The use of histotripsy as intratumoral immunotherapy beyond tissue ablation—the rationale for exploring the immune effects of histotripsy

Abstract Mechanical high-intensity focused ultrasound (M-HIFU), which includes histotripsy, is a non-ionizing, non-thermal ablation technology that can be delivered by noninvasive methods. Because acoustic cavitation is the primary mechanism of tissue disruption, histotripsy is distinct from the conventional HIFU techniques resulting in hyperthermia and thermal injury. Phase I human trials have shown the initial safety and efficacy of histotripsy in treating patients with malignant liver tumors. In addition to tissue ablation, a promising benefit of M-HIFU has been stimulating a local and systemic antitumor immune response in preclinical models and potentially in the Phase I trial. Preclinical studies combining systemic immune therapies appear promising, but clinical studies of combinations have been complicated by systemic toxicities. Consequently, combining M-HIFU with systemic immunotherapy has been demonstrated in preclinical models and may be testing in future clinical studies. An additional alternative is to combine intratumoral M-HIFU and immunotherapy using microcatheter-placed devices to deliver both M-HIFU and immunotherapy intratumorally. The promise of M-HIFU as a component of anti-cancer therapy is promising, but as forms of HIFU are tested in preclinical and clinical studies, investigators should report not only the parameters of the energy delivered but also details of the preclinical models to enable analysis of the immune responses. Ultimately, as clinical trials continue, clinical responses and immune analysis of patients undergoing M-HIFU including forms of histotripsy will provide opportunities to optimize clinical responses and to optimize application and scheduling of M-HIFU in the context of the multi-modality care of the cancer patient.


Introduction
Conventional treatment for most solid tumors generally involves surgery, radiotherapy and/or systemic therapy, which recently includes immunotherapy [1].Historically, curative intent surgical resection attempted to eliminate all tumor cells felt to be locally invading surrounding tissues.Increasingly dramatic resections of tumor and surrounding normal tissue were proposed to eliminate all tumor cells and techniques such as the 'no touch' approach were proposed in an attempt to minimize tumor manipulation, tumor embolization and systemic dissemination [2,3].Despite these technical triumphs, the morbidity and loss of function imposed by these resections were often formidable, and the impact of curative intent surgery in increasing the cure rate of cancer was disappointing, often due to the subsequent appearance of distant metastatic disease.Indeed, evidence suggests that tumor cell dissemination occurs very early in disease, even with tumors that appear to be clinically localized [4,5].In response, approaches that minimize the extent of resection and maintain tissue and organ function have been explored in combination with regional or systemic therapies to eliminate locally and distantly disseminated tumor cells [6].However, many of these therapies have a limited ability to eradicate all tumor cells.Therefore, it is likely that surgical 'cures' occur not only due to elimination of clinically localized tumor, but in some cases elimination of tumor plus systemic elimination of disseminated malignant cells, likely by immune-mediated mechanisms.This suggests that optimal curative therapy of clinically localized tumors may require both tumor elimination and optimizing immune recognition and rejection of disseminated tumor cells.
While modest successes of immune-based therapies based on systemic administration of interferon and cytokines such as interleukin-2 were observed in a few tumor types, the introduction of immune checkpoint blockade in the past 10 years has resulted in a revolution in cancer treatment [7][8][9].Immune checkpoint inhibition is associated with objective clinical response rates and prolonged overall survival in numerous tumor types.The clinical benefits of immunotherapy have been observed in patients with both advanced disease and in the adjuvant (post-surgical resection) setting.The broad-based activity of systemic immunotherapy has generated interest in the adoption of cancer immunotherapies in the early stages of a variety of diseases, but this is balanced with a concern about systemic toxicities, including immune-related toxicities which may be chronic [10,11].Thus, alternatives to systemic immunotherapies in early-stage disease are being explored.Furthermore, recognizing that the most effective immune responses ultimately require the presence of antigen-specific adaptive immunity mediated by T lymphocytes, neoadjuvant (before surgery) immunotherapy is now being pursued to capitalize on the induction of T cell immunity while tumor antigens are still abundant [12].These developments create a timely opportunity to reconsider the role of local ablative therapies in combination with intratumoral and systemic immunotherapies.

Intratumoral immunotherapy: Opportunities and challenges
While conventional systemic intravenous administration allows for the distribution of the treatment drug throughout the body, drug penetration into the tumor tissue can be inadequate owing to inherent barriers imposed by the tumor microenvironment [13].Additionally, indiscriminate exposure to the nontumorous compartment often adversely influences the agents' effectiveness or results in prohibitive toxic effects.Local intratumoral injection of anticancer therapies is a logical solution to overcoming these barriers to drug delivery [14].Direct inoculation of immune-stimulating agents into the tumor or tumor microenvironment itself or intratumoral immunotherapy, has several features that make it particularly useful, as off-target toxicities and adverse effects that can accompany global immune stimulation can be minimized.Because the toxicity of some systemic immunotherapies such as cytokines has been shown to be dose-related, the intratumoral delivery allows for an active concentration in the tumor and generally lower systemic dosages and exposure, which may enable combinations of agents that would be too toxic for combined systemic delivery [15,16].
For all intratumoral applications, the feasibility of injection into tumor lesions is a critical requirement, especially if repeated injections are needed to trigger an effective systemic immune response.Most intratumoral clinical studies have been performed in easily accessible lesions such as skin, head and neck or breast cancer.However, due to the increased availability and sophistication of interventional radiologic, endoscopic and laparoscopic procedures, most if not all lesions can now be accessed with or without the assistance of imaging modalities such as ultrasound, computed tomography (CT), etc. Sheth et al recently reported on the experiences and challenges to intratumoral delivery and found that intratumoral injections of immunotherapies are feasible across a range of histological conditions and target organs [17].As with any invasive intervention, outcomes after intratumoral drug delivery depends on procedural technique.This is corroborated by the substantial difference in intratumoral drug delivery outcome observed when the same tumor in the same patient was injected with 2 different injection techniques [18].Limited delivery of the immunotherapy drug to its target site may result in diminished efficacy, but a more immediate challenge is the potential for systemic exposure to high doses of these very potent drugs.Therefore, the challenge in intratumoral therapy now lies not in initial accessibility, but in the optimization of drug delivery technologies to enhance intratumoral delivery and retention [18].
Beyond the biochemical changes in the tumor microenvironment that work to subvert immune system detection and elimination, physical barriers that can have significant detrimental effects on therapeutic approaches are also present.Indeed, many tumors adopt organ-like structures and extensive fibrosis that build highly complex physical barriers.The tumor microenvironment remains a key barrier to effective anticancer drug delivery [19,20].Furthermore, the complex tumor microenvironment may have an even greater impact on the delivery of effective immunotherapy.Because the mechanism of action of immune checkpoint inhibitors is to improve immune system activation and targeted killing of tumor cells, it is critical that leukocytes have physical access to the malignant cells.In an attempt to enhance the effectiveness of checkpoint inhibitor therapies, additional strategies including ablation modalities have been evaluated to better understand their ability to modulate the immunological aspects of the tumor microenvironment.While debulking the tumor and reducing a patient's overall tumor burden is the primary goal of focal tumor ablation therapies, improved immune system access is proving to be an added benefit that can shift the local TME from being 'cold' and immunosuppressed to 'hot' and immunostimulated [21,22].

Tumor ablation as a component of immunotherapy: Thermal high-intensity focused ultrasound (thermal injury) versus mechanical high-intensity focused ultrasound (inertial cavitation)
Ultrasound refers to stress waves with frequencies greater than 20 kHz and its noninvasive, deep penetration and nonionized features are attractive for clinical applications including thermal and mechanical bioeffects-related therapies.High-intensity focused ultrasound (HIFU) refers to focused ultrasound waves with relatively high acoustic intensities.In terms of spatial peak pulse average intensity (I SPPA ), the corresponding acoustic intensity typically ranges from hundreds of W/cm 2 to thousands of W/cm 2 .With different electrical waveforms to drive a focused ultrasound transducer, HIFU can be categorized into the conventional thermal high-intensity focused ultrasound (T-HIFU) and the more recently described mechanical high-intensity focused ultrasound (M-HIFU).HIFU is already in clinical use for prostate [23] and other malignancies [24].
As shown in Figure 1, M-HIFU usually utilizes a shorter pulse duration of microseconds and high peak negative acoustic pressure of dozens of megapascals, while T-HIFU prefers a relatively long pulse duration of milliseconds with a relatively high duty cycle and acoustic pressure of serval megapascals [25,26].Regarding M-HIFU, a low-duty cycle (usually <2%) is usually adopted to avoid the accumulation of heating.Meanwhile, such instant and intense acoustic exposure enables microbubbles' formation, oscillation and collapsing (i.e.acoustic cavitation), which may damage tumor cells, induce cell necrosis or apoptosis and disrupt the tumor extracellular matrix [27].T-HIFU, however, is usually associated with a relatively high duty cycle (e.g.>10%) and the heat accumulation and dissipation cause tissue temperature to rise up above 65 � C but can rise to 80 � C. Such high temperatures directly ablate tumor tissue and induce coagulative necrosis [28].In addition, M-HIFU, also known as histotripsy, can be classified into three categories: intrinsic threshold histotripsy, shock-scattering histotripsy and boiling histotripsy.Further details can be found in the review manuscript by Xu [29].
Because many early HIFU devices focused on thermal ablation, research in HIFU has been historically focused on enhancing thermal ablation efficiency with more precise control of targeting and monitoring of lesion formation.Studies of the thermal effects of conventional HIFU on the immune environment are well described in other reviews of hyperthermia and recent studies using sophisticated single-cell transcriptomic analysis provide insights into the effects of these thermal changes on the tumor microenvironment [30].
Mechanical HIFU generates cavitation bubble clouds that lead to precise non-thermal tumor ablation and the biological consequences of this ablation are the creation of a large amount of tumor antigens in the form of necrotic cells and the local release of a diverse array of endogenous danger signals from the mechanical HIFU damaged tumor cells.
In 2005, we reported on the different immune effects of mechanical HIFU compared to thermal HIFU [31].We found that HIFU treatment can cause the release of endogenous danger signals (ATP and hsp60) and exposure of dendritic cells (DCs) and macrophages to the supernatants of HIFUtreated tumor cells leads to an increased expression of co-stimulatory molecules (CD80 and CD86) with enhanced secretion of IL-12 by the DCs and elevated secretion of TNFalpha by the macrophages.The potency in APC activation produced by mechanical lysis is much stronger than the thermal necrosis of the tumor cells.These findings suggested that optimization of treatment strategy may help to enhance HIFU-elicited anti-tumor immunity.
We followed up this work in 2007, by hypothesizing that these endogenous danger signals from HIFU-damaged tumor cells may trigger the activation of antigen presenting cells or DCs.This response may play a critical role in a HIFU-elicited anti-tumor immune response which can be harnessed for more effective treatment [32].Mechanical HIFU of tumors was found to increase CD11c þ cells fourfold and DC accumulation in draining lymph nodes 10-fold.Mechanical HIFU caused a reduction in the growth of primary tumors, but also provided protection against subcutaneous tumor rechallenge.Further immunological assays confirmed an enhanced cytolytic T cell (CTL) activity and increased tumorspecific IFN-c-secreting cells in the mice treated by mechanical HIFU with cytotoxicity induced by mechanical HIFU reaching as high as 27% at a 10:1 effector:target ratio.These studies present initial encouraging results confirming that focused ultrasound treatment can elicit a systemic anti-tumor immune response, and they suggest that this immunity is closely related to DC activation.Because DC activation was more pronounced when tumor cells were mechanically lysed by focused ultrasound treatment compared to thermal HIFU, mechanical HIFU in particular may be employed as a potential strategy in combination with subsequent thermal ablations for increasing the efficacy of HIFU cancer treatment by enhancing the host's anti-tumor immunity.
An appreciation of the immune effects of mechanical HIFU contributed to the development of increasingly sophisticated non-thermal HIFU ablation techniques, such as histotripsy.As these histotripsy or M-HIFU therapies have developed, the full picture of the accompanying immune response has revealed a wide range of immunogenic mechanisms that include DAMP and anti-tumor mediator release, changes in local cellular immune populations, development of a systemic immune response, and therapeutic synergism with the inclusion of checkpoint inhibitor therapies, and a realization that complete tissue liquefaction may also result in the absence of cells in the former tumor bed.
To assist in reviewing what has been reported, we summarized key sonication parameters for histotripsy-assisted cancer immunotherapies in Table 1.Currently, the main histotripsy device for cancer immunotherapy is extracorporeal, which lacks the ability to target tumor lesions behind bones and fat in an efficient way [47,48].Given the specific target and model, various ultrasound frequencies were chosen.For example, sub-megahertz ultrasound has been employed to treat bone tumors because of its effectiveness in porous structures due to low attenuation [42], while megahertz ultrasound has been used to sonicate tumors in small animal models where only a short penetration depth is needed [49].
Although the variables in sonication can be stated, it has been noted that the immunologic outcomes of any treatment of a human tumor will depend on the interactions of the treatment and the underlying tissues or organs.For example, the zone of damage created by histotripsy delivery by extracorporeal devices is determined by the size of the ultrasound beam focal zone, which is dependent on the frequency and geometry of the transducer [46,50].For many preclinical studies, small murine tumors are effectively treated by existing instruments, and a typical focal zone maybe 1-2 mm in diameter.To be clinically useful, larger tumors will need to be treated, and the focal zone needs to be directed over the course of the entire tumor.The effectiveness of the treatment resulting in either tissue ablation or tissue liquefaction will be determined by the cavitation bubbles, which depend on both the ultrasound parameters and the viscoelasticity of the target tissue.For example, to completely liquefy the tissue, it takes tens to hundreds of pulses requiring a treatment time of 0.1-60 s per focal location, depending on the mechanical property of the target tissue.The number of pulses to achieve complete tissue disruption is greater for stiffer tissue and tissue containing high concentrations of collagen.The factors will need to be considered as next-generation therapies are optimized to produce both tissue ablation and enhance systemic immunity.Nonetheless, there are a number of immunologic features that are generally altered with histotripsy.

Changes in the cellular composition of the tumor microenvironment
The effect of boiling histotripsy on tumor microenvironment was tested in in vitro study by co-incubating monocytes with the supernatant of boiling histotripsy-treated breast cancer cells [35].Based on gene expression as well as morphological and phonotypical analyses, human monocytes were found to polarize into proinflammatory anti-tumor M1-like macrophages when co-cultured with the supernatant.Furthermore, repolarization of M2-like macrophages into M1-like macrophages was observed after exposure to the supernatant, suggesting that boiling histotripsy could potentially be used to modulate the tumor immune microenvironment into more pro-inflammatory antitumor status by manipulating the polarization of macrophages.
In a murine melanoma model, histotripsy induced the increase of neutrophils, NK cells, dendritic cells and macrophages 10 days post-treatment.Helper T cells and B cells also increased while regulatory T cells decreased at this time point.The results indicate that this treatment can induce an immunostimulatory tumor microenvironment that may lead to antitumor efficacy in the treated tumors [49].Histotripsy-induced DC infiltration to tumors and draining lymph nodes in a  murine MC-38 model, suppressed the growth of metastatic tumors in a murine B16F10 tumor model and inhibited tumor growth and increased CD8þ cells in spleens and draining lymph nodes in murine RM-9 tumors [33,34].These results indicate that complete tumor destruction using histotripsy can initiate an adaptive immune response against tumors.
A recent study that compared histotripsy with unfractionated ablative radiation and thermal ablation in subcutaneous B16F10 melanoma tumors demonstrated significantly higher intratumoral CD8þ cell infiltration in histotripsytreated tumors compared to other ablation strategies [49].Furthermore, histotripsy was able to upregulate intratumoral NK cell, DC, neutrophil, B cell and T cell populations and circulating NK cells, demonstrating local and systemic inflammatory responses [49].
In our study where mechanical HIFU was compared with thermal HIFU in murine breast cancer models, we found stronger infiltration of CD4 and CD8 T cells and NK cells in M-HIFU-treated tumors.CD8þ T cells were more activated based on ICOS and Granzyme B expression.Although the proportion of DCs was not different between these treatments, M-HIFU induced more maturation of DCs than heat ablation.In addition, more M1-like macrophages and less M2-like macrophages were found in M-HIFU-treated tumors, while intratumoral macrophages relatively decreased [39].Thus, mechanical ablation appears to induce stronger infiltration of effector immune cells as well as antigen-presenting cells into the treated tumors, suggesting the modification of tumor microenvironment into more pro-inflammatory, antitumor one.

Release of tumor antigens
Compared with thermal ablation, histotripsy destroys tissue with a mechanically dominated mechanism, which may result in the release of non-thermally damaged antigens [31,32] and is thus expected to induce a stronger antigen-specific immune response.Interestingly, in an in vitro study where thermal ablation, cryoablation and irreversible electroporation (IRE) were compared for their abilities to prime the immune system, IRE released the largest amount of proteins and tumor antigen from tumor cells in vitro, which lead to stronger proliferation of antigen-specific CD8 T cells and cryoablation released the most native protein.Thermal ablation, however, released significantly smaller amount of proteins and tumor antigens, and more denatured proteins than the other two ablation methods and thus had the weakest capacity to activate T cells [51].
In a mouse model of ovalbumin-expressing lymphoma model, M-HIFU significantly enhanced DC activation in draining LNs compared to T-HIFU.When TLR9 agonist CpG was administered following HIFU procedures, lymph node cells from M-HIFU treated mice were more potent to induce proliferation of OT-1 specific CD8 T cells ex vivo compared to lymph node cells from T-HIFU group, suggesting the stronger antigen presentation by APCs in M-HIFU-treated mice [40].Interestingly, the study found that M-HIFU induced a fragmented RNA, DNA and protein profile in tumor debris, whereas T-HIFU did not.The difference in protein fragmentation status might be causing higher immunogenicity of tumor debris generated by M-HIFU compared to T-HIFU.
In addition to tumor antigens, danger signals such as calreticulin (CRT), heat-shock proteins (HSPs) and high mobilitygroup box 1 (HMGB-1) are released from BH and histotripsytreated tumors and play an important role to induce immune response against tumors.These damage-associated molecular patterns (DAMPs) are released into serum function to systemically prime the immune system and to recruit circulating leukocytes that can ultimately accumulate at the site of tissue injury following ablation [49].Our study investigated the difference in immune induction between M-HIFU and T-HIFU and found that M-HIFU can release higher levels of ATP and hsp60 and that the DAMPs released by M-HIFU treatment were more capable of stimulating downstream immunological changes, such as dendritic cell maturation [32].

Changes in cytokines and chemokines
IFN-c, a potent pro-inflammatory mediator that can activate effector cells and antigen-presenting cells, is reported as a critical cytokine increased after histotripsy [36,37,49].Boiling histotripsy causes immunogenic cell death of tumor cells, followed by significantly increased secretion of DAMPs (CRT, HSP70, HMGB-1) as well as pro-inflammatory cytokines (IFNc, IL-1a, IL-1b, IL-18) and chemokines (IL-8) that are related to M1 macrophage activation and typically involved in recruiting immune cells [35].Regarding immunosuppressive cytokines, IL-10 and TGFb were decreased in M-HIFU-generated tumor debris compared to untreated tumor tissue or T-HIFU-generated debris [40].
In a murine model of neuroblastoma, several cytokines and growth factors in the serum were tested before and after histotripsy.Significant increase in IL-2 and GM-CSF and decrease in VEGF were observed at 24h post-histotripsy.At 48h after histotripsy, IL-6 was upregulated, while IL-10 was downregulated.As an increase in GM-CSF level is known to correlate with increased APC cell differentiation and activation, and a decrease in VEGF level with better clinical outcome, histotripsy is expected to produce an anti-tumor immune status [37].
In our previous study, both M-HIFU and T-HIFU were found to stimulate the maturation of APCs, we observed enhanced secretion of IL-12 by DCs and significantly elevated secretion of TNF-alpha by macrophages.The stimulatory effect induced by M-HIFU was much stronger than that induced by T-HIFU [31].Recently, we analyzed the levels of IFN-c, TNF-a and TGF-b in HIFU treated-tumor tissues by ELISA and found a significant increase of IFN-c and a decrease in TGF-b in M-HIFU-treated tumors compared to untreated control, suggesting that M-HIFU can induce more antitumor microenvironment in treated tumors.Similar trends were observed in tumors treated with thermal ablation, but the changes were modest [39].In summary, many pre-clinical studies have demonstrated the induction of enhanced secretion of pro-inflammatory cytokines and reduced secretion of immunosuppressive cytokines in tumor tissues by mechanical ablation, which is ideal for inducing local and systemic antitumor immunity.

Functional changes in adaptive immune responses
Based on the different cellular composition, release of tumor antigens and production of cytokines induced by the different modes of focused ultrasound, histotripsy and M-HIFU would be expected to have advantages in the induction of adaptive immunity against tumors compared to thermal ablation.In the B16 melanoma model, histotripsy upregulated tumor antigen GP33-specific CD8þ cells in tumors more significantly than ablative radiation and thermal ablation [49].In our experience, the induction of tumor antigen HER2-specific T cells as analyzed by IFN-c ELISpot assay, tended to be greater in M-HIFU-treated mice compared to T-HIFU-treated mice.These results are thought to be due to the higher antigenicity of tumor debris from histotripsytreated tumors, compared to those from thermally treated tumors.Histotripsy could also induce a significant abscopal effect in remote tumors, while no such effect was observed by irradiation or radiofrequency ablation of the same B16 tumors [49].Using a bilateral breast tumor model in mice, we also observed a significant abscopal effect induced by M-HIFU, but thermal HIFU could not suppress the growth of remote tumors.In addition, when anti-PD-L1 antibody was administered in combination with M-HIFU treatment, a stronger abscopal effect was observed (Figure 2) and was dependent on both CD8 T cells and NK cells [39,49].Therefore, histotripsy and M-HIFU appear to be useful ablation strategies to induce potent adaptive immunity.Further, they can synergize with other immunotherapies.

Combining histotripsy and intratumoral immunotherapy
Although histotripsy can be combined with systemic immunotherapy, the limited tumor specificity and the immune-mediated toxicities of systemic immunotherapies have provided impetus to study intratumoral immunotherapy.Indeed, the intratumoral oncolytic virus talimogene laherperepvec (T-VEC) is commercially available to treat unresectable melanoma and numerous other intratumoral immunotherapies are under development [52].We have investigated the effects of intratumoral administration of plasmid IL-12 followed by electroporation in triple-negative breast cancer (TNBC).The treatment significantly suppressed the growth of a TNBC cell line and prolonged survival in mice [53].Moreover, the treatment demonstrated an abscopal effect in untreated lesions via the induction of systemic anti-tumor immunity.In an additional component of this published work, a clinical trial in women with TNBC tested a combination therapy comprising IT-pIL-12 treatment followed by anti-PD-L1 therapy.The results confirmed systemic antitumor immune responses in distant, untreated tumors without any evidence of cytokine storm or other toxicity.However, in these studies, the abscopal effect of treatments such as pIL12-EP on untreated, distant tumors was insufficient to eradicate them and the clinical response rates were low.We hypothesize that the physical/anatomical barrier of the tumor stroma and high interstitial tissue pressure within tumors prevent the effective distribution of intratumorally delivered therapeutic agents and minimize the infiltration of immune and T cells into tumors.Based on our experience using M-HIFU to enhance CD4 and CD8þ T cell infiltration into tumors [39], we hypothesized that systemic anti-tumor immunity induced by IT-pIL-12 would be enhanced by combining extracorporeal M-HIFU with IT-pIL-12.We observed that the combination treatment was more potent than IT-pIL-12 alone even when used to treat large, established tumors.While M-HIFU was enough to significantly slow the growth of primary tumors, the abscopal effect was only seen when a combination of M-HIFU and the ICB anti-PD-L1 was used.These data highlight both the promise of this technique and the need to improve upon it for truly robust anti-tumor immune responses that can target metastatic lesions.

Intratumoral histotripsy and immunotherapy
Currently, the main histotripsy device for cancer immunotherapy is extracorporeal, which lacks the ability to target tumor lesions behind bones and fat in an efficient way [37].To address the limitations of extracorporeal M-HIFU, it appears that intratumoral histotripsy using a miniaturized ultrasound device is a promising strategy to deliver reagents into deep-seated tumors and induce immune modulation.
Figure 3 illustrates the procedure of intratumoral histotripsy immunotherapy.A miniaturized ultrasound transducer integrated into a needle or catheter is injected intratumorally.It generates ultrasound waves with dozens of megapascals and induces cavitation bubble clouds (Figure 3A).Following that, the sonicated tumor cells undergo necrosis and release tumor antigens.It further maximizes the infiltration of T cells, dendritic cells, natural killer cells and macrophage into tumors, and eventually, triggers the immune response (Figure 3B).
To date, intratumoral histotripsy immunotherapy is still in its infancy.Only a few intratumoral ultrasound devices were reported for cancer immunotherapy.Tang et al. employed a minimally invasive ultrasound needle that can be inserted into the tumor.This kind of intracorporal ultrasonic horn device has a center frequency of 27.3 kHz and can generate acoustic pressure over a few hundred kilopascals.It can mechanically disrupt tumor tissue to increase the infiltration of CD8þ T cells and modulate the immunosuppressive tumor microenvironment [54].

Clinical results of histotripsy
While a number of clinical trials of conventional HIFU have been reported, there have been fewer reports with mechanical HIFU or histotripsy, due to the need for instrumentation capable of delivering histotripsy to patients.A phase I clinical trial (NCT03741088) of liver histotripsy in patients with multifocal hepatic malignant tumors was conducted in Barcelona in 2019 [45] using a clinical prototype liver histotripsy device (Vortex Rx) made by HistoSonics.Eleven tumors were targeted in the 8 patients who all had unresectable endstage multifocal liver tumors: colorectal liver metastases (CRLM) in 5 patients (7 tumors), breast cancer metastases in 1 (1 tumor), cholangiocarcinoma metastases in 1 (2 tumors) and hepatocellular carcinoma (HCC) in 1 (1 tumor).The average targeted tumor diameter was 1.4 cm.Local tumor regression was observed by MRI 2 months after histotripsy treatment for all tumors except for 1 mislocated tumor of 5 mm in size.Tumor markers in 1 patient with HCC and 1 patient with colorectal liver metastasis continued to decline after histotripsy treatment and the patient with colorectal liver metastasis experienced shrinkage of untreated distant colorectal cancer 8 weeks after histotripsy treatment.Based on these results, the need for more definitive clinical trials is warranted.The aim of a multi-center non-randomized phase I/II trial (#HOPE4LIVER) is to assess the initial safety and efficacy of the prototype investigational 'Edison System' in the treatment of primary and metastatic liver cancers [55] (Clinicaltrials.govidentifier-NCT04573881).

Summary and conclusions
Mechanical high-intensity focused ultrasound (M-HIFU) broadly describes the non-ionizing and non-thermal ablation technology delivered by ultrasound that can be classified into three categories: intrinsic threshold histotripsy, shockscattering histotripsy and boiling histotripsy.With acoustic cavitation as the primary mechanism, histotripsy displays unique features that distinguish it from conventional HIFUbased ablation technologies which are based on thermal ablation or thermal HIFU (T-HIFU).Characteristic of non-thermal ablation including physical tissue disruption, tissueselective ablation and a precise boundary as there a no heat sink effects.Commercial histotripsy devices have been developed that delivery ultrasound energy using extracorporeal devices and Phase I clinical trials in patients with malignant liver tumors have shown the initial safety and efficacy of histotripsy of this noninvasive approach.Nonetheless, the extracorporeal delivery of ultrasound energy has limitations due to acoustic barriers found in not only the air/fluid interfaces but also fatty tissue and bone.Consequently, the treatment of tumors in internal organs such as the lung and the gastrointestinal tract remains unfeasible, as these organs contain gas and the resulting low cavitation threshold in these organs would lead to extensive collateral damage to the normal organ tissue.Furthermore, histotripsy requires very high ultrasound pressures, and the achievable pressure is proportional to the transducer aperture size that can be applied to the targeted lesion, gas or bone may block a location that can be treated by histotripsy using extracorporeal ultrasound generators.Finally, there is a limitation on the depth of the tissue that can be treated by extracorporeal delivered ultrasound due to attenuation.Deeper lesions or extreme patient size may limit delivered energy.Fortunately, creative solutions, such as catheterbased M-HIFU devices may be used to overcome some of, if not all, of these limitations.Catheter-based devices could be used percutaneously, or with endoscopic or laparoscopic devices to delivery ultrasound energy to locations currently unachievable.
One of the unique potential benefits of histotripsy has been shown to stimulate a potent local and subsequent systemic immune which facilitates a tumor-specific abscopal response in preclinical animal tumor models [39] and potentially in the Phase I liver cancer human trial.In animals, the systemic, presumably tumor-specific immune response can result in systemic antitumor effects, observable by demonstrated regression of the residual tumor, even if untreated by M-HIFU.Furthermore, local tumor treatment would also result in the reduction of distant metastases, and finally, the local immune response, generally converting a 'cold' tumor to a 'hot' tumor would result in the enhancement of current conventional immunotherapy strategies, such as enhancing immune checkpoint blockade.Even though the current immunological mechanisms have not been completely delineated, the global treatment effect beyond the local tumor ablation appears to bring additional systemic antitumor benefits and these are likely to result in improved clinical outcomes for cancer patients.Preclinical studies combining systemic immune therapy appearing promising, but clinical studies of systemic immune therapy, including novel combination have been complicated by systemic toxicities.One potential alternative is combining histotripsy with intratumorally immunotherapy, which has been demonstrated in preclinical models and may be testing in future clinical studies.A promising alternative is to combine intratumorally histotripsy and immunotherapy using microcatheterplaced devices to delivery intratumoral histotripsy immunotherapy.
Although considered a form of hyperthermia in the past, the current technologies delivering high-intensity focused ultrasound in a non-thermal formal has offered promise in both preclinical and clinical studies.This promise results from both improved local ablation outcomes, but also promising immunologic features that can be leveraged to extend the emerging promise of cancer immunotherapy and ongoing technical advances will continue to be tested.As ongoing research continues, greater progress will be made in comparing and contrasting results if all investigators will continue to report not only the parameters of the delivery energy but also details of preclinical models to enable analysis of the immune responses.Critically, as clinical responses and immune analysis of patients undergoing histotripsy treatment are observed, opportunities to optimize clinical responses to histotripsy alone and in combination with conventional and experimental therapies will emerge.A systemic and methodical approach to clearly defining protocols and reporting results will enhance our ability to optimize application and scheduling in M-HIFU in the context of multi-modality care of cancer.
positive pressure, P _ : peak negative pressure.

Figure 2 .
Figure 2. Combination of M-HIFU and PD-L1 blockade enhanced the abscopal effect but could not eradicate distant tumors in a bilateral MM3MG-HER2 model.MM3MG-HER2 cells were implanted into the right thigh and left flank of BALB/c mice.Seven days later, M-HIFU was administered to right thigh tumors in the M-HIFU monotherapy and combination therapy groups.Anti-PD-L1 antibody (100 mg/inj) or isotype control IgG was intraperitoneally injected on days 10, 13 and 16.Tumor growth curves for treated and untreated tumors are shown.Data represent results from 2 pooled experiments.n ¼ 13-19 per group; error bar, mean ± SD; � p < 0.05, �� p < 0.01, ��� p < 0.001, ���� p < 0.0001.