Therapeutic efficacy of WS nanosheets combined with CT-guided microwave ablation in rabbit VX2 liver tumor model.

Introduction
Hepatocellular carcinoma (HCC) represents a highly prevalent and aggressive malignancy in China, exhibiting persistently high incidence and mortality rates. It is one of the most common malignant tumors of digestive system, with the sixth incidence and the third mortality.1
The therapeutic mechanism of microwave ablation (MWA) is based on thermally induced coagulative necrosis to destroy tumor tissue. Its core principle relies on the interaction between high-frequency electromagnetic energy (with a frequency exceeding 900 MHz) and the protons in water molecules, causing rapid polarity reversal within the molecules (hydrogen atoms becoming positively charged and oxygen atoms negatively charged). This polarity switching occurs at an extremely high frequency of 2–5 billion times per second, generating intense molecular friction and thereby producing sufficient heat to induce tissue coagulative necrosis.2 Computed tomography (CT)-guided MWA has been established as a minimally invasive therapeutic modality and is currently recognized as one of the radical treatment options for small HCC. This technique characterized by rapid temperature elevation, short treatment duration, excellent repeatability, and minimal tissue trauma.3,4 Comparative studies have demonstrated that MWA achieves therapeutic outcomes comparable to surgical resection in selected early-stage HCC patients. Importantly, MWA promotes tumor antigen release and enhances antigen-specific T cell immune responses, thereby activating or potentiating systemic antitumor immunity.5,6 This is because cancer development is closely associated with tumor cells evading immune surveillance and sustaining proliferation/differentiation, while postoperative tumor recurrence also involves aberrant changes in the immune system within the tumor microenvironment. Notably, cytotoxic T lymphocyte (CD8+ T cell) serve as the primary effector cells in antitumor immune responses.7,8,9 Consequently, comprehensive evaluation of tumor progression necessitates integrated assessment of host immune status.
Nevertheless, significant challenges remain in ablating tumors located in anatomically complex regions (e.g., adjacent to major blood vessels, bile ducts, or the diaphragm).10 On the one hand, energy delivery is difficult to control precisely, which may cause damage to surrounding healthy tissues. On the other hand, the temperature distribution within the ablation zone is uneven, with the highest temperature concentrated near the ablation needle.11 As the distance increases, the temperature gradually decreases, forming a transitional zone with sublethal effects that increases the risk of tumor residue. While expanding the ablation margin could address this issue, it risks damaging critical adjacent structures.
In recent years, the application of nano-synthetic materials in tumor therapy has attracted significant attention. Tungsten disulfide (WS2), as a typical two-dimensional transition metal dichalcogenide (TMD), has garnered considerable interest due to its advantages such as ease of functionalization, tunable structural morphology, adjustable specific surface area, and outstanding adsorption properties.12 In particular, WS2 exhibits excellent photothermal properties,13 Yong et al.14 demonstrated that WS2 nanosheets as photosensitizer carriers could significantly inhibit Helen Lane (HeLa) cell viability in combination with photothermal therapy, with cytotoxicity showing laser intensity-dependent enhancement. Our preliminary experiments have revealed that WS2 nanosheets can effectively absorb microwave energy, and when combined with MWA, significantly improve complete tumor necrosis rates while enhancing CD8+ T cell infiltration in vivo. Building upon these findings, this study employed a rabbit VX2 liver tumor model (a transplantable squamous cell carcinoma) to investigate whether WS2 nanosheets can enhance the therapeutic efficacy of CT-guided MWA without expanding the ablation zone. The investigation was conducted using magnetic resonance imaging (MRI), CT, and histopathological analysis.

Results

Synthesis and characterization of WS2 nanosheets
The WS2 nanosheets, predominantly monolayer, were synthesized via a liquid-phase exfoliation approach employing H2SO4 intercalation followed by ultrasonication. Transmission electron microscopy (TEM) analysis confirms that the as-prepared WS2 nanosheets exhibit a size of approximately 50 nm. The high-angle annular dark-field (HAADF) image demonstrates the uniform distribution and coexistence of W and S elements in WS2 nanosheets(Figure 1A). Energy dispersive spectroscopy (EDS) elemental analysis further reveals the composition and spatial distribution of W and S (Figure 1B). The purity and crystal structure of WS2 were characterized by X-ray diffraction (XRD). As shown in Figure 1C, all diffraction peaks correspond exclusively to the characteristic patterns of WS2 (JCPDS card no. 08-0237). In deionized water, WS2 exhibited an average particle size of approximately 606.59 nm (Figure 1E), with zeta potential measurements revealing a surface charge of −35.8 mV (Figure 1D). Under 532-nm laser excitation, the Raman spectrum of WS2 exhibits two dominant peaks: the E2g1 mode at 348 cm−1 and the A1g mode at 414 cm−1, with a peak separation of 66 cm−1 (Figure 1F). The chemical composition and valence states were further analyzed by X-ray photoelectron spectrometer (XPS). The survey spectrum (Figure 1G) verifies the presence of W and S elements, while the high-resolution spectra of W4f (Figure 1H) and S2p (Figure 1I) reveal their bonding states. Furthermore, surface functional groups on WS2 were characterized by Fourier transform infrared spectroscopy (FTIR), where the characteristic vibrational peak in the range of 800–600 cm−1 confirmed the presence of W–S bonds (Figure 1J).

Microwave thermal effect of WS2 nanosheets
Using pure water as a negative control, we systematically evaluated the thermal effects of WS2 nanosheets solutions at different concentrations (50–200 μg/mL) under microwave irradiation (12 W) for 3 min. The experimental results demonstrated that the temperature rise rate of WS2 nanosheets solutions was positively correlated with their concentration, with the 200 μg/mL WS2 solution exhibiting a significant temperature increase of 26°C after 3 min, while the pure water control group only rose by 11°C (Figures 2A and 2B). Further studies revealed that, at a fixed WS2 concentration (100 μg/mL), the final temperature of the solution increased with higher microwave power densities (5–15 W) (Figures 2C and 2D). Additionally, through five heating/cooling cycle experiments, the 100 μg/mL WS2 solution exhibited excellent thermal stability under 12 W microwave irradiation, with highly consistent temperature profiles across multiple cycles (Figure 2E).

WS2 nanosheets for CT imaging and liver tumor model preparation
We obtained a series of CT images using six different concentrations of WS2 (0–12 mg/mL) dispersed in deionized water (Figure 3A), demonstrating a clear concentration-dependent increase in Hounsfield Unit (HU) values as shown in Figure 3B.
Twenty healthy New Zealand white rabbits underwent CT-guided percutaneous implantation of VX2 tumors in liver parenchyma, with 16 successfully established VX2 liver cancer models meeting inclusion criteria (the success rate was 80%, 16/20), 3 cases showed extrahepatic metastases (15%, 3/20) and 1 rabbit died within 1 week post-implantation (5%, 1/20). All three experimental groups successfully received their designated treatments (Figure 3C). The results at 7 days post-operation showed that the body weight of rabbits in all experimental groups exhibited an increasing trend, with no significant differences between groups (Figure 3D).

Comparable ablation zone between MWA and combination group
Preoperative tumor diameters measured via picture archiving and communication system (PACS) showed comparable mean hepatic tumor sizes across groups (WS2 group: 1.79 ± 0.25 cm; MWA group: 1.78 ± 0.19 cm; MWA/WS2 group: 1.79 ± 0.24 cm), with no statistically significant intergroup differences (p > 0.05) (Figure 4A). Immediate post-ablation (Figure 4C) and 1-week follow-up (Figure 4D) non-contrast plus contrast-enhanced CT scans of rabbit livers were analyzed via PACS, revealing no statistically significant differences in ablation zone volumes between the MWA and MWA/WS2 groups at either timepoint (p > 0.05). The delta changes in ablation volumes (1-week measurement minus immediate post-ablation measurement) similarly showed no intergroup significance (p > 0.05) (Figure 4B).

Tumor necrosis and apoptosis across treatment groups
Furthermore, histopathological analysis of tumor necrosis within the ablation zones revealed significant intergroup differences in mean necrosis ratios(Figure 5A). There was also a significant difference in the level of apoptosis within the ablation zone (Figure 5C). The MWA/WS2 combination group demonstrated a significantly lower mean necrosis ratio and higher apoptosis rate compared to both WS2 and MWA monotherapy groups (p < 0.05), indicating enhanced tumor necrosis and apoptosis (Figure 5D). No significant difference was observed between WS2 and MWA groups (p > 0.05). Grading assessment (++++ to ++) showed 80% of MWA/WS2 cases achieved maximal necrosis (++++), markedly surpassing the MWA group (20%) with no ++++ cases in the WS2 group (Figure 5B).

Comparison of CD8+ T cell expression across treatment groups
Finally, we quantitatively analyzed CD8+ T cell expression levels within post-treatment ablation zones and splenic tissues across all experimental groups (Figure 6A). As demonstrated in the figures, significant differences in integrated optical density (IOD) values of CD8+ T cell were observed among the WS2, MWA, and MWA/WS2 groups in both ablation sites and spleen. The data revealed that the MWA/WS2 combination group exhibited significantly higher CD8+ T cell infiltration in ablation zones and systemic CD8+ T cell activation in splenic tissues compared to either WS2 or MWA monotherapy groups (p < 0.01), whereas no significant difference was detected between WS2 and MWA groups (p > 0.05) (Figure 6B).

In vivo biocompatibility and biosafety assessment
The results of hematological tests and hematoxylin and eosin (H&E) staining analysis of major organs in experimental rabbits showed that there were no statistically significant differences in blood routine parameters among the groups (p > 0.05) (Figure 7A). Additionally, all organs exhibited normal cellular morphology, intact tissue structure, and no obvious pathological damage (Figure 7B).

Discussion
Although CT-guided percutaneous MWA is a widely adopted minimally invasive intervention for HCC, its clinical application remains limited by high recurrence rates when treating tumors in anatomically challenging hepatic locations.15,16 To address this, we established a rabbit VX2 paravertebral tumor model to systematically evaluate the impact of WS2 nanosheets on ablation zone dimensions and therapeutic outcomes in hepatic MWA. The research results indicate that although the combined therapy did not significantly alter the ablation zone, it markedly increased the degree of tumor necrosis and apoptosis and demonstrated a clear advantage in immune response in the combination treatment group. This finding suggests that WS2 nanosheets can enhance the microwave thermal effect of MWA and, by improving local therapeutic efficacy, activate systemic antitumor immune responses, thereby providing a potential approach for HCC treatment.

Microwave thermal effect of WS2 nanosheets and their clinical application prospects
First, we characterized the synthesized WS2 nanosheets. TEM and EDS analysis reveal that the WS2 nanosheets have a diameter of approximately 50 nm, with uniform distribution of W and S elements. XRD analysis confirmed the high purity of the synthesized WS2 nanosheets with no detectable impurity peaks. The characteristic peaks observed in the Raman spectrum were consistent with previously reported data for monolayer WS2 nanosheets, further verifying the single-layer structure.17 XPS survey and high-resolution spectra confirmed the presence of W and S elements, with binding energy positions matching the standard chemical states of WS2. Additionally, FTIR spectroscopy detected characteristic W–S bond vibration peaks in the 800–600 cm−1 range, which showed excellent agreement with existing research results, providing further evidence for the successful synthesis of WS2 nanosheets.18
In vitro microwave thermal effect experiments demonstrated that WS2 exhibits remarkable microwave-sensitizing effects and excellent thermal stability under microwave irradiation. This phenomenon may be attributed to WS2 nanosheet’s ability to significantly enhance microwave absorption efficiency and effectively convert microwave energy into thermal energy, providing a theoretical foundation for its combined application with MWA.19

Balancing ablation zone and therapeutic efficacy
In vivo experiments demonstrated that incorporating WS2 material into MWA therapy did not lead to an expansion of the ablation zone. This phenomenon may be attributed to the unique microwave energy absorption characteristics of WS2 nanosheets, which enhance energy absorption in the central ablation area while reducing peripheral absorption effects.19 This property allows clinicians to precisely control the ablation range by adjusting power and duration parameters, thereby facilitating preoperative treatment planning and enabling more scientifically determined ablation technical parameters. Furthermore, the WS2-MWA combination group showed significantly higher degree of tumor necrosis and apoptosis compared to WS2-only or MWA-only control groups, demonstrating superior therapeutic efficacy that enables more complete tumor ablation and better recurrence control. Blood tests and H&E staining confirmed the material’s excellent biocompatibility in all experimental rabbits.

Activation of immune response by WS2 nanosheets combined with microwave ablation
Previous studies have established that MWA monotherapy can effectively suppress primary tumors while simultaneously activating systemic antitumor immunity, including specific T cell activation.20,21 When combined with immunotherapy, this approach demonstrates durable antitumor effects.22 Building upon these findings, our current study reveals that WS2 nanosheet-augmented MWA therapy significantly enhances CD8+ T cell infiltration in both the hepatic ablation zone and splenic tissue. This demonstrates that during treatment, the combination of WS2 nanosheets with MWA not only directly induces tumor cell death through thermal effects, but also enhances systemic antitumor immune surveillance by activating a more robust immune response. MWA induces tumor antigen release while simultaneously activating specific immune responses, and the incorporation of WS2 nanosheets likely enhances immune system reactions through their unique surface properties and functionalization.23 These results are consistent with previous studies demonstrating that hyperthermia or MWA can stimulate antitumor immune responses through tumor antigen release.24 The superior CD8+ T cell infiltration observed in the combination treatment group indicates that the combined use of WS2 nanosheets and MWA contributes to the remodeling of the tumor immune microenvironment and may open new avenues for cancer immunotherapy.
Currently, various combination strategies of MWA combined with immunotherapy have been explored in preclinical and clinical studies. MWA releases tumor antigens through thermal effects, activating dendritic cells and CD8+ T-cells to trigger an antitumor immune response. When combined with immunotherapy (e.g., programmed cell death protein 1 inhibitors), it alleviates immune suppression, enhances T cell function, and synergistically controls primary and metastatic lesions.25,26 Advantages include local and systemic synergy, potential induction of immune memory, and ease of clinical translation. However, limitations include transient immune responses, risks of immune overactivation, and variable efficacy depending on individual patient factors.
MWA combined with WS2 leverages its photothermal/microwave-sensitizing properties to enhance localized thermal efficiency, expand ablation coverage, and reduce damage. WS2 also releases metal ions and reactive oxygen species (ROS) to promote antigen release and dendritic cell (DC) activation.27 Compared to immunotherapy, WS2 directly enhances physical tumor destruction rather than relying on the immune system. Although it can modulate the tumor microenvironment, its mechanisms are less defined than immune checkpoint inhibitors, making it more suitable for tumors requiring localized therapeutic enhancement.

Advantages of nanomaterials in CT-guided theranostics and disadvantages of CT-guided tumor implantation
The study utilized CT-guided percutaneous tumor puncture for WS2 delivery, a method sharing similar targeting advantages with intratumoral chemotherapy (ITC), which effectively enhances local WS2 concentration while minimizing systemic metabolism.28 Simultaneously, experimental results revealed that the CT values of WS2 increase with concentration, primarily due to the high atomic number of tungsten (W) in the nanosheets, which significantly enhances X-ray attenuation.29 Consequently, WS2 shows potential as a CT contrast agent, enabling clear imaging localization in CT-guided MWA procedures. This property allows precise delivery of WS2 into tumors and facilitates real-time visualization, thereby substantially improving the operational accuracy of WS2-enhanced MWA therapy.
The higher incidence of extrahepatic metastasis associated with CT-guided liver tumor implantation may be related to several technical factors. First, the relatively large needle gauges commonly used in clinical practice (typically 18G–20G) make it difficult to precisely avoid small intrahepatic vessels during transhepatic puncture, increasing the risk of tumor cell dissemination through vascular pathways when small vessels are inadvertently penetrated.30 Second, while using smaller-gauge needles (e.g., 22G–24G) could theoretically reduce vascular injury rates, this approach leads to increased tumor tissue retention within the needle lumen, consequently decreasing the success rate of tumor implantation. This technical dilemma underscores the importance of meticulous puncture pathway planning during tumor implantation procedures to avoid blood vessels of all sizes while maintaining both transplantation efficiency and minimally invasive precision.

Limitations of the study
Although this study employed the rabbit VX2 liver cancer model, which effectively mimics certain biological characteristics of human HCC, several limitations remain. The molecular profile and immune response of the VX2 model differ from those of human HCC. Therefore, future research should include clinical trials to validate the therapeutic efficacy of WS2 nanosheet-enhanced MWA in HCC patients. Furthermore, additional optimization is required for the synthesis process, stability control, and concentration regulation of WS2 nanosheets, along with exploration of their synergistic application with MWA, to ensure both clinical safety and enhanced imaging performance of this technology. Finally, as this study did not clarify the specific mechanism of CD8+ T cell immune activation or its potential link to recurrence and metastasis risks, further investigations are warranted to address these questions.

Resource availability

Lead contact
Further information and requests for resources and reagents should be directed to and will be fulfilled by the lead contact, Hanfeng Yang (yanghanfeng888@163.com).

Materials availability
This study did not generate new unique reagents.

Data and code availability

•All original data have been deposited in Mendeley Data: https://doi.org/10.17632/dymrkrmns4.1 as of the date of publication. Accession numbers are listed in the key resources table.

•No new code was generated in this study.

•Any additional information required to reanalyze the data reported in this article is available from the lead contact upon request.

Acknowledgments
This study was supported by the project “Construction of Antioxidant Nanozymes Based on Tungsten Carbide and Their Mechanism in Treating Fibromyalgia Syndrome” (grant no. 23JCYJPT0068) and “Special Project for the 10.13039/501100018570Central Government to Guide the Development of Local Science and Technology in Sichuan Province” (grant no. 2024ZYD0092).

Author contributions
Y.Y. and H.Y. designed the study (conceptualization and methodology); Y.H., S.H., H.J., and C.Z. performed experiments (investigation); Y.H., M.L., Y.X., and X.X. analyzed data (formal analysis); Y.H. and B.L. wrote the original draft (writing – original draft); and H.Y. and Y.Y. supervised the project and secured funding. All authors reviewed and edited the manuscript (writing – review and editing).

Declaration of interests
There are no conflicts to declare.

STAR★Methods

Key resources table

Experimental model and study participant details
Healthy male New Zealand white rabbits (aged 4–6 months, weight 2.5–3.5 kg) and VX2 tumor-bearing rabbit (male, aged 4–6 months, weight 2.2 kg) were provided by the Animal Experiment Center of North Sichuan Medical College, along with valid animal quarantine certificates. All rabbits were acclimatized under standard conditions for 1–2 weeks prior to experiments (standard pellet diet, individual caging, constant temperature, free access to water, and a stable environment). After confirming the absence of abnormal behavior, they underwent CT screening to exclude organ malformations or underlying diseases. This study was approved by the Animal Ethics Committee of North Sichuan Medical College [Approval No.: NSMC Animal Ethics Review (2024) No. 032].

Method details

Preparation of WS2 nanosheets and characterization
Commercial WS2 powder was ground for 2 h, after which 40 mg was dispersed in 40 mL concentrated H2SO4 (95%-98%) and intercalated at 90°C for 24 h. The product was collected by centrifugation and repeatedly washed with deionized water until neutral pH. The intercalated WS2 was then dispersed in 40 mL deionized water and subjected to sequential sonication: 20 min bath sonication for preliminary exfoliation, followed by high-intensity probe sonication (325 W) for 2 h. Final centrifugation yielded mono/few-layer WS2 nanosheets.
The sample's particle diameter, elemental composition, and spatial distribution were characterized using TEM (JEOL F200). Elemental composition analysis was performed using EDS with an accelerating voltage of 200 kV (Oxford X-MaxN 80T IE250 detector). The crystal structure and phase composition of the samples were determined by XRD (Rigaku SmartLab SE). The hydrated particle size and surface potential of the WS2 nanosheets were characterized using a particle size and zeta potential analyzer (Malvern Zetasizer Pro). The Raman spectra of the materials were obtained using a Raman spectrometer (RENISHAW InVia Reflex). Additionally, the energy spectrum characteristics of the samples were tested with an XPS (Thermo Scientific K-Alpha). Finally, the samples were analyzed using FT-IR (SHIMADZU IRTracer-100).

Microwave thermal effect and CT imaging in vitro of WS2 nanosheets
To evaluate the effect of different material concentrations on temperature, WS2 nanosheets were dispersed in deionized water to prepare solutions with varying concentrations (50, 100, 150, and 200 μg/mL), using pure deionized water as a negative control. A microwave irradiation (power density: 12 W) was applied continuously for 3 minutes, and the temperature changes of the solutions were recorded every 1 minute using an infrared thermal imager (FLIR, E40) to analyze the heating curves. To assess the influence of different power densities, the concentration of WS2 nanosheets was kept constant (100 μg/mL), and microwave irradiation was applied at varying power densities (5, 8, 12 and 15 W) for 3 minutes while real-time temperature changes were monitored. Additionally, the WS2 nanosheets (100 μg/mL) were irradiated at a power density of 12 W for 2.5 minutes, followed by natural cooling to room temperature after cessation of irradiation. The process was repeated for 5 consecutive cycles to evaluate their microwave thermal stability.
WS2 nanosheets were dissolved in deionized water at varying concentrations (12, 6, 3, 1.5, 0.75, and 0.375 mg/mL) and aliquoted into 1.5 mL centrifuge tubes for subsequent phantom testing. For in vitro CT imaging, a CT imaging equipment (uCT 550+) was used with the following parameters: 120 kV, 90 mAs, and a slice thickness of 1 mm. The CT images were retrieved and analyzed using a PACS, and the HU values (CT numbers) of the solutions were measured.

Establishment of animal models and the treatments for each group
MRI was performed using a Siemens Area 3.0T MR scanner equipped with a dedicated animal coil. The scanning protocols included: axial T1-weighted imaging (T1WI) with echo time (TE) 22.0 ms, repetition time (TR) 720 ms, slice thickness 3 mm, field of view (FOV) 230×230 mm; axial T2-weighted imaging (T2WI) with TE 98 ms, TR 3000 ms, slice thickness 3 mm, FOV 180×180 mm; sagittal T2-weighted imaging (T2WI) with TE 110 ms, TR 3000 ms, slice thickness 2 mm, FOV 260×211 mm; and axial diffusion-weighted imaging (DWI) with TE 50 ms, TR 3000 ms, slice thickness 3 mm, FOV 230×230 mm. CT scanning and guidance were conducted using a Philips MX16 spiral CT system. The ablation procedure employed a KY-2000 multifunctional microwave therapy system (Jiangsu Kangyou Medical Equipment Co., Ltd.). Puncture instruments consisted of a 20G biopsy needle for modeling, a 22G biopsy needle for drug injection, and a microwave ablation antenna (Φ=1.9±0.5 mm, 2450 MHz, KY-2450B-1, Jiangsu Kangyou Medical Equipment Co., Ltd.).
VX2 tumor tissue was harvested from tumor-bearing rabbits and cut into 1 mm3 fragments, which were then ground and filtered to prepare a homogeneous tumor suspension. Experimental rabbits were anesthetized by intravenous injection of 3% sodium pentobarbital (1 mL/kg) via the marginal ear vein and positioned in supine position. Under CT guidance(Figure S1B), 0.1-0.2 mL of tumor suspension was injected into the hepatic parenchyma using a 20G biopsy needle, followed by 2-3 minutes of compression at the puncture site for hemostasis(Figure S1A). MRI scanning was performed 7 days postoperatively to confirm successful establishment of the liver tumor model. Inclusion criteria consisted of solitary focal hepatic nodules measuring 1.5-2.5 cm in diameter, while exclusion criteria included multiple hepatic tumors or tumors protruding from the liver surface(Figures S1C and S1D).
VX2 liver tumor-bearing rabbits with successful modeling were randomly assigned to three groups (n=5): the WS2 group received intratumoral injection of WS2 nanosheets suspension, the MWA group underwent microwave ablation therapy, and the combination group received both WS2 nanosheets and MWA treatment.
All experimental rabbits were fasted for 24 hours preoperatively, weighed, and anesthetized by intravenous injection of 3% sodium pentobarbital (1 mL/kg) via the marginal ear vein before being secured in supine position on the surgical table. Preoperative abdominal plain and contrast-enhanced CT scans were performed to plan puncture pathways while avoiding major blood vessels and critical structures(Figure S2A). Group-specific procedures were as follows: (1) MWA group - routine disinfection and draping were performed. Under CT guidance, the microwave ablation needle was inserted into the center of the tumor(Figure S2C). The ablation power was set at 30 W, and the ablation time was 3 minutes, aiming to simulate an incomplete ablation environment. After the ablation was completed, the needle was withdrawn while ablating at 20 W to prevent needle tract metastasis and bleeding. Postoperative CT was performed immediately to assess the ablation zone density changes, the distribution of gas bubbles, and potential complications (e.g., hemorrhage, pneumothorax, or hemoperitoneum). (2) WS2 group - after standard preparation, CT-guided needle placement into the tumor center was confirmed by negative blood aspiration before injecting 1mL WS2 solution (1 mg/mL in deionized water)(Figure S2B), followed by 1 mL saline flush and 2-3 minutes of compression. (3) Combination group - received intratumoral WS2 injection (identical to WS2 group protocol) immediately followed by MWA treatment (using MWA group parameters) with postoperative CT verification(Figure S2D). All puncture sites received pressure dressings post-procedure with continuous vital sign monitoring throughout.
All procedures were performed jointly by two interventional radiologists with over 5 years of experience, with imaging results evaluated using a double-blind method. Normal saline perfusion was employed during the procedures to protect surrounding tissues.

Sample collection, H&E staining, TUNEL staining and IF staining
All experimental groups (WS2, MWA, and combination therapy) were weighed and underwent non-contrast and contrast-enhanced CT scans at 3 and 7 days postoperatively to evaluate changes in ablation zone dimensions, residual tumor status, and complications. On postoperative day 7, all experimental rabbits underwent final CT scanning, followed by blood sample collection via the marginal ear vein. Subsequently, euthanasia was performed by intravenous administration of an overdose of pentobarbital sodium. Necropsy was immediately performed to harvest entire liver specimens containing ablation zones, which were fixed in 4% neutral buffered formaldehyde. For MWA and combination groups, tissues were sectioned longitudinally along the microwave ablation needle tract to ensure inclusion of the maximal ablation cross-section, while WS2 group specimens were sectioned along the tumor's longitudinal axis. Major organs (heart, liver, spleen, lungs, and kidneys) were systematically dissected from all groups to assess potential visceral damage.
The ablation zones were measured using the PACS system immediately after the procedure and at 1-week follow-up, with ablation volumes calculated using the modified ellipsoid volume formula (V=ab2/2), where “a” represents the long-axis diameter and “b” the short-axis diameter of the ablation zone.
Tissue samples were fixed in 4% paraformaldehyde for 3 days, paraffin-embedded, and sectioned into 3-μm-thick slices. H&E staining was performed to evaluate pathological changes, following standard experimental protocols. In this study, the TUNEL (Terminal deoxynucleotidyl Transferase dUTP Nick End Labeling) staining was employed to detect the apoptosis level of cells in the ablation zone. The specific procedures were as follows: after deparaffinization, antigen retrieval, permeabilization, and peroxidase blocking, the paraffin sections were subjected to Terminal deoxynucleotidyl Transferase (TdT) enzyme labeling reaction, followed by Streptavidin-Horseradish Peroxidase (HRP) binding, 3,3'-Diaminobenzidine (DAB) development, and finally hematoxylin counterstaining, dehydration, and mounting. This study employed immunofluorescence (IF) staining technology to detect changes in CD8+ T-cells, with the specific sequential protocol as follows: antigen retrieval to expose epitopes, primary autofluorescence quenching to reduce background interference, serum blocking to prevent nonspecific binding, specific primary antibody incubation for CD8 antigen labeling, fluorescent secondary antibody incubation for signal amplification, DAPI staining (4',6-Diamidino-2-Phenylindole) for nuclear counterstaining, secondary autofluorescence quenching to further eliminate background fluorescence, and finally fluorescence microscopy observation to analyze quantitative and distributional changes of CD8+ T-cells. All sections underwent H&E, TUNEL, and IF staining, with pathological assessments independently conducted by two certified pathologists (with 8 and 15 years of experience, respectively).

Histopathological assessment
H&E Staining: Each slide was first examined at low magnification to survey the entire tissue, followed by capturing one 200× and one 400× microscopic image. The extent of tumor cell necrosis was graded on a scale of ++ to ++++ (++++: extensive necrosis; +++: moderate necrosis; ++: minimal necrosis). Three additional 200× fields were selected for imaging. The Image-Pro Plus image analysis system was used to measure the areas of viable tumor and necrotic tumor in all acquired images, followed by calculation of the viable-to-necrotic tumor area ratio (necrosis ratio) for each image, with results expressed as mean ratios. The final necrosis ratio per sample was the average of three ratios, with lower values indicating higher necrosis.
TUNEL staining: Image acquisition was performed using a digital slide scanner. For each section, the entire tissue was first observed at low magnification, and then three fields of view were selected for microscopic image capture. Hematoxylin stained the cell nuclei blue, while DAB staining revealed positive expression as brown-yellow. The QuPath image analysis system was employed to quantify the number of positive cells and the total cell count in all captured images. The average positive cell rate for each sample was calculated as the mean of the positive cell rates (number of positive cells divided by total cell count) from the three images.
IF Staining: The image acquisition was performed using a digital slide scanner/microscope. After low-power screening, three 400× fields were captured per slide. DAPI stains nuclei blue, while positive signals appear green due to secondary antibody labeling. Image-Pro Plus analyzed fluorescence intensity and area, with results expressed as IOD.

Quantification and statistical analysis
Statistical analysis was performed using Statistical Product and Service Solutions(SPSS) 27.0. Continuous variables with normal distribution are expressed as mean ± standard deviations (SD), with comparisons made by t-test or one-way analysis of variance (ANOVA). A p-value less than 0.05 was considered statistically significant, and statistical significance was denoted as follows: ∗ for p < 0.05, ∗∗ for p < 0.01, ∗∗∗ for p < 0.001 and ns for p > 0.05.