Traditional Chinese medicine (TCM) enhances the therapeutic efficiency of a gemcitabine-loaded injectable hydrogel on postoperative breast cancer through modulating the microenvironment

Abstract

Local injection of the drug-loaded hydrogel at the surgery site is promising for postoperative breast cancer. However, the postoperative changes in the tumor microenvironment, such as inflammation, abnormal angiogenesis and hypoxia, inhibit drug perfusion and contribute to breast cancer recurrence (BCR). Normalizing the abnormal blood vessels can effectively improve perfusion and reduce hypoxia. Here, we encapsulated gemcitabine (GEM) in a PLGA–PEG–PLGA hydrogel (GEM-hydrogel) for local treatment of postoperative breast cancer. The GEM-hydrogel can be injected into the surgery cavity allowing sustained release of the drug. Meanwhile, traditional Chinese medicine (TCM) Shexiang Baoxin Pill (SBP) was given to normalize the blood vessels to enhance drug perfusion. The results suggest that the combination of SBP enhances the therapeutic efficiency of the GEM-hydrogel, inhibiting tumor recurrence. Mechanism studies reveal that SBP works by promoting PDGFB expression in macrophages, subsequently recruiting pericytes, and normalizing blood vessels, finally alleviating hypoxia. This study demonstrates that the combination of TCM and chemotherapeutics is promising for suppressing postoperative tumor recurrence.

---

Introduction

Breast cancer (BC) is the leading cause of cancer-related deaths among women worldwide. According to the Global Cancer Statistics 2020, more than 2.26 million new cases of BC were diagnosed, making it the most common cancer, surpassing lung cancer.¹ Surgery is the main treatment for BC, but residual tumor cells at the surgical margins can still lead to recurrence, impacting patient survival.² The treatment modalities for breast cancer primarily include surgical resection, chemotherapy, radiotherapy, and targeted therapy, among others.³ Chemotherapy is often used post-surgery to prevent recurrence; however, chemotherapy after surgery is always delayed which will miss the optimal window to eliminate residual tumor cells,⁴ and systemic chemotherapy can cause serious side effects.⁵ Therefore, there is a pressing need for novel postoperative treatments that are timely, effective, and minimally toxic to prevent breast cancer recurrence⁶ (BCR).

Gemcitabine (GEM) is a cell cycle-specific anti-metabolite drug, used in breast cancer, pancreatic cancer and other cancers.^7,8 However, when administrated postoperatively via intravenous infusion, its systemic circulation results in low drug concentrations at the tumor site, limiting its ability to effectively eliminate residual cancer cells, while also causing adverse side effects.⁹ Therefore, it is desirable to improve the deficiency of systemic chemotherapy by local administration, which can be directly injected into the surgical cavity, inhibiting postoperative recurrence of breast cancer by local sustained release, and reducing the inconvenience of drug administration and adverse reactions.^10,11

PLGA–PEG–PLGA is a thermosensitive material used for hydrogel preparation and drug delivery, and can remain in the liquid state at room temperature, forming a gel network at body temperature, allowing for the sustained and stable release of drugs.¹² The temperature range for hydrogel formation can be controlled by adjusting the polymer concentration, the molecular weight of PLGA, and the LA/GA ratio.¹³ Its degradation products, lactic acid and glycolic acid, are metabolized and absorbed by the body, ensuring excellent biocompatibility and biodegradability.^14,15 Studies have shown that the hydrogel loaded with gemcitabine and rapamycin can be used for the treatment of metastatic pancreatic cancer by intraperitoneal injection.¹⁶ Clinical trials have shown that intratumoral injection of this hydrogel loaded with paclitaxel is effective against esophageal cancer, reducing systemic adverse effects of paclitaxel.¹⁷ The thermosensitive properties of this hydrogel facilitate its injection as a GEM-loaded solution into the surgical cavity, where it undergoes a sol-to-gel transition upon reaching the body temperature. This enables controlled and sustained drug release, highlighting its potential as an effective strategy for postoperative BC therapy.^16,18

In addition to residual tumor cells, the postoperative tumor microenvironment (TME) also plays a critical role in tumor recurrence.¹⁹ Residual tumor cells and local inflammation stimulate the formation of surrounding blood vessels, which often exhibit abnormalities such as poor pericyte coverage, inadequate perfusion, and structural immaturity.²⁰ These abnormal tumor vasculature leads to an unfavorable TME, characterized by hypoxia and elevated interstitial fluid pressure, which hinders the delivery of chemotherapy drugs and contributes to tumor recurrence.²¹ Increasing the extent of the pericyte coverage of blood vessels in tumors is one of the approaches to achieve vascular normalization, which can improve the tumor microenvironment and enhance drug delivery.²² For instance, studies have shown that Salvianolic acid A and nitric oxide can increase pericyte coverage, and promote vascular normalization, which can improve tumor vascular dysfunction, thereby enhancing the delivery and efficacy of doxorubicin.^23,24 Therefore, promoting vascular normalization and effectively eliminating residual tumor cells could be a promising strategy to prevent postoperative BCR.

Shexiang Baoxin Pill (SBP) is a TCM formulation that has been clinically used for more than 40 years. It consists of seven ingredients: musk, ginseng extract, borneol, styrax, bezoar, cinnamon, and toad venom,²⁵ and is known for its effects in promoting blood circulation, removing blood stasis, and strengthening the heart. It is commonly prescribed for treating coronary heart disease and myocardial ischemia.²⁶ Clinical studies have shown that SBP can reduce adverse cardiovascular events in patients with coronary artery disease,^27,28 decrease myocardial infarction size and improve cardiac function in patients with myocardial infarction.²⁹ Experimental studies have further revealed that SBP can promote angiogenesis³⁰ and reduce inflammation.³¹ Our previous study shows that SBP can improve tumor blood perfusion, vascular permeability and vasodilation, thereby promoting the delivery of gemcitabine to the tumor and enhancing the anti-lung cancer effect.³² In addition, SBP can induce the expression of the vascular normalization protein PDGFB.³³ PDGFB can recruit PDGF receptor β (PDGFRβ)-positive pericytes,^34,35 which will effectively normalize tumor vasculature and facilitate drug accumulation within the tumor.³⁶ Based on this, we hypothesize that SBP may promote PDGFB secretion within TME, induce vascular normalization, enhance GEM perfusion into the tumor, and thereby improve the efficacy of the GEM-hydrogel in preventing postoperative BCR.

In this study, the GEM-hydrogel was prepared and characterized, and the thermosensitive performance, sustained drug release, and biocompatibility were evaluated. Subsequently, a mouse orthotopic breast cancer recurrence model was employed to investigate the synergistic anti-tumor effect of SBP combined with the GEM-hydrogel. Finally, through in vivo and in vitro experiments, we explored how SBP facilitates GEM delivery and modulates the TME through macrophages and synergistic effects with GEM-hydrogel to prevent BCR (Scheme 1).

Scheme 1 Schematic illustration of SBP combined with GEM-hydrogel to prevent tumor recurrence (created in Biorender.com).

Materials and methods

Materials

PLGA–PEG–PLGA (MW = 1780–1500–1780, LA/GA = 2.6 : 1) was purchased from Jinan Daigang Bioengineering Co., Ltd, and prepared as a 20 wt% polymer solution (hydrogel) in phosphate buffered saline (PBS) prior to use. SBP was supplied by Shanghai Hutchison Pharmaceuticals Co., Ltd. GEM was obtained from Shanghai Macklin Biochemical Co., Ltd. Collagenase IV and DNase I were purchased from BioFroxx. Dimethyl sulfoxide (DMSO) was purchased from Beijing Solarbio Science & Technology Co., Ltd. The BCA kit was purchased from Dalian Meilun Biotechnology Co., Ltd. DPBS, Dulbecco's modified Eagle's medium (DMEM) and Roswell Park Memorial Institute (RPMI) 1640 were obtained from Shanghai Chuan Qiu Biotechnology Co., Ltd. The Cell Counting Kit-8 (CCK-8) was purchased from TargetMol Chemicals Inc.

SBP was ground into a fine powder using a mortar and dissolved in DMSO, then filtered and diluted with the culture medium for use in experiments. We obtained the MS data of the DMSO extract of SBP using UHPLC-Q-Exactive Orbitrap MS and processed and analyzed the MS data using the mzCloud database, which contained the components reported in the literature (Fig. S1 and Table S1, ESI†).

Cell lines and animals

4T1-luc cells were obtained from the Chinese Academy of Science (Shanghai, China) and were cultured in RPMI 1640 medium supplemented with 10% fetal bovine serum (FBS), 1% penicillin and streptomycin (PS) at 37 °C with 5% CO₂. Murine macrophage cell line RAW264.7 was purchased from the Cell Bank of the Chinese Academy of Sciences (Shanghai, China). Mouse cardiac microvascular endothelial cells (MCMEC) and mouse brain vascular pericytes (MBVP) were obtained from Otwo Biotech (Guangzhou, China). RAW 264.7, MCMEC and MBVP were maintained in DMEM supplemented with 10% FBS and 1% PS at 37 °C with 5% CO₂.

Female BALB/c mice (6–8 weeks old) were purchased from the Experimental Animal Center of Shanghai University of Traditional Chinese Medicine (Shanghai, China). All animals were housed under specific pathogen-free conditions with access to sufficient food and water, and all animal experiments were approved by the Animal Ethics Committee of Shanghai University of Traditional Chinese Medicine (approval number: PZSHUTCM2308010003).

Preparation and characterization of the hydrogel

GEM was added to the hydrogel solution and thoroughly mixed at 25 °C for 10 minutes, after which the GEM-hydrogel was formed. To observe the morphology, the hydrogel was air-dried at room temperature, and transmission electron microscopy (TEM) images were obtained using a JEOL JEM-2100 Plus. PLGA–PEG–PLGA was dissolved in CDCl₃, and ¹H NMR spectra were recorded using a Bruker Avance NEO spectrometer. PLGA–PEG–PLGA were also dissolved in CH₂Cl₂, and FT-IR spectra were measured using a SHIMADZU IRAffinity-1s. The viscosity and modulus changes of the hydrogel with temperature were evaluated using an Anton Paar MCR101. The dynamic frequency sweep conditions were set as follows: angular frequency of 6.28 rad s⁻¹, strain amplitude of 1%, heating rate of 2 °C min⁻¹, and a temperature range from 20 °C to 50 °C.

Release profile of the GEM-hydrogel

0.8 mL of the GEM-hydrogel (8.17 mg mL⁻¹) was added to a 10 mL EP tube and incubated at 37 °C in a constant temperature incubator. After 10 minutes, gelation was observed. Then, 2 mL of PBS solution (pH 7.4) was added. 1 mL of PBS was withdrawn from the EP tube and replaced with 1 mL of fresh PBS at different times. The EP tube was returned to the 37 °C incubator to continue the release process. The withdrawn 1 mL PBS was filtered through a 0.22 μm membrane filter and analyzed using high-performance liquid chromatography (HPLC). The GEM concentration was calculated using the standard curve equation to determine the release rate at each time point.

Chromatographic conditions: the analysis was carried out using a Shimadzu HPLC system (Kyoto, Japan) equipped with a Thermo Scientific C18 column (5 μm, 4.6 × 250 mm) at a column temperature of 40 °C. Isocratic elution was performed using a mobile phase consisting of acetonitrile/0.1% trifluoroacetic acid in water (3 : 97, v/v) at a flow rate of 1.0 mL min⁻¹. Detection was performed at a wavelength of 275 nm.

Hemolysis test

The hydrogel was aliquoted into separate 10 mL EP tubes, and PBS buffer was added. The tubes were incubated at 37 °C for degradation. On days 1, 2, and 3, the EP tubes were removed, and the supernatant was collected by filtration. The EP tubes were not returned after sampling. After anticoagulation, the blood of rats was centrifuged (4 °C, 3000 rpm, 10 min). The supernatant was discarded, and the precipitate was washed with saline, followed by centrifugation until the supernatant became colorless. The precipitate was resuspended in saline to prepare a 2% (v/v) red blood cell suspension. Aliquots of 500 μL of the cell suspension were added to different EP tubes, with each tube receiving 500 μL of the gel extract at different time points, ddH₂O (positive control), and saline (negative control). All samples were gently vortexed and incubated at 37 °C for 3 hours, followed by centrifugation (4 °C, 5000 rpm, 10 min). Images were taken, and 100 μL of the supernatant was added to a 96-well plate. The OD value was measured at 540 nm using a microplate reader, and the hemolysis rate was calculated using the appropriate formula. Hemolysis rate (%) = (OD_sample − OD_saline)/(OD_ddH2O − OD_saline) × 100%.

In vitro biocompatibility test

The cell compatibility of the hydrogel was assessed in vitro using the CCK-8 method. The logarithmic growth phase cells, including 4T1-luc, RAW264.7, MCMEC, and MBVP, were seeded in a 96-well plate (5000 cells per well). After incubating for 12 hours, the hydrogel extract was added to the complete culture medium. Following 24 hours of incubation, the drug-containing medium was discarded, and a fresh complete medium containing CCK-8 (DMEM : CCK-8 = 10 : 1, v/v) was added for a specific incubation time. The absorbance was measured at a wavelength of 450 nm using a microplate reader to calculate cell viability.

Mouse tumor model and treatment regimens

Depilation was performed on the fourth mammary fat pad of the mice in advance. After digesting 4T1-luc cells with trypsin, the cell density was diluted to 4.0 × 10⁷ cells per mL with PBS. A volume of 50 μL was injected into the fourth mammary glands of female BALB/c mice. Two weeks later, an incomplete tumor resection was performed, leaving a volume of 20 mm³. Immediately post-surgery, the hydrogel was injected and exposed to a heating lamp for 1 minute to form a gel state. The treatment groups were as follows: control (saline), SBP (32 mg kg⁻¹, i.g.), low dose GEM-hydrogel (L-GEM, 1.63 mg/0.2 mL), SBP + L-GEM (SBP/L-GEM), and high dose GEM-hydrogel (H-GEM, 3.27 mg/0.2 mL), with SBP administered once daily. Post-surgery, tumor fluorescence changes were recorded using an in vivo imaging system (PerkinElmer, IVIS spectrum BL) in the bioluminescence mode: D-luciferin potassium salt (150 mg kg⁻¹, Meryer, M82332) was administered via intraperitoneal injection, and after 15 minutes, tumor fluorescence was recorded using the in vivo imaging system (IVIS), and fluorescence intensity was analyzed using Living Image software. Mouse body weight and tumor volume were also recorded (tumor volume = 0.5 × length × width²). After the experiment, tumor tissues were collected for immunofluorescence and western blot analyses.

Blood perfusion

After 16 days, blood perfusion in the tumor region of mice from the control group and the SBP group was measured using the IVIS fluorescence bioluminescence mode. Mice were intravenously injected with 100 μL of TRITC-Dextran (50 mg kg⁻¹, Sigma-Aldrich, T1287). After 20 minutes, the mice were anesthetized with isoflurane, and fluorescence in the tumor region was recorded using the IVIS system. Fluorescence intensity was analyzed using Living Image software.

Tumor vessel perfusion

Tumor vessel perfusion in the tumor region of mice from the control group and the SBP group was measured using small-animal ultrasound (FUJIFILM VisualSonics, Vevo 3100) in three-dimensional (3D) imaging and Doppler modes. Mice were anesthetized with 2% isoflurane (RWD Life Science) and placed in a supine position on a temperature-controlled platform. Tumor visualization was performed using a 550D probe at 32 MHz. Data along the x, y, and z axes were recorded in 3D mode with a step size of 0.152 mm. The percentage of global tumor vessel perfusion relative to the 3D tumor volume was analyzed according to the manufacturer's instructions.

HPLC analysis of GEM concentrations in the tumor

The mice were divided into a control group and a SBP group. Mice in the SBP group were orally administered SBP suspension daily. After a designated period of gavage, the GEM-hydrogel was injected near the tumor. The mice were sacrificed 24 hours later, and tumors were excised. Tumor tissues were homogenized with saline (tumor volume: saline = 2 mg: 5 μL) and centrifuged (4 °C, 5500 rpm, 10 min). 150 μL of the homogenate supernatant was mixed with 50 μL internal standard solution and 1 mL methanol–acetonitrile (1 : 9, v/v). The mixture was vortexed for 5 minutes and centrifuged (4 °C, 5500 rpm, 10 min). Next, 1 mL of the supernatant was transferred to a 1.5 mL centrifuge tube and evaporated to dryness under nitrogen at room temperature. The residue was reconstituted with 110 μL of mobile phase solution, vortexed thoroughly, and centrifuged again (4 °C, 12 000 rpm, 10 min). A 100 μL aliquot of the supernatant was injected for analysis, with an injection volume of 20 μL. The peak areas were recorded and substituted into the standard curve equation to determine the GEM concentration. The HPLC testing conditions were the same as those for the “Release profile of GEM-hydrogel”.

Flow cytometry analysis

The mice were divided into a control group and a SBP group. After 16 days of gavage, the experiment was ended and the tumor tissues were collected. The tumor tissues were placed in a dish containing 3 mL of ice-cold RPMI and minced into small pieces. A dissociation buffer (10 mL RPMI + 20 mg collagenase IV + 2 mg DNase I) was added, and the mixture was incubated at 37 °C for 30 minutes. The resulting solution was passed through a 70 μm cell strainer to prepare a single-cell suspension. Red blood cells were removed using red blood cell lysis buffer, followed by centrifugation. The cells were then stained with antibodies for 30 minutes. After washing off unbound antibodies, the cells were transferred to FACS tubes, resuspended, and analyzed by flow cytometry (BECKMAN COULTER, Cytoflex). The results were processed using FlowJo software. The flow cytometry antibodies used included 488 Rabbit anti-Mouse CD31 antibody (ABclonal, A23701), PE Rabbit anti-Mouse CD140b/PDGFR beta antibody (ABclonal, A26689), and APC Rat anti-Mouse F4/80 antibody (Elabscience, E-AB-F0995E).

Immunofluorescence analysis

After collection, the tumor tissues were fixed in 4% paraformaldehyde, embedded, and sectioned for subsequent immunofluorescence (IF) staining. The sections were placed in a light-protected humid chamber and incubated overnight at 4 °C with the primary antibody. Following this, the sections were incubated at room temperature for 2 hours with the corresponding secondary antibody, and finally, DAPI (Beyotime) was added before covering with a coverslip. Imaging was performed using a laser scanning confocal microscope (Nikon, A1R+), and analysis was conducted using NIS-Elements or ImageJ. The primary antibodies used included collagen I (1 : 500, absin, abs131984), HIF-1α (1 : 300, CST, 14179S), CD31 (1 : 500, Servicebio, GB12063), NG2 (1 : 500, Affinity, DF-12589), α-SMA (1 : 500, Servicebio, GB111364), PDGFB (upingbio, YP-Ab-15955), and F4/80 (1 : 500, Servicebio, GB113373). The secondary antibody used included 488-conjugated Goat Anti-Rabbit IgG (1 : 500, Servicebio, GB25303) and Cy3-labeled Goat Anti-Mouse IgG (1 : 500, Beyotime, A0521).

Western blot

Tumor tissues or cells were lysed using RIPA buffer (containing protease and phosphatase inhibitors). The total protein concentration was measured with a BCA kit. Western blot (WB) samples were prepared by mixing with 5× loading buffer according to the required volume. The samples were electrophoresed on polyacrylamide gels (20 V for 20 minutes, 70 V for 50 minutes, and 120 V for 50 minutes) and subsequently transferred onto PVDF membranes at 250 mA for 2 hours. The membranes were blocked with 5% BSA or 5% skimmed milk for 1.5 hours, followed by incubation with primary antibodies overnight on a shaker at 4 °C. The membranes were then incubated with corresponding secondary antibodies at room temperature for 1.5 hours. Signal detection was performed using the super sensitive ECL luminescence reagent (Meilunbio, MA0186-2), followed by exposure on an Azure Biosystem (Azure Biosystems, USA). The results were analyzed using ImageJ software. The primary antibodies used include VEGF (Santa Cruz, sc-7269), PDGFB (BOSTER, A00348), ANG-1 (Abcam, ab183701), NF-κB (Beyotime, AF0246), p-NF-κB (MCE, HY-P80839), p-PDFFRβ (CST, 4549T), and PDFFRβ (CST, 3169T). The secondary antibody used included HRP labeled Goat Anti-Rabbit IgG (H + L) (Beyotime, A0208) and HRP-labeled Goat Anti-Mouse IgG (H + L) (Beyotime, A0216).

Quantitative real-time PCR

RAW 264.7, MCMEC, and 4T1-luc cells (5 × 10⁵ cells per well) were cultured with SBP-containing medium (32 μg mL⁻¹, 64 μg mL⁻¹) in a 37 °C incubator for 3 hours. Cells were lysed using an RNA isolator reagent (Vazyme, R401) to extract RNA, and RNA concentration was measured using a nanodrop (DeNovix, United States). Reverse transcription was performed using a reverse transcription kit (Vazyme, R223), and the resulting cDNA was analyzed by quantitative PCR after adding reagents from a quantitative PCR kit (Vazyme, Q711), utilizing the ABI QuantStudio 6 Flex system (Thermo Fisher Scientific, United States). Primers were synthesized by Sangon Biotech (Shanghai, China): PDGFB (F:5′-TGTGCCCTTCAGTCTGCTCCTC3′, R:5′CAAC CTTGCTCACCCTGCTTGG-3′); β-actin (F:5′-GTCCCTCACCCTCCCA AAAG-3′, R:5′-GCTGCCTCA ACACCTCAACCC-3′).

Enzyme-linked immunosorbent assay (ELISA)

RAW 264.7 cells (2 × 10⁷ cells per well) were seeded into a six-well plate and cultured with SBP-containing DMEM (64 μg mL⁻¹) in a 37 °C incubator for 24 hours. The medium was then collected and centrifuged (4 °C, 2000 rpm, 20 min) to obtain the supernatant (not SBP treated supernatant: conditioned medium (CM); SBP treated supernatant: SBP conditioned medium (SBP-CM) for the (migration assay)). PDGFB protein expression was measured using a microplate reader to determine the OD values according to the PDGF-B ELISA kit instructions (NeoBioscience, EMC032).

Migration assay

The migration assay was performed to assess the effect of SBP-treated RAW 264.7 cells on MBVP migration. MBVP cells (5 × 10⁵ cells per well) were seeded into a six-well plate and grown to 90–95% confluence. A sterile pipette tip was used to scratch a straight line across the cell layer. The detached cells were washed away three times with DPBS. The cells were then cultured with DMEM, SBP-containing DMEM (64 μg mL⁻¹), CM, SBP-CM, or CM supplemented with PDGFB (200 pg mL⁻¹). Images of the scratch area were taken at 0 h and 12 h using an inverted microscope, and the migration rate was calculated.

Statistical analysis

Data were shown as the mean ± standard deviation (SD). Statistical analysis was performed using two-sided unpaired t-tests and an ordinary two-way ANOVA test. *P < 0.05, **P < 0.01, ***P < 0.001.

Results and discussion

Characterization of the hydrogel and GEM-hydrogel

The structure of PLGA–PEG–PLGA was analyzed using nuclear magnetic resonance (NMR) spectroscopy and Fourier transform infrared (FT-IR) spectroscopy, as shown in Fig. 1A and B. The NMR spectra revealed the chemical shifts and peak shapes of the PLGA–PEG–PLGA. The chemical shifts were observed as follows: δ = 1.55 ppm for the protons of the “–CH₃–” group in LA, δ = 3.65 ppm for the protons of the “–CH₂–” group in PEG, δ = 4.80 ppm for the protons of the “–CH₂–” group in GA, and δ = 5.20 ppm for the protons of the “–CH–” group in LA. The FT-IR spectra showed the characteristic absorption peak of C=O (1756 cm⁻¹), consistent with previous reports in the literature.³⁷ We then observed the microscopic structure of the 20 wt% PLGA–PEG–PLGA (hydrogel) using TEM (Fig. 1C), which revealed that the polymer formed spherical micelles that were uniformly dispersed in the solution. As the temperature approached 37 °C, the hydrogel transitioned from a solution to a gel state (Fig. 1D), demonstrating excellent injectability in its solution form and rapidly transforming into a gel upon increasing the temperature to 37 °C (Fig. S2A, ESI†). We further verified the temperature-dependent state changes of the hydrogel using a rheometer, which indicated that its viscosity increased with rising temperature. Near 40 °C, the viscosity began to decrease with increasing temperature (Fig. 1E), showing a similar trend for the GEM-hydrogel (Fig. S2B, ESI†). The modulus changes measured by the rheometer indicated that the hydrogel was in a gel state at approximately 35.61–38.70 °C (Fig. 1F), while the temperature range for the GEM-hydrogel in a gel state was slightly wider, also around 35.61–39.74 °C (Fig. S2C, ESI†). This observation may be related to the fact that the polymer used had an LA/GA ratio of 2.6 : 1 and was prepared at a 20 wt% PLGA–PEG–PLGA.¹³

Fig. 1 Characterization and biocompatibility of the GEM-hydrogel. (A) ¹H NMR spectra of PLGA–PEG–PLGA; (B) FT-IR spectra of blank hydrogel, GEM, and GEM-hydrogel; (C) TEM diagram of the hydrogel (scale bars, 2 μm; inset scale bars, 1 μm); (D) appearance of the hydrogel according to temperature; (E) change of complex viscosity of hydrogel solutions with temperature; and (F) change of modulus of hydrogel solutions with temperature. (G) In vitro drug release behavior of the GEM-hydrogel and (H) hemolysis test of blank hydrogel extracts. (I) In vitro cell compatibility test of the 3-day extract of the blank hydrogel (n = 6); (J) fluorescence images of calcein-AM/PI-stained cells after treatment with the blank hydrogel (live cells: green; dead cells: red; n = 6, scale bars, 200 μm).

Next, we assessed the degradation of the hydrogel by measuring its mass (Fig. S2D, ESI†). The in vitro degradation occurred over approximately 18 days, during which the hydrogel slowly absorbed water and increased in mass and volume from days 0 to 5, reaching a maximum swelling rate of about 30% (Fig. S2E, ESI†). After day 5, degradation began, and the hydrogel dispersed in PBS solution, with only a small amount of precipitate observed by day 18 (Fig. S2F, ESI†). The degradation process demonstrated that the PLGA–PEG–PLGA hydrogel exhibits excellent degradability, enabling gradual breakdown in vivo. The degradation products, lactic acid and glycolic acid, can be metabolized by the body, preventing any accumulation. We also tested the in vitro release profile of the GEM-hydrogel (Fig. 1G), which released approximately 80% of GEM over 16 days. The hydrogel effectively released GEM within the first 2 days, aiding in the rapid destruction of local tumor cells, followed by a slower release of GEM to continuously inhibit the tumor, demonstrating a good localized sustained release effect.

Biosafety of the hydrogel

To evaluate the biocompatibility of the hydrogel, we conducted in vitro hemolysis and cytotoxicity assays. In the hemolysis assay, we assessed the effect of hydrogel extracts on red blood cells, as shown in Fig. 1H, ddH₂O caused significant hemolysis, whereas hydrogel extracts collected at different time points exhibited hemolysis rates below 5%, indicating good blood compatibility. In the cytotoxicity assay, we evaluated the effect of the day-3 extract on cell proliferation and apoptosis. Four types of cells were cultured with the extracts for 24 hours, and the extracts of day 3 did not inhibit cell proliferation or promote apoptosis (Fig. 1I and J). But the extracts from days 10 and 16 showed significant cytotoxicity. The observed cytotoxicity is primarily due to the acidic degradation products of PLGA–PEG–PLGA, such as lactic acid and glycolic acid, which significantly lower the pH of the extracts from days 10 and 16 beyond the tolerance threshold of cells, thereby inducing toxicity. Importantly, lactic acid and glycolic acid are efficiently absorbed by the body and metabolized through the Krebs cycle,³⁸ preventing their accumulation in vivo and ensuring excellent biocompatibility.³⁹ Furthermore, studies have confirmed that these degradation products do not exhibit inhibitory effects on tumors, further underscoring their safety profile.¹⁶ Collectively, these results highlight the excellent biosafety of the hydrogel.

In summary, the hydrogel remains in a liquid state at 25 °C, facilitating easy injection into the post-surgical cavity after breast cancer resection. Upon contact with the body and reaching the critical gelation temperature, dispersed micelles aggregate to form a three-dimensional network that can slowly release the drug.⁴⁰ The hydrogel achieves sustained drug release through two mechanisms: diffusion and degradation.⁴¹ Our results also showed that GEM-hydrogel released GEM rapidly during the initial phase, followed by a slower release, sustaining the release over 16 days. Additionally, the hydrogel swells gradually, avoiding rapid swelling that could irritate the wound. Both the hydrogel and its degradation products exhibited good biocompatibility, suggesting that post-operative injection of the GEM-hydrogel can provide timely and effective prevention of BCR.

SBP enhances the therapeutic efficiency of the GEM-hydrogel in recurrence of breast cancer in mice

We evaluated the inhibitory effects of the GEM-hydrogel and the combination of SBP with the GEM-hydrogel on postoperative breast cancer recurrence (BCR) using a mouse model (Fig. 2A). Tumor fluorescence intensity changes were observed through IVIS imaging and tumor volume variations were recorded (Fig. 2B–D). Following surgery, tumors were excised for further studies. We noted that the SBP group exhibited a tumor growth curve similar to the control group, indicating that the SBP group did not demonstrate a significant promotion of tumor progression compared to the control group. The GEM-hydrogel effectively inhibits BCR, with H-GEM demonstrating a stronger inhibitory effect. However, one mouse in H-GEM died (Fig. 2B and E). Conversely, the SBP/L-GEM group showed a similar inhibitory effect to the H-GEM group, outperforming L-GEM, but without causing mortality in the mice, indicating an effect-enhancement of SBP. At the end of the experiment, the excised tumors corroborated the findings from tumor fluorescence and volume changes (Fig. 2E). H&E staining revealed morphological changes in tumor cells (Fig. S3, ESI†), with all groups except for the SBP and control groups showing nuclear shrinkage, cytoplasmic vacuolation, and cellular wrinkling. Additionally, we performed H&E staining on the lungs of the mice and found tumor metastasis occurred in the control, SBP, and L-GEM groups. In contrast, in the SBP/L-GEM group, no metastasis was observed, suggesting that SBP in combination with L-GEM may also inhibit breast cancer metastasis (Fig. 2F). The above results indicate that SBP can synergize with the GEM-hydrogel, not only effectively inhibiting breast cancer recurrence (BCR) and significantly reducing the dose of GEM, thereby minimizing side effects, but also demonstrating potential in preventing breast cancer metastasis.

Fig. 2 SBP promotes GEM-hydrogel to prevent postoperative recurrence of breast cancer in mice. (A) Schematic illustration of the treatment protocol; (B) bioluminescence images of tumors in various groups of mice (n = 5); (C) quantitative analysis of tumor bioluminescence; (D) the growth curves of tumors; (E) excised recurrent solid tumors; and (F) representative H&E staining images of the lungs (n = 3, scale bars, 2 mm; inset scale bars, 0.5 mm).

Synergistic effect of SBP with GEM in 4T1-luc cells

To investigate how SBP synergizes with GEM to inhibit BCR, we first explored the in vitro synergistic effects of the two drugs. The cytotoxicity of SBP and GEM on 4T1-luc cells cultured in a 2D environment was evaluated using the CCK-8 assay. The results indicated that SBP did not exhibit significant inhibitory effects on 4T1-luc cells, while the IC₅₀ of free GEM was 19.75 nM (Fig. S4A, ESI†). Next, we analyzed the synergy between SBP and GEM using SynergyFinder. The result suggested that the combination did not exhibit synergistic inhibition of 4T1-luc cells (Fig. S4B, ESI†).

Although 2D cell cultures can reflect the direct effects of drugs on cells, they do not accurately mimic the characteristics of solid tumors, such as spatial structure, extracellular matrix, and penetration barriers.⁴² Therefore, we further evaluated the synergistic effects of SBP and GEM using 3D tumor spheroids, aiming to explore their potential synergy in a more physiologically relevant setting (Fig. S4C–E, ESI†). However, the results also showed that the combination did not exhibit synergistic inhibition of tumor cells. Above all, it can be speculated, that the mechanism of how SBP enhanced GEM's therapeutic effect is not the synergistic inhibition of tumor cells.

SBP improves drug perfusion and modulates the tumor microenvironment

SBP promotes the GEM-hydrogel to prevent postoperative BCR in mice, but it does not exhibit synergistic inhibitory effects on tumor cells in vitro. We sought to explore why SBP enhances the effect of GEM-hydrogel in vivo. Previous studies suggest that SBP can improve the tumor microenvironment by enhancing vascular function, thereby promoting the efficacy of chemotherapy drugs.³² Therefore, we first evaluated the effect of SBP on tumor perfusion by injecting TRITC-dextran via the tail vein and monitoring the fluorescence in recurrent tumors (Fig. 3A). The results showed that SBP improved blood perfusion at the tumor site. Next, we investigated the impact of SBP on tumor vasculature using 3D ultrasound and color Doppler modes to assess tumor vessel perfusion (Fig. 3B). The results indicated that the blood vessel intensity in the SBP group was higher than that of the control group. Additionally, the vascular diameter and patency in the SBP group were superior to those in the control group. These results indicate that SBP promotes tumor vascular normalization, resulting in a more organized and functional tumor vasculature. This vascular normalization is expected to enhance the delivery efficiency of chemotherapeutic drugs.⁴³

Fig. 3 SBP improves blood perfusion and tumor microenvironment. (A) In vivo fluorescence imaging after TRITC dextran injection and quantitative analysis of fluorescence intensity in the tumor (n = 3); (B) representative ultrasonic 3D images of tumor and vascularity and percentages of vascularity in 3D images of the tumor (n = 3); (C) representative immunofluorescence images of collagen I (n = 6, scale bars, 100 μm) and HIF-1α (n = 6, scale bars, 100 μm); (D) quantification for immunofluorescence images of collagen I and HIF-1α; (E) schematic representation of GEM concentration determination; and (F) GEM concentration in tumor tissue (n = 3).

Hypoxia significantly affects tumor progression and treatment efficacy.⁴⁴ Aside from directly alleviating hypoxia through nanomaterials,^45,46 increasing blood perfusion at the tumor site is also an effective strategy to improve tumor hypoxia.⁴⁷ We examined the tumor oxygen levels. It was found that increased perfusion in the SBP group was associated with reduced tumor hypoxia (Fig. 3C and D). Furthermore, we observed a reduction in collagen I levels within tumors in the SBP group (Fig. 3C and D). All results indicate the SBP can modulate the vasculature and microenvironment of the residue tumor, which may promote the therapeutic outcome of the GEM-hydrogel.

Blood perfusion influences the delivery of chemotherapeutic drugs.⁴⁸ Collagen is a component of the extracellular matrix, increased collagen levels elevate interstitial fluid pressure, thereby hindering the delivery of chemotherapeutic agents.⁴⁹ SBP enhances perfusion by increasing the proportion of normalized blood vessels in tumors and reduces the production of collagen I. Therefore, we hypothesize that SBP promotes GEM delivery by improving these two factors in the tumor microenvironment. To test this, we measured the accumulation of GEM in tumors after oral administration of SBP for different durations (Fig. 3E and F). The results showed that SBP may progressively enhance the accumulation of GEM in tumor tissues over time, potentially by facilitating the delivery of GEM to these sites. In summary, these findings indicate that SBP modulates the tumor microenvironment, facilitates GEM accumulation in tumors, and enhances the inhibitory effect of the GEM-hydrogel on BCR.

SBP induces the secretion of PDGFB, recruits pericytes, and promotes the normalization of tumor blood vessels

It is widely recognized that abnormal blood vessels induced by tumors are structurally irregular, immature, and poorly permeable, impairing tumor perfusion and exacerbating hypoxia and drug delivery challenges.²⁰ SBP can enhance GEM delivery by improving vascular function and modulating the tumor microenvironment. To explore whether these effects are related to vascular normalization, we performed flow cytometry to analyze the proportion of cells associated with vascular normalization (Fig. 4A). The results revealed that SBP-treated mice exhibited an increased proportion of endothelial cells (CD31), macrophages (F4/80), and pericytes (PDGFRβ) in total cells, with respective increases of 3%, 8%, and 7% compared to the control group. These findings indicate that SBP not only promotes angiogenesis but also induces vascular normalization, alongside an increase in macrophage populations. To further investigate the interaction between endothelial cells and pericytes following SBP treatment, we used NG-2 and α-SMA to label pericytes and CD31 to label endothelial cells (Fig. 4B–D). The results showed that SBP treatment increased the number of pericytes and their coverage, confirming that SBP promotes vascular normalization.

Fig. 4 SBP promotes tumor vascular normalization. (A) Flow cytometry analysis of pericytes, macrophages and endothelial cells in the tumor (n = 3); (B) representative immunofluorescence images of co-staining of CD31 (red) with NG2 (green) and CD31 (red) with α-SMA (green); (C) quantification of the NG2 area and pericyte coverage (n = 6, large scale bars, 50 μm; inset scale bars, 5 μm); (D) quantification for α-SMA area and pericyte coverage (n = 6, large scale bars, 50 μm; insert scale bars, 5 μm); and (E) and (F) the protein expression levels of PDGFB, ANG-1 and VEGF by western blot analysis (n = 6).

We then explored how SBP enhances angiogenesis and recruits pericytes by evaluating the expression of three angiogenesis-related proteins in tumor tissues using western blotting: vascular endothelial growth factor (VEGF), platelet-derived growth factor B (PDGFB), and angiopoietin 1 (ANG-1). The results showed that SBP treatment increased the expression of pro-angiogenic VEGF and vascular normalization factors PDGFB and ANG-1 in tumor tissues (Fig. 4E and F).

VEGF is a growth factor with pro-angiogenic activity that promotes endothelial cell proliferation and regulates pathological angiogenesis.⁵⁰ PDGFB, a member of the PDGF family, plays a crucial role in recruiting PDGFRβ-positive pericytes to vessels, thereby maintaining vascular integrity in the tumor microenvironment.⁵¹ ANG-1, a ligand for TIE2, competes with ANG-2 to bind TIE2, promoting vascular stability and maturation⁵² and is secreted by pericytes.⁵³ In summary, SBP promotes VEGF expression, facilitating angiogenesis in tumor tissues, while also enhancing PDGFB levels to recruit pericytes to the tumor. This recruitment increases ANG-1 expression and promotes vascular stability and maturation, ultimately leading to tumor vascular normalization. Therefore, SBP promotes tumor angiogenesis and facilitates vascular normalization, thereby providing a more efficient pathway for GEM delivery.

SBP stimulates macrophages to secret PDGFB

Research has shown that SBP can promote angiogenesis by activating macrophages to release VEGF,⁵⁴ but the mechanism by which it promotes PDGFB secretion remains unclear. PDGFB can be secreted by endothelial cells,⁵⁵ cancer cells,⁵⁶ and macrophages.⁵⁷ To investigate which type of cells SBP promotes for PDGFB secretion, we first analyzed the mRNA expression of Pdgfb in Raw 264.7, MCMEC and 4T1-luc cells after SBP treatment using qRT-PCR. Notably, we found that Pdgfb mRNA expression significantly increased in Raw 264.7 cells following SBP treatment (Fig. 5A), meanwhile, SBP showed no cytotoxicity towards any of the three cell types (Fig. S4A and S5A, B, ESI†). Subsequently, we measured the PDGFB levels in the supernatant of Raw 264.7 cells treated with SBP for 24 hours (Fig. 5B), which was significantly higher than that in the control group. Previous studies indicate that PDGFB expression is associated with NF-κB. After treating cells with an NF-κB activator, PDGFB expression was found to be positively correlated with the concentration of the activator.^56,58 Therefore, we assessed the phosphorylation status of NF-κB in Raw 264.7 cells after SBP treatment (Fig. 5C), and the results showed that NF-κB phosphorylation levels were higher in the SBP group compared to the control group. We also performed immunofluorescence staining with F4/80 and PDGFB to observe the secretion of PDGFB by macrophages in tumor tissues, and the results were consistent with the cell experiments (Fig. 5D and E), showing an increase in PDGFB secretion by macrophages in the tumor tissues. The above results indicate that SBP activates NF-κB in macrophages, promoting PDGFB secretion and thereby increasing PDGFB expression in tumors.

Fig. 5 SBP promotes PDGFB secretion from macrophages to recruit pericytes. (A) The Pdgfb mRNA expressions of different cells after treatment for 3 h (n = 3); (B) PDGF-BB concentration in the supernatants of raw 246.7 after treatment for 24 h (n = 6); (C) the protein expression levels of p-NF-κB by western blot analysis; (D) and (E) representative immunofluorescence images of co-staining of F4/80 (red) with PDGFB (green), quantification for PDGFB/F4/80 (n = 6, large scale bars, 50 μm; insert scale bars, 5 μm); (F) and (G) representative images of MBVP cell migration, quantification for migration (n = 6); (H) the protein expression levels of p-PDGFRβ by western blot analysis.

Next, we aimed to explore the effects of PDGFB secreted by macrophages on pericytes. First, we generated conditioned media from Raw 264.7 cells treated with or without SBP (SBP-CM or CM) to observe the effect of these conditioned media on the migration of PDGFRβ⁺ pericytes in vitro (Fig. 5F and G). Scratch assays showed that both SBP-CM and exogenous PDGFB significantly promoted pericyte migration compared to the CM group; however, the SBP group did not enhance MBVP pericyte migration. Additionally, SBP and SBP-CM did not inhibit MBVP cell proliferation, and SBP-CM even promoted MBVP proliferation (Fig. S5C and D, ESI†). It can be speculated that SBP promotes the activation of PDGFRβ on pericytes by driving PDGFB expression in Raw 264.7 cells (Fig. 5H), thereby enhancing pericyte migration. We also observed that SBP could promote the adhesion of MBVP to MCMEC through cell adhesion assays, which may also be due to the activation of PDGFRβ on pericytes (Fig. S6A, ESI†). These experimental results demonstrate that SBP promotes pericyte migration and adhesion through macrophages, leading to increased pericytes coverage in tumors.

Overall, the role of SBP in modulating macrophages within the tumor microenvironment has been further validated. We found that SBP activates the NF-κB signaling pathway in macrophages, promoting PDGFB secretion, which enhances pericyte migration. Consequently, this contributes to the maturation and stabilization of tumor vasculature.

Toxicity evaluation of SBP in combination with the GEM-hydrogel

At the end of the experiment, blood samples and major organs were collected from the mice. Changes in body weight indicated that all mice experienced weight loss post-surgery (Fig. 6A), suggesting that the surgery had an impact on the body. However, there were no significant differences in weight changes between the SBP/L-GEM group and the other groups. Analysis of ALT, AST, UREA, and CREA levels in the mice's blood (Fig. 6B) showed no significant differences in UREA and CREA among the groups; however, the ALT and AST levels in the SBP/L-GEM group were lower than those in the SBP and L-GEM groups, indicating reduced toxicity with the combination treatment. Additionally, we weighed the major organs to calculate organ coefficients (Fig. 6C), while no obvious changes were observed. Finally, H&E staining of the major organs (Fig. 6D) revealed no significant damage to the liver, heart, spleen, and kidneys after SBP treatment. These results demonstrate that the combination of SBP and GEM-hydrogel presents no significant safety issues for the postoperative treatment of breast cancer and could be considered as a treatment option for preventing breast cancer recurrence.

Fig. 6 Systemic toxicity evaluation of SBP combined with the GEM-hydrogel. (A) Body weight changes of different groups of mice; (B) biochemical analysis of liver and kidney damage; (C) organ coefficients of the liver, heart, spleen, lungs and kidneys; and (D) representative H&E staining images of liver, heart, spleen, and kidneys (n = 3, scale bars, 2 mm).

Conclusions

In summary, we developed a GEM-hydrogel that can be conveniently injected into the surgical cavity and provides timely, effective, and low-toxicity release of GEM, making it an ideal candidate for postoperative chemotherapy in breast cancer. SBP enhances the inhibitory effect of the GEM-hydrogel on BCR. The specific mechanism involves SBP activating NF-κB in tumor-associated macrophages, which promotes the secretion of the pro-vascular normalization factor PDGFB. PDGFB recruits PDGFRβ+ pericytes into the tumor area, further facilitating tumor vascular normalization. Additionally, SBP reduces collagen deposition in tumor tissue, improves blood perfusion through vascular normalization, and ultimately remodels the TME. This process not only enhances the drug delivery efficiency of the GEM-hydrogel within the tumor but also significantly alleviates tumor hypoxia, thereby strengthening the inhibitory effect of the GEM-hydrogel on BCR. The combination of SBP and the GEM-hydrogel is both safe and effective, offering a novel strategy for improving drug delivery and preventing BCR.

Data availability

Data for this article, including drug release, quantitative analysis of fluorescence, Q-PCR and western blotting, are available at the Science Data Bank at https://doi.org/10.57760/sciencedb.18537.

Conflicts of interest

There are no conflicts to declare.

Acknowledgements

The work was supported by the State Administration of Traditional Chinese Medicine of the Peoples Republic of China (GZYYGJ2019059), the New Interdisciplinary Subject Funding Program for Shanghai Traditional Chinese Medicine (E2-F18003), the Shanghai Municipal Health Commission/Shanghai Municipal Administration of Traditional Chinese Medicine (ZY (2021-2023)-0501), the Shanghai Science and Technology Development Fund from Central Leading Local Government (YDZX20223100001004), the Shanghai Municipal Education Commission (SMEC) (2019-01-07-00-10-E00072), the Open Project of National Major Scientific and Technological Infrastructure for Translational Medicine (Shanghai) (TMSK-2021-405), the National Natural Science Foundation of China (82003800, 82473849), and the Science and Technology Commission of Shanghai Municipality (20ZR1473200), the Expenditure Budget Program of Shanghai University of Traditional Chinese Medicine (23KFL001), and the Shanghai Pujiang Programme (23PJD085).

References

1. H. Sung, J. Ferlay, R. L. Siegel, M. Laversanne, I. Soerjomataram, A. Jemal and F. Bray, Ca-Cancer J. Clin., 2021, 71, 209–249.
2. J. G. Hiller, N. J. Perry, G. Poulogiannis, B. Riedel and E. K. Sloan, Nat. Rev. Clin. Oncol., 2018, 15, 205–218.
3. N. Harbeck and M. Gnant, Lancet, 2017, 389, 1134–1150.
4. M. Chavez-MacGregor, C. A. Clarke, D. Y. Lichtensztajn and S. H. Giordano, JAMA Oncol., 2016, 2, 322–329.
5. A. G. Waks and E. P. Winer, JAMA, 2019, 321, 288–300.
6. S. Shaha, D. Rodrigues and S. Mitragotri, J. Controlled Release, 2024, 367, 737–767.
7. L. de Sousa Cavalcante and G. Monteiro, Eur. J. Pharmacol., 2014, 741, 8–16.
8. E. Mini, S. Nobili, B. Caciagli, I. Landini and T. Mazzei, Ann. Oncol., 2006, 17(Suppl. 5), v7–v12.
9. M. A. Tempero, U. Pelzer, E. M. O'Reilly, J. Winter, D. Y. Oh, C. P. Li, G. Tortora, H. M. Chang, C. D. Lopez, T. Bekaii-Saab, A. H. Ko, A. Santoro, J. O. Park, M. S. Noel, G. L. Frassineti, Y. S. Shan, A. Dean, H. Riess, E. Van Cutsem, J. Berlin, P. Philip, M. Moore, D. Goldstein, J. Tabernero, M. Li, S. Ferrara, Y. Le Bruchec, G. Zhang, B. Lu, A. V. Biankin and M. Reni, J. Clin. Oncol., 2023, 41, 2007–2019.
10. J. Li, Q. Chen, S. Li, X. Zeng, J. Qin, X. Li, Z. Chen, W. Zheng, Y. Zhao, Z. Huang, X. Yang and L. Gan, Chem. Eng. J., 2023, 473, 145212.
11. H. Liu, X. Shi, D. Wu, K. F. Kahsay, L. Deng, A. Dong, W. Wang and J. Zhang, ACS Appl. Mater. Interfaces, 2019, 11, 19700–19711.
12. Q. Pan, Z. Zong, H. Li, L. Xie, H. Zhu, D. Wu, R. Liu, B. He and Y. Pu, Int. J. Biol. Macromol., 2025, 284, 137893.
13. M. S. Shim, H. T. Lee, W. S. Shim, I. Park, H. Lee, T. Chang, S. W. Kim and D. S. Lee, J. Biomed. Mater. Res., 2002, 61, 188–196.
14. G. M. Zentner, R. Rathi, C. Shih, J. C. McRea, M. H. Seo, H. Oh, B. G. Rhee, J. Mestecky, Z. Moldoveanu, M. Morgan and S. Weitman, J. Controlled Release, 2001, 72, 203–215.
15. J. Shi, L. Yu and J. Ding, Acta Biomater., 2021, 128, 42–59.
16. S. Y. Kim, M. J. Jo, M. S. Yoon, C. E. Jin, Y. B. Shin, J. M. Lee, H. J. Shin, J. G. Oh, J. M. Cho, H. Kim, H. Park, Y. W. Choi, C. W. Park, J. S. Kim and D. H. Shin, J. Controlled Release, 2023, 360, 796–809.
17. N. L. Elstad and K. D. Fowers, Adv. Drug Delivery Rev., 2009, 61, 785–794.
18. X. Chen, M. Wang, X. Yang, Y. Wang, L. Yu, J. Sun and J. Ding, Theranostics, 2019, 9, 6080–6098.
19. K. E. de Visser and J. A. Joyce, Cancer Cell, 2023, 41, 374–403.
20. C. Viallard and B. Larrivee, Angiogenesis, 2017, 20, 409–426.
21. R. Shen, L. Peng, W. Zhou, D. Wang, Q. Jiang, J. Ji, F. Hu and H. Yuan, J. Controlled Release, 2022, 349, 550–564.
22. Y. M. Meng, X. Jiang, X. Zhao, Q. Meng, S. Wu, Y. Chen, X. Kong, X. Qiu, L. Su, C. Huang, M. Wang, C. Liu and P. P. Wong, Nat. Commun., 2021, 12, 6011.
23. C. Qian, Y. Zhou, T. Zhang, G. Dong, M. Song, Y. Tang, Z. Wei, S. Yu, Q. Shen, W. Chen, J. P. Choi, J. Yan, C. Zhong, L. Wan, J. Li, A. Wang, Y. Lu and Y. Zhao, Acta Pharm. Sin. B, 2024, 14, 2077–2096.
24. Y. Sung, P. Jin, L. Chu, F. Hsu, M. Wang, C. Chang, S. Chiou, J. T. Qiu, D. Gao, C. Lin, Y. Chen, Y. Hsu, J. Wang, F. Wang, P. Yu, A. Chiang, A. Y. Wu, J. J. Ko, C. P. Lai, T. Lu and Y. Chen, Nat. Nanotechnol., 2019, 14, 1160.
25. L. Lu, X. Sun, C. Chen, Y. Qin and X. Guo, Front. Pharmacol., 2018, 9, 1161.
26. Y. Yu, S. Liu, Y. Zhou, K. Huang, B. Wu, Z. Lin, C. Zhu, Y. Xue and K. Ji, Phytomedicine, 2022, 104, 154336.
27. H. Shi, J. Zhou, C. Ma, F. Ji, Y. Wu, Y. Zhao, J. Qian and X. Wang, Front. Cardiovasc. Med., 2022, 9, 1002400.
28. J. Wei, T. Ma, C. Zhou, P. Hao, B. Li, X. Wang, R. Yu, M. Zhu and Y. Wang, Front. Pharmacol., 2022, 13, 1002713.
29. H. Qin, S. Li and Z. Liu, Front. Pharmacol., 2021, 12, 721011.
30. J. Zhang, Q. Cui, Y. Zhao, R. Guo, C. Zhan, P. Jiang, P. Luan, P. Zhang, F. Wang, L. Yang, X. Yang and Y. Xu, Atherosclerosis, 2020, 292, 99–111.
31. X. Xu, S. Li, T. Wang, P. Zhen, Q. Wei, F. Yu and J. Tong, Phytomedicine, 2024, 128, 155390.
32. L. Yang, R. Li, X. Yang, Q. Cui, F. Wang, G. Lin and J. Zhang, Front. Pharmacol., 2019, 10, 565.
33. S. Goel, D. G. Duda, L. Xu, L. L. Munn, Y. Boucher, D. Fukumura and R. K. Jain, Physiol. Rev., 2011, 91, 1071–1121.
34. A. N. Stratman, A. E. Schwindt, K. M. Malotte and G. E. Davis, Blood, 2010, 116, 4720–4730.
35. S. S. Kemp, K. N. Aguera, B. Cha and G. E. Davis, Arterioscl. Throm. Vas., 2020, 40, 2632–2648.
36. S. Ma, Z. Tian, L. Liu, J. Zhu, J. Wang, S. Zhao, Y. Zhu, J. Zhu, W. Wang, R. Jiang, Y. Qu, J. Lei, J. Zhao and T. Jiang, Small, 2024, 20, e2308638.
37. B. Yuan, Y. Zhang, Q. Wang, G. Ren, Y. Wang, S. Zhou, Q. Wang, C. Peng and X. Cheng, Int. J. Pharm., 2022, 627, 122225.
38. F. Danhier, E. Ansorena, J. M. Silva, R. Coco, A. Le Breton and V. Preat, J. Controlled Release, 2012, 161, 505–522.
39. X. Chen, J. Zhang, K. Wu, X. Wu, J. Tang, S. Cui, D. Cao, R. Liu, C. Peng, L. Yu and J. Ding, Small Methods, 2020, 4, 2000310.
40. X. Yang, X. Chen, Y. Wang, G. Xu, L. Yu and J. Ding, Chem. Eng. J., 2020, 396, 125320.
41. M. Qiao, D. Chen, X. Ma and Y. Liu, Int. J. Pharm., 2005, 294, 103–112.
42. E. C. Costa, A. F. Moreira, D. de Melo-Diogo, V. M. Gaspar, M. P. Carvalho and I. J. Correia, Biotechnol. Adv., 2016, 34, 1427–1441.
43. Y. C. Sung, P. R. Jin, L. A. Chu, F. F. Hsu, M. R. Wang, C. C. Chang, S. J. Chiou, J. T. Qiu, D. Y. Gao, C. C. Lin, Y. S. Chen, Y. C. Hsu, J. Wang, F. N. Wang, P. L. Yu, A. S. Chiang, A. Y. Wu, J. J. Ko, C. P. Lai, T. T. Lu and Y. Chen, Nat. Nanotechnol., 2019, 14, 1160–1169.
44. V. Aggarwal, O. Miranda, P. A. Johnston and S. Sant, Cancer Lett., 2020, 490, 124–142.
45. J. Liu, X. Li, Y. Li, Q. Gong and K. Luo, Theranostics, 2025, 15, 993–1016.
46. Q. Pan, H. Tang, L. Xie, H. Zhu, D. Wu, R. Liu, B. He and Y. Pu, J. Mater. Chem. B, 2024, 12, 11560–11572.
47. N. Zhang, R. Yin, P. Zhou, X. Liu, P. Fan, L. Qian, L. Dong, C. Zhang, X. Zheng, S. Deng, J. Kuai, Z. Liu, W. Jiang, X. Wang, D. Wu and Y. Huang, Proc. Natl. Acad. Sci. U. S. A., 2021, 118, 2020057118.
48. P. P. Wong, F. Demircioglu, E. Ghazaly, W. Alrawashdeh, M. R. Stratford, C. L. Scudamore, B. Cereser, T. Crnogorac-Jurcevic, S. McDonald, G. Elia, T. Hagemann, H. M. Kocher and K. M. Hodivala-Dilke, Cancer Cell, 2015, 27, 123–137.
49. T. R. Cox, Nat. Rev. Cancer, 2021, 21, 217–238.
50. C. S. Melincovici, A. B. Bosca, S. Susman, M. Marginean, C. Mihu, M. Istrate, I. M. Moldovan, A. L. Roman and C. M. Mihu, Rom. J. Morphol. Embryol., 2018, 59, 455–467.
51. Y. Zhang, J. Cedervall, A. Hamidi, M. Herre, K. Viitaniemi, G. D'Amico, Z. Miao, R. Unnithan, A. Vaccaro, L. van Hooren, M. Georganaki, A. Thulin, Q. Qiao, J. Andrae, A. Siegbahn, C. H. Heldin, K. Alitalo, C. Betsholtz, A. Dimberg and A. K. Olsson, Cancer Res., 2020, 80, 3345–3358.
52. H. G. Augustin, G. Y. Koh, G. Thurston and K. Alitalo, Nat. Rev. Mol. Cell Biol., 2009, 10, 165–177.
53. D. Chakroborty, C. Sarkar, H. Yu, J. Wang, Z. Liu, P. S. Dasgupta and S. Basu, Proc. Natl. Acad. Sci. U. S. A., 2011, 108, 20730–20735.
54. J. Zhang, Q. Cui, Y. Zhao, R. Guo, C. Zhan, P. Jiang, P. Luan, P. Zhang, F. Wang, L. Yang, X. Yang and Y. Xu, Atherosclerosis, 2020, 292, 99–111.
55. M. Reis, C. J. Czupalla, N. Ziegler, K. Devraj, J. Zinke, S. Seidel, R. Heck, S. Thom, J. Macas, E. Bockamp, M. Fruttiger, M. M. Taketo, S. Dimmeler, K. H. Plate and S. Liebner, J. Exp. Med., 2012, 209, 1611–1627.
56. Y. Lei, J. Xu, M. Xiao, D. Wu, H. Xu, J. Yang, X. Mao, H. Pan, X. Yu and S. Shi, Br. J. Cancer, 2024, 130, 1505–1516.
57. Y. Onogi, T. Wada, A. Okekawa, T. Matsuzawa, E. Watanabe, K. Ikeda, M. Nakano, M. Kitada, D. Koya, H. Tsuneki and T. Sasaoka, Sci. Rep., 2020, 10, 670.
58. M. Mooring, G. A. Yeung, P. Luukkonen, S. Liu, M. W. Akbar, G. J. Zhang, O. Balogun, X. Yu, R. Mo, K. Nejak-Bowen, M. V. Poyurovsky, C. J. Booth, L. Konnikova, G. I. Shulman and D. Yimlamai, Sci. Transl. Med., 2023, 15, e3157.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: https://doi.org/10.1039/d4tb02776k

---