Triple‐Combination Immunogenic Nanovesicles Reshape the Tumor Microenvironment to Potentiate Chemo‐Immunotherapy in Preclinical Cancer Models

Abstract Immune checkpoint blockade (ICB) therapies have had a tremendous impact on cancer therapy. However, most patients harbor a poorly immunogenic tumor microenvironment (TME), presenting overwhelming de novo refractoriness to ICB inhibitors. To address these challenges, combinatorial regimens that employ chemotherapies and immunostimulatory agents are urgently needed. Here, a combination chemoimmunotherapeutic nanosystem consisting of a polymeric monoconjugated gemcitabine (GEM) prodrug nanoparticle decorated with an anti‐programmed cell death‐ligand 1 (PD‐L1) antibody (αPD‐L1) on the surface and a stimulator of interferon genes (STING) agonist encapsulated inside is developed. Treatment with GEM nanoparticles upregulates PD‐L1 expression in ICB‐refractory tumors, resulting in augmented intratumor drug delivery in vivo and synergistic antitumor efficacy via activation of intratumor CD8+ T cell responses. Integration of a STING agonist into the αPD‐L1‐decorated GEM nanoparticles further improves response rates by transforming low‐immunogenic tumors into inflamed tumors. Systemically administered triple‐combination nanovesicles induce robust antitumor immunity, resulting in durable regression of established large tumors and a reduction in the metastatic burden, coincident with immunological memory against tumor rechallenge in multiple murine tumor models. These findings provide a design rationale for synchronizing STING agonists, PD‐L1 antibodies, and chemotherapeutic prodrugs to generate a chemoimmunotherapeutic effect in treating ICB‐nonresponsive tumors.


Characterization of nanoparticle size and zeta potential
The hydrodynamic diameter (D H ), polydispersity index (PDI), and ζ potential of the drugloaded micelles were analyzed by DLS using a Malvern Nano-ZS 90 laser particle size analyzer (Malvern Instruments, Malvern, UK) at 25°C.

Transmission electron microscopy analysis
The morphological characteristics of the nanoparticles were observed by TEM (Philips, 120 kV, Netherlands). GEM NPs, αPD-L1/GEM NPs and GPS were prepared at 1 mg/mL GEM-equivalent concentration. A droplet of each solution was dripped on a 400 mesh copper grid. After 5 min of deposition, the excess water on the surface was absorbed by a filter paper and dried in the air. Aqueous solution containing 2 wt % uranyl acetate was used for positive staining.

Quantitative real time PCR (qRT-PCR) assay
Total RNA was extracted using Trizol reagent (Takara, Dalian, China). Then cDNA was obtained by reverse transcription using a cDNA transcription kit (Takara, Dalian, China) according to manufacturer's instructions. Quantitative real time PCR (qRT-PCR) was performed using SYBR Green mix from Takara on Light Cycler 96 (Roche, USA). β-actin was selected as internal control. Primer sequences were listed as follows: For qPCR analysis of BMDCs or BMDMs, the cells were first seed at a concentration of 10 4 cells/ well in 6-well plates. After 24 hours of adherence, the cells were treated with GPS or 3 Free for 4 hours. Gene expression of Gadph was defined as internal reference; all treatments were compared to untreated BMDCs or BMDMs (control).
For qPCR analysis of 4T1 cells, the tumor cells were first seed at a concentration of 10 4 cells/ well in 6-well plates. After 24 hours of adherence, the cells were treated with GPS or 3 Free for 3 hours. Gene expression of Gadph was defined as internal reference; all treatments were compared to untreated 4T1 (control).

In vivo bioluminescence analysis
Mice bearing Panc02-Luci pancreatic cancer or 4T1-Luci breast tumor cancer were injected intraperitoneally with 0.1 ml of D-luciferin at a dose of 75mg/kg per mouse) and were then anesthetized with 1.5% isoflurane in oxygen in a ventilated anesthesia chamber and imaged 7 min after the injection with an in vivo imaging system (IVIS, PerkinElmer).

Pharmacokinetic assay
Sprague-Dawley (SD) rats were randomly divided into two groups (4 rats per group) and treated as follows: (1) GPS (15 mg/kg GEM-equivalent dose, 3 mg/kg diABZI, and 200 µg αPD-L1 per rat) or (2) free drug combination. Prior to administration, blood was collected from each rat as baseline sample. Following drug administration, blood samples were collected at predetermined time intervals (5 min, 30 min, 1 h, 2 h, 4 h, 6 h, 24 h and 48 h). All blood samples were collected in anticoagulant tubes and centrifuged for 10 min at 12,000 rpm to obtain serum.
All samples were treated with sodium hydroxide (0.1 mol L -1 ) for 2 h and neutralized with S5 hydrogen chloride. Finally, the hydrolyzed sample was extracted by acetonitrile, and the drug concentration was determined by analytical high-performance liquid chromatography (HPLC).
Drug concentrations were quantified using a standard curve. HPLC was carried out on a Hitachi Chromaster system, and a YMC-Pack ODS-A C18 column (5 μm, 250 × 4.6 mm) at a flow rate of 1.0 mL/min. The UV detection wavelengths for GEM and diABZI are 280 nm and 325 nm, respectively.

Biodistribution study
Panc02 orthotopic pancreatic model was used for investigation of drug tissue biodistribution. The mice were randomly divided into two groups (4 mice per group) and treated with a single injection as follows: (1) GPS (15 mg/kg GEM-equivalent dose, 3 mg/kg diABZIs and 20 µg αPD-L1 per mouse); (2) free drug combination at the same doses of each drug. At 24 h post-injection, the mice were sacrificed, and major organs and primary tumors were collected. The tissues were frozen in liquid nitrogen and manually pulverized with a mortar and pestle. All samples were treated with sodium hydroxide (0.1 mol L -1 ) for 2 h and neutralized with hydrogen chloride. Finally, the hydrolyzed tissue samples extracted by acetonitrile and subjected to HPLC analysis.

Isolation of tumor-infiltrating leukocytes
After 48h of the third injection, the mice were sacrificed and tumors were collected. The tumors were snipped into multiple pieces and digested in thermo shaker at 37°C for 30 min.
The digested tissues were then squashed by the plunge of syringes through cell strainers (70 μm). The cell suspension was centrifuged at 400 g for 5 min. The supernatant was discarded and the cells were resuspended in 4 ml of RPMI. 3 ml of 100% Ficoll-hypaque, 3ml of 75% Ficoll-Hypaque and 4ml of cell suspension was layered in sequence for density gradient centrifugation (300 g for 30 min). Tumor-infiltrating leukocytes were harvested between the two layers of Ficoll-hypaque for flow cytometry analysis.
To analyze the DC maturation, TILs were stained with Alexa Fluor 700 anti-mouse CD45  Figure S10C.

Quantification of IFN-γ in pancreatic tumor tissues
An orthotopic pancreatic tumor model was established according to the protocol described in the experimental section (orthotopic pancreatic tumor model). After 2 weeks of inoculation, the tumor-bearing mice were randomly divided into the following six groups (5 mice per group): (1) saline (control group); (2) αPD-L1 (20 µg per mouse); (3) free GEM (5 mg/kg); (4) GEM NPs (5 mg/kg GEM equivalent); (5) GEM NPs (5 mg/kg GEM equivalent) combined with αPD-L1 (20 µg per mouse) and (6) αPD-L1/GEM NPs (5 mg/kg GEM equivalent and 20 µg αPD-L1 per mouse). The above treatments were intravenously injected every other day for a total of three times. Two days after the last injection, the mice were sacrificed and the tumor tissues were collected and homogenized. The concentration of IFN-γ in tumor tissues was determined by Mouse IFN-γ ValukineTM ELISA Kit (Novus, Bio-Techne China, catalog no. VAL607).

Flow cytometry analysis for TAA-specific T cell
After stimulated with SIINFKEL peptides as described in the experimental section (TAA-     CD8a + T cell activation was analyzed by flow cytometry.