Engineering ROS‐Responsive Bioscaffolds for Disrupting Myeloid Cell‐Driven Immunosuppressive Niche to Enhance PD‐L1 Blockade‐Based Postablative Immunotherapy

Abstract The existence of inadequate ablation remains an important cause of treatment failure for loco‐regional ablation therapies. Here, using a preclinical model, it is reported that inadequate microwave ablation (iMWA) induces immunosuppressive niche predominated by myeloid cells. The gene signature of ablated tumor presented by transcriptome analyses is highly correlated with immune checkpoint blocking (ICB) resistance. Thus, an in situ scaffold with synergistic delivery of IPI549 and anti‐programmed death‐ligand 1 blocking antibody (aPDL1) for postablative cancer immunotherapy is designed and engineered, in which IPI549 capable of targeting myeloid cells could disrupt the immunosuppressive niche and subsequently improve ICB‐mediated antitumor immune response. Based on five mouse cancer models, it is demonstrated that this biomaterial system (aPDL1&IPI549@Gel) could mimic a “hot” tumor‐immunity niche to inhibit tumor progression and metastasis, and protect cured mice against tumor rechallenge. This work enables a new standard‐of‐care paradigm for the immunotherapy of myeloid cells‐mediated “cold” tumors after loco‐regional inadequate practices.


A: Experimental section
Study design. Percutaneous thermal ablation (PTA) has been well established as one of the most widely utilized therapeutic interventions for at least 12 distinct cancer types and showed significant magnitude of benefit supported by several international clinical guidelines.
However, inadequate ablation still remains a therapeutic dilemma for any loco-regional treatment. Accordingly, the objective of this study was to investigate the potential principles responsible for such pro-oncogenic effects derived from inadequate ablation and further provide a precise combined immunotherapy strategy to maximize the clinical effects of PTA treatment. We first utilized a preclinical colon adenocarcinoma murine model to confirm that inadequate ablation indeed stimulated outgrowth of residual tumor and we discovered the mechanism that tumors were exposed to myeloid cell-mediated tumor immune microenvironment (TIME) after inadequate microwave ablation (iMWA). Then, a myeloid cell-targeted combined strategy was designed based on a ROS-responsive scaffold and its anticancer efficacy was evaluated through multiple types of mouse models. Mice from different treatment groups were measured and imaged to detect tumor progression and metastasis, and were also rechallenged with tumor cells to evaluate immune memory effects.
Animals were euthanized by carbon dioxide for signs of cachexia or when the tumor volume reached 1500 mm 3 .
Synthesis of PVA-TSPBA hydrogels. PVA (72 kDa; 98% hydrolyzed; 1 g) and deionized water (20 mL) were mixed and stirred together in a water bath with magnetic stirring. The temperature was raised slowly up to 95°C in order to achieve a clear solution. TSPBA (5 weight % (wt %) in H2O, 2 mL)) and PVA (5 wt % in H2O, 2 mL) were mixed to fabricate a tough hydrogel. The gels were divided into several copies for in vitro experiments. For the fabrication of aPDL1-and IPI549-loaded gel, aPDL1 (50 µg per sample) and IPI549 (25 µg dissolved at 5% 1-methyl-2-pyrrolidinone in polyethylene glycol 400 per sample) were dissolved in PVA aqueous solution. For in vivo experiments, PVA (with or without drugs) and TSPBA solutions were first kept at room temperature for 30 minutes, then double tube syringe with 0.45 mm-diameter needle was used to form gels by injecting PVA and TSPBA solutions in a volume ratio of 1:1.
Characterization of PVA-TSPBA hydrogels. After hydrogels were frozen at -80°C overnight, the microstructure of gels was performed by cryo-scanning electron microscopy (Cryo-SEM) (SU8010, Japan). The dynamic rheological behavior of PVA aqueous solution before and after gelation was measured at 25°C using a Thermo HAAKE MARS 60 stress-controlled rheometer with 20 mm parallel plates.
In vitro and in vivo hydrogels degradation. For in vitro experiment, gels were put in PBS with or without H2O2 (1 mM, 0.5 mM, 0.25 mM) and pictured on day 0, 3 and 7 to observe their morphology changes. For in vivo experiment, gels (200 µL per sample) were injected subcutaneously into BALB/c mice. Pictures were taken and mice were sacrificed on day 0, 3, 7 and 14 for hematoxylin and eosin (H&E) staining of the surrounding skin in order to record degradation behavior and tissue biocompatibility.
In vitro and in vivo release of aPDL1 and IPI549 from hydrogels. The in vitro drug release studies from ROS-responsive hydrogels were performed in PBS with or without H2O2 (1 mM) at room temperature. The released IPI549 and Cy5.5-aPDL1 were analyzed using ultraviolet spectrum and SpectraMax iD5, respectively. To evaluate the in vivo release of aPDL1, free Cy5.5-aPDL1 or Cy5.5-aPDL1@Gel was injected subcutaneously into CT26 tumor-bearing BALB/c mice. On day 0, 3 and 7, fluorescence imaging of Cy5.5-aPDL1 release was monitored by the IVIS imaging system. Likewise, for the assessment of IPI549 release, representative fluorescence imaging of a drug-loaded gel in which indocyanine green was used as a fluorescent surrogate for IPI549 that was taken via the IVIS imaging system. Bioluminescence signals from cancer cells were taken via the IVIS imaging system to assess the residual tumor progression. The size of the tumors was measured every second day with a caliper, and the volume was calculated using the formula, (L × W 2 )/2, where L is the longest diameter of tumor and W is the perpendicular diameter. Mice were weighed and arranged to evaluate survival curves. Mice that no visible tumor could be measured on continuous measurement days were considered as complete regressions. For the rechallenged study, on day 50 since primary tumor implantation, cured mice after aPDL1&IPI549@Gel treatment were rechallenged with 5×10 5 fLuc-CT26 cells tumor cells on their opposite flanks.
For the distant tumor model, one day after 1×10 6 fLuc-CT26 cells suspended in PBS were inoculated into the right flank of mice, a second tumor as the mimic distant tumor (1×10 6 fLuc-CT26 cells) was subcutaneously inoculated into the left flank of each mouse. Ten days later, tumors in the right flank were received iMWA treatment and then tumor-bearing mice were divided randomly into two groups, and Gel or aPDL1&IPI549@Gel was injected peritumorally in the right site, respectively, while no treatment was performed for the left tumor site. The subsequent monitoring of mouse bilateral tumors and survival duration was the same as aforementioned procedures.
To establish lung metastases, 1 × 10 6 of CT26 tumor cells were injected intradermally into the BALB/c mouse in the right flank, and sex-and age-matched healthy mice were chosen as controls. Nine days later, all the mice were inoculated intravenously with fLuc-CT26 cells (1 × 10 5 ) via tail vein infusion. On the following day, when the longest diameter of tumor reached about 0.8 cm, the primary tumor of each mouse was treated with iMWA. The followed immunotherapy strategy of treatment group was the same as above. The bioluminescence imaging was carried out using the IVIS imaging system with 60 seconds exposure time to record the state of lung metastases. At the end of this experiment, lungs were collected and fixed in Bouin's solution for 24 h. Pictures of lung tissues were taken with a digital camera, and then the lungs were studied by pathological analysis.
In vivo bioluminescence and imaging. The progression of tumor was observed using the IVIS Smart Imaging System (Vieworks Co. Ltd. Korean). Ten minutes after intraperitoneal injection of d-luciferin dissolved in PBS (15 mg mL -1 ) at a dose of 10 µL g -1 , mice were anesthetized and photographed by the imaging system for 60 seconds of exposure time.
Bioluminescence images were then analyzed using IVIS Living Image software. Regions of interest were quantified as average radiance (photons s −1 cm −2 sr −1 ).

RNA-sequencing.
Tumors were isolated from the mice 3 days after treated with iMWA and RNA was extracted using HiPure Unviersal RNA Mini Kit (Magen Biotechnology Co. Ltd, China). RNA sequencing was performed using RNA Nano 6000 Assay Kit of the Agilent Bioanalyzer 2100 system (Agilent Technologies, CA, USA). log2(Count) was calculated for each gene, and the data were mean centred for display in the heatmaps. Western blotting. Equal amounts of protein were mixed with an equal volume of 2 × Laemmli buffer and heated at 95°C for 5 min. Afterward, four samples were loaded into the well of 8% SDS-PAGE gel and ran at 90 V for 60 minutes. After protein transformation, anti-p110γ antibody at a 1:1,000 dilution (Cell Signaling Technology, Catalog No. 4252) was used as primary antibody and incubated overnight at 4°C. The secondary antibody was used for these blots at room temperature for 1 hour. Figure S1. Representative in vivo bioluminescence images of mice post inadequate microwave ablation (iMWA).              Figure S15. Serum biochemistry data. Female BALB/c mice were sacrificed at day 10 and day 20 after local injection of aPDL1&IPI549@Gel. Healthy mice were identified as controls. Serum biochemistry data including blood urea nitrogen (BUN), alanine aminotransferase (ALT), alkaline phosphatase (ALP), creatinine (Crea) and aspartate aminotransferase (AST) were measured. Data are presented as means ± SD (n = 3). aPDL1, anti-programmed deathligand 1 blocking antibody.      after various treatment. Data are presented as means ± SD (n = 7). Statistical significance was calculated by one-way ANOVA with a Tukey post-hoc test. ***p < 0.001.