Development of Histotripsy as a Local-Regional Liver Cancer Therapy: Preclinical to Clinical Translation

Abstract

Histotripsy is the first noninvasive, nonthermal, and nonionizing focused ultrasound ablative therapy that effectively destroys targeted tissue. Histotripsy research began as an early preclinical prototype that eventually underwent small- and large-animal preclinical testing to ensure safe clinical translation to human use. Small-animal studies demonstrated the efficacy of histotripsy in destroying liver tumors and generating immune responses, while large-animal studies in human-sized livers further refined the technology for clinical translation. Preclinical studies revealed that histotripsy is a safe ablative modality that has tissue selectivity, meaning tissues of different structural composition have different thresholds for cavitation-induced destruction, with relative sparing of collagenous structures such as vessels and bile ducts. The noninvasive nature of histotripsy potentially allows for treatment of patients with coagulation abnormalities or those receiving anticoagulation therapy. Additional findings include the creation of well-defined treatment zones that are seen to rapidly resorb over time, a feature allowing for easier treatment response evaluation. This article reviews key observations and results from preclinical and early clinical histotripsy studies leading to U.S. Food and Drug Administration’s approval for the treatment of liver cancer. A thorough understanding of the key findings and biologic effects of histotripsy in animal models will facilitate the clinical adoption of histotripsy.

Keywords: Ultrasound, Ablation Techniques, Animal Studies, Abdomen/GI, Liver

© RSNA, 2025

Summary

Preclinical and clinical histotripsy studies demonstrating feasibility, safety, efficacy, and antitumor immunomodulatory effects support the clinical translation of histotripsy, making it a unique local-regional treatment option for primary and metastatic liver cancer.

Essentials

• ■ Histotripsy is a noninvasive, nonthermal US ablation modality delivering high-pressure, short-duration ultrasound pulses to generate cavitation bubbles, which mechanically destroy the cells in the targeted region.

• ■ Initial safety and feasibility studies in human-sized porcine models have demonstrated that histotripsy can safely and effectively create prescribed ablation volumes within the liver in a clinically acceptable time frame, while sparing large vessels and bile ducts, and the ablated region is resorbed by the body over time.

• ■ Preclinical studies in rodent tumor models have demonstrated a reduced local tumor burden, local and systemic immune responses, and abscopal effects in distant, untargeted tumors.

• ■ The first histotripsy clinical trial in humans provided early evidence for the involution of targeted tumors, as well as abscopal effects resulting in tumor burden reduction in untargeted tumors in select patients.

• ■ #HOPE4LIVER was a phase I/II clinical trial, which reported a technical success rate of 95% (42 of 44 lesions) and a procedure-related major complication rate of 7% (three of 44 participants), encouraging clinical adoption given the recent U.S. Food and Drug Administration’s approval of histotripsy for liver tumor treatment.

Introduction

As millions of people worldwide continue to be diagnosed with primary and metastatic liver cancer, the development of new treatment options is crucial. Current treatment options can be divided into five pillars of cancer therapy, which include surgery, chemotherapy, radiation, immunotherapy, and interventional oncology (minimally invasive/noninvasive local-regional ablative therapies and transarterial ablative therapies). Yet, 5-year survival rates in patients with liver cancer remain low (<10% to approximately 50%) (1,2). Ablative local-regional therapies including radiofrequency ablation, microwave ablation, cryoablation, irreversible electroporation, and high-intensity focused ultrasound thermal ablation (3–6) have inherent limitations, including dependence on tumor stage, liver function, physician expertise, inaccessible tumor locations within the liver, and the heat sink effect caused by blood flow adjacent to the tumor. These factors result in the limitation of the ablation zone size and often unpredictable treatment effects along the tumor margins (7,8).

Histotripsy is an ablation technique that uses a noninvasive, nonradiation, nonthermal mechanism of action. Early clinical studies have shown high rates of treatment efficacy for liver tumors, with low rates of complications. Histotripsy uses high-pressure, short ultrasound pulses to generate acoustic cavitation at a focal point in the target tissue, which results in high-precision mechanical destruction of the target tissue, turning the tissue into acellular homogenate consisting of subcellular components (9,10). This homogenate is then absorbed by the body over time (11,12). Histotripsy is fundamentally different from high-intensity focused ultrasound, a thermal ablation technique that deposits ultrasound energy to cause thermal necrosis in the target tissue (13). The low duty cycle of the histotripsy pulses minimizes bulk tissue heating at the focal zone and allows more efficient and homogeneous ablation while avoiding undesirable thermal biologic effects (14).

Here, we review the mechanism of histotripsy and its biologic effects. We discuss the development and evaluation of histotripsy as a liver cancer therapy from ex vivo and in vivo preclinical assessments to human clinical trials. Finally, we address current limitations and future considerations for continued safe and effective liver cancer treatments.

Histotripsy Mechanism

The first histotripsy study was published in 2004 by a group from the University of Michigan, who coined the name histotripsy—originating from the Greek words histo, which refers to soft tissue, and tripsy, which refers to breakdown (9). Histotripsy applies high-pressure, short duration ultrasound pulses to generate bubbles causing cellular destruction via three different mechanisms. Many studies have shown that tissues contain intrinsic nanometer-sized endogenous gas pockets, which serve as the nuclei for histotripsy-induced cavitation (9,15,16). One technique is intrinsic threshold histotripsy, which uses two or fewer cycles of ultrasound pulses with a single negative pressure phase, such that the peak negative amplitude exceeds a cavitation threshold intrinsic to the target tissue for a fraction of a microsecond, generating cavitation bubbles (15,17). This threshold has been measured in ex vivo experiments to be 26–30 MPa for water-based tissues and 14–17 MPa for adipose tissue (15,18). A second technique is shock-scattering histotripsy, which employs ultrasound pulses of three to 20 cycles with peak negative pressures lower than the intrinsic threshold to create nonlinear shockwaves and generate a cavitation bubble cloud (10,19). In both methods, cavitation microbubbles then expand from an original size of 2–5 nm to more than 100 mm, followed by collapse within a few hundred microseconds, disrupting the cellular structures in the target tissue (10,19). A third technique is boiling histotripsy, developed at the University of Washington in 2010; however, this method is not used clinically. Millisecond-length, high-amplitude, ultrasound pulses are used to create a boiling vapor bubble due to a localized superheating effect at the focus (20,21).

In this review, we focus on the cavitation-based (intrinsic threshold or shock-scattering histotripsy) preclinical histotripsy studies and clinical trials (Table 1).

Table 1:

Summary of Key Findings from Preclinical and Clinical Histotripsy Studies

Study	Year	Key Findings
Feasibility
Vlaisavljevich et al (22)	2013	Approximately 1-cm3 ablation zones were successfully created in the in vivo porcine liver through intact chest and rib blockage.
Smolock et al (30)	2018	Clinical sized 3-cm ablation zones were created in the in vivo porcine liver; the ablation zone was resorbed 4 weeks posttreatment.
Longo et al (32)	2020	Modified ablation volumes were prescribed to overcome the respiratory motion effects on the ablation zone dimensions in the craniocaudal direction.
Safety
Kim et al (33)	2014	Transcostal histotripsy ablation was successfully performed in pig livers with no damage to the ribcage or overlying tissue, and the average temperature increase was <4°C.
Smolock et al (30)	2018	Thermal damage was observed on the body wall in subjects with severely limited acoustic windows. Thrombosis was reported involving hepatic vein and portal vein branches posthistotripsy liver ablation but resolved within 2–4 weeks without treatment.
Longo et al (31)	2019	Pulse modifications were used to decrease the time-average intensity of histotripsy energy >60% to successfully mitigate thermal body wall damage.
Knott et al (35)	2021	Pulse modification used prolonged cooling time during the delivery of histotripsy pulses to prevent severe thermal body wall damage during transcostal ablations.
Vlaisavljevich et al (36)	2017	Liver histotripsy was performed surrounding the hepatic vein with no observable differences in ablation sizes between anticoagulated versus normal pigs.
Mauch et al (37)	2023	There was no elevated risk of hemorrhage following histotripsy in an anticoagulated in vivo porcine model.
Tissue selectivity
Vlaisavljevich et al (41)	2014	Stiffer tissues (higher Young modulus) have higher cavitation thresholds for initiating the histotripsy bubble cloud.
Vlaisavljevich et al (42)	2015	There was a significant decrease in maximum bubble radius and collapse time with both increasing Young modulus and increasing transducer frequency.
Vlaisavljevich et al (43)	2014	Tissues with higher density, higher tensile strength, higher fractional strain, and lower water content were more resistant to damage from histotripsy, that is, collagenous structures such as blood vessels and bile ducts.
Vlaisavljevich et al (22)	2013	Histotripsy ablation in porcine liver showed that large blood vessels >1 mm remained intact, while all vessels <0 μm were disrupted.
Smolock et al (30)	2018	Histotripsy ablation in porcine liver showed that major vessels and bile ducts remained intact, as observed both at MRI and on histology.
Resorption of ablation zone
Vlaisavljevich et al (11)	2016	Complete cell death was observed in histotripsy ablation zones in normal rat livers on acute histology. At 4 weeks, the ablation zone was completely resorbed with only a fibrous scar remaining at the treatment site.
Smolock et al (30)	2018	Ablation zones in pig livers demonstrated complete cell death surrounded by a region of partial necrosis on acute histology. At 4 weeks, the ablation zone size reduced by >64% with fibrous tissue surrounding it.
Tumor burden reduction
Worlikar et al (24)	2018	Human-derived Hep3B hepatocellular carcinoma (HCC) subcutaneous tumors in mice were treated with histotripsy. Acute histology revealed complete acellularization in the ablation zone. In the survival cohort, tumor burden reduced over time initially and eventually recurred by 12 weeks, likely due to the lack of treatment margin.
Worlikar et al (25)	2020	Orthotopic N1S1 rat HCC tumors were treated completely with a 2-mm margin or partially (50%–75% tumor volume targeted) with histotripsy. Complete regression of the tumor was observed posthistotripsy in >93% of treated rats, including those with partial tumor treatment. In these animals, only <5 mm of fibrous tissue remained at the treatment site at 3 months.
Local and systemic immune effects
Qu et al (51)	2020	Histotripsy ablation in bilateral subcutaneous murine melanoma and HCC tumors induced significant innate and adaptive immune response. It was shown that histotripsy could release tumor antigens with retained immunogenicity. Histotripsy ablation of one of the tumors resulted in abscopal effect (reduction of untreated contralateral tumor burden) when used as a standalone therapy and in combination with immune checkpoint inhibitors. Histotripsy ablation of primary tumor demonstrated inhibition of tail vein–induced pulmonary metastases.
Pepple et al (52)	2023	In a bilateral subcutaneous murine liver tumor model, histotripsy ablation of one of the tumors resulted in local immunogenic cell death, leading to antigen-specific abscopal inhibition of the contralateral untreated tumor.
Impact on metastasis
Worlikar et al (53)	2022	Orthotopic McA-RH7777 rat HCC tumors were treated partially (50%–75% tumor volume targeted) with histotripsy. Complete tumor regression (including the untargeted tumor volume) was observed in 81.82% of rats, with no metastases or recurrence at 12 weeks. In comparison, all untreated controls showed tumor progression and developed intrahepatic metastases. Increased immune infiltration and improved survival were observed in the histotripsy-treated animals compared with untreated controls.
Dosimetry
Hendricks-Wenger et al (54)	2021	In a subcutaneous, human-derived cholangiocarcinoma mouse model, three histotripsy doses (250, 500, and 1000 pulses/point) were found to achieve complete ablation. The highest dose created the sharpest ablation boundary, while intact cells were observed in the ablation periphery for the lower doses. An ex vivo experiment evaluated histotripsy doses in human HCC, cholangiocarcinoma, and colorectal liver metastasis (CLM) specimens. Mechanically stiffer CLM tumors needed >1500 pulses/point to achieve complete ablation in comparison to the softer HCC tumors, which required fewer than 500 pulses/point. The fibrotic cholangiocarcinoma tumors could not demonstrate complete ablation of the entire treatment volume even with 4000 pulses/point.
Hendricks-Wenger et al (55)	2022	In a subcutaneous, human-derived cholangiocarcinoma mouse model, histotripsy dose of 500 pulses/point showed improved tumor progression–free and overall survival compared with untreated controls.
Clinical trials and posttrial clinical safety data
Vidal-Jove et al (56)	2022	The first phase I human trial, THERESA, of histotripsy of primary or metastatic hepatic tumors was conducted in Barcelona, Spain, in 2019 (NCT03741088) and reported no procedure-related significant adverse events.
Vidal-Jove et al (59)	2021	In the THERESA trial, abscopal effects were observed in two patients, one each with HCC and CLM, with continuous postprocedure decline in tumor markers. One patient with CLM had shrinkage of nontargeted tumor lesions 8 weeks after the procedure.
Mendiratta-Lala et al (57)	2024	The first phase I/II histotripsy clinical trial #HOPE4LIVER was conducted in Europe and the United States. The coprimary end points were >70% technical success rate (entire tumor treated with margins as preplanned) and <25% major complication rate (no procedure-related major complication within 30 days). Within 36 hours of the procedure, 42 of 44 lesions (95.5% [95% CI: 83.72, 100]) achieved technical success. Three of 44 (6.8% [95% CI: 2.35, 18.23]) participants experienced index procedure-related major complications (common terminology criteria for adverse events [CTCAE] ≥3) within 30 days after the procedure.
Wehrle et al (60)	2025	Real-world clinical safety outcomes from 230 patients with liver cancer treated with histotripsy across 18 centers were reported. Only 12 of 230 patients (5.2%) reported complications within 30 days; of these, nine patients experienced minor complications (Clavien-Dindo grade II), while three patients had disease progression leading to death.
Future directions
Yeats et al (62)	2023	A technique was developed to correct soft tissue aberrations in ex vivo bovine liver and increase the focal pressure amplitude by using acoustic emissions from cavitation cloud initiation and collapse.
Wagner et al (63)	2023	An alternative histotripsy-targeting approach guided by an x-ray C-Arm was developed using cone beam CT (CBCT) imaging to identify the target lesion and two-dimensional fluoroscopy and deep learning to estimate the three-dimensional position of the histotripsy transducer for performing histotripsy without using US guidance.
Wagner et al (64)	2025	Pseudotumors were created using US-guided histotripsy in in vivo pig livers. These pseudotumors were then targeted with an additional margin using CBCT-guided histotripsy to demonstrate feasibility and targeting accuracy.
Lundt et al (65)	2017	Electronic focal steering was used instead of mechanical steering to achieve rapid, homogeneous volumetric ablation at a speed of 0.9–3.3 mL/min, which exceeds any current clinical volumetric ablation method (approximately 1–2 mL/min).

Treatment Setup

Previous studies have used different custom-built and commercially available focused ultrasound therapy transducers for in vivo and ex vivo preclinical histotripsy research. The complete image-guided histotripsy system consists of a therapy transducer and associated electronic driver, an ultrasound imaging system, an ultrasound imaging probe, a three-axis motorized positioner or a robotic arm, and a coupling medium for ultrasound transmission from the transducer to the skin. Figure 1 shows Edison, the current clinical histotripsy system manufactured by HistoSonics. The ultrasound imaging probe (GE HealthCare C1-6 with frequency range of 1.5–6.0 MHz) is coaxially aligned by placing it at the center of the histotripsy treatment transducer to ensure adequate visualization and targeting of the treatment zone (Fig 1A). The robotic arm precisely moves the treatment transducer to ensure adequate coverage of the intended target volume using overlapping focal zones (22). Once the target is identified, the ultrasound therapy pulses are delivered by gradually increasing the applied voltage until a cavitation bubble cloud is observed at US imaging (23). The applied voltage necessary to generate cavitation depends on the acoustic path and ultrasound attenuation, which can vary due to overlying tissue structure across patients. The entire treatment can be observed in real time via simultaneous B-mode US imaging (Fig 1B, 1C) (22,24,25). During treatment, the bubble cloud appears as hyperechoic twinkles at the focal point (Fig 1C). Immediately posthistotripsy, the treatment zone is liquified and appears hypoechoic compared with the pretreatment appearance due to the reduction in the quantity and size of the sound scatterers (Fig 1D) (26,27).

Figure 1:

Histotripsy instrumentation and targeting. (A) The clinical histotripsy treatment system with the therapy transducer mounted on a robotic arm and a coaxially mounted imaging probe. The robotic positioner moves therapy and imaging transducers to perform volumetric treatment using a water bath as an ultrasound coupling medium. (B) Real-time US imaging is used for guiding histotripsy; the tumor (yellow arrows) is visualized pretreatment using B-mode US and contrast-enhanced US (CEUS). (C) Yellow arrow shows the bubble cloud generated at the focus of the therapy transducer. Crosshairs depict the volume to be targeted with histotripsy pulses. (Adapted, under a CC BY 4.0 license, from reference 57.) (D) The targeted region appears anechoic after histotripsy.

Feasibility, Safety, Efficacy, and Biologic Effects of Histotripsy in ex Vivo and in Vivo Studies

Rigorous evaluation has been performed using preclinical small- and large-animal models to determine the safety, feasibility, efficacy, and biologic effects of histotripsy. To date, specific cancer types studied include liver, prostate, renal, breast, pancreatic, musculoskeletal, brain, oral, and skin cancer. A review of the literature on preclinical efficacy for these cancers is beyond the scope of this article and has been discussed previously (28,29). Herein, we will provide an overview of ex vivo and in vivo preclinical studies leading to successful clinical translation of histotripsy for liver cancer treatment.

Initial in Vivo Feasibility Studies

For successful clinical translation, preclinical validation in large-animal models whose anatomy is closer to humans is necessary. Multiple in vivo porcine studies have shown the ability of histotripsy to create effective and safe ablation zones in porcine livers (22,23,30,31).

The first proof-of-concept feasibility study investigating histotripsy for noninvasive liver ablation through intact chest and rib blockage was performed at the University of Michigan in 2013 in an in vivo porcine model (22). In that study, completely ablated lesions measuring approximately 1 cm³ were successfully created in the livers (22).

Thereafter, Smolock et al performed an in vivo survival study to create larger, clinically relevant ablation zones in healthy pig livers (30). Treatment zones of 3.0 cm in size were prescribed to the targeted liver parenchyma through 3–12 cm of overlying tissue, mimicking a human scenario, and average ablation zone sizes ranged from 3.2 to 3.4 cm (30). No significant distress was observed following histotripsy ablation (30). Immediate and 4-week posttreatment MRI of the ablation cavity showed eventual resorption of the treatment zone (Fig 2A, 2B) (30). Histologic examination confirmed uniform tissue disruption at the ablation site with no remaining intact cells (Fig 2C) (30). Of note, larger than expected sizes of the ablation zone in the craniocaudal direction were attributed to respiratory motion, which was studied by Longo et al, who successfully modified the prescribed ablation zone size to compensate for respiratory motion (32).

Figure 2:

Feasibility of creating ablation zones noninvasively in a human-scale, porcine model. (A) Histotripsy ablation zone visualized on posttreatment postcontrast T1-weighted axial MRI scan matched the prescribed ablation volume. (B) MRI scan obtained 4 weeks posttreatment demonstrates a reduction in the volume of the ablation zone, indicating resorption. (C) Photomicrograph (hematoxylin-eosin; original magnification, x10) shows that the ablation zone is acellularized, and fibrous tissue surrounding the boundary of the ablation zone is observed. (Reprinted, with permission, from reference 30.)

Initial in Vivo Safety Studies

Although multiple in vivo studies have shown the efficacy of histotripsy ablation in human-sized livers, there are technical limitations to ultrasound propagation through the body wall into the liver secondary to attenuation from the ribcage. An early study showed comparable ablation zone sizes when created through an intercostal technique with full ribcage coverage without aberration correction versus through only soft tissue (33). No damage to the ribcage or overlying tissue was observed, and the temperature increase near the ribcage remained less than 4°C. However, later large-animal studies employing an intercostal approach reported body wall damage, likely secondary to prefocal cavitation (ie, generation of the bubble cloud proximal to the focal point) on the ribs (30). In the study by Smolock et al, the histotripsy treatments were applied at a pulse repetition frequency (PRF) of 300 Hz (30). There is evidence showing that histotripsy pulses applied at PRF greater than 100 Hz can lead to enhanced prefocal cavitation due to microscopic remnant bubbles (which can persist on the order of 1 second even if histotripsy cavitation cloud persists <100 µsec), attenuating subsequent histotripsy pulses and impacting bubble cloud dynamics (34). Thermal damage on the body wall was typically observed in subjects with severely limited acoustic windows. This limitation was due to overlying rib coverage and/or an air-filled stomach, necessitating an increase in amplitude to create a cavitation bubble cloud (Fig 3A) (30). This increased amplitude, combined with a pulse sequence that was not specifically designed to limit thermal energy deposition, likely contributed to the body wall thermal injuries. To overcome this limitation, modifications to decrease the time-average intensity of histotripsy energy delivered by greater than 60% successfully mitigated body wall damage (Fig 3B) (31). Another modification used prolonged cooling time during the delivery of histotripsy pulses, resulting in a treatment time of 75 minutes (35) compared with a treatment time of 24 minutes (30) to successfully create transcostal ablations with sizes close to the prescribed volumes in porcine livers without severe body wall damage. These studies illustrate how histotripsy parameter modifications overcame technical limitations and minimized thermal energy deposition to safely produce consistent ablation in the liver through abdominal or transcostal acoustic windows (23,31–33).

Figure 3:

Mitigation of body wall damage in a porcine model. (A) Body wall injury (arrowhead) was observed as regions of T2 hyperintensity on T2-weighted MRI scan in muscle layers during transcostal histotripsy. The arrow shows the ablation zone. (Reprinted, with permission, from reference 30.) (B) Using a modified pulse sequence to reduce the energy deposition, no body wall injury or edema was observed on axial short tau inversion recovery MRI scan. Arrow indicates the ablation zone. (Adapted, with permission, from reference 31.)

Another potential safety concern or adverse effect of histotripsy treatment is the development of vascular thrombosis, possibly due to cavitation-induced platelet activation and aggregation. In a proof-of-concept study by Smolock et al, thrombosis was reported involving hepatic vein (nine of 10 animals) and portal vein (two of 10 animals) branches adjacent to or within the histotripsy treatment zone in the immediate period following histotripsy liver ablation, which resolved within 2–4 weeks without treatment and without consequence (atrophy of the ipsilateral lobe and hypertrophy of the contralateral lobe) at follow-up imaging (Fig 4) (30).

Figure 4:

Resolution of histotripsy-related vascular thrombosis in a porcine model. (A) Portal vein thrombosis was observed on a postcontrast T1-weighted MRI scan posthistotripsy ablation. (B) Thrombosis resolved within 4 weeks posttreatment. (Reprinted, with permission, from reference 30.)

Many studies have evaluated the risk of bleeding following histotripsy, even in anticoagulated animals. Vlaisavljevich et al performed a safety study in the porcine model with two treatment groups, one of which received systemic heparinization (36). Histotripsy was applied successfully to the liver parenchyma surrounding the hepatic vein and within the hepatic vein (36). No observable differences were noted in histotripsy treatment sizes between the subjects who received heparin and those who did not, suggesting that bleeding risks are nominal even in subjects receiving anticoagulation (36). A study by Mauch et al reported no elevated risk of hemorrhage in animals receiving warfarin at 7 days posthistotripsy treatment, even though five of seven ablations extended to the surface of the liver (37). This is in contrast to invasive procedures such as needle-based thermal ablation, in which anticoagulation is a contraindication to treatment because of elevated risk of bleeding from needle puncture.

Another common limitation in thermal local-regional therapies is the heat sink effect, observed in tumors adjacent to large hepatic or portal vessels in which rapidly flowing blood can prevent the development of sustained temperatures required for cell death, thus limiting ablation size next to these structures (38–40). Histotripsy is not affected by the heat sink effect due to the purely mechanical, nonthermal mechanism of tissue destruction.

Tissue Selectivity

A unique property of histotripsy is its differential action on various tissue types. Numerous ex vivo studies have shown that tissues with different mechanical properties have different susceptibility to histotripsy-induced mechanical damage (41,42). In an ex vivo porcine study, Vlaisavljevich et al demonstrated that stiffer tissues (higher Young modulus) have higher cavitation thresholds for initiating the histotripsy bubble cloud (Table 2) (41). Vlaisavljevich et al performed another landmark study in which histotripsy was applied to 43 different types of ex vivo porcine tissue to investigate which tissue properties are more susceptible to histotripsy-induced damage (43). They found that tissues with higher density, higher tensile strength, higher fractional strain, and lower water content (ie, collagenous structures such as tendon, cartilage, blood vessels, bile ducts, and urothelium) were more resistant to damage from histotripsy and required higher mechanical strain to break compared with noncollagenous tissue (43). One potential reason for tissue selectivity is that microbubbles in stiffer collagen-based tissues expand to a smaller maximum diameter compared with parenchymal and tumor tissue, resulting in less strain produced by the bubble cloud (41,42). Combined with a higher mechanical strain necessary to break these tissues, a greater number of histotripsy pulses are required to destroy collagen-based tissue than noncollagenous tissue (41,42). This tissue selectivity can be exploited to specifically target tissue while preventing damage to critical structures such as vessels and bile ducts, as shown in multiple in vivo animal studies (23,43). However, there is a caveat. If the target tumor has higher mechanical strength (due to high fibrin, collagen, or calcium content) compared with the vessels and nerves, then those critical structures would not be preserved. In such cases, advanced targeting strategies that use the high-precision and image-guided features of histotripsy would be required to spare the critical tissue structures by treating near, but not directly on, that region of the tissue. In general, the number of ultrasound pulses (ie, dose) and/or pressure level (ie, amplitudes) can be modified to induce a dose-dependent effect of tissue destruction (16). Transducer parameters such as frequency and f-number also impact the density of the bubbles within the histotripsy bubble cloud, with a higher density of cavitation hypothesized to induce larger strain on the tissue (44–46). Finally, pulsing parameters such as the PRF impact tissue selectivity, as higher PRF results in greater strain on tissue structures, thereby reducing the potential to preserve critical tissue structures (47).

Table 2:

Histotripsy Cavitation Cloud Threshold in ex Vivo Porcine Tissue

Tissue	Threshold (MPa) PRF 100 Hz	Threshold (MPa) PRF 1000 Hz	Young Modulus (MPa)
Lung	1.578 ± 0.89	13.42 ± 1.08	0.0026
Fat	17.13 ± 1.41	13.26 ± 1.85	0.0032
Kidney	17.84 ± 1.48	14.56 ± 0.95	0.0061
Liver	19.97 ± 0.77	17.75 ± 1.07	0.0087
Heart	20.03 ± 0.36	17.06 ± 1.28	0.0042
Muscle	21.01 ± 0.48	19.12 ± 0.57	0.0062
Skin	25.10 ± 0.69	23.21 ± 1.01	0.014
Tongue	26.54 ± 0.88	24.27 ± 0.44	0.025
Tendon	26.41 ± 0.52	24.47 ± 0.49	380
Cartilage	no cloud	27.28 ± 0.85	0.90
Bone	no cloud	no cloud	18 600

Note.—Reprinted, with permission, from reference 41. Experimentally observed histotripsy cavitation initiation threshold in different ex vivo porcine tissues treated at pulse repetition frequencies (PRFs) of 100 and 1000 Hz. The corresponding Young modulus (tension) values were obtained from the literature (66–75).

Effects of Histotripsy on Vessels and Adjacent Critical Structures

Histotripsy demonstrates vessel-sparing effects as a direct consequence of tissue selectivity, where major hepatic blood vessels and bile ducts remain intact within the ablation zone even though the surrounding liver tissue is completely fractionated (Fig 5) (22,31). Collagenous tissues such as large vessels, bile ducts, nerves, and renal collecting system are mechanically strong with a higher tensile strength compared with parenchymal structures such as solid organs and tumors, resulting in vessel-sparing during histotripsy (43). Vessels and bile ducts with lower tensile strength than the tumor will not remain intact within the treated tumor volume. In a porcine study, Vlaisavljevich et al showed that large vessels (>1 mm) remained intact and were observable on MRI scan (22). Acute histology of the ablation zone revealed that no vessel less than 50 μm remained intact, while a significant decrease was observed in the count of small blood vessels (between 50 and 100 μm) compared with surrounding liver parenchyma (P < .01) (22). The small vessels that remained intact were likely protected due to their proximity to large hepatic structures and connective tissue within the ablation zone (22). An in vivo porcine study by Smolock et al showed that major vessels and bile ducts remained intact (30). This was seen both radiologically on hepatobiliary phase MRI and histologically due to the preservation of bile duct mucosa and submucosa with histotripsy necrosis on both sides of the intact bile duct (30). In contrast, thermal ablation and radiation can result in biliary strictures with resultant bilomas and abscesses, while intra-arterial therapies can result in biliary necrosis and acute cholangitis (48).

Figure 5:

Vessel-sparing effects in a porcine model. (A) Photomicrographs (hematoxylin-eosin) show intact blood vessels on histology. (Adapted, with permission, from reference 22.) (B) Photomicrograph (hematoxylin-eosin) shows patent bile ducts within the ablation zone. (Reprinted, with permission, from reference 31.)

Resorption of Ablation Zone

In vivo survival studies have revealed that the histotripsy ablation zone is resorbed by the body over time. In a rodent study, acute histology showed acellular homogenate (11). After 28 days, pathology revealed the acellular homogenate was almost completely replaced by regenerated liver parenchyma, with a microscopic fibrous lesion at the treatment site (Fig 6) (11). Similarly, in a pig study, ablation zones demonstrated complete cell death surrounded by a region of partial necrosis in the acute phase, and after 4 weeks there was greater than 64% reduction in the size of the ablation zone with fibrous tissue surrounding the ablation zone (Fig 2) (30). These results suggest that histotripsy has potential as a noninvasive liver tissue–debulking technique for downstaging larger tumors to meet resection or transplant criteria by reducing the number and volume of tumors. Recently, Uysal et al reported the first clinical case where histotripsy was used as a bridging therapy before liver transplant in a patient with hepatocellular carcinoma (HCC) (49).

Figure 6:

Resorption of ablation zone in a rat model. Normal rat livers were treated with histotripsy. (A–E) Photomicrographs (hematoxylin-eosin) revealed the resorption of ablation zone over time on histology. Immediately after treatment at day 0, the ablation zone consisted mostly of acellular debris, fibrin, and extravasated blood cells. Within 3–7 days, fibroblasts infiltrated the ablation zone. By day 14, only a small area of residual scar with calcification zones remained at the site of ablation. (Adapted, with permission, from reference 11.) BD = bile ducts, GC = giant cells, HL = histotripsy lesion.

Tumor Burden Reduction

To establish feasibility of histotripsy for HCC treatment, an immunocompromised human–derived Hep3B subcutaneous HCC mouse model was used by Worlikar et al (24). To our knowledge, this was the first study to use MRI to characterize the ablation zone changes over time (50). Before histotripsy, tumors appeared heterogeneous; however, immediately posttreatment the tumors were noted to be homogeneously hyperintense at T2-weighted MRI (Fig 7A) (24,50). The ablation zone revealed an acellular homogenate at acute histology with no intact cells or structural components in the targeted region, corresponding to the MRI appearance (Fig 7B) (50). In the survival cohort, MRI evaluation revealed a gradual decrease in the size of the treatment cavity over time; however, tumors eventually recurred by 12 weeks, likely due to the lack of a treatment margin surrounding the tumor, which prevented complete ablation; this is an inherent limitation of the subcutaneous tumor model (24). This initial study suggested that histotripsy could effectively treat HCC.

Figure 7:

Characterization of MRI and histology of human-derived hepatocellular carcinoma tumor before and posthistotripsy in a subcutaneous mouse model. (A) Within 1 day posthistotripsy, the ablated volume (yellow dashed lines) appeared T2 hyperintense compared to the heterogeneous appearance of the nontargeted tumor on T2-weighted MRI scan. (Adapted, with permission, from reference 50.) (B) Photomicrograph (hematoxylin-eosin) shows that the ablation zone consisted of completely fractionated cells with a sharp boundary separating the ablated region from the nontargeted tumor. (Reprinted, with permission, from reference 50.) H&E = hematoxylin-eosin.

To overcome the limitations of the subcutaneous tumor model, an orthotopic N1S1 rat liver tumor model was used (25). Real-time US-guided histotripsy was applied to ablate the entire tumor volume or 50%–75% of tumor volume (partial ablation) (25). MRI evaluation revealed effective posthistotripsy reduction of tumor burden with complete tumor regression in 14 of 15 (93.3%) treated rats, including those with partial histotripsy treatment, suggesting a possible immune response posthistotripsy of HCC (25). Histopathology at 3 months demonstrated fibrous tissue (<5 mm) at the treatment site, with no viable tumor (25). This study suggested effective tumor response posthistotripsy, even in partially treated tumors, with minimal risk of local tumor progression or recurrence.

Local and Systemic Immune Effects

During preclinical studies, a unique phenomenon was discovered: histotripsy-induced local and systemic immune stimulation. In a subcutaneous murine melanoma and HCC tumor study by Qu et al, histotripsy ablation induced significant immune infiltration of innate and adaptive immune cell populations in the tumor (51). This was one of the first studies demonstrating the abscopal effect with histotripsy, as evidenced by the growth restriction of untreated contralateral tumors, as well as inhibition of tail vein–induced pulmonary metastases (Fig 8A) (51). It was observed that histotripsy could release tumor antigens with retained immunogenicity, and this immunostimulatory effect was associated with calreticulin translocation to the cellular membrane and local and systemic release of high mobility group box protein 1. When combined with immune checkpoint inhibition, histotripsy ablation inhibited abscopal tumor growth even more than that of histotripsy alone or immune checkpoint inhibition alone (Fig 8B) (51). A follow-up study by Pepple et al in a bilateral subcutaneous murine liver tumor model showed that histotripsy ablation of one of the tumors resulted in local immunogenic cell death (Fig 8C), leading to antigen-specific abscopal inhibition of the contralateral untreated tumor (52).

Figure 8:

Immune effects of histotripsy in immunocompetent mouse tumor models. (A) Histotripsy treatment of the primary subcutaneous tumor (melanoma) resulted in inhibition of distant tail vein–induced pulmonary metastasis compared with controls. (B) Histotripsy treatment of the primary subcutaneous liver tumor resulted in abscopal effect such as the growth inhibition of a contralateral untreated tumor. This effect was further augmented by combining histotripsy with anticytotoxic T-lymphocyte associated protein 4 (anti-CTLA-4), an immune checkpoint inhibitor. (Adapted, under a CC BY 4.0 license, from reference 51.) (C) Multiplex immunohistochemistry revealed co-localization of necroptosis markers of phosphorylated receptor-interacting protein kinase 3 (pRIPK3) and phosphorylated mixed lineage kinase domain-like protein (pMLKL) in the periphery of histotripsy-treated tumors, indicating immunogenic cell death compared with untreated controls. (Reprinted, under a CC BY 4.0 license, from reference 52.)

Impact on Metastasis

A study by Worlikar et al evaluated the impact of partial histotripsy ablation (50%–75% tumor volume targeted) on immune infiltration, survival outcomes, and metastasis using the McA-RH7777 orthotopic rat liver tumor model with spontaneous metastases (53). Posttreatment MRI evaluation revealed complete local tumor regression in nine of 11 histotripsy-treated rats, with no recurrence or intrahepatic metastasis and only submillimeter residual scar tissue on histology (Fig 9) (53). Increased immune infiltration in the histotripsy-treated tumors compared with the controls may have contributed to the regression of the untargeted tumor, suggesting an abscopal response (53). In comparison, all controls demonstrated local tumor progression and developed distant intrahepatic metastases (53). Survival outcomes in the histotripsy-treated animals were significantly better than controls (P < .0001) (53).

Figure 9:

Impact of histotripsy on the risk of metastasis in an orthotopic rodent hepatocellular carcinoma model. (A) Control animals demonstrated rapid tumor progression and intrahepatic metastasis on T2-weighted MRI scans and histology photomicrographs (hematoxylin-eosin). (B) In comparison, tumors treated partially with histotripsy demonstrated complete regression of both the targeted (yellow arrow) and untargeted (blue arrow) tumor regions with no sign of viable tumor on MRI scans or histology photomicrographs (hematoxylin-eosin). (Adapted, under a CC BY 4.0 license, from reference 53.) A = ablation, L = liver, T = tumor.

Dosimetry Studies

Hendricks-Wenger et al performed a histotripsy dosimetry study to investigate the impact of different histotripsy doses in a subcutaneous human-derived cholangiocarcinoma mouse model (histotripsy dose was arbitrarily defined as the number of pulses/point delivered within the target volume) (54). The researchers investigated three doses (250, 500, and 1000 pulses/point). Although all doses achieved complete ablation (complete disruption of cellular structure with no evidence of intact tumor cells) within the core of the ablation zone, the highest dose (1000 histotripsy pulses/point) created the sharpest boundary differentiating the ablation zone from surrounding intact tissue (54). In contrast, intact cells were observed in the periphery of the ablation zone for the lower doses (250 and 500 pulses/point) (Fig 10) (54). A survival study in the same tumor model with a dose of 500 pulses/point showed improved tumor progression–free and overall survival in histotripsy-treated versus control mice (55). Another ex vivo experiment evaluating histotripsy doses required for completely ablating different types of human liver tumor specimens including HCC, cholangiocarcinoma, and colorectal liver metastasis (CLM) revealed that mechanically stiffer CLM tumors needed more than 1500 pulses/point to achieve complete ablation compared with the softer HCC tumors, which required fewer than 500 pulses/point (54). The fibrotic cholangiocarcinoma tumors did not exhibit complete ablation of the entire treatment volume even with 4000 pulses/point (54). Further studies are necessary to identify optimal histotripsy doses for different tumor types to maximize the treatment outcomes, including stimulation of the immune-mediated response, while minimizing the energy deposited.

Figure 10:

Characterization of histotripsy dose effects at US imaging and on histology in a subcutaneous, patient-derived cholangiocarcinoma mouse model. (A–D) US scans show the approximate ablation zone with increasing dose. (E–H) Corresponding photomicrographs (hematoxylin-eosin) show the untargeted tumor region and ablation zone with increasing dose. A sharper boundary was observed for a dose of 1000 pulses/point compared with 250 pulses/point and 500 pulses/point. (Adapted, under a CC BY 4.0 license, from reference 54.)

Clinical Trials and Posttrial Clinical Safety Data

The THERESA and #HOPE4LIVER clinical trials evaluated the safety and efficacy of histotripsy treatment of primary and metastatic tumors of the liver. In brief, both trials demonstrated that hepatic histotripsy can safely and effectively destroy a targeted tissue volume without major procedure-related complications, supporting early clinical adoption (56–58).

The phase I human trial, THERESA, investigating histotripsy of primary or metastatic hepatic tumors was conducted in Barcelona, Spain, in 2019 (NCT03741088) (56). A clinical prototype hepatic histotripsy device, VORTX Rx, manufactured by HistoSonics, was used. Eleven tumors (range, 0.5–2.1 cm) were targeted in eight patients, including one HCC and 10 metastases (colorectal, breast, and gallbladder cancer) in the liver (56). No procedure-related significant adverse events were reported, revealing the early safety and tolerability of histotripsy. The primary end point, defined as the creation of an ablation volume as prescribed (evaluated with MRI 1-day posthistotripsy), was met in all patients (56). A single 5-mm tumor was mistargeted due to limitations in diagnostic US imaging, while in all other tumors, local tumor regression was observed using MRI at 2 months (Fig 11A) (56). Volume contraction of the ablation zone averaged 36.0% at 1 week, 53.6% at 1 month, and 71.8% at 2 months, measured with T1-weighted contrast-enhanced MRI (56). Interestingly, abscopal effects were observed in two patients (two of eight, 25%), one each with HCC and CLM, in which continuous postprocedure decline in tumor markers (α-fetoprotein and carcinoembryonic antigen) was observed (59). Additionally, the patient with CLM had shrinkage of nontargeted tumor lesions 8 weeks after the procedure (Fig 11B) (59). This phase I trial established the initial safety and efficacy of histotripsy for treatment of liver cancer and provided the earliest clinical evidence of a potential histotripsy-induced abscopal effect (56,59).

Figure 11:

Effects of histotripsy on targeted and nontargeted liver tumors in the first-in-human clinical trial. (A) Involution of the targeted tumor region was observed by 8 weeks after treatment, as shown by red arrows on contrast-enhanced US scans. (Adapted, under a CC BY 4.0 license, from reference 56.) (B) Abscopal effects were observed in a patient with colorectal metastases as evidenced by the involution of the untargeted tumors on MRI (red circles). (Adapted, with permission, from reference 59.)

Shortly after completion of the THERESA trial, the first phase I/II histotripsy clinical trial #HOPE4LIVER began. This was a prospective multicenter single-arm investigational device exemption trial conducted in Europe and the United States to evaluate the safety and efficacy of histotripsy ablation in the liver. Forty-four participants (21 from the United States, 23 from the European Union or United Kingdom; 50% female individuals; mean age, 63.9 years ± 12.3 [SD]) with 49 tumors were enrolled and treated (57). Of the treated tumors, 18 (40.9%) were HCC, and 26 (59.1%) were liver metastases (colorectal [n = 10], breast [n = 4], pancreas [n = 5], brain [n = 1], neuroendocrine [n = 1], ovary [n = 1], uveal melanoma [n = 1], bronchial carcinoma [n = 1], and large cell lung neuroendocrine [n = 1]). Patients who met inclusion criteria with up to three tumors each less than 3 cm in diameter were treated with histotripsy in one session. All patients had pre- and posttreatment evaluation with contrast-enhanced CT or MRI. The co-primary end points were greater than 70% technical success rate (entire tumor treated with margins as preplanned) and less than 25% major complication rate (no procedure-related major complication within 30 days). The greater than 70% performance goal for the primary efficacy end point was met in 42 of 44 lesions (95.5% [95% CI: 83.72, 100]); technical success was achieved within 36 hours of the procedure. The lack of technical success in two lesions was due to user error and not device failure. The less than 25% performance goal for the primary safety end point was met with three of 44 (6.8% [95% CI: 2.35, 18.23]) participants experiencing index procedure–related major complications (common terminology criteria for adverse events [CTCAE] ≥3) within 30 days after the procedure. Two of these three complications were likely related to cancer; however, they occurred within the first 30 days and were thus considered procedure-related complications. Two participants experienced CTCAE grade 3 events (sepsis and pleuritic pain, respectively); and one participant with innumerable metastatic lesions experienced a grade 5 event (hepatic failure at 12 days after the procedure, which resulted in the participant’s death at 37 days after the procedure). Overall, a total of 101 adverse events occurred within the first 30 days after the procedure, of which 94 (93.1%) were considered nonserious. Of the seven (6.9%) reported events classified as serious, three were considered primary safety end point failures (as reported previously). The remaining four were splenic hematoma, melena, procedural pain, and metastatic colorectal cancer, three of which were likely not treatment-related. There were no procedure-related bleeding events or deaths in the first 30 days (57). The results of the #HOPE4LIVER trial led to the U.S. Food and Drug Administration’s approval of Edison, the second-generation histotripsy device.

Recently, multisite posttrial clinical safety outcomes were reported from 230 patients with different types of liver cancer (CLM, HCC, pancreatic, neuroendocrine, and breast metastases) treated with histotripsy across nine centers (60). Only 12 of 230 patients (5.2%) experienced complications within 30 days of treatment; of these, nine patients experienced minor complications (Clavien-Dindo grade ≤2), while three patients experienced major complications (disease progression leading to death) (60). It should be noted that these patients had known advanced intra- and extrahepatic diseases and received histotripsy with palliative intent only, hence, these complications were considered as the progression of the natural history of the disease (60).

Current Limitations and Future Directions

Preclinical studies have shown that histotripsy can treat malignant tumors in small-animal models, create clinically relevant treatment zones in large-animal models, avoid heat sink effect, and spare collagenous structures such as blood vessels and bile ducts due to the mechanical mechanism of tissue destruction; similar effects have also been seen in human patients (28,29). While thermal ablation induces protein denaturing leading to coagulative necrosis, histotripsy avoids thermal damage and is capable of releasing tumor-associated antigens (61), likely facilitating local and systemic immune responses. The first-in-human clinical trial suggested abscopal effects in untargeted tumors in two patients. However, further clinical research on the true extent of histotripsy-induced immune and abscopal effects is imperative before treating patients with metastases for an indication of abscopal response. Figure 12 summarizes the important milestones in the development of histotripsy as a local-regional liver cancer therapy. The overall promising results from the THERESA and #HOPE4LIVER trials resulted in the U.S. Food and Drug Administration’s approval for the treatment of liver cancer in humans, but research is still ongoing to further optimize this technology.

Figure 12:

Timeline depicts the important milestones in the development of histotripsy as a local-regional liver cancer therapy. UM = University of Michigan.

One of the major considerations in performing noninvasive ablations with histotripsy is the acoustic window. Histotripsy requires high focal peak pressures for successful cavitation. The actual focal pressure generated depends on both the acoustic access available to the transducer and acoustic aberration caused by soft tissue heterogeneity. Target locations within the body that are blocked by gas or bones have limited acoustic access through the intact body wall and may be difficult to noninvasively treat with histotripsy due to insufficient focal pressure achieved for sustained cavitation. Ultrasound attenuation is directly proportional to tissue depth; hence, the maximum depth of the tissue that can be targeted is also limited. Yeats et al developed a technique to correct soft tissue aberrations in ex vivo bovine liver and increase the focal pressure amplitude by using acoustic emissions from cavitation cloud initiation and collapse (62). This work will help increase the usability of the clinically approved histotripsy device in human patients.

Another limitation is that currently only US imaging guidance can be used for histotripsy. However, to improve tumor visualization, efforts are underway to coregister US imaging with MRI or CT imaging performed before treatment using fusion technology. Wagner et al are developing an alternative x-ray C-arm–guided, histotripsy-targeting approach using cone beam CT imaging to identify the target lesion and two-dimensional fluoroscopy and deep learning to estimate the three-dimensional position of the histotripsy transducer in order to perform histotripsy without using US guidance (63). In a follow-up study, Wagner et al used pig livers to create in vivo pseudotumors (ie, histotripsy lesions) using US guidance and then targeted those pseudotumors with additional margins by using histotripsy with cone beam CT to demonstrate feasibility and targeting accuracy (64).

A third limitation is the overall treatment time, which depends on tumor location, acoustic pathway, and the cooling time between histotripsy pulse delivery. Lundt et al investigated the use of electronic focal steering instead of mechanical steering to achieve rapid, homogeneous volumetric ablation at a speed of 0.9–3.3 mL/min, which exceeds any current clinical volumetric ablation method (approximately 1–2 mL/min) (65). This technique may help decrease ablation times in the clinical setting, where patients are under general anesthesia. Adjusting the number of treatment pulses based on the mechanical properties of the tumor may also help reduce treatment time. The ex vivo human liver tumor experiments by Hendricks-Wenger et al showed that the number of pulses necessary for achieving complete ablation is relatively lower in softer tumors like HCC compared with stiffer tumors like cholangiocarcinoma and CLM (54). While the current clinical system uses a high dose to ensure the capability of ablating all tumor types, tumor-specific refinements could be made to reduce treatment times for softer tumors such as HCC.

Finally, most of the preclinical studies and the clinical trials to date have investigated histotripsy as a standalone cancer therapy. Future studies are necessary to investigate the combination of histotripsy with surgery, immunotherapy, chemotherapy, and/or transarterial therapies for improving patient outcomes.

Conclusion

The clinical translation of histotripsy began with preclinical small- and large-animal studies to prove feasibility, safety, and efficacy. Promising results for treating liver cancers in preclinical studies as well as in clinical trials led to the U.S. Food and Drug Administration’s approval of histotripsy for treating liver cancer in humans. The noninvasive and nonthermal mechanism of action and potential for proinflammatory immune stimulation makes histotripsy a unique and powerful local-regional treatment option for primary and metastatic liver cancer. Future investigations are needed to optimize histotripsy dosing protocols to maximize treatment outcomes, including induction of abscopal effects. We hope that the key findings and insights discussed in this review will facilitate the clinical adoption of histotripsy.