High-intensity focused ultrasound applied to the placenta using a toroidal transducer: a preliminary ex-vivo study.

OBJECTIVE: To demonstrate in an ex-vivo model the feasibility of applying high-intensity focused ultrasound (HIFU) using a toroidal transducer for the creation of placental lesions. METHODS: In this study we used a toroidal transducer, composed of 32 ring-shaped emitters with an ultrasound probe at the center, operating at a frequency of 2.5 MHz. We examined 45 human placentae, following either normal vaginal delivery or medical termination of pregnancy between 17 and 40 gestational weeks. First, the attenuation coefficients of 12 human placentae were measured and integrated into a numerical model for simulating HIFU lesions. Then, using acoustic parameters from this preliminary study, we performed ex-vivo experiments with 33 human placentae, each overlain with an animal abdominal wall to simulate the maternal wall. We created single HIFU lesions in 25 of these placentae, and a series of six juxtaposed lesions in eight, studying these both sonographically and macroscopically. RESULTS: Human placental attenuation coefficients of the 12 human placentae ranged from 0.072 to 0.098 Np/cm/MHz, according to gestational age. The 25 single HIFU lesions created had an average diameter of 7.1 ± 3.2 mm and an average depth of 8.2 ± 3.1 mm. The average diameter of the eight series of six juxtaposed HIFU lesions was 23.0 ± 5.0 mm and the average depth was 11.0 ± 4.7 mm. The average thickness of the abdominal walls was 10.5 ± 1.8 mm. No lesions or damage were observed in intervening tissues. CONCLUSION: This study demonstrates, using an ex-vivo model, the feasibility, reproducibility, harmlessness and effectiveness of HIFU applied to the human placenta.

Imaging of shear waves induced by Lorentz force in soft tissues.

This study presents the first observation of elastic shear waves generated in soft solids using a dynamic electromagnetic field. The first and second experiments of this study showed that Lorentz force can induce a displacement in a soft phantom and that this displacement was detectable by an ultrasound scanner using speckle-tracking algorithms. For a 100 mT magnetic field and a 10 ms, 100 mA peak-to-peak electrical burst, the displacement reached a magnitude of 1 µm. In the third experiment, we showed that Lorentz force can induce shear waves in a phantom. A physical model using electromagnetic and elasticity equations was proposed. Computer simulations were in good agreement with experimental results. The shear waves induced by Lorentz force were used in the last experiment to estimate the elasticity of a swine liver sample.

Salvage high-intensity focused ultrasound ablation for prostate cancer local recurrence after external-beam radiation therapy: prognostic value of prostate MRI.

AIM: To assess the prognostic value of magnetic resonance imaging (MRI) before salvage high-intensity focused ultrasound (HIFU) for locally recurrent prostate cancer after external-beam radiotherapy (EBRT). MATERIALS AND METHODS: Forty-six patients who underwent prostate MRI before salvage HIFU for locally recurrent prostate cancer after EBRT were retrospectively studied. HIFU failure was defined as a prostate-specific antigen (PSA) value >nadir + 2 ng/ml (Phoenix criteria) or positive follow-up biopsy or initiation of any other salvage therapy. The following prognostic parameters were assessed: neoadjuvant hormone therapy, clinical stage and Gleason score of recurrence, PSA level and velocity at HIFU treatment, and six MRI-derived parameters (prostate volume, tumour volume, extracapsular extension, seminal vesicle invasion, tumour extension into the apex or anterior to the urethra). RESULTS: Two factors were significant independent predictors of salvage HIFU failure: the PSA level at HIFU treatment (p < 0.012; risk ratio: 1.15, 95% CI: 1.03-1.29) and the tumour extension anterior to the urethra, as assessed by MRI (p = 0.046, risk ratio: 2.51, 95% CI: 1.02-6.16). CONCLUSION: The location of cancer recurrence anterior to the urethra on MRI is an independent significant predictor of salvage HIFU failure for locally recurrent prostate cancer after EBRT. Therefore, MRI may be useful for patient selection before post-EBRT salvage HIFU ablation.

Assisted hepatic resection using a toroidal HIFU device: an in vivo comparative study in pig.

PURPOSE: Bleeding is the main cause of postoperative complications during hepatic surgery. Blood loss and transfusions increase tumor recurrence in liver metastases from colorectal cancer. A high intensity focused ultrasound (HIFU) device with an integrated ultrasound imaging probe was developed for the treatment of colorectal liver metastasis. METHODS: The HIFU toroidal-shaped transducer contains 256 elements (working frequency: 3 MHz) and can create a single conical lesion of 7 cm3 in 40 s. Then, the volume of treatment can be significantly increased by juxtaposing single lesions. Presented here is the use of this device in an animal model as a complementary tool to improve surgical resection in the liver. Before transecting the liver, a wall of coagulative necrosis was performed using this device in order to minimize blood loss and dissection time during hepatectomy. Resection assisted by HIFU was compared to classical dissections with clamping [intermittent Pringle maneuver (IPM) group] and without clamping (control group). For each technique, 14 partial liver resections were performed in seven pigs. Blood loss per dissection surface area and resection time were the main outcome parameters. RESULTS: Conserving liver blood inflow during hepatic resection assisted by HIFU did not increase total blood loss (7.4 +/- 3.3 ml cm(-2)) compared to hepatic resection performed during IPM and controlled blood inflow (11.2 +/- 2.2 ml cm(-2)). Lower blood loss was measured on average when using HIFU, even though difference with clamping (IPM) was not statistically significant (p = 0.09). Resection assisted by HIFU reduced blood loss by 50% compared to control group (14.0 +/- 3.4 ml cm(-2), p = 0.03). The duration of transection when using HIFU (13 +/- 3 min) was significantly lower compared to clamping (23 +/- 4 min, p < 0.01) and control (18 +/- 3 min, p = 0.02). Precoagulation also resulted in sealing blood vessels with a diameter of less than 5 mm, and therefore the number of clips needed in the HIFU group was significantly lower (0.8 +/- 0.2 cm(-2)) when compared to clamping (1.6 +/- 0.2 cm(-2), p < 0.01) and control (1.8 +/- 0.4 cm(-2), p < 0.01). CONCLUSIONS: This method holds promise for future clinical applications in resection of liver metastases.

[Outcomes of HIFU for localised prostate cancer using the Ablatherm Integrate Imaging® device]. / Traitement du cancer localisé de la prostate avec l'appareil Ablatherm Imagerie Intégrée® : résultats carcinologiques et fonctionnels.

OBJECTIVES: To report the functional and oncological outcomes of HIFU for prostate cancer using the Ablatherm Integrate Imaging(®) device. METHODS: Between January 2005 and June 2009, all patients treated with HIFU as a primary care option for localized prostate cancer and fulfilling the French Urological Association (AFU) guideline were included in this study. Validated questionnaires were used to assess continence, potencies and quality of life. RESULTS: A total of 297 patients met the inclusion criteria: 149 were low risk and 148 were intermediate risk according to d'Amico's risk group. The median prostate specific antigen (PSA) nadir was 0.12ng/ml with 65% of patients reaching a nadir less than 0.3 ng/ml. Systematic control biopsies were performed on 175 patients with 89% of negative biopsies. The disease free survival rate at 40 months was 79% for low risk group and 62% for intermediate risk group. The pre and post-HIFU treatment International Prostate Symptoms Score (IPSS) score and quality of life questionnaire were not statistically different. In the opposite, the pre and post-HIFU erection function and continence status were significantly different. CONCLUSION: Local control and Biochemical Free Survival Rate achieved with HIFU were similar to those expected with conformal external radiation beam therapy. Among the functional outcomes, potency was the most impacted by the treatment.

Robot-assisted ultrasound navigation platform for 3D HIFU treatment planning: Initial evaluation for conformal interstitial ablation.

Interstitial Ultrasound-guided High Intensity Focused Ultrasound (USgHIFU) therapy has the potential to deliver ablative treatments which conform to the target tumor. In this study, a robot-assisted US-navigation platform has been developed for 3D US guidance and planning of conformal HIFU ablations. The platform was used to evaluate a conformal therapeutic strategy associated with an interstitial dual-mode USgHIFU catheter prototype (64 elements linear-array, measured central frequency f = 6.5 MHz), developed for the treatment of HepatoCellular Carcinoma (HCC). The platform included a 3D navigation environment communicating in real-time with an open research dual-mode US scanner/HIFU generator and a robotic arm, on which the USgHIFU catheter was mounted. 3D US-navigation was evaluated in vitro for guiding and planning conformal HIFU ablations using a tumor-mimic model in porcine liver. Tumor-mimic volumes were then used as targets for evaluating conformal HIFU treatment planning in simulation. Height tumor-mimics (ovoid- or disc-shaped, sizes: 3-29 cm3) were created and visualized in liver using interstitial 2D US imaging. Robot-assisted spatial manipulation of these images and real-time 3D navigation allowed reconstructions of 3D B-mode US images for accurate tumor-mimic volume estimation (relative error: 4 ± 5%). Sectorial and full-revolution HIFU scanning (angular sectors: 88-360°) could both result in conformal ablations of the tumor volumes, as soon as their radii remained ≤ 24 mm. The presented US navigation-guided HIFU procedure demonstrated advantages for developing conformal interstitial therapies in standard operative rooms. Moreover, the modularity of the developed platform makes it potentially useful for developing other HIFU approaches.

New integrated imaging high intensity focused ultrasound probe for transrectal prostate cancer treatment.

The present study proposes a new integrated imaging (II) high-intensity focused ultrasound (HIFU) probe intended as an improvement to the Ablatherm prostate cancer treatment. Because of a perforation in the center of the II probe, the expected lesion differs from the one obtained for the original Ablatherm probe. In this paper, the new geometry and the strategy followed to establish the treatment parameters are presented. The original probe has a 40-mm focal length, a 50-mm aperture and is truncated at 31 mm. The II probe has a 45-mm focal length, a 61-mm aperture, a central perforation of 25 mm and is truncated at 31 mm. Both probes operate at 3 MHz. A mathematical model for lesion prediction was used for setting the treatment parameters for the II probe. These parameters should ensure equivalence between the lesions obtained with the original and II probes. Simulation-obtained parameters were validated by in-vitro and in-vivo (on liver of 70 New Zealand rabbits) experiments. The new II probe was used clinically to treat 30 patients. The mean age was 70.9 +/- 5.3 years (SD), the mean prostate volume 26.9 +/- 7.7 mL and the mean serum prostate specific antigen (PSA) concentration before treatment was 9.2 +/- 5.5 ng/mL. Simulations showed that for the II probe acoustical power and duration when the transducer is inactive should be reduced of 14% and 1s. In-vitro and in-vivo experiments confirmed the equivalence between the lesions obtained with the two probes. The lesion volume obtained under in-vitro conditions (for a traversed tissue depth of 16 mm to the focus) was 5 +/- 0.4 cm(3) and 5.1 +/- 0.5 cm(3) for the original and II probes, respectively. Under in-vivo conditions, the lesion volume (for a traversed tissue depth of 18 mm) was 5.3 +/- 1.1 cm(3) and 5.1 +/- 1.1 cm(3) for the original and II probes, respectively. During the clinical trial, a correction of + 1s in the exposure time was required to recreate the same degree of efficacy observed with the original probe (p = 0.97): 66.7 % of negative biopsies and 75% of patients with PSA at 3 mo < or =1 ng/mL. The morbidity observed was minimal and identical to that observed with the original probe.

Morphological analysis of the interstitial ultrasonic ablation in porcine liver in vivo.

BACKGROUND: The shape of the induced thermal ablation area is as important as its dimension. The aim of this study was to analyze the size reproducibility and the aspect of the interstitial ultrasonic ablation obtained by a planar transducer in porcine liver in vivo. METHODS: Five pigs were used. Two complete ultrasonic lesions were made in each animal under pedicle clamping. All the lesions underwent MR examination on day 7 and then a histological analysis. RESULTS: The tested probe has the advantage of providing a step-by-step and highly directional treatment in the target zone. The ultrasonic lesions presented as well-defined and homogenous areas of tissue coagulation. The lesion volumes ranged from 8.1 to 92.3 cm3 with an averaged lesion length of 56 mm at gross examination. Three-dimensional reconstruction of the lesions from the MR images showed cylindrical and conical shapes. Large intrahepatic vessels distorted the lesion shape, and the vicinity of the application to the liver surface increased significantly the volume of the ultrasonic necrosis. Histological examination showed complete necrosis in the area of damage. CONCLUSION: The ultrasonic ablation has a regular shape, always with sharply defined borders. However, it showed some variability in the size of the induced lesions.

[MR detection of local prostate cancer recurrence after transrectal high-intensity focused US treatment: preliminary results]. / Détection par IRM des récidives locales du cancer de prostate après traitement par ultrasons focalisés de haute intensité (HIFU) transrectaux: étude préliminaire.

PURPOSE: To assess T2W and dynamic: contrast-enhanced (DCE) MR imaging in the detection of local tumor recurrence after transrectal high-intensity focused US (HIFU) treatment. MATERIALS AND METHODS: Fifteen patients treated by HIFU for prostate cancer were referred for MR due to biological evidence of tumor recurrence. Axial, sagittal and coronal T2W images and DCE images (12 3-mm thick axial images, temporal resolution: 15 seconds) were obtained first. Transrectal biopsies were then obtained under US guidance. MR findings were compared to biopsy results for 10 prostate sectors. RESULTS: Biopsies demonstrated tumor recurrence in 13/15 patients (23/108 sectors). On T2W images, the treated prostate tissue was diffusely hypointense which interfered with interpretation. Three patients (5 sectors) had suspicious areas of T2W signal abnormality and 15 patients (29 sectors) had suspicious areas on DCE scans. An analysis per sector for T2W and DCE imaging showed sensitivity, specificity, positive predictive and negative predictive values respectively of 0.13, 0.98, 0.6 and 0.81 and 0.70, 0.85, 0.55 and 0.91. DCE MR was strongly predictive of positive biopsy results (Odds ratio: 12.8 (95% confidence interval: 4.4-37.3)) whereas T2W imaging was not (Odds ratio: 4.0 (95% confidence interval: 0.5-30)). CONCLUSION: MR, especially DCE MR, is promising for the detection and localization of local prostate cancer recurrence after transrectal HIFU treatment.

[Locally recurrent prostatic adenocarcinoma after exclusive radiotherapy: results of high intensity focused ultrasound]. / Adénocarcinome prostatique en récidive locale après radiothérapie exclusive: résultats du traitement par ultrasons focalisés.

OBJECTIVES: To determine the efficacy and adverse effects of high intensity focused ultrasound (HIFU) for the treatment of local recurrence of prostate cancer after exclusive external beam radiotherapy. MATERIAL AND METHODS: Seventy-two patients with histologically and biologically documented local recurrence after radiotherapy were treated by HIFU. The mean age was 68.27+/-5.93 years, and mean PSA was 6.64+/-7.26ng/ml. Thirty patients were treated according to standard parameters and 42 according to specific parameters. ASTRO 2005 criteria, specific for salvage therapy (Phoenix consensus), were used to define recurrence. Progression-free survival was calculated by the Kaplan-Meier method. RESULTS: Mean follow-up was 39+/-28 months. The negative biopsy rate was 80% and the median nadir PSA was 0.10ng/ml. Specific survival was 94% at three years and 90% at five years, and progression-free survival was 50% at three years and 44% at five years. The urinary incontinence rate was 44% (grade 1 : 12%, grade 2/3 : 32%) and the urethral stricture or bladder neck stenosis rate was 30%. The use of specific parameters reduced the incidence of severe incontinence (19% versus 50, P=0.005) and stenosis (24% versus 40). CONCLUSIONS: Treatment with HIFU achieved a five-year progression-free survival of 44%, but patients must be clearly informed about the high rate of adverse effects.

[A tumour-mimic pig liver model for guiding focused ultrasound thermal ablation]. / Modèle de pseudotumeur pour guider l'ablation thermique par ultrasons focalisés: étude sur le porc.

There is no established liver tumour model in pigs to study the efficacy of ablative treatment options available for the treatment of liver tumours by physical agents. A tumour-mimic model visible with high contrast on sonograms and on gross pathology has been studied at mid-term on 20 pigs. The aim was to determine if these tumour-mimics are well tolerated and can be used to validate the use of thermal therapies at a preclinical stage. The dimensions of the tumour-mimics measured on sonograms were reproducible (diameter: 9.6 +/- 1.9 mm) and correlated with those performed in gross pathology (R(2)=0.73). The accuracy of focused ultrasound thermal therapy can be evaluated preclinically using these tumour-mimics.

[Indications, techniques and outcomes of high-intensity focused ultrasound (HIFU) for the treatment of localized prostate cancer]. / Indications, techniques et résultats du traitement par ultrasons focalisés (HIFU) du cancer localisé de la prostate.

High-intensity focused ultrasound (HIFU) is a minimally invasive alternative for patients with localized prostate cancer, not suitable for radical prostatectomy because of a life expectancy less than 10 years or because of major co-morbidities precluding surgery. HIFU can be performed in patients with LUTS (associated TURP) or with a previous history of BPH surgery. HIFU is repeatable after the initial procedure if a recurrent cancer is diagnosed on control biopsies. Furthermore, this therapy is a viable option for patients with a local relapse after external beam radiation therapy: oncologic efficacy is conversely related to the initial prostate cancer stage before radiation therapy.

Cytotoxic effects of acoustic cavitation on HT-29 cells and a rat peritoneal carcinomatosis in vitro.

Damage to cells and tissues exposed to shock waves (SWs) is thought to be secondary to cavitation phenomena involving the collapse of gas bubbles in a fluid. Using HT-29 cells and DHDK12PROb tumors, we tried to enhance SW-related damage by the simultaneous administration of gas microbubbles. Bubbles resulted from a mixture of air and gelatin (HT-29 cells) or from a carbonated NaCl solution (tumors). HT-29 cells in suspension received either SW (50, 250, or 1000 SWs) alone or in association with bubbles. Trypan blue-negative cells decreased as the number of SWs increased. Exposure to SWs and bubbles resulted in not only an increased but also a delayed mortality as compared to SWs only. One thousand SWs with bubbles induced a complete inhibition of cell growth, with cytoplasmic vacuolae, ruptured membranes, and abnormal nuclear shape and chromatin. Exponential and confluent cells exhibited a similar mortality and growth. DHDK12PROb tumors received either SWs only (50, 100, 250, 500, or 1000 SWs) or SWs with bubbles in vitro. Thymidine incorporation was significantly lower after exposure to SWs with bubbles as compared with controls and SWs only; it was nil by 1000 SWs with bubbles. Histopathological features of tumors exposed to SWs with bubbles included erosion and hemorrhage, disorganized structure, pyknotic nuclei, and cytoplasmic vacuolae. We conclude that cavitation, as produced by a combination of SWs and gas microbubbles, can achieve bioeffects which are relevant to cancer therapy.

In vivo effects of high-intensity ultrasound on prostatic adenocarcinoma Dunning R3327.

High-intensity ultrasound has been used to treat Dunning R3327 prostatic adenocarcinoma implanted s.c. in Fischer Copenhagen rats. Focused ultrasound was generated with a 1-MHz transducer and energy was provided by a 7.5-kW power amplifier. Seventy-four rats were treated using two different sublines of Dunning tumor. Study 1 dealt with 49 rats with the Mat-Ly-Lu subline, treated with acoustic intensities ranging from 300 to 2750 W/cm2. Of the 49 rats in Study 1, 30 had complete tumor necrosis and 19 had no effect; of the 30 who had complete local tumor necrosis, 14 had local relapse, 9 had distance metastases to lung and nodes without local occurrence, and 7 remained free of tumor and were still alive 12 months after treatment. In Study 2, 25 rats with AT2 subline were treated with an intensity of 820 W/cm2. Similarly for Study 2, there was complete local tumor necrosis in 24 of 25 animals, with local regrowth in 7 of 24 and no recurrence of metastasis in the remaining 16 after a follow-up of 3 months. These results suggested that high-intensity focused ultrasound could be useful for the treatment of small localized cancerous tumors such as low-grade prostate carcinoma.

Elastography for the follow-up of high-intensity focused ultrasound prostate cancer treatment: initial comparison with MRI.

We previously developed an ultrasonic elastography imaging system that may provide a simple and cost-effective solution to monitor high-intensity focused ultrasound (HIFU) treatments. The objective of this clinical study was to evaluate the reliability of our system in assessing the volume of HIFU lesions in the prostate, using a comparison with magnetic resonance imaging (MRI). Elastograms were obtained in 20 patients after HIFU treatment for prostate cancer and gadolinium-enhanced T1- and T2-weighted MRI was performed. Lesion boundaries were manually outlined and the volume was calculated. A statistically significant correlation of rho = 0.62 (p = 0.022) was found between elastographic and MRI measurements of lesion volume, with elastographic measurements that generally underestimated the volume measured in MRI. Some basic physics (hypoechoic areas) and instrumentation (frame rate and band width) issues that were detrimental to image quality in vivo are reported, along with propositions to improve the technique. Because of these issues and, although good correspondence between elastographic and MRI measurements was found in some patients, elastographic measurements were unable to predict MRI measurements in a single individual. Nevertheless, the results confirmed the potential of elastography for monitoring HIFU treatment of the prostate. Further investigation will be conducted using better suited ultrasound equipment and performing real-time elastogram calculations.

Electronic beam steering of shock waves.

Shock-wave generators are now currently used for the treatment of renal stones. In all these generators the focal zone is determined by their geometrical parameters. We propose, for the first time, a piezocomposite shock-wave generator with electronic focusing. The system is composed of a two-dimensional array and its electronic hardware. The array is composed of 121 independent piezocomposite transducers arranged in a spherical shell 20 cm in diameter and focused at 190 mm. The electronic hardware includes 121 x 6 kV-impulse generators. The interdelay of each channel can be adjusted between 10 ns to 100 microseconds by steps of 10 ns. The results show: the use of composite material is possible for the generation of high amplitude pressure waves; the pressure-voltage relationship is linear up to a pressure of about 28 x 10(5) Pa at the transducer front face; the material can be used for a long period of time; i.e., after one million shocks, no decrease in sensitivity, no alteration in its electrical behaviour and no time wave form distortion were observed. Electronic focusing is efficient in an ellipsoidal region of about 4 cm in diameter and 6 cm in length. The pressure in the focal zone is about 600 x 10(5) Pa.

Ultrasound-induced tissue ablation: studies on isolated, perfused porcine liver.

Minimally invasive methods for the treatment of cancers, such as high-intensity focused ultrasound (HIFU) and high-energy shock waves (SW), have been proposed recently. Their feasibility for treatment of human cancer needs to be confirmed. A simplified model of isolated perfused pig liver that is close to the human liver in vivo has been proposed. The objective was to study the feasibility of deep focused tissue ablation with HIFU and SW in large organs approaching the size of the human liver. The model was demonstrated to be physiologically valid during the first 2 h of anoxic perfusion with a composite saline solution; arterial and portal pressure, enzymes, urea levels and bile secretion remained stable. It can simulate the major effects of perfusion and physical phenomena that occur in vivo during treatment. Histological analysis revealed no major changes. Previous results obtained in vivo in animal models at a depth of 2-3 cm were successfully reproduced and deeper lesion arrays at 4, 6, 8 and 9 cm from the surface were produced using the same principles. The depth of 9 cm from the liver surface is consistent with an extracorporeal treatment of most of the liver segments in man. Other applications of the model are proposed, particularly for the study of the role of interferences such as ribs and intestinal gas, blood perfusion and respiratory movements.

Development of an interstitial ultrasound applicator for endoscopic procedures: animal experimentation.

Biliary cancer is very difficult to treat, mainly because of the advanced stage at which such tumours are detected and the low efficacy of systemic therapeutic modalities like radiotherapy. Palliative measures designed to clear the duct (either by means of surgery or an endoscopic procedure) are presently performed. A relatively noninvasive alternative could be developed to fill this gap in the therapeutic arsenal. To this end, we have designed an interstitial ultrasound (US) applicator suitable for use with a digestive endoscope. This applicator is based on a water-cooled plane transducer that operates at 10 MHz. Although, because the target zone is cylindrical in shape, it might have seemed more logical to use a cylindrical transducer. Nevertheless, a plane transducer was chosen because the pressure field from this kind of emitter decreases less quickly, which means faster and deeper heating. However, to generate coagulation necrosis all around the duct, the applicator has to be rotated around its axis; this is achieved by means of a flexible metallic shaft (2 m in length and 3. 8 mm in diameter) that joins the device's active head (which contains the transducer) to the casing with all the connectors. A holder is fixed at the endoscope channel inlet; it controls the rotation of the applicator. Trials were conducted on pigs. The duodenoscope was introduced via the oesophagus down through the duodenum as far as the hepatopancreatic ampulla. Using a guide wire, the applicator was navigated into the duct via the endoscope instrument channel. Well defined, reproducible volumes of coagulation necrosis with a diameter of 20 mm were generated in the biliary tissue and the liver. These promising results indicate that this kind of endoscopic US delivery system may represent an effective tool for the treatment of biliary tumours in humans. An Independent Ethics Committee recently approved preliminary clinical trials of this applicator.