values were calculated using one-way ANOVA with Tukey’s post-tests (A and D) or Kruskal-Wallis with Dunn’s multiple comparisons tests (A) depending on the normality of the distribution, or log-rank test (C and F). **** P <0.0001, *** P <0.001, ** P <0.01, * P <0.05. ns, not significant.

Fig. 5.. Tumor and host STING signaling contribute to anti-tumor immune responses to therapy.

(A) Schematic of orthotopic transplantation of sg. Scramble or sg. Sting KPC1 cells into WT or Sting^−/− mice and treatment. (B) Immunoblots of control or sg.Sting KPC1 PDAC cells. (C) Representative STING IHC staining in sg. Scramble or sg. Sting KPC1 transplant PDAC tumors from WT or Sting^−/− mice treated with trametinib (1 mg/kg) and palbociclib (100 mg/kg) (T/P) for 2 weeks and immuno-NPs for 48 hr. Scale bar, 100μm. (D) Representative H&E staining of sg. Scramble or sg. Sting KPC1 transplant tumors from WT or Sting^−/− mice treated as in (C) (left). Scale bar, 500μm. Right, quantification of percent of tumor area covered in necrosis (n = 8 mice per group). (E to I) Flow cytometry analysis of total CD45⁺ immune cell and CD8⁺ T cell (E), NK cell (F), CD4⁺ T cell (G), B cell (H), and DC (I) frequencies and activation markers in sg. Scramble or sg. Sting KPC1 transplant tumors from WT or Sting^−/− mice treated as in (C) (n = 8 to 9 mice per group). Error bars, mean ± SEM. P values were calculated using two-tailed, unpaired Student’s t-test (D,E,F,G,H,I,) or Mann-Whitney test (E,F,H) depending on the normality of the distribution. **** P <0.0001, *** P <0.001, ** P <0.01, * P <0.05. ns, not significant.

Fig. 6.. Immuno-NP and T/P therapy efficacy driven by IFNAR-dependent NK and CD8⁺ T cell immune surveillance.

(A) Kaplan-Meier survival curve of mice harboring KPC1 transplant PDAC tumors treated with trametinib (1 mg/kg) and palbociclib (100 mg/kg) (T/P) 4 times per week, immuno-NPs weekly, and blocking antibodies against NK1.1 (PK136; 250 μg), CD8 (2.43; 200 μg), or IFNAR-1 (MAR15A3; 200 μg) twice per week (n = 7 mice per group). (B) Waterfall plot of the response of KPC1 transplant PDAC tumors to 2 weeks of treatment as in (A) (n = 4 to 7 mice per group). (C) Representative H&E staining of KPC1 transplant PDAC tumors from mice treated with trametinib (1 mg/kg) and palbociclib (100 mg/kg) (T/P) for 2 weeks and with immuno-NPs for 48 hr in the presence or absence of IFNAR-1 (MAR15A3; 200 μg) blocking antibodies administered biweekly for 2 weeks. Scale bar, 500μm. Quantification of percent of tumor area covered in necrosis is shown in the inset (n = 3 to 4 mice per group). (D to H) Flow cytometry analysis of CD8⁺ T cell (D), NK cell (E), CD4⁺ T cell (F), B cell (G), and DC (H) frequencies and activation markers in KPC1 transplant PDAC tumors from mice treated as in (C) (n = 16 to 17 mice per group). Error bars, mean ± SEM. P values were calculated using log-rank test (A), Kruskal-Wallis with Dunn’s multiple comparisons tests (B), two-tailed, unpaired Student’s t-test (C,D,E,F,H), or Man-Whitney test (D,E,G) depending on the normality of the distribution. **** P <0.0001, *** P <0.001, ** P <0.01, * P <0.05. ns, not significant.

Fig. 7.. STING, TLR4, and type I interferon pathway gene expression correlate with NK and T cell signatures in human PDAC.

(A and B) Pearson’s correlation analysis plots are shown comparing NK and T cell signatures with expression of STING (TMEM174), TLR4, and downstream interferon signaling pathway genes in human PDAC transcriptomic data from Bailey et al. (37) (A) and Moffitt et al. (38) (B) (n = 91 to 145 biological replicates). Shading represents confidence interval. Pearson’s correlation coefficient (R) values are displayed. P values were calculated using a two-tailed, unpaired Student’s t-test.