Drug repurposing: isosorbide mononitrate enhances tumor accumulation to augment sonodynamic therapy for hepatocellular carcinoma

Abstract

Hepatocellular carcinoma (HCC) remains a leading cause of cancer death worldwide. Sonodynamic therapy (SDT) offers a non-invasive, deep-penetrating approach by using ultrasound to activate sonosensitizers and generate cytotoxic reactive oxygen species (ROS). Yet poor intratumoral delivery and low ROS quantum yields of existing agents have stalled clinical translation. Here, we present a synergistic SDT platform that overcomes these barriers by combining transient vasodilation of tumor microvessels with the clinically widely used Antianginal drug isosorbide mononitrate and an acceptor-donor-acceptor-donor-acceptor type organic nanosonosensitizer (BTz) engineered for a narrow bandgap and enhanced ultrasound responsiveness. Isosorbide mononitrate increases nanosonosensitizer accumulation by ~ 1.8-fold. Under ultrasound irradiation, nanosonosensitizer produced high ROS generation, resulting in 78% tumor growth inhibition in murine HCC models—nearly double that of SDT alone—without detectable systemic toxicity. Crucially, the near-infrared fluorescence of nanosonosensitizer enabled real-time, image-guided tracking of sonosensitizer uptake and therapeutic response. By repurposing a safe vasodilator and integrating it with a high-performance organic sonosensitizer, this work establishes a readily translatable, minimally invasive paradigm for precise SDT of localized, inoperable or metastatic HCC.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12951-025-03674-7.

Introduction

Hepatocellular carcinoma (HCC) is the predominant form of primary liver cancer and a major global health burden, causing over 800 000 deaths each year and ranking as the fourth leading cause of cancer mortality [1–3]. Although early-stage HCC patients can achieve a 5-year survival rate exceeding 70% following surgical resection or ablation, more than 60% present with multifocal lesions, vascular invasion or impaired liver function at diagnosis. For these advanced cases, systemic therapies—such as tyrosine kinase inhibitors (e.g., lenvatinib), immune checkpoint inhibitors, and loco-regional interventions (e.g., transarterial chemoembolization, radiofrequency ablation)—remain the mainstays of treatment [4–10]. Yet despite combination approaches, objective response rates hover below 30% and median overall survival plateaus at 12–16 months, hampered by tumor microenvironment immunosuppression, genomic heterogeneity and acquired resistance. This therapeutic bottleneck highlights an urgent need for innovative locoregional modalities that can enhance tumor specificity while minimizing systemic toxicity [11, 12].

Sonodynamic therapy (SDT) has emerged as a promising non-invasive approach for treating deep-seated tumors [13–18]. By using ultrasound (US) to induce cavitation, SDT activates sonosensitizers to generate cytotoxic reactive oxygen species (ROS), resulting in oxidative damage and tumor cell death [19–23]. Unlike photodynamic therapy, SDT circumvents light-penetration limits and facilitates treatment in deep-seated tumors. Preclinical studies have validated its capacity for deep-seated tumor treatment with minimal off-target effects [24–30]. Despite these advantages, clinical translation is hampered by two key obstacles: poor intratumoral delivery of sonosensitizers and suboptimal ROS yields under ultrasound [31]. Conventional agents fall into two categories: inorganic sonosensitizers (e.g., TiO₂, ZnO) exhibit strong ultrasonic responsiveness but suffer from limited biodegradability and potential metal-ion leakage. Organic sonosensitizers (e.g., porphyrins, chlorin derivatives) are biocompatible and structurally tunable but often undergo rapid clearance, require high doses and produce low ROS quantum yields [32–39]. Consequently, insufficient tumor uptake and inadequate ROS generation based on organic sonosensitizers combine to limit SDT’s therapeutic efficacy and highlight the need for new strategies to enhance both tumor enrichment and sonodynamic performance [13, 40–46].

To overcome these barriers, we report a synergistic SDT platform that combines (1) transient vasodilation of tumor vessels using the clinically approved antianginal drug ISMN and (2) a novel acceptor-donor-acceptor-donor-acceptor (A-D-A′-D-A) type organic nanosonosensitizer with a reduced bandgap for enhanced ROS generation. ISMN transiently enlarges tumor microvessels to boost nanosonosensitizer delivery. Compared to the no ISMN group, the delivery efficiency of the nanosonosensitizer was increased by ~ 1.8-fold after ISMN conviction. Meanwhile, ultrasound irradiation of nanosonosensitizer elicits high levels of ROS generation. In murine HCC models, this combination achieves up to 78% tumor growth inhibition—nearly double that of SDT alone—without observable systemic toxicity. Moreover, near-infrared fluorescence from nanosonosensitizer enables real-time image-guided monitoring of sonosensitizer accumulation and therapeutic response. By repurposing a safe, widely available vasodilator and integrating it with a high-performance organic sonosensitizer, our approach addresses the dual limitations of SDT and establishes a readily translatable, minimally invasive treatment paradigm for localized, inoperable or metastatic HCC (Scheme 1).

Scheme 1.

Mechanism diagram of isosorbide mononitrate-enhanced sonodynamic effect in treating liver cancer

Results

Synthesis and characterization of nanosonosensitizer

In order to construct near-infrared sonodynamic molecule (BTz), we extended the absorption and emission wavelengths through the strategy of multiple donor-acceptor (D-A) interactions between molecules. Specifically, we chose dithiophenopyrrole-benzotriazole-dithiophenopyrrole (compound 1) as the D-A′-D core, and further linked the 2-(3-oxo-2,3-dihydro-1 H-inden-1-ylidene) malononitrile (compound 3) as the strong A unit at the end of D-A′-D core to form the A-D-A′-D-A skeleton. The synthetic routes and characterization of the BTz molecule were detailed in Fig. 1A and Figure S1, S2 and S3. Firstly, two aldehyde groups were introduced to both ends of compound 1 by Vilsmeier-Haack Reaction, resulting in compound 2. Then, the 2-(3-oxo-2,3-dihydro-1 H-inden-1-ylidene) malononitrile (compound 3) was attached to both aldehyde group of the compound 2 by the Knoevenagel reaction to obtain BTz molecule with larger conjugation systems and multiple D-A interactions. The corresponding compounds were confirmed via [¹] H nuclear magnetic resonance (NMR), aligning well with anticipated structures.

Fig. 1.

Synthesis and characterization of nanosonosensitizer. (A) Synthetic route and molecular structure of BTz. (B) Illustration of BTz-based nanoparticles preparation via nanoprecipitation. (C) Transmission electron microscopy (TEM) and (D) dynamic light scattering (DLS) of nanosonosensitizer. (E) The UV-visible absorption and (F) fluorescence spectra of BTz molecule in THF solution and nanosonosensitizer in water solution

Next, the BTz molecule was processed into water-soluble nanoparticles using nanoprecipitation with DSPE-PEG as a surfactant (Fig.1B). The nanosonosensitizer were further examined morphology and size using transmission electron microscopy (TEM) and dynamic light scattering (DLS). The results showed a spherical morphology and hydrodynamic diameter of 60–70 nm (Fig. 1C and D). Subsequently, the UV/Vis absorption and fluorescence spectra of BTz molecule in tetrahydrofuran (THF) solution and nanosonosensitizer in water solution were measured to evaluate photophysical properties. The BTz molecule in THF solution showed the typical absorption peak at 720 nm and fluorescence peak at 940 nm. Compared to THF solution, the nanosonosensitizer in water solution showed a maximum absorption peak with red-shifts 60 nm and a maximum fluorescence peak with red-shifts 55 nm (Fig. 1E and F). Thus, the long emission wavelength of nanosonosensitizer indicates their potential for effective NIR fluorescence imaging.

Sonodynamic properties of the nanosonosensitizer

To evaluate the sonodynamic effects of nanosonosensitizer, we measured the reactive oxygen species (ROS) generation under ultrasound conditions. To measure the ROS generation, we used 1,3-diphenylisobenzofuran (DPBF) as a ROS indicator. Firstly, we evaluated the ROS generation at different ultrasound times. The DPBF group and DPBF adding nanosonosensitizer group were respectively placed on ultrasound probes and then treated with different ultrasound times (0, 1, 2, 3, 4 and 5 min, 0.3 W/cm²). As shown in Fig. 2B and C, the results showed a gradual decrease in absorption peak at 420 nm after different ultrasound time, indicating ROS production. Compared to DPBF group, the DPBF adding nanosonosensitizer group showed the 2 times higher DPBF degradation, indicating higher ROS generation and excellent sonodynamic properties (Fig. 2D). Figure S5 showed degradation rates of DPBF, compounds 1, 2, and BTz-based nanoparticles at 8%, 7%, 11%, and 20%, respectively (see Figures S5A-S5E).

Fig. 2.

Sonodynamic properties of nanosonosensitizer. (A) Schematic illustration of ROS generation of nanosonosensitizer under ultrasonic conditions. UV–vis absorption spectra for ROS using DPBF as an indicator under different ultrasonic time of (B) DPBF group and (C) DPBF + nanosonosensitizer group. (D) Quantification of ROS generation in (B) and (C). UV–vis absorption spectra for ROS using DPBF as an indicator under different concentration of (E) 5 µg/mL, (F) 10 µg/mL (G) 15 µg/mL and (H) 20 µg/mL. (I) Quantification of ROS generation in (B) and (E)-(H)

Then, we optimized the ROS generation at different concentrations of nanosonosensitizer. The DPBF adding different concentrations of nanosonosensitizer (5 µg/mL, 10 µg/mL, 15 µg/mL and 20 µg/mL) were respectively placed on ultrasound probes and then treated with different ultrasound times (0, 1, 2, 3, 4 and 5 min, 0.3 W/cm²). As shown in Fig. 2E and H, the results showed a gradual decrease in absorption peak at 420 nm after ultrasound with different concentrations of nanosonosensitizer, indicating ROS production. Meanwhile, as the nanosonosensitizer concentration increases, the DPBF degradation also gradually increases (Fig. 2I). The results showed that ROS generation was positively correlated with nanosonosensitizer concentration. The excellent ROS generaton of nanosonosensitizer under different ultrasound time and nanosonosensitizer concentrations indicates their potential for sonodynamic treatment.

Sonodynamic treatment of the nanosonosensitizer at the cellular level

To validate the sonodynamic effects of nanosonosensitizer at the cellular level, we performed the cell viability experiment with standard in vitro CCK8 analysis using Hepa1-6 mouse liver cancer cells (Fig. 3A). After 24 h incubation of Hepa1-6 cancer cells with different concentrations of nanosonosensitizer, even at concentrations as high as 100 µg/mL, more than 90% of Hepa1-6 cells remained viable, indicating that nanosonosensitizer have low cytotoxicity and excellent biocompatibility (Fig.3B). In contrast, after 6 h of incubation with nanosonosensitizer, the well plates were placed on an ultrasound probe coated with ultrasound coupling agent. After ultrasound excitation (0.6 W/cm², 10 min), standard CCK8 analysis using a microplate reader to detect relative cell viability, cells treated with nanosonosensitizer + ultrasound showed the decreased viability as the nanosonosensitizer concentration increased. At a concentration of 100 µg/mL of nanosonosensitizer, cell viability decreased to 20% (Fig. 3B). These results showed that the cytotoxicity of SDT exceeded 80% in cell experiments.

Fig. 3.

Biocompatibility, liver cancer therapeutic performance in Hepa1-6 cell line. (A) Schematic diagram of acoustic dynamic effects in cell experiments. Hepa1-6 cells were co-cultured with nanosonosensitizer and then treated with US using 0.6 w/cm² for 10 min to generate ROS to induce apoptosis. (B) CCK8 assay of the nanosonosensitizer in Hepa1-6 cells at different concentrations with or without US conditions (n = 3). (C) Fluorescence images of DCFH-DA in Hepa1-6 cells excited at US with different treatment (first panel, DCFH-DA fluorescence; second panel, merged images). (D) Fluorescence microscope imaging of Hepa1-6 cells after US (scale bar = 100 μm). Viable cells are stained green with Calcein-AM, and the dead/later apoptotic cells are stained red with PI. Data were shown as mean ± SD

The excellent sonodynamic effect could originate from the elevated ROS production during the ultrasound treatment process. To verify the increase in ROS generation in SDT, we utilized 2’,7’-dichlorodihydrofluorescein diacetate (DCFH-DA, as a ROS indicator) staining assay to monitor intracellular ROS levels in Hepa1-6 cells. Six hours after treatment, cells were incubated with DCFH-DA (10 µM) and Hoechst 33,342 (10 µg/mL) for 30 min. Imaging was then performed for four experimental groups: (i) PBS, (ii) PBS + US, (iii) nanosonosensitizer, and (iv) nanosonosensitizer + US. Compared to PBS, PBS + US and nanosonosensitizer groups, the nanosonosensitizer + US group showed significantly increased red fluorescence intensity (Fig. 3C). The results showed that nanosonosensitizer triggered a substantial ROS production in ultrasound, which were in agreement with the cell viability experiment.

A high level of ROS production could lead to cell death. Subsequently, we further investigated the live/dead cell ratio during SDT using Calcein-AM/PI staining. Hepa1-6 cells treated with different treatments were stained with calcein AM/PI to determine the live/dead cell ratio. Cells treated with PBS, US alone, and nanosonosensitizer groups remained largely viable with minimal cell death (dead cells are shown in red, live cells are shown in green). However, the cells were killed after treatment with nanosonosensitizer and ultrasound (Fig. 3D). These findings demonstrate that nanosonosensitizer, when activated by ultrasound, generate substantial ROS, leading to extensive cell death and highlighting their potent sonodynamic effects.

Enhanced accumulation of nanosonosensitizer for liver tumors in vivo

Encouraged by the excellent NIR fluorescence in solution and impressive sonodynamic effects in cell levels, we further investigated the potential application of nanosonosensitizer in vivo. Firstly, the biological toxicity of the probe was evaluated by hemocompatibility test, and the material was suitable for in vivo experiment (Figure S6). The C57/BL6 mice with subcutaneous liver tumors were intravenously injected with nanosonosensitizer (1 mg/mL, 100 µL). After intravenously injecting 1, 6 and 12 h, the fluorescence imaging was immediately captured using an IVIS imaging system. As shown in Fig. 4B, the fluorescence signal intensity increased progressively over time, suggesting that the nanosonosensitizer had successfully accumulated in the tumor tissue via the enhanced permeability and retention (EPR) effect.

Fig. 4.

NIR-I FL images of subcutaneous Hepa1-6 tumor-bearing mice at different time points after intravenous injection of the nanosonosensitizer. (A) Subcutaneous tumor-bearing mice enhance enrichment of tail vein-injected nanosonosensitizer in tumors by oral administration of isosorbide mononitrate vasodilator. (B) NIR-I FL images of subcutaneous tumor models at various time points after intravenous injection of the nanosonosensitizer (n = 5, scale bar = 1 cm). (C) NIR-I FL images of subcutaneous tumor models at various time points after intragastric administration of the isosorbide mononitrate and intravenous injection of the nanosonosensitizer (n = 5, scale bar = 1 cm). (D) Quantification of FL intensities of different groups in (B) and (C) (n = 5). (Please provide Graphical Abstract for the article as per journal instructions.) Quantitative analysis of tumor fluorescence intensity after 12-hour ex vivo enrichment: PBS gavage vs. isosorbide mononitrate gavage. (F) Quantitative comparison of tumor fluorescence enrichment (12-hour) following PBS and isosorbide mononitrate gavage treatments in (E) (n = 3). (G) Immunofluorescence staining of tumour slices of each group with CD31 staining (n = 5 independent experiments). (H) Quantitative comparison of each group with CD31 staining (n = 5). Data were shown as mean ± SD

ISMN, a nitrite drug commonly used for treating angina and hypertension, mainly improves blood flow by dilating blood vessels. Increased tumor vascular blood flow facilitates enhanced nanoparticle accumulation. ISMN was administered as a vasodilator to promote nanoparticle enrichment in tumor-bearing mice [47–49] (Fig. 4A, Figure S7).

Subsequently, to confirm the enhanced accumulation of the nanosonosensitizer, experimental mice first received an oral administration of ISMN prior to intravenous administration of the nanosonosensitizer formulation. Fluorescence imaging was performed at designated time points to quantify tumor-targeted nanosonosensitizer distribution. First, we administered ISMN to the mice via oral gavage at a dosage of 5 mg per kg body weight. Subsequently, nanosonosensitizer probes were injected intravenously into mice. Imaging to assess probe accumulation was then performed at designated time points: 1, 6 and 12 h post-injection. As demonstrated in Fig.4C, oral administration of ISMN to mice resulted in a time-dependent escalation of fluorescence signal intensity, with peak accumulation observed at 12 h post-injection of the nanosonosensitizer. Notably, the nanosonosensitizer + ISMN combination group exhibited markedly higher fluorescence intensity compared to nanosonosensitizer-only treatment groups, indicating IMN-enhanced vascular permeability promoted nanosonosensitizer targeting. The fluorescence intensity was higher in the highest accumulation observed between 6 and 12 h post-administration (Fig. 4B-C). Following the 12-hour administration of nanosonosensitizer, imaging data showed an obvious increase in fluorescence signal of in the nanosonosensitizer + ISMN group compared to the nanosonosensitizer group (Fig. 4D).

After intravenous injection of nanosonosensitizer for 12 h, dissection of the mice and ex vivo NIR-I Fluorescence imaging of tumor tissues further assessed the biodistribution of the nanosonosensitizer (Fig. 4E). Tumor tissues in the ISMN-treated group showed stronger NIR-I Fluorescence intensity than those in the nanosonosensitizer group, confirming that ISMN enhanced the accumulation of nanosonosensitizer in the tumor regions for 1.8 fold (Fig. 4F).

Furthermore, we conducted CD31 staining analysis to evaluate tumor angiogenesis (Fig.4G). The CD31 staining demonstrated that ISMN oral administration significantly expanded blood vessels, with observable changes observed at 6 h compared to the control group, while no such effect was observed in the control group. This aligns with the pro-angiogenic response. Quantitative maps also showed a vasodilatory effect of more than 10 times (Fig. 4H). The distribution of kidney and lung in ISMN group was higher than that in PBS group (Figure S8).

Imaging guided enhanced sonodynamic therapy for liver tumors

Building on the enhanced the accumulation from the subcutaneous tumor model, we further evaluate effects in the subcutaneous liver cancer tumor mouse model (Fig. 5A). The mice were divided into three groups: group I received PBS, group II received nanosonosensitizer + US, group III nanosonosensitizer + ISMN + US. Mice were injected intravenously with nanosonosensitizer (1 mg/mL, 100 µL) followed by oral administration of ISMN (5 mg/Kg). After 12 h injection, mice were coated with ultrasound coupling agent, and subjected to US excitations (2.7 W/cm², 10 min). We assessed the SDT efficacy by recording tumor growth during 15 days after US treatment. Throughout the experimental duration, there was no significant weight change in the three groups.There was no significant growth inhibition curve in the tumor growth curve of the pure ISMN group compared with that of the PBS group (Fig. 5B). The PBS group showed rapid tumor growth, while the nanosonosensitizer + ISMN + US group exhibited threefold higher tumor suppression compared to the PBS control (Fig. 5B). Compared to the nanosonosensitizer treatment group, the nanosonosensitizer + ISMN + US group exhibited twofold higher tumor suppression. After 15 days of monitoring and treatment, tumors were subjected to ex vivo weighing. The average weight of the excised tumor tissue at the end of the treatment showed a trend consistent with tumor suppression (Fig. 5C). The results demonstrated that nanosonosensitizer exhibited sonodynamic effects, and the addition of the ISMN further enhanced the efficacy of SDT.

Fig. 5.

Efficacy evaluation of SDT in subcutaneous Hepa1-6 tumor-bearing mice. (A) The timeline schedule of drug administration and SDT treatment in C57/BL6 mice. (B) Correlative analysis of tumor volume in subcutaneous tumor-bearing mice at 15 days post-treatment (n = 5). (C) Ex vivo tumor gravimetric analysis across experimental groups in subcutaneous tumor-bearing mice at 15 days post-treatment (n = 5). (D) ROS and (E) TUNEL staining of tumor tissue after different treatment (scale bar = 50 μm). (F) H&E staining images of subcutaneous tumor after different treatment (scale bar = 10 μm). Data were shown as mean ± SD

The outstanding therapeutic effects could originate from higher levels of ROS production during the sonodynamic process. To evaluate the level of ROS production during the treatment process, we performed DCFH-DA staining of tumors. Slices of tumor tissue were incubated with DAPI (10 µM) and DCFH-DA (10 µM) for 30 min. Then, the immunofluorescence imaging of these slices was performed on Nikon Confocal Microscopy. Nanosonosensitizer + ISMN + US treatment group demonstrated significant ROS generation in tumor post-treatment, with levels markedly higher than those in the control and nanosonosensitizer + US groups. These findings indicate that SDT induces ROS production in tumors, while ISMN promotes intratumoral accumulation of sonosensitizers, elevating ROS levels, and resulting in significantly improved therapeutic efficacy (Fig. 5D).

To further evaluate cell damage in tumor sections after ablation, TUNEL staining was performed. TUNEL staining showed no significant apoptosis in the PBS group, while the nanosonosensitizer + US group exhibited higher cell apoptosis, and the nanosonosensitizer + ISMN + US group showed a highest increase in apoptotic cells (Fig. 5E). Hematoxylin and eosin (H&E) staining revealed tumor necrosis in the tumor sections of mice treated with nanosonosensitizer + US and nanosonosensitizer + ISMN + US groups: extensive coagulative necrosis was observed in the nanosonosensitizer + ISMN + US group, characterized by homogeneous eosinophilic necrotic foci with nuclear pyknosis, karyorrhexis, and karyolysis. In contrast, unused ISMN group displayed densely packed tumor cells, alongside sporadic focal necrosis (Fig. 5F). These findings demonstrate that oral administration of ISMN enhances intratumoral accumulation of nanosonosensitizers. Combined with SDT, this approach achieved superior therapeutic outcomes.

Blood was taken from mice at each time period to evaluate liver and kidney function. No significant differences were found among the groups, indicating that no obvious damage was caused to liver and kidney after probe and drug acoustic dynamics (Figure S9). Histopathological analysis via H&E staining of ex vivo heart, liver, spleen, lungs, and kidneys revealed no significant histological differences across groups treated with the nanosonosensitizer or ISMN formulations, demonstrating minimal organ toxicity (Figure S10).

Discussion

Sonodynamic therapy (SDT) leverages ultrasound-activated sonosensitizers to generate cytotoxic reactive oxygen species (ROS), offering a non-invasive alternative with fewer systemic side effects compared to conventional chemotherapy therapies. However, while SDT demonstrates precision in targeting tumors, its clinical utility is constrained by the limited penetration depth of ultrasound in larger or deeply situated lesions, such as those in the liver or pancreas [19, 26, 50–52]. Tumor heterogeneity—manifested as variations in density, vascularity, and echogenicity—further complicates the uniform distribution of ultrasound energy, often resulting in incomplete ablation or collateral damage to healthy tissues. These limitations underscore the need for innovative strategies to enhance sonosensitizer delivery and optimize ultrasound-triggered ROS generation [53, 54].

To address these challenges, we focused on isosorbide mononitrate (ISMN), a clinically approved nitrates. ISMN transiently expands tumor-associated blood vessels, improving nanoparticle accumulation. This approach circumvents the risks of increasing ultrasound power (e.g., thermal injury) while synergizing with nanoparticle-based sonosensitizers. Clinically, ISMN is used as a long-term treatment of coronary heart disease, mainly for the prevention of vasospastic and mixed angina pectoris, as well as after myocardial infarction and the treatment of chronic heart failure [55–57]. The principle of mononitrate in the treatment of angina pectoris is mainly to increase the blood supply to the heart by dilating the blood vessels. It acts on the nitrate receptors on the smooth muscle of the blood vessels, prompting the release of nitric oxide, which causes vasodilation, especially of the coronary arteries. This not only reduces the anterior and posterior loads on the heart and decreases myocardial oxygen consumption, but also increases myocardial blood and oxygen supply, which in turn relieves angina symptoms. The vasodilatory effect of mononitrate reduces myocardial ischemia and is particularly effective in patients with stable angina [58–60]. The concept of repurposing clinically established drugs to enhance precision therapy in HCC represents a transformative strategy to overcome the limitations of conventional treatments. Our approach, which integrates ISMN—a vasodilator with decades of clinical use for cardiovascular diseases—into SDT, exemplifies the potential of “old drugs, new tricks” in oncology. By transiently dilating tumor vasculature, ISMN significantly amplifies the intratumoral accumulation of sonosensitizers, achieving a 1.8-fold enrichment in preclinical models. This pharmacological enhances enrichment, as evidenced by the 78% tumor growth inhibition observed in nanosonosensitizer + ISMN + SDT cohorts. Crucially, this strategy preserves the inherent biosafety of ISMN, a drug already proven safe in millions of patients worldwide, thereby minimizing the regulatory hurdles typically associated with novel therapeutics. The oral administration route of ISMN further aligns with clinical practicality, enabling outpatient-based protocols that could be seamlessly integrated into existing HCC management frameworks. By leveraging the safety, cost-effectiveness, and scalability of repurposed drugs like ISMN, our work provides a blueprint for accelerating the translation of nanotechnology-driven therapies into clinical practice.

The clinical translation of this integrated therapeutic system is underpinned by three synergistic advantages: drug accessibility, material safety, and ultrasound compatibility. First, ISMN, as a clinically approved vasodilator, holds immediate applicability in HCC patients, particularly those with comorbid cardiovascular conditions—a common scenario given the overlapping risk factors (e.g., metabolic syndrome, chronic inflammation) [61, 62]. By repurposing ISMN to transiently dilate tumor vasculature, our strategy leverages its established safety profile and broad patient eligibility, circumventing the need for novel drug development. Second, the organic sonosensitizer demonstrated negligible systemic toxicity in both cellular and preclinical models, addressing a critical barrier to clinical adoption. Unlike inorganic sonosensitizers prone to metal ion leakage, this organic molecule-based design ensures biodegradability and minimizes long-term biosafety concerns, aligning with regulatory requirements for human trials. Third, ultrasound—a non-ionizing, real-time image-guided modality—is inherently suited for abdominal tumors like HCC due to its deep penetration, adaptability to lesion topography, and compatibility with cirrhotic livers. By tailoring ultrasound parameters, we achieved spatiotemporally controlled ROS generation in ISMN-preconditioned tumors, enhancing tumor suppression while preserving surrounding tissue integrity. This integrated platform combines clinically available vasoactive agents for sonosensitizer accumulation, organic sonosensitizers for optimized ROS production, and programmable ultrasound protocols to ensure targeted ROS production. This synergy positions the system as a frontline candidate for advanced HCC patients ineligible for surgery, with potential extensions to other abdominal malignancies (e.g., pancreatic cancer) where vascular abnormalities and deep tumor locations challenge conventional therapies. By bridging drug repositioning, nanotechnology, and interventional radiology, our approach exemplifies a translatable blueprint for precision oncology.

Conclusions

In conclusion, our study presents a clinically translatable strategy that synergistically integrates pharmacologically enhanced vascular modulation, nanotechnology, and ultrasound precision to overcome the therapeutic limitations of sonodynamic therapy (SDT) in advanced hepatocellular carcinoma (HCC). By repurposing the vasodilator isosorbide mononitrate (ISMN), we amplified tumor-specific accumulation of organic sonosensitizers by 1.8-fold, addressing the critical challenge of insufficient drug delivery. Subsequent ultrasound activation achieved robust ROS generation, yielding a 78% tumor growth inhibition rate—a significant improvement over conventional SDT. The biosafety profile of ISMN ensures compatibility with cirrhotic livers and comorbid cardiovascular conditions, which are common in HCC patients. Furthermore, the non-thermal nature and deep-penetrating capabilities of ultrasound enable precise ablation of abdominal tumors while preserving peri-tumoral tissue integrity. This triad of drug repositioning, material design, and imaging-guided technology not only redefines minimally invasive HCC therapy but also establishes a blueprint for treating other deep-seated malignancies, such as pancreatic tumor.

Author contributions

Yu Peng and Zhe Li performed the data; Yu Peng, Zhe Li, Lei Zhou, Qian Jian and Baoli Yin contributed significantly to analysis and manuscript preparation; Bo Sun, Yinghui Song, Hao Chen, Xianzheng Tan, Xiaohui Duan and Chuang Peng performed the data analyses; Yu Peng, Zhe Li, Guosheng Song and Sulai Liu wrote the manuscript; Guosheng Song and Sulai Liu contributed to the conception of the study; Guosheng Song, Sulai Liu and Chuang Peng helped perform the analysis with constructive discussions.All authors reviewed the manuscript.

Funding

This work was financially supported by National Natural Science Foundation of China (NO. 22374045/82303186); Key R&D funding of Hunan Province (2024JK2110/2023SK2060); Natural Science Fund for Outstanding Young Scholars of Hunan Province (2024JJ2037); Frontiers diagnosis and treatment of liver cancer innovation team (KCT202404); The Science and Technology Innovation Program of Hunan Province (2024RC1051); Hunan Province People's Hospital benevolence fund key cultivation project (RS2022A07); Project of Hunan Provincial Health Commission (Z2023031); .National Clinical Key Specialty Construction Project (20191127-1001).

Data availability

The data that support the findings of this study are available from the corresponding author upon reasonable request.

Declarations

Competing interests

The authors declare no competing interests.