Quercetin-loaded PCL scaffold suppresses bone metastatic tumors and augments in situ bone formation

Abstract

This study developed a dual-functional 3D-printed poly(ε-caprolactone) (PCL) scaffold loaded with quercetin (Q-PCL) for synergistic inhibition of bone metastatic tumors and promotion of bone regeneration. The scaffold was fabricated via melt extrusion, exhibiting a uniform porous structure conducive to sustained quercetin release. In vitro, Q-PCL significantly suppressed renal carcinoma (RENCA) cell proliferation and induced apoptosis, while enhancing the osteogenic differentiation and mineralization of bone marrow mesenchymal stem cells (BMSCs). In a murine subcutaneous tumor model, Q-PCL implantation effectively inhibited tumor growth via apoptosis induction without systemic toxicity. In a rat femoral defect model, the scaffold markedly accelerated bone repair, showing increased bone volume, improved trabecular morphology, and mature bone formation. The Q-PCL scaffold demonstrates great potential as a localized co-therapy strategy for treating osteolytic bone metastases and facilitating bone regeneration.

Supplementary Information

The online version contains supplementary material available at https://doi.org/10.1186/s12896-026-01198-y.

Introduction

The skeletal system, despite its robust and highly mineralized nature, represents a common and fertile site for metastasis for a range of primary carcinomas, most notably breast, prostate, lung, and renal cancers [1]. This predilection is largely attributed to the unique microenvironment of the bone marrow, which is rich in growth factors, cytokines, and possesses an extensive vascular network, creating a conducive soil for circulating tumor cells to seed and proliferate [2, 3]. The ensuing osteolytic or osteoblastic lesions severely compromise bone structural integrity, leading to devastating sequelae such as severe pain, hypercalcemia, spinal cord compression, and pathological fractures [4, 5]. These skeletal-related events not only drastically diminish the quality of life for patients but also signify a progressive, often incurable stage of the disease, presenting a formidable challenge in clinical oncology.

The current clinical paradigm for managing bone metastases primarily revolves around a multi-modal approach including radiotherapy, systemic chemotherapy, bisphosphonates, and the RANKL inhibitor denosumab [6, 7]. While these strategies can be effective in palliating pain and slowing tumor progression to some extent, they exhibit significant limitations. A critical, often unaddressed issue is the profound disconnect between anti-tumor therapy and bone regeneration. Radiotherapy and chemotherapy are cytotoxic and indiscriminately affect both cancerous and healthy cells, including the vital osteoprogenitors and mesenchymal stem cells (MSCs) essential for bone repair [8, 9]. Consequently, these treatments effectively create a void—inhibiting the tumor but simultaneously impairing the body’s innate capacity to heal the resultant bone defect [10]. This failure to achieve concomitant bone regeneration leaves the skeletal architecture weakened and susceptible to fractures, necessitating subsequent surgical interventions that are complex in the context of compromised bone and systemic disease. Therefore, developing a therapeutic strategy that can synchronously suppress tumor recurrence and orchestrate functional bone tissue repair at the metastatic site is an urgent, unmet medical need.

In the quest for such a dual-function agent, the natural flavonoid quercetin has emerged as a highly promising candidate. Quercetin, ubiquitously found in fruits and vegetables, has garnered considerable attention in recent years for its broad-spectrum anti-cancer properties. Its potent anti-tumor efficacy is mediated through a multitude of mechanisms [11, 12]. Induction of Cell Cycle Arrest and Apoptosis. Quercetin modulates key regulatory proteins leading to cell cycle arrest primarily at the G1/S and G2/M checkpoints. It concurrently upregulates pro-apoptotic proteins while downregulating anti-apoptotic ones, triggering mitochondrial-mediated apoptosis. Nhibition of Metastasis and Angiogenesis. It suppresses epithelial-to-mesenchymal transition by modulating signaling pathways, thereby reducing cell migration and invasion. Quercetin also inhibits VEGF expression and signaling, curtailing the formation of new tumor vasculature. Simultaneously and quite remarkably, quercetin demonstrates potent pro-osteogenic capabilities, making it an ideal agent for addressing the bone healing aspect [13]. Quercetin promotes the proliferation and differentiation of MSCs and osteoblastic cells. It upregulates the expression of critical osteogenic markers. Meanwhile, it enhances bone formation by activating the BMP-2/Smad and Wnt/β-catenin pathways, which are master regulators of osteogenesis. Moreover, Quercetin inhibits the formation and function of bone-resorbing osteoclasts by suppressing RANKL-induced signaling pathways, and by downregulating key osteoclastogenic factors. This dual action boosting bone formation while inhibiting bone resorption positions quercetin uniquely for restoring bone homeostasis in metastatic lesions.

However, the clinical translation of free quercetin is severely hampered by its inherent physicochemical limitations. Its poor aqueous solubility, low bioavailability, chemical instability, rapid systemic metabolism and clearance, and lack of target specificity prevent it from accumulating effectively at the bone metastatic site at a therapeutically relevant concentration for a sustained duration. Systemic administration would require impractically high doses, increasing the risk of off-target side effects without guaranteeing efficacy at the desired location.

This is where advanced biomaterial-based local drug delivery systems offer a paradigm-shifting solution. By encapsulating quercetin within a biocompatible scaffold, these systems can provide localized, sustained release directly at the pathological site, maintaining an optimal drug concentration, minimizing systemic exposure, and overcoming the pharmacokinetic drawbacks of the free drug. Among various biomaterials, poly(ε-caprolactone) (PCL) stands out as a particularly suitable candidate for this application [14]. PCL, an FDA-approved, synthetic polyester, possesses an exceptional combination of properties. PCL can provide the necessary mechanical support and load-bearing capacity required for bone tissue engineering, helping to restore structural integrity to weakened bone. Its degradation via hydrolysis of ester linkages is slow, ensuring mechanical stability throughout the critical bone regeneration period and enabling long-term, sustained release of the encapsulated therapeutic agent [15]. PCL is compatible with a wide range of drugs and is amenable to various processing techniques. And it has a long history of safe use in numerous medical devices and drug delivery systems [16].

The fabrication methodology of the PCL scaffold is paramount, as it dictates its architectural, mechanical, and biological properties. While numerous techniques exist, such as solvent casting/particulate leaching, electrospinning, and gas foaming, to achieve precise, predetermined, and complex scaffold geometries with high reproducibility, 3D printing is unequivocally the most advanced and suitable technique [17]. Bioprinting allows for the creation of scaffolds with meticulously controlled pore size, porosity, interconnectivity, and internal architecture, which are critical for facilitating nutrient diffusion, waste removal, cell infiltration, vascularization, and ultimately, effective tissue integration and regeneration. Furthermore, 3D printing enables the fabrication of patient-specific implants based on clinical CT scans, perfectly matching the bone defect’s contour. Most importantly for this study, it allows for the precise incorporation of bioactive molecules like quercetin into the printing process, enabling the development of a scaffold with spatiotemporally controlled release kinetics tailored for simultaneous anti-tumor therapy and osteogenesis.

Therefore, this study aims to design, fabricate, and evaluate a 3D-printed PCL scaffold loaded with quercetin for the combined therapy of bone metastatic tumors and the promotion of in situ bone regeneration. We hypothesize that the localized and sustained release of quercetin from the mechanically robust PCL scaffold will effectively inhibit tumor cell proliferation and induce apoptosis while concurrently stimulating the differentiation of osteogenic cells, thereby addressing the critical dual challenges of tumor recurrence and bone defect repair in a single, integrated platform.

Materials and methods

Fabrication of quercetin-loaded PCL (Q-PCL) scaffolds

The Q-PCL scaffolds were fabricated using a melt extrusion-based 3D printing system (Bioplotter, EnvisionTEC, Germany). Briefly, medical-grade Poly(ε-caprolactone) (PCL, Mn 80,000, Sigma-Aldrich) pellets and quercetin (≥ 95% purity, Sigma-Aldrich) were physically blended at a designated weight ratio (10:1 PCL: Quercetin) to ensure a homogenous mixture. The mixture was then loaded into a stainless-steel syringe cartridge and heated to 70 °C to achieve a viscous, printable melt. A calibrated nozzle (diameter: 200 μm) was used for extrusion. The printing process was conducted under a pressure of 5–6 bar and a printing speed of 4 mm/s. A pre-designed 3D model (a grid-like structure with 0/90° laydown pattern) was sliced using the manufacturer’s software, and scaffolds were printed layer-by-layer onto a cooled build platform to ensure rapid solidification. The pure PCL control scaffolds were printed under identical parameters. The printed scaffolds were subsequently freeze-dried for 24 h to remove any residual moisture and stored in a desiccator at -20 °C until further use. The macroscopic and microscopic morphology of the scaffolds was characterized using scanning electron microscopy (SEM).

Cell culture

Mouse renal adenocarcinoma cells (RENCA) and bone marrow-derived mesenchymal stem cells (BMSCs) were purchased from the American Type Culture Collection (ATCC). RENCA cells were maintained in RPMI-1640 medium, and BMSCs were cultured in α-MEM medium. Both media were supplemented with 10% fetal bovine serum (FBS, Gibco) and 1% penicillin/streptomycin. Cells were incubated at 37 °C in a humidified atmosphere of 5% CO₂.

Subcutaneous RENCA tumor model and treatment

All animal procedures were approved by the Institutional Animal Care and Use Committee of Jianyang People’s Hospital. 15 female Balb/c mice (6 weeks old, provided by Beijing Weitong Lihua Laboratory Animal Technology Co., Ltd.) were randomly divided into 3 groups (n = 5): (1) Tumor + No implant, (2) Tumor + PCL implant, (3) Tumor + Q-PCL implant. RENCA cells (1 × 10⁶ cells in 100 µL Matrigel) were injected subcutaneously into the right flank of the mice. When the tumor volume reached approximately 100 mm³, the mice in groups 2 and 3 were implanted with sterilized scaffolds adjacent to the tumor mass. Tumor volume (V=length×width²/2) and body weight were measured every 2 days.

After 16 days post-implantation, all mice were euthanized. Mice were euthanized by carbon dioxide inhalation in their home cages. Compressed CO₂ gas from a cylinder equipped with a pressure-reducing regulator and flow meter was introduced into the cage at a displacement rate of 50% of the cage volume per minute. CO₂ flow was maintained for at least one additional minute after the complete cessation of respiration to ensure death. Tumors and major organs (heart, liver, spleen, lungs, kidneys) were harvested for further analysis.

Rat femoral defect model and treatment

12 Sprague-Dawley rats (provided by Liaoning Changsheng Biotechnology Co., Ltd.) were used and randomly divided into three groups (n = 4). (1) Defect only (empty), (2) Defect + PCL implant, (3) Defect + Q-PCL implant. A critical-sized segmental defect (~ 3 mm) was created in the right femur of each rat under general anesthesia and sterile conditions. The defects in groups 2 and 3 were filled with the corresponding 3D-printed scaffold. The defect in group 1 was left empty. The incision was closed, and post-operative analgesia was administered.

Once the bone repair observation period was reached, the rats were euthanized. Sodium pentobarbital was administered at a dose of 200 mg/kg body weight via intraperitoneal injection. Rats were manually restrained, and pentobarbital solution was injected into the lower quadrant of the abdomen. Animals were then placed in a quiet, warm environment and observed until cessation of vital signs. Death was confirmed by ensuring the absence of a palpable apical heartbeat and the loss of corneal and pupillary reflexes for a minimum observation period of 10 min.

Statistical analysis

All quantitative data were obtained from at least three independent experiments and are presented as mean ± standard deviation (SD). Statistical analysis was performed using one-way or two-way analysis of variance (ANOVA) followed by Tukey’s post-hoc test in GraphPad Prism software. A value of p < 0.05 was considered statistically significant.

Results and discussion

Successful fabrication and characterization of 3D-printed Q-PCL scaffolds

To develop a localized co-therapy system for bone metastatic tumors, quercetin-loaded poly(ε-caprolactone) (Q-PCL) scaffolds were successfully designed and fabricated via a precise melt extrusion-based 3D printing technique. The overall conception of this strategy is illustrated in Fig. 1A. A homogeneous mixture of PCL and quercetin was first prepared, which was then utilized as the bio-ink for printing. The printing process yielded scaffolds with a well-defined, grid-like architecture, as confirmed by their general appearance shown in Fig. 1B. Both the pure PCL control scaffolds and the Q-PCL scaffolds exhibited a uniform macroscopic structure with consistent shape and size, demonstrating the robustness of the printing process even with the incorporation of the bioactive compound.

Fig. 1.

Design, fabrication, and morphological characterization of the 3D-printed Q-PCL scaffold. (A) Schematic illustration of the overall concept and fabrication workflow: homogeneous blending of PCL and quercetin followed by melt extrusion-based 3D printing to construct a dual-functional scaffold for inhibiting bone metastatic tumors and promoting in situ bone regeneration. (B) Gross macroscopic appearance of the PCL scaffold and the quercetin-loaded (Q-PCL) scaffold, demonstrating a well-defined grid-like architecture and structural uniformity. (C) Representative scanning electron microscopy (SEM) images showing the surface micro-topography and interconnected porous architecture of PCL and Q-PCL scaffolds. (D) Porosity and average pore size of the scaffolds. (E) Compressive modulus of the scaffolds. (F) FTIR spectra of pure PCL, pure quercetin, and Q-PCL scaffold. (G) In vitro degradation profiles of the scaffolds over 21 days, evaluated by weight loss percentage. (H) In vitro cumulative release profile of quercetin from the Q-PCL scaffold. Data are presented as mean ± SD (n = 3)

Further morphological characterization using scanning electron microscopy (SEM, Fig. 1C) revealed the detailed micro-topography of the scaffolds. The images confirmed that the 3D-printed scaffolds possessed a highly porous and interconnected network. The struts of both PCL and Q-PCL scaffolds were smooth and continuous, indicating stable extrusion during the printing process. No obvious structural defects were observed, and the incorporation of quercetin did not adversely affect the printability or the intrinsic micro-architecture of the PCL matrix. The pores, critical for cell infiltration, nutrient exchange, and vascularization, were clearly visible and consistent in size. This successful fabrication establishes a solid foundation for the subsequent in vitro and in vivo functional evaluations, providing a platform capable of local, sustained quercetin release while maintaining the mechanical integrity required for bone defect repair.

The quantitative physicochemical properties of the fabricated scaffolds were systematically characterized. The porosity of the scaffolds was measured using a solution-wetting method, revealing a porosity of 69.3 ± 2.05% for the Q-PCL scaffold with an average pore size of 1.32 ± 0.06 mm. The compressive modulus of the Q-PCL scaffold was measured using a universal testing machine, showing a value of 254.3 ± 7.77 MPa, which falls within the range of cancellous bone and confirms its ability to provide mechanical support for bone defects.

FTIR spectroscopy was performed to confirm successful drug loading. The FTIR spectrum of the Q-PCL scaffold showed characteristic peaks of quercetin alongside the typical PCL peaks, indicating successful incorporation without chemical degradation. An in vitro degradation study was conducted in PBS (pH 7.4, 37 °C) over 21 days. The results showed a gradual decrease in scaffold weight, with less than 10% weight loss at 21 days, indicating that the scaffold maintains structural integrity throughout the critical bone healing period. The in vitro quercetin release profile of the Q-PCL scaffold was evaluated over 21 days under physiological conditions. As shown in Fig. 1H, the release curve exhibited a biphasic pattern. An initial burst release of approximately 20% of the total loaded quercetin was observed within the first 24 h, which can be attributed to the rapid dissolution of drug molecules located on or near the scaffold surface. This was followed by a sustained and controlled release phase, during which the cumulative release gradually increased to 41% at day 21. The sustained release profile over the 21-day period indicates that the PCL matrix effectively serves as a reservoir for quercetin, providing continuous drug elution without an undesirable burst effect that could cause local toxicity. Importantly, this release kinetics aligns well with the therapeutic windows required for both sustained tumor suppression and long-term osteogenic stimulation, as the released quercetin maintains effective local concentrations throughout the critical period of bone regeneration and anti-tumor therapy.

The Q-PCL scaffold effectively suppresses RENCA cell proliferation in vitro

The antitumor efficacy of the quercetin-eluting scaffold was rigorously evaluated in vitro using mouse renal adenocarcinoma (RENCA) cells. A Live/Dead viability assay was first employed to provide a direct visual assessment of cell survival and cytocompatibility. As shown in Fig. 2A, after 24 h of culture, RENCA cells in pure PCL scaffolds group exhibited robust growth, evidenced by a confluent layer of cells fluorescing intensely green (calcein-AM, live cells), with only sporadic red fluorescence (ethidium homodimer-1, dead cells). In stark contrast, cells in the Q-PCL scaffolds group displayed a significantly different morphology and viability profile. The population of live cells was markedly reduced, and a substantial increase in red fluorescence was observed, indicating widespread cell death. This clear visual evidence demonstrated the potent cytotoxic effect of quercetin released from the Q-PCL scaffold on RENCA cells.

Fig. 2.

In vitro dual-functional evaluation: Antitumor efficacy and osteogenic induction of the Q-PCL scaffold. (A) Live/Dead staining of RENCA cells cultured on PCL and Q-PCL scaffolds for 24 h. Viable cells are stained green (Calcein-AM) and dead cells red (Ethidium homodimer-1). (B) Immunofluorescence staining for the proliferation marker Ki67 in RENCA cells. (C) Cytoskeletal organization and cell adhesion of BMSCs visualized by F-actin staining (red) after 24 h of culture. (D) Wound healing assay assessing BMSC migration capacity. Gap closure at 0,12, and 24 h. (E) Alkaline Phosphatase (ALP) staining of BMSCs cultured under different conditions for3 and 7 days. The intensified blue-purple staining indicates elevated ALP activity and early osteogenic commitment. (F) Alizarin Red S (ARS) staining for extracellular matrix mineralization at days 7 and 14. Significantly enhanced staining intensity and coverage area, indicative of accelerated late-stage osteogenic differentiation and matrix calcification. Data are presented as mean ± SD (n = 3). ***p<0.001; n.s., not significant

To further quantify the anti-proliferative effect, we performed immunofluorescence staining for Ki67, a nuclear protein universally expressed in all active phases of the cell cycle and a definitive marker of cellular proliferation. Representative fluorescence images (Fig. 2B) showed a high density of Ki67-positive nuclei within the confluent cell in the PCL scaffolds group. Conversely, the number of Ki67-positive cells in the Q-PCL scaffolds group was drastically diminished. Confirmed this observation, revealing a statistically significant reduction in the proliferation index of RENCA cells cultured in Q-PCL scaffolds group compared to those in PCL and control group. That conclusively demonstrates that the sustained release of quercetin from the 3D-printed scaffold effectively inhibits the proliferation of RENCA tumor cells in vitro.

The Q-PCL scaffold demonstrates excellent osteogenic promotion capacity in vitro

The dual-functionality of our design necessitates that the scaffold not only inhibits tumor growth but also actively promotes bone regeneration. We therefore investigated the osteo-inductive potential of the Q-PCL scaffold using bone marrow-derived mesenchymal stem cells (BMSCs).

The initial cell-material interaction was assessed by examining cell adhesion and cytoskeletal organization through F-actin staining. As depicted in Fig. 2C, BMSCs cultured in Q-PCL adhered and spread well after 24 h. Cells on the Q-PCL scaffolds exhibited a more pronounced stretched and elongated morphology with well-organized stress fibers, suggesting enhanced adhesion and early activation, potentially primed by the released quercetin.

The effect on cellular migration, a critical determinant in bone defect bridging and tissue integration, was further evaluated using a wound healing assay. As illustrated in Fig. 2D, the Q-PCL group exhibited a pronounced acceleration of BMSC migration compared to the Con and pristine PCL groups. Notably, at the 12-hour time point, cells exposed to Q-PCL scaffolds had already commenced efficient gap closure, demonstrating a superior migration rate. This pro-migratory effect became substantially more pronounced at the 24-hour interval, wherein the Q-PCL group achieved near-complete confluence of the denuded area, significantly outpacing the limited lateral motility observed in the control and unmodified PCL counterparts. This trend suggests that the sustained bioavailability of quercetin from the scaffold matrix not only promotes initial adhesion but also actively orchestrates the chemotactic recruitment of osteoprogenitors essential for subsequent osteogenesis.

We next evaluated the early osteogenic differentiation by measuring Alkaline Phosphatase (ALP) activity, a key early-stage marker, after 3 and 7 days in different group. Visual observation of ALP staining (Fig. 2E) revealed that ALP activity was significantly enhanced in the BMSCs cultured on the Q-PCL scaffolds compared to those on the pure PCL scaffolds. This indicates that the quercetin released from the scaffold effectively accelerates the commitment of BMSCs to the osteoblastic lineage.

To further substantiate the osteoinductive potential of the Q-PCL scaffolds, Alizarin Red S (ARS) staining was performed to evaluate extracellular matrix mineralization and calcium nodule formation, a definitive marker of late-stage osteogenic differentiation. As shown in Fig. 2F, a discernible enhancement in mineral deposition was observed in the Q-PCL group as early as day 7, in stark contrast to the negligible staining detected in both the pristine PCL and the non-treated Control groups. This early promotion of calcified nodule formation underscores the potent bioactivity conferred by the sustained release of quercetin from the fibrous matrix. Notably, this disparity in osteogenic efficacy became markedly more pronounced by day 14. The Q-PCL group exhibited extensive, densely packed calcium nodules with significantly elevated ARS staining intensity and coverage area, indicating robust progression into the terminal differentiation phase. Conversely, while the PCL scaffold alone demonstrated a modest intrinsic capacity to support mineralization over time, the magnitude of calcium accretion remained substantially inferior to that observed in the Q-PCL constructs. These quantitative and qualitative findings in mineralization align closely with the enhanced cytoskeletal organization noted earlier, suggesting that the quercetin-primed adhesion and spreading state may potentiate downstream osteogenic cascades, culminating in accelerated and augmented matrix calcification. Collectively, these data provide compelling histochemical evidence that the incorporation of quercetin into PCL scaffolds effectively orchestrates a favorable microenvironment for expedited osteogenic maturation.

Tumor inhibition efficacy

Digital caliper measurements revealed a dramatic difference in tumor growth kinetics among the groups over the 16 days treatment period (Fig. 3A). As shown in the tumor growth curves (Fig. 3B), tumors in the Con group grew rapidly and aggressively. Implantation of a PCL group resulted in a modest, statistically insignificant reduction in tumor volume compared to Con group, which can be attributed to minor physical interference or the innate biological activity of PCL degradation products. In stark contrast, local implantation of the Q-PCL group effectively suppressed tumor growth throughout the observation period. The final tumor volume in the Q-PCL group was significantly smaller than those in both the Con and PCL groups, providing direct visual and quantitative evidence of the potent in vivo antitumor capability of the quercetin-eluting scaffold.

Fig. 3.

In vivo evaluation of the Q-PCL scaffold in a subcutaneous RENCA tumor model. (A) Representative macroscopic images of excised tumors from the Control (tumor only), PCL implant, and Q-PCL implant groups at the experimental endpoint (Day 16). (B) Quantitative analysis of tumor growth kinetics over the 16-day treatment period. Tumor volume was measured every two days. Data are presented as mean ± SD. (C) qPCR analysis of gene expression profiles in tumor tissues. (D) Hematoxylin and Eosin (H&E) staining of tumor tissue sections. (E) H&E staining of major organs (heart, liver, spleen, lung, kidney) harvested post-treatment. Data are presented as mean ± SD (n = 3). **p<0.01;***p<0.001; n.s., not significant

The therapeutic efficacy at the molecular level was further substantiated by qPCR analysis. As shown in Fig. 3C, treatment with this regimen resulted in a pronounced downregulation of Ki67 and VHL gene expression, indicating effective suppression of cellular proliferation and modulation of the hypoxia-sensing pathway. In stark contrast, the expression levels of the tumor suppressor genes Sirt6 and p53 were significantly elevated. This reciprocal transcriptional shift towards a tumor-suppressive phenotype suggests a restoration of genomic stability and activation of senescence-associated pathways, thereby corroborating the observed therapeutic outcome.

Histopathological and apoptotic analysis

Histological examination via H&E staining of the excised tumors (Fig. 3D) provided further insight. Tumors from Con and PCL exhibited typical malignant histoarchitecture, characterized by high cellular density, large nuclei, and frequent mitotic figures. In contrast, tumor tissues from the Q-PCL group showed extensive areas of necrosis, characterized by pyknotic nuclei, eosinophilic cytoplasm, and loss of cellular structure, alongside visible regions of apoptosis.

Systemic safety evaluation

The systemic biocompatibility and potential off-target toxicity of the treatment were assessed by monitoring body weight and conducting histopathological analysis of major organs. H&E staining of vital organs (heart, liver, spleen, lungs, and kidneys) harvested at the endpoint (Fig. 3E) revealed normal tissue architecture, clear cellular boundaries, and no signs of pathological lesions, necrosis, or inflammatory infiltration in any of the treatment groups, including Q-PCL. This confirms that the local implantation and drug release from the Q-PCL scaffold achieved potent tumor inhibition in situ without causing detectable damage to distant organs, underscoring its excellent safety profile.

The Q-PCL scaffold promotes excellent bone repair in a rat femoral condyle defect model

The critical osteogenic potential of the Q-PCL scaffold for bone regeneration was further validated in a critical-sized rat femoral condyle defect model in vivo. The bone repair process was systematically evaluated through radiographic, micro-architectural, and histological analyses.

Representative digital radiography (DR) images acquired at 12 weeks post-implantation are presented in Fig. 4A. In the empty control group (Con), the defect boundary remained clearly visible with minimal radio-opaque signal within the defect site, indicating a failure of substantial bone regeneration. The group implanted with the pure PCL scaffold showed a slightly increased radio-opacity at the periphery of the defect, suggesting a moderate level of new bone formation, but the central region of the defect remained largely radiolucent. In striking contrast, the defects treated with the Q-PCL scaffold exhibited the most pronounced radiopacity. The defect margins were largely obscured, and a significant amount of continuous, highly radio-opaque tissue bridged the defect site, providing primary evidence of successful bone repair.

Fig. 4.

In vivo assessment of bone regeneration capacity in a rat femoral defect model. (A) Digital Radiography (DR) images of the femoral defect site at 12 weeks post-implantation. (B) Three-dimensional micro-CT reconstruction images. (C) Quantitative morphometric analysis of the newly formed bone microarchitecture. Parameters include Bone Volume/Tissue Volume (BV/TV), Trabecular Thickness (Tb.Th), Trabecular Number (Tb.N), and Trabecular Separation (Tb.Sp). (D) Histological evaluation via Hematoxylin and Eosin (H&E) staining. (E) Masson’s Trichrome staining differentiating mature bone (red) from immature osteoid and collagen (blue). (F) qPCR analysis of gene expression profiles in osteogenic. Data are presented as mean ± SD (n = 3). **p<0.01;***p<0.001; n.s., not significant

Subsequent three-dimensional micro-CT reconstruction images (Fig. 4B) provided a more detailed visualization of the newly formed bone microstructure within the defects. The 3D renderings clearly showed that the Q-PCL group possessed the most substantial and continuous bone tissue growing into and throughout the defect area, closely integrating with the host bone. The morphology of the new bone appeared more mature and structurally organized compared to the other groups.

Quantitative morphometric analysis of the region of interest (ROI) (Fig. 4C) provided robust statistical evidence to support the observations. The Q-PCL group demonstrated a significantly higher Bone Volume/Tissue Volume (BV/TV) ratio compared to both the PCL and control groups, indicating a greater volume of mineralized tissue deposited within the defect. Furthermore, the trabecular parameters were significantly improved in the Q-PCL group. The Trabecular Thickness (Tb.Th) and Trabecular Number (Tb.N) were markedly increased, while the Trabecular Separation (Tb.Sp) was significantly decreased (p < 0.05 for all parameters compared to Control and PCL groups). These data conclusively demonstrate that the Q-PCL scaffold not only enhances the quantity of new bone formation but also improves its quality by promoting a denser and more interconnected trabecular network.

Histological analysis

Histological assessment via Hematoxylin and Eosin (H&E) and Masson’s Trichrome staining provided further insight into the nature and maturity of the regenerated tissue at the cellular level. H&E staining (Fig. 4D) revealed that the defect area in the control group was predominantly filled with fibrous tissue (FT). The PCL group showed islands of newly formed woven bone (NB) mainly at the defect edges, but the central region still contained abundant fibrous tissue. In the Q-PCL group, the defect was almost entirely bridged by a significant amount of new bone tissue with a more mature, lamellar-like structure. The presence of numerous osteoblasts (OB) lining the bone surfaces and osteocytes (OCY) embedded within the bone matrix indicated active, ongoing osteogenesis. Masson’s Trichrome staining (Fig. 4E), which distinguishes mature bone (red) and newly formed collagenous matrix/osteoid (blue), provided complementary evidence. The Q-PCL group exhibited the most extensive area of staining, with a large proportion of mature red bone and less blue osteoid, indicating advanced remodeling and maturation of the newly formed bone. In contrast, the control and PCL groups showed predominantly blue-stained fibrous tissue or immature osteoid, with minimal mature bone formation. The integrated results from DR, micro-CT, and histological analyses unanimously demonstrate that the localized and sustained release of quercetin from the 3D-printed PCL scaffold significantly accelerates and enhances the process of bone regeneration, leading to the repair of critical-sized bone defects with mature, well-structured bone tissue.

Q-PCL scaffold promotes osteogenic differentiation of BMSCs at the molecular level

To provide quantitative evidence for the osteogenic capacity of the Q-PCL scaffold at the molecular level, we performed qPCR analysis to evaluate the mRNA expression levels of key osteogenic markers in bone tissue from the PCL and Q-PCL scaffold groups. As shown in Fig. 4F, the expression levels of Runx2, OCN, OPN, and BMP2 were all significantly upregulated in the Q-PCL group compared to the pure PCL group. Specifically, BMP2 (by 3.6-fold, p < 0.01), Runx2 (by 4.0-fold, p < 0.01), OCN (by 2.5-fold, p < 0.01), and OPN (by 13.1-fold, p < 0.001) compared to the pure PCL group at day 14. Among these, Runx2 serves as a master transcription factor governing osteoblast differentiation and bone formation, while OCN and OPN are late-stage markers of osteogenic maturation and matrix mineralization. BMP2 is a critical morphogen that initiates the osteogenic signaling cascade. The significant and time-dependent upregulation of these genes in the Q-PCL group provides robust molecular evidence that the sustained release of quercetin from the scaffold effectively promotes the osteogenic differentiation of BMSCs, which is consistent with the enhanced ALP activity and mineralized matrix formation observed in the previous experiments. These quantitative gene expression data strongly support the conclusion that the Q-PCL scaffold creates a favorable microenvironment for bone regeneration.

Limitations and future directions

It should be acknowledged that the present study did not include a free quercetin control group for direct comparison with the Q-PCL scaffold. This omission was primarily based on the consideration that free quercetin, due to its poor aqueous solubility, rapid metabolism, and short circulation half-life, would be unlikely to achieve sustained therapeutic concentrations at the target site following direct administration. Furthermore, free quercetin would not provide the mechanical support required for bone defect repair. However, we recognize that a direct comparative study would more rigorously demonstrate the advantages of scaffold-based localized delivery over conventional drug administration. Future work will incorporate such comparisons using optimized formulation strategies, such as liposomal encapsulation or nanoparticle-based delivery, to enable a fair assessment of scaffold-based versus free drug efficacy. We also note that several recent studies have similarly highlighted the challenges of translating free flavonoid compounds into effective in vivo therapies, underscoring the value of biomaterial-based delivery systems.

The Q-PCL scaffold offers several strengths, including precise architectural control via 3D printing, mechanical properties matching cancellous bone, sustained quercetin release over 21 days, and a favorable safety profile. Compared to advanced systems reported in the literature, however, our system has limitations. It relies solely on single-agent quercetin delivery without incorporating additional therapeutic modalities (e.g., photothermal or stimuli-responsive release), PCL’s hydrophobicity may limit initial cell adhesion compared to more bioactive materials, and the absence of imaging capabilities prevents non-invasive tracking of scaffold performance [18–24].

Nevertheless, our system offers distinct advantages in clinical translatability, including the use of FDA-approved materials and patient-specific customization via 3D printing, while achieving the critical dual functionality of tumor suppression and osteogenic promotion.

Conclusion

In conclusion, this study successfully designed and fabricated a dual-functional quercetin-loaded PCL (Q-PCL) scaffold via melt extrusion 3D printing for the combined therapy of bone metastatic tumors and subsequent bone regeneration. The scaffold exhibited a uniform and interconnected porous structure, ensuring mechanical support and facilitating sustained quercetin release. In vitro experiments demonstrated that the Q-PCL scaffold effectively inhibited the proliferation of RENCA tumor cells by inducing apoptosis while simultaneously promoting the adhesion, osteogenic differentiation, and mineralization of BMSCs. In vivo evaluations further confirmed its potent efficacy: in a subcutaneous tumor model, local implantation of the Q-PCL scaffold significantly suppressed tumor growth through enhanced apoptosis without inducing systemic toxicity. In a critical-sized rat femoral defect model, the Q-PCL scaffold markedly enhanced bone repair, as evidenced by increased bone volume, improved trabecular architecture, and extensive formation of mature bone tissue. These findings collectively indicate that the 3D-printed Q-PCL scaffold serves as an efficient localized co-therapy system, effectively addressing the dual challenges of inhibiting tumor recurrence and promoting bone regeneration. This work provides a promising biomaterial-based strategy for the clinical treatment of osteolytic bone metastases and critical bone defects.

Supplementary Information

Below is the link to the electronic supplementary material.

Abbreviations

3D

Three-dimensional

ALP

Alkaline phosphatase

ARS

Alizarin Red S

BMP-2

Bone morphogenetic protein-2

BMSCs

Bone marrow mesenchymal stem cells

BV/TV

Bone volume/tissue volume

CO₂

Carbon dioxide

DAPI

4′,6-diamidino-2-phenylindole

DR

Digital radiography

F-actin

Filamentous actin

FBS

Fetal bovine serum

H&E

Hematoxylin and eosin

Micro-CT

Micro-computed tomography

PCL

Poly(ε-caprolactone)

Q-PCL

Quercetin-loaded poly(ε-caprolactone)

qPCR

Quantitative polymerase chain reaction

RENCA

Renal adenocarcinoma

ROI

Region of interest

SEM

Scanning electron microscopy

Tb.N

Trabecular number

Tb.Sp

Trabecular separation

Tb.Th

Trabecular thickness

VEGF

Vascular endothelial growth factor

Author contributions

L.Y. and J.Z. conceived and designed the study. D.Z, X.X. and R.Q. performed the experiments and data collection. X.Y. and M.W. analyzed and interpreted the data. M.W., C.J. and J.Z. provided critical revision of the manuscript for important intellectual content. All authors reviewed the manuscript.

Funding

This study was supported by the the second batch of clinical Scientific Research Fund project of Chengdu Medical College in 2024 (24LHJGTY-01); Sichuan Pharmaceutical Association New Quality In review Pharmacy Start-up Plan Youth Project (scyxh20240709); Research Project of Jianyang People’s Hospital (JY202349); Research and Development Project of Chengdu Science and Technology Bureau (2024-YF05-02542-SN).

Data availability

No datasets were generated or analysed during the current study.

Declarations

Ethics approval and consent to participate

This study was approved by the Institutional Animal Care and Use Committee (IACUC) of Jianyang People’s Hospital. All animal procedures were performed in compliance with the ARRIVE guidelines and the National Institutes of Health Guide for the Care and Use of Laboratory Animals. All efforts were made to minimize animal suffering and to reduce the number of animals used. This study does not contain any studies with human participants performed by any of the authors; therefore, consent to participate is not applicable.

Consent for publication

Not applicable. This manuscript does not contain any individual person’s data in any form, including individual clinical details, images, or videos.

Competing interests

The authors declare no competing interests.