*2.9. Statistical Analysis*

In this study, the statistical differences between each group were analyzed through a one-way ANOVA in the Origin 2020 software (OriginLab Corporation, Northampton, MA, USA). The significant difference was marked with an asterisk (\*) in the figures.

#### **3. Results and Discussion**

#### *3.1. In Vitro Characterization of DOX-Loaded Glycol Chitosan Nanoparticles (DOX-CNPs)*

The amphiphilic glycol chitosan-5β-cholanic acid conjugates were prepared by direct coupling between hydrophobic 5β-cholanic acid and hydrophilic glycol chitosan, resulting in the formation of nanoparticles in an aqueous condition (Figure 1b). The glycol chitosan-5β-cholanic acid conjugates contained 162 ± 6.5 molecules of 5β-cholanic acid per glycol chitosan backbone, confirmed by a colloidal titration method [39,40]. For the in vitro and in vivo near-infrared fluorescence (NIRF) imaging, 3.9 molecules of Cy5.5 were chemically conjugated to glycol chitosan-5β-cholanic acid conjugates. Next, the anticancer drug, DOX, was encapsulated into the hydrophobic cores of CNPs via a dialysis method. The amount of DOX in DOX-CNPs was 10 ± 1.5 wt%, calculated by the DOX-standard curve measured at 490 nm in the UV–Vis spectrum. The sizes of CNPs and DOX-CNPs in the aqueous solution measured by DLS were 283.7 ± 5.3 nm and 265.9 ± 35.5 nm. The volume-weighted distribution of CNPs and DOX-CNPs ranged from 260 to 300 nm and 140 to 300 nm, respectively, showing a wider size distribution after DOX encapsulation (Figure 2a). Additionally, the TEM images showed that CNPs and DOX-CNPs had spherical nanoparticle structures with diameters of 262 ± 12 nm and 250 ± 17 nm, respectively (Figure 2b). The size histogram from the TEM images of CNPs and DOX-CNPs showed 200 to 260 nm of distribution, resulting in the similar size distribution of CNPs and DOX-CNPs (Figure S1a and S1b). The surface charges of CNPs and DOX-CNPs were measured to +15.45 ± 0.90 mV and +15.3 ± 0.45 mV, indicating that the surface of CNPs and DOX-CNPs were covered with positively charged glycol chitosan polymers (Figure 2c). Although CNPs and DOX-CNPs had a positive surface, the volume-weighted size distribution of CNPs and DOX-CNPs showed no changes in the size distribution under 1% FBS-containing medium (Figure S1c and S1d). Furthermore, the size of CNPs and DOX-CNPs was stable for 3 days in both PBS (pH 7.4) and 1% FBS-containing PBS (pH 7.4) conditions (Figure 2d). This is because glycol moiety can act as poly(ethylene glycol) (PEG), resulting in the inhibition of size changes by interrupting protein adsorption [41]. This biocompatible glycol chitosan polymer outer surface of CNPs and DOX-CNPs could prevent the adsorption of non-specific proteins in vivo and enables high accumulation at targeted tumor tissues through EPR in vivo [37]. Based on these characterization results, we expect that CNPs can sufficiently encapsulate the anticancer drug DOX into the hydrophobic inner cores of 5β-cholanic acid. Prior to the in vivo study of the DOX-CNPs, in vitro drug release profiles were evaluated. In the case of DOX-CNPs without ultrasound treatment, it was confirmed that the drug was released gradually up to 24 h, but 5 min of HIFU treatment (destruction mode, power: 10 MHz, mechanical index: 0.235) substantially increased the drug release amount by 1.6 and 2.2 times after 30 min or 1 h post-incubation in PBS at 37 ◦C, compared to untreated DOX-CNPs, due to the ultrasound-triggered rapid drug release from DOX-CNPs (Figure 2e).

*Pharmaceutics* **2020**, *12*, x 8 of 16

**Figure 2.** Physicochemical properties of DOX-CNPs in vitro. (**a**) The volume-weighted size distribution of CNPs and DOX-CNPs (1 mg/mL) in PBS (pH 7.4), measured using dynamic light scattering. (**b**) TEM image of CNPs and DOX-CNPs. (**c**) Characteristic table of CNPs and DOX-CNPs summarized diameter, zeta potential, and DOX contents. (**d**) Size stability of CNPs and DOX-CNPs in PBS (pH7.4) and 1% FBS-containing PBS (pH 7.4) for 3 days. (**e**) In vitro release behavior of DOX from the DOX-CNPs (mean ± SD, *n* = 5). HIFU treatment was carried out for 5 min. **Figure 2.** Physicochemical properties of DOX-CNPs in vitro. (**a**) The volume-weighted size distribution of CNPs and DOX-CNPs (1 mg/mL) in PBS (pH 7.4), measured using dynamic light scattering. (**b**) TEM image of CNPs and DOX-CNPs. (**c**) Characteristic table of CNPs and DOX-CNPs summarized diameter, zeta potential, and DOX contents. (**d**) Size stability of CNPs and DOX-CNPs in PBS (pH 7.4) and 1% FBS-containing PBS (pH 7.4) for 3 days. (**e**) In vitro release behavior of DOX from the DOX-CNPs (mean ± SD, *n* = 5). HIFU treatment was carried out for 5 min.

#### *3.2. In Vitro Cellular Uptake and Cytotoxicity of HIFU-Triggred DOX-CNPs 3.2. In Vitro Cellular Uptake and Cytotoxicity of HIFU-Triggred DOX-CNPs*

To observe the cellular uptake of HIFU-triggered Cy5.5-labeled DOX-CNPs, A549 cells were treated with 100 μg of DOX-CNPs and the cells were exposed to HIFU in destruction mode (power: 10 MHz and mechanical index: 0.235) for 5 min and the cellular uptake of HIFU-triggered Cy5.5 labeled DOX-CNPs were visualized with confocal microscopy (Figure 3a). In the control, without HIFU exposure, Cy5.5-labeled DOX-CNPs (red colors) slowly bound to cell membranes after 10 min post-incubation and then they were internalized into the cytoplasm at up to 30 min, wherein the bright red colors of Cy5.5-labeled DOX-CNPs were clearly observed in the cytoplasm compartment of A549 cells. It is reported that CNPs show a fast uptake into cancer cells via diverse nanoparticlederived endocytic pathways [42–44]. Interestingly, the HIFU-triggered cellular uptake of Cy5.5 labeled DOX-CNPs (red color) was clearly observed after 5 min of HIFU pre-treatment (destruction mode; power: 10 MHz, mechanical index: 0.235), compared to untreated Cy5.5-labeled DOX-CNPs. After 10 min post-incubation, a large amount of Cy5.5-labeled DOX-CNPs in the cell membrane and cytoplasm were observed in A547 cells. Furthermore, most DOX-CNPs were rapidly internalized into the cytoplasm compartment of A549 cells after 30 min post-incubation, due to the HIFU-triggered rapid cellular uptake mechanism. More importantly, the rapid cellular uptake of DOX (green color) in CNPs was clearly observed via the HIFU-triggered fast cellular uptake of DOX-CNPs, compared to free DOX. In the case of HIFU exposure, the cellular uptake of Cy5.5-labeled DOX-CNPs increased by To observe the cellular uptake of HIFU-triggered Cy5.5-labeled DOX-CNPs, A549 cells were treated with 100 µg of DOX-CNPs and the cells were exposed to HIFU in destruction mode (power: 10 MHz and mechanical index: 0.235) for 5 min and the cellular uptake of HIFU-triggered Cy5.5-labeled DOX-CNPs were visualized with confocal microscopy (Figure 3a). In the control, without HIFU exposure, Cy5.5-labeled DOX-CNPs (red colors) slowly bound to cell membranes after 10 min post-incubation and then they were internalized into the cytoplasm at up to 30 min, wherein the bright red colors of Cy5.5-labeled DOX-CNPs were clearly observed in the cytoplasm compartment of A549 cells. It is reported that CNPs show a fast uptake into cancer cells via diverse nanoparticle-derived endocytic pathways [42–44]. Interestingly, the HIFU-triggered cellular uptake of Cy5.5-labeled DOX-CNPs (red color) was clearly observed after 5 min of HIFU pre-treatment (destruction mode; power: 10 MHz, mechanical index: 0.235), compared to untreated Cy5.5-labeled DOX-CNPs. After 10 min post-incubation, a large amount of Cy5.5-labeled DOX-CNPs in the cell membrane and cytoplasm were observed in A547 cells. Furthermore, most DOX-CNPs were rapidly internalized into the cytoplasm compartment of A549 cells after 30 min post-incubation, due to the HIFU-triggered rapid cellular uptake mechanism. More importantly, the rapid cellular uptake of DOX (green color) in CNPs was clearly observed via the HIFU-triggered fast cellular uptake of DOX-CNPs, compared to free DOX. In the case of HIFU exposure, the cellular uptake of Cy5.5-labeled DOX-CNPs

1.6 and 2.0 times after 10 and 30 min post-incubation, compared to untreated DOX-CNPs (Figure S2).

increased by 1.6 and 2.0 times after 10 and 30 min post-incubation, compared to untreated DOX-CNPs (Figure S2). Finally, after 30 min post-incubation, the cellular uptake of DOX molecules in CNPs increased by 5.1 times via the HIFU-triggered cellular uptake of DOX-CNPs, compared to free DOX. This is because HIFU exposure to live cells can increase the permeability of nanoparticles into cell membranes via the sonoporation effect as well as the mechanical effects of ultrasound [45–47]. *Pharmaceutics* **2020**, *12*, x 9 of 16 Finally, after 30 min post-incubation, the cellular uptake of DOX molecules in CNPs increased by 5.1 times via the HIFU-triggered cellular uptake of DOX-CNPs, compared to free DOX. This is because HIFU exposure to live cells can increase the permeability of nanoparticles into cell membranes via the sonoporation effect as well as the mechanical effects of ultrasound [45–47].

**Figure 3.** In vitro cellular uptake and cell viability of DOX-CNPs in cultured cells. (**a**) HIFU-triggered (US +) cellular uptake mechanism of DOX-CNPs. A549 cancer cells were incubated with free DOX (1 μg/mL), DOX-CNPs (10 μg/mL) and HIFU-triggered (US +) DOX-CNPs (10 μg/mL) for 10 min and 30 min. DOX-CNP-treated A549 cells were exposed to HIFU in destruction mode (power: 10 MHz, mechanical index: 0.235) for 5 min. (**b**) Cell viability of free DOX, CNPs and DOX-CNPs in cultured A549 cells (mean ± SD, *n* = 5). (**c**) Cell viability of HIFU-triggered DOX-CNPs (10 μg/mL) in A549 cells. CNPs (US+) and DOX-CNPs (US+) groups were treated with HIFU in destruction mode (power: 10 MHz, mechanical index: 0.235) for 5 min and the cell viability was measured 24 h post-incubation (mean ± SD, *n* = 5). (\*) indicates difference at the *p* < 0.05 significance level. **Figure 3.** In vitro cellular uptake and cell viability of DOX-CNPs in cultured cells. (**a**) HIFU-triggered (US+) cellular uptake mechanism of DOX-CNPs. A549 cancer cells were incubated with free DOX (1 µg/mL), DOX-CNPs (10 µg/mL) and HIFU-triggered (US+) DOX-CNPs (10 µg/mL) for 10 min and 30 min. DOX-CNP-treated A549 cells were exposed to HIFU in destruction mode (power: 10 MHz, mechanical index: 0.235) for 5 min. (**b**) Cell viability of free DOX, CNPs and DOX-CNPs in cultured A549 cells (mean ± SD, *n* = 5). (**c**) Cell viability of HIFU-triggered DOX-CNPs (10 µg/mL) in A549 cells. CNPs (US+) and DOX-CNPs (US+) groups were treated with HIFU in destruction mode (power: 10 MHz, mechanical index: 0.235) for 5 min and the cell viability was measured 24 h post-incubation (mean ± SD, *n* = 5). (\*) indicates difference at the *p* < 0.05 significance level.

To evaluate the cytotoxicity of DOX-CNPs in A549 tumor cells, the cell viability of A549 cells was assayed using cell counting kit-8 (CCK-8) at various concentrations of free DOX, CNPs, and DOX-CNPs. CNPs did not show severe cytotoxicity at a high concentration (500 μg/mL) in culture To evaluate the cytotoxicity of DOX-CNPs in A549 tumor cells, the cell viability of A549 cells was assayed using cell counting kit-8 (CCK-8) at various concentrations of free DOX, CNPs, and DOX-CNPs. CNPs did not show severe cytotoxicity at a high concentration (500 µg/mL) in culture media. However, the cell viability of DOX-CNPs-treated A549 cells gradually decreased, due to the release of free DOX

from DOX-CNPs (Figure 3b). Furthermore, the cell viability of free DOX-treated A549 cells decreased in a DOX concentration-dependent manner. Interestingly, HIFU exposure could increase the cytotoxicity of DOX-CNPs in cultured cells, compared to untreated DOX-CNPs (Figure 3c). When the A549 cells were treated with 100 µg of DOX-CNPs (10 µg of DOX) for 24 h, the cell viability was measured at 67.68 ± 5.07%. However, after HIFU exposure for 5 min, the cell viability of DOX-CNP-treated A549 decreased to 51.99 ± 1.79%. It is deduced that the rapid cellular uptake and the rapid drug release of HIFU-triggered DOX-CNPs could increase the cytotoxicity of drug-loaded nanoparticles in cultured cells, compared to untreated DOX-CNPs.

#### *3.3. In Vivo Biodistribution and Therapeutic E*ffi*cacy of HIFU-Triggered DOX-CNPs*

The in vivo biodistribution of Cy5.5-labeled DOX-CNPs without or with HIFU treatment was monitored in ECM-rich A549 tumor-bearing mice. This is because A549 tumor tissues with stiff ECMs composed of dense collagen and hyaluronan could prevent the deep tissue penetration of nanosized drug carriers [15,28]. Prior to monitor-targeted tumor accumulation, we firstly confirmed the collagen contents of tumor tissues using A549 and SCC7 tumor-bearing mice. When the tumor volume reached to 250 <sup>±</sup> 50 mm<sup>3</sup> , tumor tissues were excised from the mice, followed by staining using Masson's trichrome staining solution. Compared to SCC7 tumor tissue images, A549 tumor tissue images showed a widely dispersed collagen area which was blue-stained by Masson's trichrome staining solution (Figure S3a). Furthermore, the blue-colored collagen fibers were intricately connected to each other throughout A549 tumor tissues, indicating ECM-rich tumor tissues. However, in the case of SCC7 tumor tissues, almost no collagen fibers were seen in ECM-less tumor tissues. In particular, the amount of collagen fibers in A549 tumor tissues was eight times higher than that of SCC7 tumor tissues (Figure S3b).

Next, we confirmed a tumor tissue collagen destruction effect by HIFU treatment. The A549 tumor model was established by a subcutaneous injection of 1 <sup>×</sup> <sup>10</sup><sup>7</sup> cells into the Balb-c/nude mice. When the tumor was grew up 250 <sup>±</sup> 50 mm<sup>3</sup> , A549 tumor tissues were treated with HIFU (intensity: 5 W/cm<sup>2</sup> , frequency: 1.5 MHz, duty cycle: 10%, pulse repetition frequency: 1 Hz, and time per spot: 30 s) for 5 min. Then, A549 tumor tissues were excised from the mice, followed by staining using Masson's trichrome staining solution. A549 tumor tissue images showed a widely dispersed blue-stained collagen area whereas the blue-stained collagen area was dramatically reduced after HIFU treatment (Figure S4a). Furthermore, the blue-stained collagen area of the HIFU-treated A549 tumor tissue observed was six times lower than that of the A549 tumor tissue, resulting in a significant reduction in collagen in the tumor tissue by HIFU treatment (Figure S4b).

To monitor the targeted tumor accumulation of DOX-CNPs, the A549 tumor model was made by a subcutaneous injection of 1 <sup>×</sup> <sup>10</sup><sup>7</sup> cells into the Balb-c/nude mice. When the tumor volume reached <sup>250</sup> <sup>±</sup> 50 mm<sup>3</sup> , Cy5.5-DOX-CNPs (20 µg/kg, 100 µL) were intravenously injected into the A549 tumor-bearing mice (*n* = 3). After an intravenous injection of Cy5.5-DOX-CNPs, the tumor was treated with HIFU (intensity: 5W/cm<sup>2</sup> , frequency: 1.5 MHz, duty cycle: 10%, pulse repetition frequency: 1 Hz, and time per spot: 30 s) for 5 min and the tumor-targeting ability of Cy5.5-DOX-CNPs were visualized using non-invasive NIRF imaging. As we expect, 6 h post-injection, HIFU-treated groups showed a high NIRF intensity of Cy5.5-labeled DOX-CNPs at targeted tumor tissues (white dotted circle), compared to untreated DOX-CNP groups (Figure 4a). The tumor accumulation of HIFU-treated DOX-CNPs groups increased noticeably via the nanoparticle-derived EPR effect for 24 h, whereas untreated DOX-CNP groups slightly accumulated at targeted tumor tissues up to 24 h post-incubation, due to the dense ECM structure of A549 tumor tissues. To observe ex vivo NIRF images, the tumor tissues and the major organs, including liver, lung, spleen, kidney, and heart, were excised from the tumor-bearing mice at 24 h post-injection (Figure 4b). The bright NIRF signal of HIFU-treated DOX-CNPs was clearly observed at targeted tumor tissues. In the control, both DOX-CNPs without and with HIFU treatment showed the similar non-specific accumulations in normal tissues, such as liver, lung, and kidney. In the HIFU-treated groups, the NIRF signal intensity of Cy5.5-labeled DOX-CNPs in

the tumor tissue was 1.84 times higher than that of the untreated groups (Figure 4c). It is deduced that HIFU treatment at targeted tumor tissues could destroy dense ECMs composed of collagen and hyaluronan, resulting in the deep tumor penetration of DOX-CNPs at ECM-rich A549 tumor tissues [28,48]. Surprisingly, intravenously injected DOX-CNPs could be successfully accumulated at in ECM-rich tumors exposed to HIFU treatments. Lastly, the excised tumor tissues were further observed using fluorescence microscopy. The NIRF microscopic images showed that untreated DOX-CNPs mainly localized in the boundary region of ECM-rich tumor tissues, indicating that dense ECM structures inhibited the deep tumor penetration of drug-loaded nanoparticles (Figure 4d). However, HIFU-treated DOX-CNPs localized substantially in the deep inner part of ECM-rich tumor tissues via the HIFU-derived destruction of the dense ECM structure. These results indicate that HIFU treatment on ECM-rich tumor tissues helped the deep tumor penetration of DOX-CNPs in vivo. *Pharmaceutics* **2020**, *12*, x 11 of 16 It is deduced that HIFU treatment at targeted tumor tissues could destroy dense ECMs composed of collagen and hyaluronan, resulting in the deep tumor penetration of DOX-CNPs at ECM-rich A549 tumor tissues [28,48]. Surprisingly, intravenously injected DOX-CNPs could be successfully accumulated at in ECM-rich tumors exposed to HIFU treatments. Lastly, the excised tumor tissues were further observed using fluorescence microscopy. The NIRF microscopic images showed that untreated DOX-CNPs mainly localized in the boundary region of ECM-rich tumor tissues, indicating that dense ECM structures inhibited the deep tumor penetration of drug-loaded nanoparticles (Figure 4d). However, HIFU-treated DOX-CNPs localized substantially in the deep inner part of ECM-rich tumor tissues via the HIFU-derived destruction of the dense ECM structure. These results indicate that HIFU treatment on ECM-rich tumor tissues helped the deep tumor penetration of DOX-CNPs in vivo.

**Figure 4.** In vivo near-infrared fluorescence (NIRF) imaging of Cy5.5-labeled DOX-CNPs in ECM-rich A549 tumor animal model. (**a**) Biodistribution of Cy5.5-labeled DOX-CNPs without and with HIFU treatment (intensity: 5 W/cm2, frequency: 1.5 MHz, duty cycle: 10%, pulse repetition frequency: 1 Hz, time per spot: 30 s, interval: 2 mm, expose time: 5 min). The red and white dot circles indicate tumor site. (**b**) Ex vivo NIRF imaging of liver, lung, spleen, kidney, heart, and tumor at 24 h post-injection. (**c**) Mean NIRF signal intensity of ex vivo NIRF image (Spl.; spleen, Kid.; kidney). (**d**) Ex vivo NIRF microscopic images of deep tumor penetration of untreated Cy5.5-labeled DOX-CNPs and HIFUtreated Cy5.5-labeled DOX-CNPs in ECM-rich tumor tissues. **Figure 4.** In vivo near-infrared fluorescence (NIRF) imaging of Cy5.5-labeled DOX-CNPs in ECM-rich A549 tumor animal model. (**a**) Biodistribution of Cy5.5-labeled DOX-CNPs without and with HIFU treatment (intensity: 5 W/cm<sup>2</sup> , frequency: 1.5 MHz, duty cycle: 10%, pulse repetition frequency: 1 Hz, time per spot: 30 s, interval: 2 mm, expose time: 5 min). The red and white dot circles indicate tumor site. (**b**) Ex vivo NIRF imaging of liver, lung, spleen, kidney, heart, and tumor at 24 h post-injection. (**c**) Mean NIRF signal intensity of ex vivo NIRF image (Spl.; spleen, Kid.; kidney). (**d**) Ex vivo NIRF microscopic images of deep tumor penetration of untreated Cy5.5-labeled DOX-CNPs and HIFU-treated Cy5.5-labeled DOX-CNPs in ECM-rich tumor tissues.

#### *3.4. In Vivo Therapeutic E*ffi*cacy Using HIFU-Triggered DOX-CNPs in A549 Tumor-Bearing Mice Pharmaceutics* **2020**, *12*, x 12 of 16

The in vivo therapeutic efficacy of HIFU-triggered DOX-CNPs in tumors was monitored up to 24 days. When tumors grew to approximately 60 <sup>±</sup> 5 mm<sup>3</sup> , saline, DOX (2 mg/kg), DOX-CNPs (2 mg/kg of DOX) were injected into the A549 tumor-bearing mice through the tail vein. At 1, 3, 5 and 7 days after injection, tumor tissues were treated with HIFU (+HIFU) (intensity: 5 W/cm<sup>2</sup> , frequency: 1.5 MHz, duty cycle: 10%, pulse repetition frequency: 1 Hz, and time per spot: 30 s) for 5 min. In the control, saline-injected mice without and with HIFU treatment did not show any therapeutic efficacy during the experiment. However, free DOX, DOX (+HIFU), DOX-CNPs showed a mild inhibitory effect on tumor growth, indicating free DOX was not enough to kill A549 tumor cells. Moreover, DOX-CNPs did not present an enhanced therapeutic efficacy of drug-loaded nanoparticles in ECM-rich tumor tissues due to the limited deep tumor penetration effect. In particular, HIFU-treated DOX-CNPs showed an improved therapeutic efficacy, compared to free DOX and DOX-CNPs without HIFU treatment. At 22 days post-treatment, the mean tumor volumes of DOX-CNPs (+HIFU) were greatly suppressed to 110.46 <sup>±</sup> 18.52 mm<sup>3</sup> , compared to free DOX (+HIFU) (197.01 <sup>±</sup> 21.22 mm<sup>3</sup> ) and DOX-CNPs without HIFU treatment (187.77 <sup>±</sup> 18.30 mm<sup>3</sup> ) (Figure 5a). To demonstrate the organ toxicity of DOX-CNPs after HIFU treatment, H&E-stained tissue images of liver and kidney after 22 days post-injection confirmed that there was no organ toxicity (Figure S5). Furthermore, all animal groups showed no changes in survival rate during the treatment (Figure 5b). These results indicate that the HIFU treatment of DOX-CNPs could greatly increase the antitumor efficacy in ECM-rich tumor models, resulting in the deep tumor penetration of drug-loaded nanoparticles at targeted tumor tissues. *3.4. In Vivo Therapeutic Efficacy Using HIFU-Triggered DOX-CNPs in A549 Tumor-Bearing Mice*  The in vivo therapeutic efficacy of HIFU-triggered DOX-CNPs in tumors was monitored up to 24 days. When tumors grew to approximately 60 ± 5 mm3, saline, DOX (2 mg/kg), DOX-CNPs (2 mg/kg of DOX) were injected into the A549 tumor-bearing mice through the tail vein. At 1, 3, 5 and 7 days after injection, tumor tissues were treated with HIFU (+ HIFU) (intensity: 5 W/cm2, frequency: 1.5 MHz, duty cycle: 10%, pulse repetition frequency: 1 Hz, and time per spot: 30 s) for 5 min. In the control, saline-injected mice without and with HIFU treatment did not show any therapeutic efficacy during the experiment. However, free DOX, DOX (+ HIFU), DOX-CNPs showed a mild inhibitory effect on tumor growth, indicating free DOX was not enough to kill A549 tumor cells. Moreover, DOX-CNPs did not present an enhanced therapeutic efficacy of drug-loaded nanoparticles in ECMrich tumor tissues due to the limited deep tumor penetration effect. In particular, HIFU-treated DOX-CNPs showed an improved therapeutic efficacy, compared to free DOX and DOX-CNPs without HIFU treatment. At 22 days post-treatment, the mean tumor volumes of DOX-CNPs (+HIFU) were greatly suppressed to 110.46 ± 18.52 mm3, compared to free DOX (+HIFU) (197.01 ± 21.22 mm3) and DOX-CNPs without HIFU treatment (187.77 ± 18.30 mm3) (Figure 5a). To demonstrate the organ toxicity of DOX-CNPs after HIFU treatment, H&E-stained tissue images of liver and kidney after 22 days post-injection confirmed that there was no organ toxicity (Figure S5). Furthermore, all animal groups showed no changes in survival rate during the treatment (Figure 5b). These results indicate that the HIFU treatment of DOX-CNPs could greatly increase the antitumor efficacy in ECM-rich tumor models, resulting in the deep tumor penetration of drug-loaded nanoparticles at targeted tumor tissues.

**Figure 5.** In vivo therapeutic efficacy of HIFU-triggered DOX-CNPs in ECM-rich A549 tumor-bearing mice. (**a**) Antitumor efficacy of DOX-CNPs with HIFU treatment (*n* = 5 per group). The arrows indicate DOX-CNP injection and HIFU treatment (intensity: 5 W/cm2, frequency: 1.5 MHz, duty cycle: 10%, pulse repetition frequency: 1 Hz, time per spot: 30 s, interval: 2 mm, expose time: 5 min). (\*) indicates **Figure 5.** In vivo therapeutic efficacy of HIFU-triggered DOX-CNPs in ECM-rich A549 tumor-bearing mice. (**a**) Antitumor efficacy of DOX-CNPs with HIFU treatment (*n* = 5 per group). The arrows indicate DOX-CNP injection and HIFU treatment (intensity: 5 W/cm<sup>2</sup> , frequency: 1.5 MHz, duty cycle: 10%, pulse repetition frequency: 1 Hz, time per spot: 30 s, interval: 2 mm, expose time: 5 min). (\*) indicates difference at the *p* < 0.05 significance level. (**b**) Survival rate of A549 tumor-bearing mice treated with saline, saline (HIFU+), DOX, DOX (HIFU+), DOX-CNPs, and DOX-CNPs (HIFU+). The arrows indicate HIFU treatment.
