DUS system and acoustic detection

A commercial DUS system (VINNO70, VINNO Technology Co. Ltd., Suzhou, China) connected to an X4-12 L linear array transducer was used for both therapeutic US exposure and US imaging. The system was equipped with contrast bubble imaging (CBI), integrated contrast-enhanced ultrasound (CEUS) imaging software and flash mode for microbubble destruction. The flash mode was specifically modified to deliver customized pulse sequences for regulating microbubble cavitation or so-called ultrasound stimulated microbubble (USMB), named Vflash. The Vflash pulses can be operated with adjustable frequency, MI, pulse length (PL), PRF and destruction (on)/replenish (off) time as described previously [24]. In addition, the Vflash US beams can be weakly focused to a trapezoid region of interest (ROI) using the electronic focusing method (Fig. 1). The size of the ROI is also adjustable and can cover the tumour body, similar to a colour doppler sample volume.

The peak negative pressure (PNP) within a designated ROI of 1 × 1 cm was measured by a membrane hydrophone (HMB-0500, ONDA Corp., Sunnyvale, CA, USA) positioned 2 cm away from the probe surface. The probe was placed above the hydrophone separated with degassed water in a sink (AIMS III, ONDA Corp., Sunnyvale, CA, USA).

Fig. 1

Schematic illustration of the USMB treatment in rats. A: Tumour perfusion enhancement is related to vasodilator release and promotes Dox uptake. B: Treatment plan 1: 56 SD rats received USMB treatment for 10 minutes and CEUS imaging at baseline, immediately after and 4 hours later. The control group received sham DUS. C: Treatment plan 2: Another 25 rats were enrolled for the Dox delivery study. The animals were treated with the selected USMB parameters and sham. After treatment, Dox solution was injected at approximately 3 hours 20 minutes.

Microbubbles

SonoVue® microbubbles (Bracco Sine Pharmaceutical Corp., Ltd., Shanghai, China) are a commercially available US contrast agent that contains 108 mL− 1 microbubbles with a mean diameter range from 2.0 to 4.0 μm after preparation in 4.0 mL of saline solution. The SonoVue® suspension was then used in both CEUS as an US contrast agent and in therapeutic US as cavitation nuclei. CEUS is a reliable method to assess tumour perfusion. According to the EFSUMB guidelines, the mean time-intensity curves within the tumours after bolus injection of a contrast agent were qualified to reflect microbubble wash-in and washout, thus representing the condition of tumour perfusion [25].

Animal model and experimental design

A total of 81 Sprague‒Dawley (SD) rats bearing subcutaneous Walker-256 tumours were used. The tumour model was made by injecting 0.2 mL of Walker-256 cell suspension (approximately 1 × 107/mL) into the inner thigh of the rat. Then, the model was included in the study when the tumour size reached approximately 1 cm in diameter. All of the animal experimental procedures were approved by the Institutional Animal Care and Use Committee of the university.

Among the 81 SD rats, 56 rats were randomly divided into seven groups, including six experimental groups (A-F) according to different treatment parameter combinations and one control group. The treatment parameters of all groups are illustrated in Table 1. All of the treatments were performed with an X4-12 L linear array transducer operating at a central frequency of 4 MHz. Two PNP outputs of 0.26 MPa and 0.32 MPa (equal to MI values of 0.13 and 0.16, respectively), which were measured by the hydrophone, were selected to test the PNP variable in the experimental groups. Under the fixed PL of 10.5 cycles, we selected 50 Hz, 1 and 2 kHz as three PRF variables (Table 1).

In treatment plan 2 (Fig. 1C), Dox served as a chemotherapeutic agent because it was detectable by fluorescent imaging and quantified by high-performance liquid chromatography (HPLC). For Dox delivery, another 25 rats were randomly divided into one experimental group (n = 14) and one control group (n = 11).

Table 1 The treatment parameters of all groups. Experimental procedures

The animals were anaesthetized by intraperitoneal injection of 2% pentobarbital sodium at 2 ml/kg, and the tumour surface was shaved and depilated. A catheter connected to a 22G needle was inserted into the caudal vein to establish the channel for intravenous injection. High-resolution two-dimensional (2-D) DUS was performed with the same VINNO70 system and the X4-12 L transducer to find the maximal dimension of the tumour section (Fig. 1). Then, a standard CEUS was conducted staying on the section using low MI contrast mode and an intravenous bolus injection of 0.15 mL SonoVue®. Ten minutes after the CEUS study, the hand-held transducer was placed in contact with the tumour surface but separated with a 2-cm-thick gel pad while the Vflash treatment was turned on for 10 minutes. The parameters for USMB treatment were different in each group (Table 1). During the USMB treatment, 0.4 mL of SonoVue® suspension was slowly and constantly injected into the caudal vein during the treatment. After treatment, CEUS performance was repeated twice on the same 2-D section, immediately and 4 hours later (Fig. 1B). The control group received only sham US exposure without MB injection.

For the Dox study, the experimental animals were treated with PNP of 0.26 MPa and PRF of 1.0 kHz combination based on previous results of the best tumour perfusion enhancement, while the control received sham US exposure. Three hours and 20 minutes after treatment, 10 mg/kg Dox solution (Meilun, Dalian, China) was injected through the tail vein (Fig. 1C).

Tumour perfusion quantitation

The dynamic video clips of CEUS before treatment, immediately after treatment and 4 hours after treatment were analysed by the perfusion parametric imaging software of the machine. After manual drawing of the tumour borderline, the machine could automatically generate a time-intensity curve (TIC) of tumour contrast intensity, including the peak intensity (PI) and area under curve (AUC) data. The PI is the peak value of the TIC, and the AUC is integrated by the area under the TIC within 60 s starting from TIC elevation.

For the calculation of the tumour perfusion area rate, the images of the largest tumour contrast perfusion area in the clip were intercepted. Then, the tumour perfusion area was manually delineated using Adobe Photoshop CC (Adobe), and the rate of tumour perfusion area was calculated by the perfusion area/entire tumour area ×100%. The increment of the tumour perfusion area rate was calculated by the percentage of perfusion rate after treatment minus the percentage of perfusion rate before treatment.

Vasodilators and inflammatory factors

Immediately after the experimental procedures, 56 animals in the perfusion study were sacrificed by inhalation of carbon dioxide with exposure to 100% CO2 at a filling rate of 20% cv/min. The tumours from Groups B and E and the control were harvested. The tumour tissues were minced into small pieces and homogenized. Then, the homogenates were centrifuged to obtain the supernatant for enzyme-linked immunosorbent assays (ELISAs).

The contents of eNOS, PGE2, PGD2, PGF2, PGI2, C3a, C5a, LTC4 and TNF-α in tumour tissues were determined by the Rat eNOS-3 ELISA Kit, Rat PGE2 ELISA Kit, Rat PGD2 ELISA Kit, Rat PGF2α ELISA Kit, Rat PGI2 ELISA Kit, Rat C3a ELISA Kit, Rat C5a ELISA Kit, Rat LTC4 ELISA Kit and Rat TNF-α ELISA Kit, respectively (MEIMIAN Industrial Co., Ltd., Jiangsu, China). The absorbance optical density (OD) of each well was measured at 450 nm. The levels of ATP, NO and ROS in tumour tissues were determined by an ATP assay kit, NO assay kit and reactive oxygen species assay kit (Nanjing Jiancheng Bioengineering Institute, China), and the OD values were measured by Microplate Reader according to the instructions.

Dox concentration

For the quantification of the Dox concentration, the rats in the treated group (n = 14) and the control group (n = 11) were sacrificed 40 minutes after Dox infusion. Approximately half of the tumour bulk tissues were taken, and the Dox content was determined by HPLC.

Another half of the tumour sample was frozen and sliced, the nuclei were stained with DAPI, and the sections were examined under a fluorescence microscope (Nikon Eclipse C1, Nikon, Japan). Dox can spontaneously emit red light, while the nuclei appeared blue under UV excitation.

Histological examination

One tumour sample from the treated group or the control was stained with haematoxylin and eosin (H&E) for morphological observation. Under a light microscope, tumour cells are surrounded by connective tissue in a disordered arrangement.

Statistical analysis

SPSS 25.0 software was used for statistical analysis. Multifactor repeated-measures ANOVA was used to determine the influence of different groups on the blood perfusion of Walker-256 tumours at different time points for the PI, AUC and tumour perfusion area of CEUS. If there was an interaction, it was necessary to test the separate effects, and the Bonferroni method was used for pairwise comparison. The contents of ATP, eNOS, PGF2, PGI2, LTC4, TNF-α and ROS in tumour tissues were analysed by one-way ANOVA with a completely random design, and the LSD method was used for further comparison between groups. The variance of NO, PGE2, PGD2, C3a and C5a in tumour tissues was uneven. The independent sample Kruskal‒Wallis rank sum test was used, and Bonferroni correction was used for further comparison between groups. The concentration of Dox in tumour tissues was determined by an independent sample T test. A p value less than 0.05 was considered statistically significant.