Advertisement

Transluminal Approach with Bubble-Seeded Histotripsy for Cancer Treatment with Ultrasonic Mechanical Effects

Open AccessPublished:March 07, 2018DOI:https://doi.org/10.1016/j.ultrasmedbio.2018.01.023

      Abstract

      Bubble-seeded histotripsy (BSH) is a newly developed ultrasound-based mechanical fractionation technique using locally injected phase change nanodroplets (PCNDs) as sensitizers. The PCNDs are a kind of microbubble precursor compressed into submicron-size in droplets form, which were designed for local administration and will expand into microbubbles under ultrasound exposure. Previously, we reported that a combination of PCNDs injection and pulsed high-intensity focused ultrasound (pHIFU) with an acoustic intensity as low as about 3 kW/cm2 at 1.1 MHz, which is similar to the acoustic intensity of currently available HIFU coagulation therapy, was enough to induce tissue fractionation after significant antitumor effects in an in vivo study. Toward therapeutic application of BSH to deep-seated tissues such as the pancreas, the transluminal approach, using endoscopic ultrasound was thought to be ideal. Therefore, for a preliminary examination, we developed a new transducer with a small aperture (20- × 20-mm square) and long focal length (35 mm), operating at 2.1 MHz that could be attached to an EUS-mimicking probe. With the newly developed transducer and locally injected PCNDs, predictable tissue mechanical fractionation was observed in both ex vivo and in vivo studies at acoustic intensities that were too low to induce any significant bioeffects (around 4 kW/cm2) without using PCNDs. For in situ monitoring of the treatment site during a procedure, the degree of attenuation of microbubble motions after exposing the microbubbles to pHIFU was monitored, using ultrafast echographic imaging. Microbubble movements were observed to be largest at 25–30 s after pHIFU exposure. On the contrary, after 40 s, the movement of microbubbles decreased to the same level as at the start of the procedure, suggesting that an overdose of pHIFU exposure causes coagulation attributable to the thermal effect caused by absorption of the energy. Those results were promising for expanding the application of BSH for a transluminal approach, using a small transducer under real-time monitoring.

      Key Words

      Introduction

      Histotripsy (
      • Hall T.L.
      • Hempel C.R.
      • Wojno K.
      • Xu Z.
      • Cain C.A.
      • Roberts W.W.
      Histotripsy of the prostate: Dose effects in a chronic canine model.
      ,
      • Lake A.M.
      • Hall T.L.
      • Kieran K.
      • Fowlkes J.B.
      • Cain C.A.
      • Roberts W.W.
      Histotripsy: Minimally invasive technology for prostatic tissue ablation in an in vivo canine model.
      ,
      • Roberts W.W.
      • Hall T.L.
      • Ives K.
      • Wolf Jr., J.S.
      • Fowlkes J.B.
      • Cain C.A.
      Pulsed cavitational ultrasound: A noninvasive technology for controlled tissue ablation (histotripsy) in the rabbit kidney.
      ,
      • Xu Z.
      • Ludomirsky A.
      • Eun L.Y.
      • Hall T.L.
      • Tran B.C.
      • Fowlkes J.B.
      • Cain C.A.
      Controlled ultrasound tissue erosion.
      ) has made the mechanical effects of ultrasound a main part of research for therapeutic application and developed it into an ultrasonic minimally invasive therapy system. Histotripsy is a controllable mechanical tissue fractionation induced by repetitive short ultrasound pulses at extremely high intensities, which was originally reported by a research group at the University of Michigan (
      • Xu Z.
      • Ludomirsky A.
      • Eun L.Y.
      • Hall T.L.
      • Tran B.C.
      • Fowlkes J.B.
      • Cain C.A.
      Controlled ultrasound tissue erosion.
      ). A clinical trial for the treatment of benign prostate hyperplasia (
      • Roberts W.W.
      Development and translation of histotripsy: Current status and future directions.
      ) is now ongoing, and research on other applications, such as transcranial clot liquefaction (
      • Zhang X.
      • Owens G.E.
      • Gurm H.S.
      • Ding Y.
      • Cain C.A.
      • Xu Z.
      Noninvasive thrombolysis using histotripsy beyond the intrinsic threshold (microtripsy).
      ), prostate cancer and renal stones, are being carried out intensively (
      • Roberts W.W.
      Development and translation of histotripsy: Current status and future directions.
      ). Before the development of histotripsy, ultrasound bioeffects used for therapeutic applications had mainly been thermal effects that raise the targets' temperatures enough to induce tissue coagulation (
      • Hill C.R.
      • ter Haar G.R.
      Review article: High intensity focused ultrasound–potential for cancer treatment.
      ). Therapies using such thermal effects are referred to as high-intensity focus ultrasound (HIFU) therapies (
      • Aus G.
      Current status of HIFU and cryotherapy in prostate cancer—A review.
      ), and now they are clinically available for both benign (
      • Ebert T.
      • Graefen M.
      • Miller S.
      • Saddeler D.
      • Schmitz-Drager B.
      • Ackermann R.
      High-intensity focused ultrasound (HIFU) in the treatment of benign prostatic hyperplasia (BPH).
      ,
      • Hynynen K.
      • Pomeroy O.
      • Smith D.N.
      • Huber P.E.
      • McDannold N.J.
      • Kettenbach J.
      • Baum J.
      • Singer S.
      • Jolesz F.A.
      MR imaging-guided focused ultrasound surgery of fibroadenomas in the Breast: A Feasibility Study.
      ) and malignant diseases (
      • Hill C.R.
      • ter Haar G.R.
      Review article: High intensity focused ultrasound–potential for cancer treatment.
      ,
      • Hou A.H.
      • Sullivan K.F.
      • Crawford E.D.
      Targeted focal therapy for prostate cancer: A review.
      ,
      • Khokhlova T.D.
      • Hwang J.H.
      HIFU for palliative treatment of pancreatic cancer.
      ).
      Histotripsy has many potential advantages over HIFU therapy as follows: (i) It is supposed to be less dependent on the vascularity of the tissues and more suitable to treat sites adjacent to large vessels. Applying HIFU therapy to such sites may result in insufficient therapeutic effects attributable to a mechanism referred to as the heat sink effect (
      • Jiang F.
      • He M.
      • Liu Y.J.
      • Wang Z.B.
      • Zhang L.
      • Bai J.
      High intensity focused ultrasound ablation of goat liver in vivo: Pathologic changes of portal vein and the “heat-sink” effect.
      ). (ii) The therapeutic process of histotripsy can easily be monitored as the acoustic images change during the procedure (
      • Vlaisavljevich E.
      • Kim Y.
      • Allen S.
      • Owens G.
      • Pelletier S.
      • Cain C.
      • Ives K.
      • Xu Z.
      Image-guided non-invasive ultrasound liver ablation using histotripsy: Feasibility study in an in vivo porcine model.
      ). In HIFU therapy, it is difficult to monitor the temperature change caused in conjunction with the change of acoustic parameters because acoustic parameters such as sound speed and attenuation coefficient have different temperature dependencies (
      • Ghoshal G.
      • Luchies A.C.
      • Blue J.P.
      • Oelze M.L.
      Temperature dependent ultrasonic characterization of biological media.
      ) and are difficult to measure separately. (iii) Mechanical tissue fractionation induced by histotripsy allows anticancer drugs to penetrate cancer cells, while the thermal effects of HIFU cause tissue coagulation followed by fibrosis, which makes it difficult for anticancer drugs to reach the cancer cells.
      Although histotripsy has much potential as described earlier in this report, it needs a very high amplitude and negative peak pressure of more than 20 MPa (
      • Hall T.L.
      • Hempel C.R.
      • Wojno K.
      • Xu Z.
      • Cain C.A.
      • Roberts W.W.
      Histotripsy of the prostate: Dose effects in a chronic canine model.
      ), which roughly corresponds to 40 kW/cm2 (
      • Xu J.
      • Bigelow T.A.
      • Nagaraju R.
      Precision control of lesions by high-intensity focused ultrasound cavitation-based histotripsy through varying pulse duration.
      ). Therefore, transducers with large aperture sizes are necessary, which makes it difficult to apply histotripsy to a lesion by an intercostal approach, and the treatment of deep-seated regions is thought to be difficult because of the attenuation of acoustic energy inside the body.
      Therefore, considering the histotripsy characteristics discussed earlier in this report, we previously developed a therapeutic system referred to as bubble-seeded histotripsy (BSH) that combines pulsed ultrasound and a locally injected sensitizer to induce mechanical fractionation that is similar to histotripsy with a reduced acoustic amplitude that is as low as that of HIFU therapy (several kW/cm2 intensity), utilizing conventional HIFU exposure systems. For BSH, we use a special contrast agent as a sensitizer that changes its phase from liquid to vapor (microbubbles), referred to as phase change nanodroplets (PCNDs) (
      • Asami R.
      • Ikeda T.
      • Azuma T.
      • Umemura S.
      • Kawabata K.
      Acoustic signal characterization of phase change nanodroplets in tissue-mimicking phantom gels.
      ). Moreover, we aimed to develop a new system for cancer treatment, using direct low-dose antitumor agent administration (
      • Ashida R.
      • Kawabata K.
      • Maruoka T.
      • Asami R.
      • Yoshikawa H.
      • Takakura R.
      • Ioka T.
      • Katayama K.
      • Tanaka S.
      New approach for local cancer treatment using pulsed high-intensity focused ultrasound and phase-change nanodroplets.
      ,
      • Kawabata K.
      • Maruoka T.
      • Asami R.
      • Ashida R.
      2013 Controlled induction of mechanical bioeffects with pulsed ultrasound and chemical agents.
      ,
      • Kawabata K.
      • Maruoka T.
      • Asami R.
      • Ashida R.
      Acousto-chemical manipulation of drug distribution: In vitro study of new drug delivery system.
      ). No other studies on cancer treatments, using tissue fractionation at such a low intensity with the aid of PCNDs seem to have succeeded, although the effects of nanodroplets on histotripsy have been investigated with tissue-mimicking phantoms for drug delivery purposes (
      • Vlaisavljevich E.
      • Aydin O.
      • Yuksel Durmaz Y.
      • Lin K.-W.
      • Fowlkes B.
      • ElSayed M.
      • Xu Z.
      Effects of ultrasound frequency on nanodroplet-mediated histotripsy.
      ,
      • Vlaisavljevich E.
      • Aydin O.
      • Durmaz Y.Y.
      • Lin K.-W.
      • Fowlkes B.
      • Xu Z.
      • ElSayed M.E.H.
      Effects of droplet composition on nanodroplet-mediated histotripsy.
      ).
      Regarding the transluminal approach, several studies report on HIFU therapy using the endoscopic approach for the liver or pancreas. Hwang's group successfully ablated porcine liver and pancreatic tissues with an endoscopic ultrasound (EUS)-guided HIFU system (
      • Li T.
      • Khokhlova T.
      • Maloney E.
      • Wang Y.-N.
      • D'Andrea S.
      • Starr F.
      • Farr N.
      • Morrison K.
      • Keilman G.
      • Hwang J.H.
      Endoscopic high-intensity focused US: Technical aspects and studies in an in vivo porcine model (with video).
      ). However, histotripsy-like mechanical effects does not seem to occur in their situation. Therefore, it is worthwhile to develop therapeutic systems for the transluminal approach, utilizing mechanical effects, given the advantages of the mechanical effects as described earlier in this report.
      In this report, based on the previous successful feasibility test of BSH, we further discuss the feasibility of BSH being applied in the transluminal approach. A transluminal approach would be suitable for treating deep-seated tissues such as the pancreas, but it is challenging to develop a system because a small aperture and long focal length is essential, and the system needs to be very small. The conceptual differences between transducers for conventional and transluminal approaches are presented in Figure 1. In addition to showing a transducer for conventional extracorporeal approach, Figure 1 describes a transducer for micro histotripsy (
      • Woodacre J.
      • Landry T.
      • Brown J.
      Real-time imaging, targeting, and ablation of ex-vivo tissue using a handheld histotripsy transducer and coregistered 64-element high-frequency endoscopic phased array.
      ), which aims to treat very small surface regions.
      Fig. 1
      Fig. 1Conceptual difference in transducer shape for histotripsy for transluminal approach and other conventional transducers.
      We also developed a real-time BSH monitoring method during the treatment. Previously we used the strain ratio to evaluate the stiffness of the tissue after the procedure. However, it is difficult to monitor the strain ratio in real time because it cannot be measured during the pHIFU exposure. Therefore, in this study, we utilized real-time movement evaluation of microbubbles derived from PCNDs, using ultrafast imaging during the procedure.

      Materials and Methods

      Chemicals

      Perfluoro-n-pentane (PFP), perfluoro-n-hexane (PFH) and chloroform were purchased from Wako Pure Chemical Industries (Osaka, Japan) and used without further purification. Dipalmitoyl-phosphatidylcholine (DPPC), dipalmitoyl-phosphatidic acid (DPPA) and PEGylated dipalmitoyl-phosphatidylethanolamine (PEG-DPPE) were purchased from NOF Corporation (Tokyo, Japan).

      PCND preparation

      PCNDs were prepared as described previously (
      • Asami R.
      • Ikeda T.
      • Azuma T.
      • Umemura S.
      • Kawabata K.
      Acoustic signal characterization of phase change nanodroplets in tissue-mimicking phantom gels.
      ) with minor modifications in composition as follows. DPPC, DPPA and PEG-DPPE were mixed at a molar ratio of 8:1:1, and phospholipids were added to adjust the total concentration of 8 mM The mixture was then combined with 4.5% (v/v) of PFP and PFH. A laser-diffraction particle-size analyzer (Beckman Coulter LS13320, Beckman Coulter, Inc., Brea, CA, USA) was used to measure the particle-size distribution of the PCNDs suspensions. The final concentration of perfluorocarbon was measured with a gas chromatograph (Hitachi High-Technologies G-6000, Hitachi, Ltd., Tokyo, Japan). The concentration of PFP and PFH (4.5%) is used as the original concentration of PCNDs in this study.

      Experimental setup

      The experimental setup for ultrasound exposure used in this study is presented schematically in Figure 2. All the experiments were performed in a poly (methyl 2-methylpropenoate) tank (inner bottom: 400 × 300 mm2, height: 200 mm) filled with continuously degassed water kept at 37 ± 1 °C. The samples were either fresh chicken breast tissues or CDF1 mice with subcutaneously inoculated colon 26 tumor tissues. The oxygen concentration of the water was verified before each series of experiments and confirmed to be less than 2.0 g/m3. A 2.1-MHz focused transducer (with an aperture of 20- × 20-mm square and a 35-mm focal length) was used as the source of ultrasound (Fig. 3). The transducer consists of an air-backed piezoelectrically active lead zirconate titanate (PZT) element (Fuji Ceramics, Shizuoka, Japan) and an aluminum housing. A Wavetek 195 arbitrary wave generator and an E&I 2100 L RF amplifier were used to generate ultrasound waves. In the experiments, echographic monitoring was performed with an EUP-F331 micro-convex diagnostic ultrasound probe (center frequency = 6.5 MHz, Radius = 10 mm, Hitachi Ltd.), which was aligned adjacent to the previously described transducer coaxially facing at the focal point of the transducer. As shown in Figure 2, the probe and the transducer are facing each other with an angle of 24°. The probe was connected to a Vantage 256 system (Verasonics, Kirkland, WA, USA), which was used for the monitoring. On monitoring, the pulse transmission and reception were synchronized with the transducer firing. Upon the firing, a triggering signal is sent to the scanner and transmitting and following receiving pulses were performed.
      Fig. 3
      Fig. 3Outlook of ultrasound transducer in this study.

      Acoustic-pressure measurement

      The acoustic pressure distribution of the ultrasound transmitted by the transducer at the focus was measured with a calibrated hydrophone (Onda HGL-0085, Onda Corporation, Sunnyvale, CA, USA) in degassed water by scanning the hydrophone in depth and lateral directions with a pitch of 0.1 mm at an acoustic power of 5.5 W. The measured pressure distributions are presented in Figure 4. From the pressure distribution, the −6 dB focal acoustic intensity was calculated and used in this study by assuming that the acoustic intensity is a quadratic function of the input voltage applied toward the transducer within the range in this study.
      Fig. 4
      Fig. 4Distribution of ultrasound pressure amplitude at focus of transducer: (a) lateral direction and (b) longitudinal direction.

      Total acoustic power measurement

      The total acoustic power of the transducer was estimated by measuring the radiation force. The details of the measurement are described elsewhere (
      • Umemura S.
      • Kawabata K.
      • Sasaki K.
      Enhancement of sonodynamic tissue damage production by second-harmonic superimposition: Theoretical analysis of its mechanism.
      ). Briefly, we used a modified steel-ball method that involves using an air-filled aluminum plate instead of a steel ball to receive the radiation force.

      Echographic monitoring

      With the Vantage 256 system, two types of monitoring were performed at 6.5 MHz with an EUP-F331 probe attached to the transducer for inducing fractionation. (i) For positioning monitoring, we used conventional scanlines with focused beams at a frame rate of 30 Hz. (ii) For treatment process monitoring, we used diverging-wave imaging, utilizing a single plane-wave imaging mode at a frame rate of 6,250 Hz. In both cases, the applied voltage was 50 V. For the latter case, the obtained raw echo-intensity data were utilized for further analysis offline. The pulse transmission and reception with the probe were performed synchronizing with the pulse exposure with the transducer in a different manner. In the positioning monitoring, the pulse transmission is performed 0.01 s after each.

      Ultrasound-exposure sequence

      A pulsed regime was used for the ultrasound exposure sequence, as shown in Figure 5. Three parameters—pulse duration, pulse interval and acoustic intensity—were investigated. For the monitoring of the treatment process, echographic images were taken at 160 µs intervals and were synchronized to the pulsed ultrasound for the treatment, as shown in Figure 6.
      Fig. 5
      Fig. 5Pulsed ultrasound exposure regime in this study.
      Fig. 6
      Fig. 6Sequence for monitoring treatment process.

      Ex vivo experiments

      Freshly prepared chicken breast tissues purchased on the day of experiments were used for the ex vivo experiments. Tissues were cut into blocks of about 10 × 10 × 3 cm. After being packed into a 0.03-cm-thick polyethylene bag with degassed saline, the sample was submerged into a degassed water tank to be set in front of the transducer as shown in Figure 2. One of the smallest planes of the sample was placed to face the transducer. A 1.0-mL PCNDs aliquot was injected into the tissue with a 24-gauge syringe while the tissue was echographically monitored. The focus of the transducer was set to the injection site. After pulsed ultrasound exposure, the sample was immediately cut and the lesions were determined. The research protocol on the animal experiments were approved by the Institutional Animal Care and Use Committee (IACUC) of Research & Development Group, Hitachi, Ltd. On the experiments, all institutional and national guidelines for the care and use of laboratory animals were followed.

      In vivo experiments

      CDF1 mice with subcutaneously implanted colon 26 tumor tissues were used for the in vivo experiments. The colon 26 tumor cells originated from those obtained from the Cell Resource Center for Biomedical Research, Tohoku University, Sendai, Japan, were passaged through subcutaneous implantation in mice. Male CDF1 mice (5 wk of age) were purchased from Charles River Japan (Tokyo, Japan). The mice were anesthetized and the hair around the tumor tissues was shaved with a razor. After having been shaved, the mice were fixed to a plastic-made clamp with an acoustic window for the ultrasound exposure. After the mice were put into the water tank, a 1.0-mL PCNDs aliquot was injected into the tumor tissue with a 24-gauge syringe while the tissue was echographically monitored. The focus of the transducer was set to the injection site. The concentration of PCNDs used was 1/10 of that of the stock solution prepared, which corresponds to 0.45% (v/v) of the perfluorocarbon concentration. Immediately after the pulsed ultrasound exposure, the mice were sacrificed while anesthetized, and the tumor tissues were sampled.

      Fractionation volume measurement

      In this study, fractionation volumes generated by the treatment were measured as follows: First, immediately after treatment, tissue samples were cut to expose the site at the focal plane in the longitudinal direction. Then, the tissues at the focus were removed very gently by spatula, and the long and short diameters were measured, using calipers assuming that the lesion is a spheroid. Finally, the lesion volume was calculated with eqn (1), where V is the fractionation volume, A is the diameter at the long axis and B is the long axis.
      V=43A2B
      (1)


      Correlation coefficient calculation

      In this study, the bubble movement after an ultrasound pulse was applied for monitoring. The 2-D correlation coefficient, a value that is generally used to represent the degree of correlation between images, was calculated between two temporally adjacent echographic images obtained by treatment process monitoring.
      The definition of the 2-D correlation coefficient R is shown in eqn (2). In the equation, A and B are the pixel data of echographic images at a certain time and pixel data of echographic images successively obtained, respectively. The x and y denote coordinates in the horizontal and vertical directions, respectively. The A¯and B¯ are the mean values of A and B.
      The calculations were performed within the ROI set in the 12- × 12-mm area centered on the predetermined focal point of the ultrasound pulse.
      Not conventional log-compressed data, but raw echo signal intensity data were used for the calculation.
      R=x=x_minx_maxy=y_miny_max(A(x,y)A¯)(B(x,y)B¯)x=x_minx_maxy=y_miny_max(A(x,y)A¯)2)(x=x_minx_maxy=y_miny_max(B(x,y)B¯)2)
      (2)


      Fractionation index

      In this study, an index was used to determine how advanced the tissue fractionation is, which is based on correlation coefficient r, which is obtained as described earlier in this report. Fraction index I was calculated with eqn (3), where F0 and F1 are the first and last frame for the calculation, α is a scaling factor, Δt is time between frames and R(f) is the correlation coefficient calculated between frame number f and f+1. The Δt is included in the equation to obtain an independent I value for calculation with a different frame rate. In this study, F0 and F1 were set to 10 (1.2 ms after pulse exposure is stopped) and 30 (4.5 ms after pulse exposure is stopped), and α was set to 106 to set I value reasonably high to handle with ease.
      I=αf=F0F1Δt(1R(f))
      (3)


      Results

      Ex vivo experiments with chicken breast tissues

      To study the feasibility of BSH with a small prototype transducer, experiments were performed using chicken breast tissues. The ultrasound intensity investigated in this study was in the range of conventional HIFU coagulation therapy; thus, the pulsed ultrasound we used in this study is described as pHIFU hereinafter.
      Figure 7 shows a typical echographic change of chicken breast tissues during pHIFU exposure after local PCNDs injection. The acoustic parameters described in Figure 6 are as follows: Pulse duration = 140 µs, pulse interval = 5 ms and acoustic intensity = 4.7 kW/cm2.
      Fig. 7
      Fig. 7Typical echographic change induced in chicken breast tissue exposed to pulsed ultrasound. (a) Before and (b) during exposure.
      During pHIFU exposure, besides bright three lines that correspond to exposed pHIFU, a brightness change appears at the focus area of the pHIFU transducer. No such brightness change at the focus was observed in the absence of PCNDs injection alone as previously reported (
      • Ashida R.
      • Kawabata K.
      • Maruoka T.
      • Asami R.
      • Yoshikawa H.
      • Takakura R.
      • Ioka T.
      • Katayama K.
      • Tanaka S.
      New approach for local cancer treatment using pulsed high-intensity focused ultrasound and phase-change nanodroplets.
      ).
      The change of the gross appearance of the tissue sample after 30-s pHIFU exposure is presented in Figure 8. The middle of the tissue where the ultrasonic wave was focused turned white and was discolored, as presented in Figure 8 (upper). However, this lesion was easily removed with a spatula, as presented in Figure 8 (lower), suggesting BSH was generated successfully even with the small transducer. Figure 9 presents the gross appearance of the tissue sample after pHIFU exposure in the same acoustic conditions as Figure 8 but replacing PCNDs with saline. No significant gross change was observed. Thus, it is obvious that the non-coagulated mechanical fractionation is formed by the combination of PCNDs and pHIFU.
      Fig. 8
      Fig. 8Typical gross image of chicken breast tissues after exposure to pulsed ultrasound in presence of PCNDs. (a) Before and (b) after removing tissues at lesion with spatula. PCNDs = phase change nanodroplets.
      Fig. 9
      Fig. 9Typical gross image of chicken breast tissues after exposure to pulsed ultrasound in presence of saline.

      Mechanical fractionation with various acoustic parameters

      The degree of tissue fractionation with various acoustic parameters was investigated. Figure 10 shows the change of fractionation volume with various PCND concentrations. The acoustic parameters are as follows: acoustic intensity = 4.7 kW/cm2, pulse duration = 140 µs and pulse interval = 5 ms. The mean value from four independent experiments and standard deviations are plotted. In the concentration range investigated (dilution rate of 1/10, 1/30 and 1/100 and concentration of PCNDs = 0.45, 0.15 and 0.045, respectively), the larger fractionation volume was obtained as the concentration was increased. From these results, the highest concentration (dilution rate of 1/10 and concentration of PCNDs = 0.45%) was used in both the ex vivo and in vivo experiments.
      Fig. 10
      Fig. 10Dependence of fractionation volume on PCNDs concentration. Pulse duration = 140 µs, pulse interval = 5 ms, exposure time = 30 s, acoustic intensity = 4.7 kW/cm2. PCNDs = phase change nanodroplets.
      Figure 11 presents the degree of fractionation volume generated by pHIFU exposure with various acoustic intensities of pHIFU. The acoustic parameters are as follows: Pulse duration = 140 µs and pulse interval = 5 ms. The fractionation volume was measured as described in the Materials and Method section of this report. The mean value from four independent experiments and standard deviations are plotted. There is a threshold at around 4 kW/cm2, and the fractionation volume seems to reach a plateau at 4.7 kW/cm2 in this study. In Figure 11, the results with the injection of saline instead of PCNDs are also plotted (open circle). In this study, no significant biological changes were observed without using PCNDs.
      Fig. 11
      Fig. 11Dependence of fractionation volume on acoustic intensity. Pulse duration = 140 µs, pulse interval = 5 ms, exposure time = 30 s. Solid circle: With PCND injection, Open circle: With saline injection. PCND = phase change nanodroplet.
      Figure 12 presents the degree of fractionation volume generated by pHIFU exposure with various exposure times. The acoustic parameters are as follows: acoustic intensity = 4.7 kW/cm2, pulse duration = 140 µs and pulse interval = 5 ms. The mean value from four independent experiments and standard deviations are plotted. There is a threshold at around 20 s, and lesion generation was induced when the exposure time was longer than this threshold. The maximum fractionation volume was observed at around 30 s. Of interest, even with a longer exposure time than 30 s, the fractionation volume did not increase.
      Fig. 12
      Fig. 12Dependence of fractionation volume on exposure time. Acoustic intensity = 4.7 kW/cm2, pulse duration = 140 µs, pulse interval = 5 ms.
      Figure 13 presents the degree of fractionation volume generated by pHIFU exposure with various sets of acoustic parameters while using the same acoustic intensity of 4.7 kW/cm2. The mean values from four independent experiments and standard deviations are plotted. To emphasize that biological effects discussed in this experiment are not thermal coagulative, temporal mean power, not intensity, is shown. As presented in Figure 13, no significant lesion generation was observed when using a set of acoustic parameters such as pulse duration = 70 µs, pulse interval = 5 ms and mean temporal power = 1.05 W. However, in other cases, with various combinations of pulse duration (70–280 µs) and pulse interval (2.5–10 ms) but with the same temporal mean power of 2.1 W, mechanical fractionation was successfully generated, and the degrees of fractionation volume were similar. From these obtained results, we believe the creation of mechanical tissue fractionation depended on the acoustic power, not the pulse duration or pulse interval itself.
      Fig. 13
      Fig. 13Dependence of fractionation volume on pulse duration and temporal mean power. Acoustic intensity = 4.7 kW/cm2, exposure time = 30 s.

      In vivo experiments with murine tumor tissues

      To further evaluate the feasibility of BSH with a small prototype transducer, using a set of parameters that was determined in the previous ex vivo study, experiments were performed using murine tumor tissue as an in vivo study.
      Figure 14 presents a typical optical change of colon 26 tumor tissues induced by pHIFU exposure with locally injected PCNDs. The acoustic parameters are as follows: pulse duration = 140 µs, pulse interval = 5 ms, exposure time = 30 s, and acoustic intensity = 4.7 kW/cm2. During pHIFU exposure, a brightness change appeared at the focus point of pHIFU that was similar to the echographic change, as presented in Figure 7. The gross change after the procedure is presented in Figure 14 (upper). The treated area was easily removed by spatula, as presented in Figure 14 (lower).
      Fig. 14
      Fig. 14Typical gross image of tumor tissues after exposure to pulsed ultrasound in presence of PCNDs. (a) Before and (b) after removing tissues at lesion with spatula. PCNDs = phase change nanodroplets.
      We also investigate the change of colon 26 tumor tissues with the same acoustic conditions as in Figure 14 but using various exposure times. With an exposure time of 20 s, no significant fractionation (0 cases in 4 experiments) was observed but when the exposure time was either 30 s or 45 s, significant fractionation was observed (4 cases in 4 experiments). However, the longer exposure time of 60 s induced tissue changes that could not be removed by spatula (2 cases in 4 experiments), suggesting the generation of tissue coagulation. Therefore, it was suggested that tissue fractionation begins after the 20-s exposure, but thermal denaturation occurs in reverse when the irradiating time becomes too long, such as 60 s. A typical tissue section is presented in Figure 15 after exposures of 20, 45 and 60 s. Fractionated areas in tissue with the 45-s exposure were easily removed by spatula.
      Fig. 15
      Fig. 15Typical gross image of tumor tissues after exposure to pulsed ultrasound in presence of PCNDs with different exposure time. (a) 20 s, (b) 45 s (after removing tissues at lesion with spatula) and (3) 60 s. PCNDs = phase change nanodroplets.
      Based on the obtained results and previously reported results, we further investigated the possibility of enlarging the fractionation volume by shifting the exposure point of pHIFU with a single PCNDs injection. For this study, the PCNDs injection was performed at the first focus position only. The focus point of pHIFU was shifted 5 times with a pitch of 2 mm after a 30-s exposure at each position. The interval between ultrasound exposure from a site to another was 1 s. The focus shift was performed from the bottom to the top of the tumor. In Figure 16, a typical gross tissue change after pHIFU exposure is presented. As presented in Figure 16, the fractionated tissue was easily removed with a spatula. The results obtained clearly suggest that the lesion where BSH was generated was enlarged by the focus shift of pHIFU, although PCNDs were injected at only one point.
      Fig. 16
      Fig. 16Typical gross image of tumor tissues after 5 exposures to pulsed ultrasound in presence of PCNDs for 30 s. Each 30-s exposure were performed with shifting the focus at a pitch of 2 mm from bottom to top. PCNDs = phase change nanodroplets.

      Real-time monitoring of BSH with mice tumor tissues

      Regarding real-time monitoring of BSH, observation of the movement of microbubbles from PCNDs was thought to be the best way. Regarding evaluation of the movement of the microbubbles, the correlation coefficient and fractionation index were used for measurement in this study.
      Figure 17 shows the temporal changes of correlation coefficient R that were calculated as described in the Materials and Method section of this report during and after a pulse of HIFU was exposed. Three lines are plotted to indicate the temporal changes of R at 2, 30 and 45 s after the start of exposure of pulses. In this series of experiments, 200 pulses were shot to tissues. Thus, 2, 30 and 45 s after the start of exposure, 400, 6000 and 9000 pulses per second are exposed, respectively. Figure 17a presents the temporal change of R and Figure 17b presents a part of Figure 17a. In Figure 17a, in any of these cases, correlation coefficient R rapidly decreases by the pHIFU exposure and gradually comes back to the original value after pHIFU is stopped. In Figure 17b, it can be seen that in any of cases, R reaches a plateau at around 4.5 ms. However, during time between about 1.5–4 ms, R differs depending on the time after the start of exposure. Values of R are the lowest at 30 s followed by 45 s and 2 s, suggesting microbubble movements occurred most often at 30 s after pHIFU exposure started.
      Fig. 17
      Fig. 17Correlation coefficient R as a function of time after ultrasound pulse is stopped. Data at exposure time of 2 s (dashed line), 30 s (solid line), 45 s (dotted line) are plotted. Pulse duration = 140 µs, pulse interval = 5 ms, acoustic intensity = 4.7 kW/cm2. (a) Temporal change of R and (b) enlarged plot of (a).
      Figure 18 presents the fractionation index, as defined in the Materials and Method section of this report, for various exposure times. The mean value from 4 independent experiments and standard deviations are plotted. After exposure begins, the index starts rising at 20 s, reaches a plateau at about 25 s and then goes down again at around 40 s. These results indicate that the movement of the microbubbles will decrease if the irradiation continues for too long.
      Fig. 18
      Fig. 18Fractionation index as a function of exposure time. Pulse duration = 140 µs, pulse interval = 5 ms, acoustic intensity = 4.7 kW/cm2. Exposure time = 50 s.

      Discussion

      This study is a preliminary feasibility test for cancer treatment using BSH and specifically targeting deep-seated tissues such as the pancreas by the transluminal approach. Previous studies suggested that tissue fractionation and further tumor treatment was practicable by using a combination of HIFU range ultrasound intensities, locally injected PCNDs, microbubble precursors as the sensitizers and murine tumor tissues in vivo (
      • Ashida R.
      • Kawabata K.
      • Maruoka T.
      • Asami R.
      • Yoshikawa H.
      • Takakura R.
      • Ioka T.
      • Katayama K.
      • Tanaka S.
      New approach for local cancer treatment using pulsed high-intensity focused ultrasound and phase-change nanodroplets.
      ,
      • Kawabata K.
      • Maruoka T.
      • Asami R.
      • Ashida R.
      2013 Controlled induction of mechanical bioeffects with pulsed ultrasound and chemical agents.
      ,
      • Kawabata K.
      • Maruoka T.
      • Asami R.
      • Ashida R.
      Acousto-chemical manipulation of drug distribution: In vitro study of new drug delivery system.
      ).
      In our previous in vivo study with tumor-bearing mice, it was found that the mechanical effects were induced with locally injected PCNDs by using 1.1-MHz ultrasound pulses at an intensity of several kW/cm2, although no significant biological events were observed without PCNDs. Those mechanical effects uniformly induced cell death in the treated areas and were effective in suppressing tumor growth (
      • Ashida R.
      • Kawabata K.
      • Maruoka T.
      • Asami R.
      • Yoshikawa H.
      • Takakura R.
      • Ioka T.
      • Katayama K.
      • Tanaka S.
      New approach for local cancer treatment using pulsed high-intensity focused ultrasound and phase-change nanodroplets.
      ,
      • Kawabata K.
      • Maruoka T.
      • Asami R.
      • Ashida R.
      2013 Controlled induction of mechanical bioeffects with pulsed ultrasound and chemical agents.
      ,
      • Kawabata K.
      • Maruoka T.
      • Asami R.
      • Ashida R.
      Acousto-chemical manipulation of drug distribution: In vitro study of new drug delivery system.
      ). Moreover, the presence of a low dose of Adriamycin, an anti-tumor agent, resulted in complete tumor disappearance. However, no significant reduction in tumor size was found when pulsed ultrasound, PCNDs and Adriamycin were applied separately. Regarding monitoring for treatment effects, the lesion made by BSH can be visualized as a soft tissue by using ultrasonic strain imaging.
      Encouraged by those results, we also began to investigate whether tissue fractionation is practicable with a transluminal approach by using a transducer with small apertures and a long focal length assuming the EUS approach (Fig. 3).
      The transducer and setup for the experiments were designed as described in a previous study on the endoscopic approach of HIFU therapy on porcine liver and pancreas by
      • Li T.
      • Khokhlova T.
      • Maloney E.
      • Wang Y.-N.
      • D'Andrea S.
      • Starr F.
      • Farr N.
      • Morrison K.
      • Keilman G.
      • Hwang J.H.
      Endoscopic high-intensity focused US: Technical aspects and studies in an in vivo porcine model (with video).
      . Our ex vivo experiments were performed with the same focal length as
      • Li T.
      • Khokhlova T.
      • Maloney E.
      • Wang Y.-N.
      • D'Andrea S.
      • Starr F.
      • Farr N.
      • Morrison K.
      • Keilman G.
      • Hwang J.H.
      Endoscopic high-intensity focused US: Technical aspects and studies in an in vivo porcine model (with video).
      , and most of the acoustic path was filled with breast tissue. Thus, the tissue fractionation induced in this study is expected to occur in our future planning of endoscopic experiments with swine liver or pancreas if the intensity threshold for tissue fractionation does not change significantly from tissue to tissue. In this study, it was shown that murine tumor tissues and swine breast tissues could be fractionated with the same acoustic conditions. Also, according to an ex vivo porcine study, the threshold for histotripsy, the pressure threshold for fractionation of liver tissue and muscle was about the same (
      • Vlaisavljevich E.
      • Cain C.A.
      • Xu Z.
      The effect of histotripsy on tissues with different mechanical properties.
      ). Therefore, we hope we can perform tissue fractionation with BSH in swine experiments. Basic researches such as the dependence of BSH effects on stiffness of tissue-mimicking phantom should be performed before those animal investigations.
      Regarding the fractionation generation, the results obtained in this study were basically consistent with our previous studies. Previous studies were performed with a transducer with an aperture and focal length of 58 mm and frequency of 1.1. MHz, while the current studies use an aperture of 20- × 20-mm square and 35-mm focal distance at 2.1 MHz. In this study, about 4 kW/cm2 acoustic intensities were the threshold to induce BSH, using the current small prototype transducer with a pulse duration of about 150 µs. The values of the acoustic intensities and pulse duration were similar to the value in previous results (3 kW/cm2 and 100 µs), using an extracorporeal transducer. These acoustic intensities are an order of magnitude lower (
      • Xu J.
      • Bigelow T.A.
      • Nagaraju R.
      Precision control of lesions by high-intensity focused ultrasound cavitation-based histotripsy through varying pulse duration.
      ) than those of conventional histotripsy. In contrast, the required pulse duration was about an order of magnitude higher than conventional histotripsy. Generally, generating pulses at a longer duration is easier in an HIFU coagulation treatment system. Thus, these results suggest that BSH-based cancer treatment is practical with minor modifications of the current HIFU system and could be performed in both extracorporeal or transluminal approaches.
      In in vivo experiments, promising results were also obtained, using the current small transducer. As presented in Figure 14, fractionation was induced in murine tumor tissues, using the same acoustic conditions as in ex vivo experiments. Moreover, as presented in Figure 16, it was possible to enlarge the fractionated zone by shifting the focus position of the transducer, which was similar to that of our previous results (
      • Ashida R.
      • Kawabata K.
      • Maruoka T.
      • Asami R.
      • Yoshikawa H.
      • Takakura R.
      • Ioka T.
      • Katayama K.
      • Tanaka S.
      New approach for local cancer treatment using pulsed high-intensity focused ultrasound and phase-change nanodroplets.
      ). The fact that it is unnecessary to carry out multiple PCNDs injections is the most advantageous point of BSH.
      The mechanism of the phenomenon involved in BSH was speculated as presented in Figure 19 (
      • Ashida R.
      • Kawabata K.
      • Maruoka T.
      • Asami R.
      • Yoshikawa H.
      • Takakura R.
      • Ioka T.
      • Katayama K.
      • Tanaka S.
      New approach for local cancer treatment using pulsed high-intensity focused ultrasound and phase-change nanodroplets.
      ). The bubble wall is first generated from PCNDs at the focused area of the ultrasound-creating tissue fractionation, and then this wall advances toward the transducer side because the ultrasound energy reflected by the bubble wall induces the next fractionation on the side of the transducer. After that, it is also suggested that locally injected PCNDs could travel into the fractionated area (
      • Kawabata K.
      • Maruoka T.
      • Asami R.
      • Ashida R.
      Acousto-chemical manipulation of drug distribution: In vitro study of new drug delivery system.
      ); thus, the lesion can be enlarged by shifting the ultrasound focus. According to our previous in vitro results, such a phenomenon should be discussed separately from so-called bubble shielding effect (
      • Zderic V.
      • Foley J.
      • Luo W.
      • Vaezy S.
      Prevention of post-focal thermal damage by formation of bubbles at the focus during high intensity focused ultrasound therapy.
      ). The bubble shielding discussed here is the prevention of HIFU damage at the post-focal region induced by bubbles generated at the focal region through HIFU exposures. With the shielding, HIFU damages at pre-focal regions are widened toward the HIFU transducer as a function of time along the HIFU beam path and HIFU damages are induced widely within or without the focal region of HIFU. Thus, it is very difficult to control the size of lesion with the bubble shielding. However, the bubble-advancing phenomenon discussed in this study is the forming and widening of bubble-rich regions toward the transducer. It was found that such a phenomenon is induced only within the focal region (
      • Kawabata K.
      • Maruoka T.
      • Asami R.
      • Ashida R.
      Acousto-chemical manipulation of drug distribution: In vitro study of new drug delivery system.
      ) and will not be extended beyond the focal region, thus it is a very controllable effect.
      Fig. 19
      Fig. 19Proposed tissue fraction mechanism in bubble-seeded histotripsy.
      Moreover, our results in this study showed the creation of mechanical effects and the coagulation effect were controllable at the fractionated area, which is very crucial for our purpose. The concern before the investigation was that other than a cavitation sensitizer (
      • Kawabata K.
      • Asami R.
      • Azuma T.
      • Yoshikawa H.
      • Sugita N.
      • Umemura S.
      Novel highly-selective and imaging-based ultrasound caner treatment with phase-change nano droplet.
      ,
      • Kawabata K.
      • Maruoka T.
      • Asami R.
      • Umemura S.
      Phase change nanodroplets and microbubbles generated from them as sources of chemically active cavitation.
      ), PCNDs are excellent sensitizers for HIFU thermal therapy as well (
      • Kawabata K.
      • Asami R.
      • Azuma T.
      • Yoshikawa H.
      • Sugita N.
      • Umemura S.
      Novel highly-selective and imaging-based ultrasound caner treatment with phase-change nano droplet.
      ,
      • Kawabata K.
      • Maruoka T.
      • Asami R.
      • Umemura S.
      Phase change nanodroplets and microbubbles generated from them as sources of chemically active cavitation.
      ), and the frequency was increased from 1.1 to 2.1 MHz to make a small transducer, which increases the tissue absorption of ultrasonic energy causing thermal effects. However, under certain circumstances, it was possible to create only mechanical fractionation, although further studies are needed to investigate the combined advantage of thermal coagulation and tissue fractionation. Additionally, the involvedness of the process similar to boiling histotripsy, a type of mechanical tissue fractionation initiated by boiling of tissues with absorbed acoustic energy (
      • Khokhlova T.D.
      • Canney M.S.
      • Khokhlova V.A.
      • Sapozhnikov O.A.
      • Crum L.A.
      • Bailey M.R.
      Controlled tissue emulsification produced by high intensity focused ultrasound shock waves and millisecond boiling.
      ), is not likely to be significant. The tissue fractionation in this study was performed within tissues where PCNDs had been injected. In such a condition, to induce the boiling, extra energy is to be supplied to compensate the energy absorbed by PCNDs other than the energy for tissue boiling. The pulse duration needed for tissue boiling in conventional boiling histotripsy is considered to be more than a millisecond order. However, the duration used in this study is an order of shorter than the threshold. Thus, we speculate that tissue boiling does not play a major role in this study. Moreover, according to an ex vivo study (
      • Khokhlova T.D.
      • Haider Y.A.
      • Maxwell A.D.
      • Kreider W.
      • Bailey M.R.
      • Khokhlova V.A.
      Dependence of boiling histotripsy treatment efficiency on hifu frequency and focal pressure levels.
      ), boiling histotripsy is induced independent of the total HIFU on time. As presented in Figure 13, a clear dependence on total HIFU on time was observed with BSH. In the case of pulse duration of 70 µs, the change of pulse interval from 2.5 to 5 ms, which halved the total HIFU on time, resulted in the decrease of fractionation volume.
      In this study, significant progress was made regarding in situ real-time monitoring of the fractionation process. In general, coagulation and fractionation make tissues stiff and soft, respectively. Therefore, acoustic measurements on stiffness are considered to be effective in judging the therapeutic effect. In histotripsy, there are several reports of acoustic monitoring of the process of fractionation, such as brightness change in echography (
      • Vlaisavljevich E.
      • Kim Y.
      • Allen S.
      • Owens G.
      • Pelletier S.
      • Cain C.
      • Ives K.
      • Xu Z.
      Image-guided non-invasive ultrasound liver ablation using histotripsy: Feasibility study in an in vivo porcine model.
      ), changes in Doppler signals (
      • Miller R.M.
      • Maxwell A.
      • Wang T.-Y.
      • Fowlkes J.B.
      • Cain C.A.
      • Xu Z.
      2012 Real-time elastography-based monitoring of histotripsy tissue fractionation using color Doppler.
      ) and changes in shear wave elastography (
      • Wang T.Y.
      • Hall T.L.
      • Xu Z.
      • Fowlkes J.B.
      • Cain C.A.
      Imaging feedback for histotripsy by characterizing dynamics of acoustic radiation force impulse (ARFI)-induced shear waves excited in a treated volume.
      ). Among them, the last method is most directly related to the stiffness change.
      In BSH, we previously reported that tissue stiffness measured by strain imaging can clearly show the fractionated area in murine tumor after the procedure (
      • Ashida R.
      • Kawabata K.
      • Maruoka T.
      • Asami R.
      • Yoshikawa H.
      • Takakura R.
      • Ioka T.
      • Katayama K.
      • Tanaka S.
      New approach for local cancer treatment using pulsed high-intensity focused ultrasound and phase-change nanodroplets.
      ). However, it was not easy to monitor the fractionation process in a real-time setting, especially in the presence of locally injected PCNDs because they stay at a high concentration, causing echogenic change by producing microbubbles under ultrasound exposure.
      In this study, therefore, we tried to assess the degree of attenuation of the microbubble movements that can be seen immediately after the pulsed ultrasound was turned off. In our previous study, it was found that the movements of microbubbles generated from PCNDs were less attenuated when the surrounding media were softer (
      • Asami R.
      • Ikeda T.
      • Azuma T.
      • Umemura S.
      • Kawabata K.
      Acoustic signal characterization of phase change nanodroplets in tissue-mimicking phantom gels.
      ). Thus, it was predicted that the movement of the microbubbles would be less attenuated if the surrounding tissues become softer because of mechanical fractionation. As described in the Materials and Method of this report, we monitored the bubble motion by using ultrafast imaging at a frame rate of 6,250 Hz.
      To measure the degree of bubble movement, the correlation coefficient of successive frames was used in this study. By measuring the variations of the whole position of microbubbles, using ultrafast imaging, the correlation coefficient can be calculated and then plotted as presented in Figure 17. Although it is difficult to measure the movement of individual bubbles, this method does not require high computing costs; thus, it is advantageous for developing a real-time monitoring method. We also developed an index, which is a simple one that uses temporal summation of the correlation coefficient so that processing will be easier, to show the degree of fractionation quantitatively, using a correlation coefficient.
      As presented in Figure 17, the correlation coefficient of successive frames shows the lowest at 30 s after pHIFU exposure began, suggesting that microbubble movement is the most intense at 30 s, which matches the fact that tissue fractionation occurs most efficiently at 30 s.
      Moreover, as presented in Figure 18, the movements of microbubbles began at around 15 s and reached the peak at around 25 s. Then, the movement showed the same level for a while but decreased after 40 s. These results suggested that the longer exposure of pHIFU will cause coagulation even in the same area where the fractionation was first generated. These results are consistent with the results of the in vivo study using colon 26 tumor tissues that showed tissue denaturation after pHIFU exposure of 60 s, as presented in Figure 15.
      These results suggest two things: (i) the process of fractionation generation by BSH can be monitored by bubble movement and (ii) decrease of the movement index after reaching a plateau can be utilized as a sign, showing that the tissue is exposed enough for BSH.
      With these findings, we believe that the preliminary feasibility test using a small prototype transducer for the transluminal approach showed promising results. It is necessary to create a more practical experiment setup to study the clinical usefulness of BSH including the combined treatment of BSH and local injection of anti-cancer drugs.
      Although designing and fabricating a prototype transducer for practical experiments could be the most significant issue, local injection could also be an issue for the development of BSH. Local injection of drugs to tumor tissues through endoscopy is an established procedure and is performed clinically (
      • Shirley L.A.
      • Aguilar L.K.
      • Aguilar-Cordova E.
      • Bloomston M.
      • Walker J.P.
      Therapeutic endoscopic ultrasonography: Intratumoral injection for pancreatic adenocarcinoma.
      ). However, for BSH, the amount of PCNDs or concentration of PCNDs is not fully understood for clinical application. Moreover, the effects of anti-cancer drug administration need to be investigated for future cancer treatment.
      Tissue coagulation attributed to thermal effects of pHIFU might happen during the BSH when the pHIFU exposure lasts too long. However, it is possible to monitor and control the effects of both the thermal effects and mechanical effects during BSH by using an in-site monitoring method as described earlier in this report. Instead, it might be very useful to have both effects in one therapeutic setting as far as it is controllable because both thermal and mechanical effects have their own advantages and disadvantages, and we can use both effects under the various conditions suited to each effect by changing the pulse exposure parameters (
      • Yoshizawa S.
      • Matsuura K.
      • Takagi R.
      • Yamamoto M.
      • Umemura S.
      Detection of tissue coagulation by decorrelation of ultrasonic echo signals in cavitation-enhanced high-intensity focused ultrasound treatment.
      ); however, further study on this is needed.

      Conclusion

      For developing a novel tumor treatment that utilizes ultrasonic mechanical effects and a locally injected sensitizer, a feasibility study was performed both ex vivo and in vivo. It was found that the transluminal approach is practicable with a small transducer by using a pHIFU of 2 MHz and an acoustic intensity of around 4–5 kW/cm2. It was also found that the fractionation index, using microbubble movement, was useful to monitor the real-time therapeutic process acoustically. These findings are promising for our new treatment method, BSH, to apply for deep-seated tissues.

      Acknowledgments

      This work was supported in part by the Japan Society of Ultrasound in Medicine .

      References

        • Asami R.
        • Ikeda T.
        • Azuma T.
        • Umemura S.
        • Kawabata K.
        Acoustic signal characterization of phase change nanodroplets in tissue-mimicking phantom gels.
        Jpn J Appl Phys. 2010; 49: 1-6
        • Ashida R.
        • Kawabata K.
        • Maruoka T.
        • Asami R.
        • Yoshikawa H.
        • Takakura R.
        • Ioka T.
        • Katayama K.
        • Tanaka S.
        New approach for local cancer treatment using pulsed high-intensity focused ultrasound and phase-change nanodroplets.
        J Med Ultrason. 2015; 42: 557-566
        • Aus G.
        Current status of HIFU and cryotherapy in prostate cancer—A review.
        Eur Urol. 2006; 50 (discussion 934): 927-934
        • Ebert T.
        • Graefen M.
        • Miller S.
        • Saddeler D.
        • Schmitz-Drager B.
        • Ackermann R.
        High-intensity focused ultrasound (HIFU) in the treatment of benign prostatic hyperplasia (BPH).
        Keio J Med. 1995; 44: 146-149
        • Ghoshal G.
        • Luchies A.C.
        • Blue J.P.
        • Oelze M.L.
        Temperature dependent ultrasonic characterization of biological media.
        J Acoust Soc Am. 2011; 130: 2203-2211
        • Hall T.L.
        • Hempel C.R.
        • Wojno K.
        • Xu Z.
        • Cain C.A.
        • Roberts W.W.
        Histotripsy of the prostate: Dose effects in a chronic canine model.
        Urology. 2009; 74: 932-937
        • Hill C.R.
        • ter Haar G.R.
        Review article: High intensity focused ultrasound–potential for cancer treatment.
        Br J Radiol. 1995; 68: 1296-1303
        • Hou A.H.
        • Sullivan K.F.
        • Crawford E.D.
        Targeted focal therapy for prostate cancer: A review.
        Curr Opin Urol. 2009; 19: 283-289
        • Hynynen K.
        • Pomeroy O.
        • Smith D.N.
        • Huber P.E.
        • McDannold N.J.
        • Kettenbach J.
        • Baum J.
        • Singer S.
        • Jolesz F.A.
        MR imaging-guided focused ultrasound surgery of fibroadenomas in the Breast: A Feasibility Study.
        Radiology. 2001; 219: 176-185
        • Jiang F.
        • He M.
        • Liu Y.J.
        • Wang Z.B.
        • Zhang L.
        • Bai J.
        High intensity focused ultrasound ablation of goat liver in vivo: Pathologic changes of portal vein and the “heat-sink” effect.
        Ultrasonics. 2013; 53: 77-83
        • Kawabata K.
        • Asami R.
        • Azuma T.
        • Yoshikawa H.
        • Sugita N.
        • Umemura S.
        Novel highly-selective and imaging-based ultrasound caner treatment with phase-change nano droplet.
        Drug Deliv Syst. 2011; 25: 456-465
        • Kawabata K.
        • Maruoka T.
        • Asami R.
        • Umemura S.
        Phase change nanodroplets and microbubbles generated from them as sources of chemically active cavitation.
        Jpn J Appl Phys. 2011; 50: 1-7
        • Kawabata K.
        • Maruoka T.
        • Asami R.
        • Ashida R.
        2013 Controlled induction of mechanical bioeffects with pulsed ultrasound and chemical agents.
        in: 2013 IEEE international ultrasonics symposium (IUS). IEEE, Piscataway, NJ2013: 2106-2109
        • Kawabata K.
        • Maruoka T.
        • Asami R.
        • Ashida R.
        Acousto-chemical manipulation of drug distribution: In vitro study of new drug delivery system.
        Jpn J Appl Phys. 2014; 53 (07 KF28 1–07 KF28 6)
        • Khokhlova T.D.
        • Hwang J.H.
        HIFU for palliative treatment of pancreatic cancer.
        J Gastrointest Oncol. 2011; 2: 175-184
        • Khokhlova T.D.
        • Canney M.S.
        • Khokhlova V.A.
        • Sapozhnikov O.A.
        • Crum L.A.
        • Bailey M.R.
        Controlled tissue emulsification produced by high intensity focused ultrasound shock waves and millisecond boiling.
        J Acoust Soc Am. 2011; 130: 3498-3510
        • Khokhlova T.D.
        • Haider Y.A.
        • Maxwell A.D.
        • Kreider W.
        • Bailey M.R.
        • Khokhlova V.A.
        Dependence of boiling histotripsy treatment efficiency on hifu frequency and focal pressure levels.
        Ultrasound Med Biol. 2017; 43: 1975-1985
        • Lake A.M.
        • Hall T.L.
        • Kieran K.
        • Fowlkes J.B.
        • Cain C.A.
        • Roberts W.W.
        Histotripsy: Minimally invasive technology for prostatic tissue ablation in an in vivo canine model.
        Urology. 2008; 72: 682-686
        • Li T.
        • Khokhlova T.
        • Maloney E.
        • Wang Y.-N.
        • D'Andrea S.
        • Starr F.
        • Farr N.
        • Morrison K.
        • Keilman G.
        • Hwang J.H.
        Endoscopic high-intensity focused US: Technical aspects and studies in an in vivo porcine model (with video).
        Gastrointest Endosc. 2015; 81: 1243-1250
        • Miller R.M.
        • Maxwell A.
        • Wang T.-Y.
        • Fowlkes J.B.
        • Cain C.A.
        • Xu Z.
        2012 Real-time elastography-based monitoring of histotripsy tissue fractionation using color Doppler.
        in: 2012 IEEE international ultrasonics symposium (IUS). IEEE, Piscataway, NJ2012: 196-199
        • Roberts W.W.
        Development and translation of histotripsy: Current status and future directions.
        Curr Opin Urol. 2014; 24: 104-110
        • Roberts W.W.
        • Hall T.L.
        • Ives K.
        • Wolf Jr., J.S.
        • Fowlkes J.B.
        • Cain C.A.
        Pulsed cavitational ultrasound: A noninvasive technology for controlled tissue ablation (histotripsy) in the rabbit kidney.
        J Urol. 2006; 175: 734-738
        • Shirley L.A.
        • Aguilar L.K.
        • Aguilar-Cordova E.
        • Bloomston M.
        • Walker J.P.
        Therapeutic endoscopic ultrasonography: Intratumoral injection for pancreatic adenocarcinoma.
        Gastroenterol Res Pract. 2013; 2013: 207129
        • Umemura S.
        • Kawabata K.
        • Sasaki K.
        Enhancement of sonodynamic tissue damage production by second-harmonic superimposition: Theoretical analysis of its mechanism.
        IEEE Trans Ultrason Ferroelectr Freq Control. 1996; 43: 1054-1062
        • Vlaisavljevich E.
        • Cain C.A.
        • Xu Z.
        The effect of histotripsy on tissues with different mechanical properties.
        in: 2011 IEEE international ultrasonics symposium (IUS). IEEE, Piscataway, NJ2011: 1490-1493
        • Vlaisavljevich E.
        • Kim Y.
        • Allen S.
        • Owens G.
        • Pelletier S.
        • Cain C.
        • Ives K.
        • Xu Z.
        Image-guided non-invasive ultrasound liver ablation using histotripsy: Feasibility study in an in vivo porcine model.
        Ultrasound Med Biol. 2013; 39: 1398-1409
        • Vlaisavljevich E.
        • Aydin O.
        • Yuksel Durmaz Y.
        • Lin K.-W.
        • Fowlkes B.
        • ElSayed M.
        • Xu Z.
        Effects of ultrasound frequency on nanodroplet-mediated histotripsy.
        Ultrasound Med Biol. 2015; 41: 2135-2147
        • Vlaisavljevich E.
        • Aydin O.
        • Durmaz Y.Y.
        • Lin K.-W.
        • Fowlkes B.
        • Xu Z.
        • ElSayed M.E.H.
        Effects of droplet composition on nanodroplet-mediated histotripsy.
        Ultrasound Med Biol. 2016; 42: 931-946
        • Wang T.Y.
        • Hall T.L.
        • Xu Z.
        • Fowlkes J.B.
        • Cain C.A.
        Imaging feedback for histotripsy by characterizing dynamics of acoustic radiation force impulse (ARFI)-induced shear waves excited in a treated volume.
        IEEE Trans Ultrason Ferroelectr Freq Control. 2014; 61: 1137-1151
        • Woodacre J.
        • Landry T.
        • Brown J.
        Real-time imaging, targeting, and ablation of ex-vivo tissue using a handheld histotripsy transducer and coregistered 64-element high-frequency endoscopic phased array.
        in: 2016 IEEE international ultrasonics symposium (IUS). IEEE, Piscataway, NJ2016: 1-4
        • Xu J.
        • Bigelow T.A.
        • Nagaraju R.
        Precision control of lesions by high-intensity focused ultrasound cavitation-based histotripsy through varying pulse duration.
        IEEE Trans Ultrason Ferroelectr Freq Control. 2013; 60: 1401-1411
        • Xu Z.
        • Ludomirsky A.
        • Eun L.Y.
        • Hall T.L.
        • Tran B.C.
        • Fowlkes J.B.
        • Cain C.A.
        Controlled ultrasound tissue erosion.
        IEEE Trans Ultrason Ferroelectr Freq Control. 2004; 51: 726-736
        • Yoshizawa S.
        • Matsuura K.
        • Takagi R.
        • Yamamoto M.
        • Umemura S.
        Detection of tissue coagulation by decorrelation of ultrasonic echo signals in cavitation-enhanced high-intensity focused ultrasound treatment.
        J Ther Ultrasound. 2016; 4: 15
        • Zderic V.
        • Foley J.
        • Luo W.
        • Vaezy S.
        Prevention of post-focal thermal damage by formation of bubbles at the focus during high intensity focused ultrasound therapy.
        Med Phys. 2008; 35: 4292-4299
        • Zhang X.
        • Owens G.E.
        • Gurm H.S.
        • Ding Y.
        • Cain C.A.
        • Xu Z.
        Noninvasive thrombolysis using histotripsy beyond the intrinsic threshold (microtripsy).
        IEEE Trans Ultrason Ferroelectr Freq Control. 2015; 62: 1342-1355