Emerging HIFU applications in cancer therapy

Abstract

High intensity focused ultrasound (HIFU), is a promising, non-invasive modality for treatment of tumours in conjunction with magnetic resonance imaging or diagnostic ultrasound guidance. HIFU is being used increasingly for treatment of prostate cancer and uterine fibroids. Over the last 10 years a growing number of clinical trials have examined HIFU treatment of both benign and malignant tumours of the liver, breast, pancreas, bone, connective tissue, thyroid, parathyroid, kidney and brain. For some of these emerging indications, HIFU is poised to become a serious alternative or adjunct to current standard treatments – including surgery, radiation, gene therapy, immunotherapy, and chemotherapy. Current commercially available HIFU devices are marketed for their thermal ablation applications. In the future, lower energy treatments may play a significant role in mediating targeted drug and gene delivery for cancer treatment. In this article we introduce currently available HIFU systems, provide an overview of clinical trials in emerging oncological targets, and briefly discuss selected pre-clinical research that is relevant to future oncological HIFU applications.

• Cancer
• emerging
• HIFU
• ultrasound

Introduction

This year, high intensity focused ultrasound (HIFU), also called focused ultrasound surgery (FUS), enters its seventh decade as a clinical, interventional technology – dating back to the first demonstrations of its therapeutic potential in disorders of the central nervous system by Lindstrom [1] and Fry [2] in the 1950s. The most significant advances in clinical applications of HIFU have occurred over the last 20 years, as the development of modern magnetic resonance imaging (MRI) and diagnostic ultrasound allowed for accurate targeting and monitoring of HIFU interventions. At present there are both validated clinical applications of HIFU as well ongoing clinical trials examining new therapeutic opportunities. In this article we will provide a brief overview of HIFU delivery systems and discuss the current clinical status of HIFU in emerging oncological targets (excluding the more established targets of prostate cancer and uterine leiomyomas) as well as potential future applications.

HIFU systems

Ultrasound causes both thermal and non-thermal effects in biological tissues. Absorption of high intensity ultrasound energy results in elevated tissue temperatures and ablation at specific temperature thresholds. Non-thermal effects related to cavitation (the formation and collapse of gas-filled bubbles in an ultrasound field) or generation of boiling bubbles can also result in mechanical tissue disruption. To achieve these effects with HIFU, a piezoelectric transducer generates a pulse of ultrasound energy that propagates via a coupling medium (typically oil or degassed water) into the body, converging on a focal region. Within the focal region a small volume of tissue is exposed to high amplitude pressure waves. Depending on the protocol selected for HIFU delivery, the effect achieved at the focal point can be tailored to be predominantly thermal or non-thermal. The desired tissue effect occurs within the focal region, and the intervening tissue is not damaged. The minimally invasive nature of this ‘trackless’ therapy is an appealing alternative to percutaneous or surgical approaches (Figure 1).

Figure 1. Simplified cross-sectional image of an MRI-guided focused ultrasound (MRgHIFU) system. The tissue between the HIFU beam focus within the target lesion and the HIFU transducer remains undamaged, as does the tissue distal to the target lesion. A temperature map within the target region can be obtained in near-real time in multiple imaging planes.

There are several commercially available HIFU devices that are manufactured worldwide for treatment of both benign and malignant tumours. In clinical practice both MRI and diagnostic ultrasound are used to safely guide HIFU therapy, and therapeutic HIFU transducers are designed for pairing with one of these imaging modalities. Ultrasound-guided HIFU (USgHIFU) systems allow real-time visualisation of gas bubbles caused by cavitation and boiling of the targeted soft-tissue [3]. The active focal point appears as a hyperecho at the focus in the field of view of the diagnostic ultrasound probe. USgHIFU is relatively low cost and portable in comparison with MRI-guided HIFU (MRgHIFU) systems. There is active, preliminary research in the field of temperature mapping using diagnostic ultrasound [4–6]; however, to date, a reliable and clinically accepted means of real-time temperature mapping with diagnostic ultrasound is not widely available. In contrast, MRgHIFU systems are large and expensive, but offer fine anatomic detail and high sensitivity for tumour detection (useful for therapy planning), as well as near real-time thermal monitoring of HIFU treatment with accuracy to approximately 1 °C [7,8] (Figure 1).

Oncological applications of HIFU

HIFU in liver tumours

HIFU has been used for ablation of both primary and secondary liver tumours. More than 80% of patients with hepatocellular carcinoma (HCC) are poor candidates for curative surgical resection due to advanced underlying cirrhosis, and thus minimally invasive alternatives to surgery are desirable [9–11]. Early trials conducted in the late 1990s in China demonstrated safety and feasibility of extracorporeal USgHIFU therapy for patients who either had unresectable HCC, or refused surgery [12]. These results were subsequently confirmed in small trials in the UK [13,14]. USgHIFU ablation has also been successfully employed as adjunctive therapy to transarterial chemoembolisation (TACE) for non-surgical cases of HCC [15–18] and hepatoblastoma [19]. Chok et al. offered USgHIFU as a bridging therapy to liver transplant for patients with HCC in the context of Child-Pugh C cirrhosis who were not candidates for TACE or radiofrequency ablation (RFA), and successfully bridged a small cohort without significant adverse events [20]. Other groups have successfully performed USgHIFU treatments for patients with liver metastases and observed complete lesion ablation, with minimal adverse events [21,22]. These preliminary data are promising and future randomised controlled trials will help to define survival benefit.

There are challenges to performing HIFU within the liver. Limited therapeutic windows through the ribs contribute to reflection of high energy ultrasound waves from the bone that can result in unintended burns to soft tissues between the transducer and its target [23]. Respiratory motion is an additional complicating factor that can be mitigated via selective intubation and control of respiratory motion via mechanical ventilation, as well as saline injection to augment liver distance from the diaphragm [24]. MRgHIFU for liver lesions remains in its infancy, but MR may offer additional solutions to current therapeutic obstacles encountered with ultrasound guidance. Active research continues in the development of motion gating for both MRgHIFU therapy and MR thermometry [25,26].

HIFU in breast tumours

The breast is ideally suited for HIFU therapy due to its superficial position, with no intervening vital structures. Ablation of both benign and malignant breast tumours has been performed using USgHIFU [27,28] and MRgHIFU [29–33]. HIFU is a particularly attractive option for patients who desire breast conserving therapy since it is completely non-invasive, and could theoretically reduce the number of adverse events and improve cosmetic outcome compared to surgical intervention. Hynynen et al. [32] conducted a pilot clinical trial in the USA using MRgHIFU for treatment of benign fibroadenomas of the breast, and demonstrated feasibility without significant adverse events. Overall tumour size was significantly smaller on follow up MRI, with subjective tissue softening over an average follow up of 6 months (range 1.5 months to 4 years). Early progress has also been made towards demonstrating non-inferiority of HIFU ablation to local surgical excision for solitary breast cancer. In 2005, Wu et al. [28] reported a series of 22 patients with biopsy-proven, solitary breast cancer who refused surgery and underwent USgHIFU with adjuvant chemotherapy. Five-year disease-free survival and recurrence-free survival were 95% and 89%, respectively. Furusawa et al. [29] reported results from a clinical trial in Japan involving 21 patients with biopsy-proven invasive or non-invasive ductal carcinoma. Patients were treated with MRgHIFU and followed over an average of 14 months (range 3–26 months), with one case of recurrence during that timeframe. Napoli et al. [33] recently reported a small series of 10 patients in Italy with biopsy-proven invasive ductal breast cancer (stage T1M0N0) who underwent MRgHIFU of their tumours prior to routine breast conserving surgery. Surgical pathology revealed absence of residual cancer in 90% of cases with at least 5 mm of normal breast tissue around the target tissue. Further studies with long-term observation are necessary to draw definitive conclusions regarding the possibility of improved cosmesis, fewer adverse events, and non-inferior survival with HIFU vs. surgical standard of care in treatment of solitary breast cancer.

Of note, there is active research in pre-clinical, murine models examining MRgHIFU in delivery of adjuvant nanoparticle chemotherapy for breast cancer. Wong et al. reported on the use of a thermally activated chemotherapy agent as adjuvant treatment with MRgHIFU ablation of implanted breast cancer in mice [34]. They found a significant increase in median survival time for mice who received the two therapies concurrently, vs. those who did not – suggesting the selective activation of the chemotherapy drug at the focus of the ultrasound treatment provides synergistic destruction of residual tumour cells, and may increase the rate of sustained remission.

HIFU in pancreatic cancer

Chemotherapy, radiotherapy and targeted drugs have all been relatively unsuccessful to date in significantly altering the course pancreatic cancer – which, untreated, carries a 16% 12-month survival rate [35]. Most patients are inoperable at the time of diagnosis due to locally advanced disease or metastasis, and experience severe abdominal pain as the disease progresses [36]. Ablative HIFU has been used as both primary and palliative therapy for pancreatic cancer. The majority of reported treatments have taken place in China since the late 1990s, and more recently case series have emerged from Korea, Japan, and a few European groups. Xiong et al. retrospectively examined 89 patients with inoperable pancreatic cancer (TNM stage II–IV) who underwent ablative USgHIFU for pain palliation – pain relief was achieved in 81% of patients [37], similar to results achieved in other studies, including a small series from Italy [38–41], with no significant adverse events. Zhao et al. [42] conducted a phase II trial to evaluate the safety and efficacy of concurrent gemcitabine and USgHIFU therapy for 39 patients with locally advanced, unresectable pancreatic cancer. Pain was relieved in 79% of patients, the overall oncological response rate was 44% (determined by CT scan and the Response Evaluation Criteria in Solid Tumours), and four responders were subsequently able to undergo surgery with negative resection margins; however, the overall survival at 12 months was 51% – similar to the previously reported 53% 12-month survival rate derived from gemcitabine therapy alone [43]. Further randomised controlled clinical trials are needed to evaluate any significant survival benefit achieved with ablative HIFU therapy alone. The mechanism of pain relief is not fully understood, but is hypothesised to involve direct, thermally induced damage of nerve fibres within the tumour, as well as indirect tumour shrinkage, removing mass effect from adjacent nerves [44].

There are several areas of active HIFU research that are applicable to pancreatic cancer. Endoluminal HIFU transducers are currently in pre-clinical development for use with both ultrasound [45] and MR [46] guidance. These devices are particularly appealing for use in pancreatic cancer therapy, as they allow endoscopic placement of the HIFU transducer within the stomach or duodenum, directly adjacent to the pancreas. This allows for easier lesion targeting and lower risk of damage to overlying soft tissue structures in this delicate anatomical area. The use of HIFU-induced hyperthermia for localised chemotherapy drug release also shows great potential for treatment of pancreatic cancer [47]. Conventional chemotherapy drugs are relatively ineffective in penetrating the fibrotic, hypovascular stroma of pancreatic adenocarcinoma, independent of the target specificity of the drug [48]. Preliminary data from murine pancreatic cancer experiments utilising MRgHIFU hyperthermia to release doxorubicin from a temperature-sensitive liposome within the tumour resulted in a 2–8-fold increase in doxorubicin concentration within the tumour [47].

HIFU in bone metastases

Bone metastases are the greatest contributor to cancer-related pain, and often severely diminish patient quality of life [49]. Radiation therapy, in compliment to systemic and analgesic therapy, is currently the standard of care for localised, painful bony metastases. Overall response (pain reduction of ≥2 on the Numeric Rating Scale pain score at the treated site, without analgesic increase, or analgesic reduction of ≥25% without increased pain) to this treatment is approximately 60%, with complete response (a pain score of zero at the treated site) in approximately 25% [50]. Patients with residual or recurrent pain following radiotherapy have limited treatment options, including: re-irradiation (resulting in some response, which may or may not be temporary, in approximately 60% of patients) [51], surgical intervention (invasive, with complications observed in 13–20% of cases) [52,53], percutaneous cryoablation (minimally invasive, 49% of patients have at least partial response 1 week post-treatment, 69% at 24 weeks) [54], and HIFU. Relevant advantages of HIFU over other modalities in this context include its non-invasive nature and the repeatability of the treatment until the clinical goal is reached – without issues related to radiation accumulated dose, or post-procedural tissue changes limiting access in repeated percutaneous intervention.

MRgHIFU is clinically approved in the European Union for palliative treatment of bone lesions, and is garnering clinical attention in western healthcare for treatment of radiotherapy-refractory painful bony metastases. Cortical bone has high acoustic resorption and low thermal conduction, and thus ultrasound energy is absorbed by the cortical surface, with minimal penetration into the medullary cavity [55]. In cases of painful bone metastases, this relationship is advantageous for HIFU therapy – with most energy depositing over the periosteal surface of the bone, causing thermal injury to the periosteal nerves responsible for nociception [56]. Preliminary clinical studies on the safety and efficacy of MRgHIFU for ablation of painful bone metastases demonstrated excellent response rates and safety – with 72–92% of patients reporting pain reduction and no significant adverse events [56–58]. A recent phase III trial by Hurwitz et al. [59] involving 147 patients from 17 medical centres across the USA, Canada, Israel, Italy and Russia found overall response rate in patients with painful bony metastases undergoing MRgHIFU ablation to be 64%. The most common adverse event was sonication pain, which occurred in 32% of patients, with 60% of adverse events resolving on the day of treatment. This study has established the safety and efficacy of MRgHIFU as a palliative treatment for painful bone metastases.

HIFU in primary bone tumours

HIFU ablation has been used in China to treat primary bone malignancy. Li et al. [60] demonstrated safety of USgHIFU treatment for a small series of patients with osteosarcoma. Transient skin burns and local nerve damage were identified as the most common complications. Chen et al. conducted a prospective clinical study of limb-sparing USgHIFU ablation, in combination with appropriate chemotherapy, for 80 patients with primary bone tumours who refused or were not candidates for surgery [61]. The treated tumours included osteosarcoma (78.8%), chondrosarcoma (12.5%), Ewing sarcoma (3.7%), giant cell tumour (1.3%) and unknown histological type (3.7%). 75% of patients had US Musculoskeletal Tumor Society stage IIb disease (high grade, extracompartmental, no metastases), 25% had stage III disease (metastatic). Complete tumour necrosis was achieved in 86% of patients (obstacles included tumour size >10 cm, or pelvic tumours with overlying bowel), using 1–2-cm margins of normal soft tissue and 3–5-cm of normal bone margins; 7% of patients with complete tumour necrosis had local progression at mean follow-up of 36.8 months (range 5–87 months), with 11.8% recurrence at 5 years (similar to results reported with limb-salvaging surgery). The disease-free survival rate was 86.4% for stage IIb patients at 5 years who had complete tumour ablation and completed their concurrent chemotherapy regimen. These results are comparable to those reported in western literature for limb-sparing surgical intervention combined with radiotherapy [62,63]; however, well-powered randomised control trials have yet to be published to compare HIFU ablation with limb-sparing surgery.

Osteoid osteoma (a benign bone tumour most common in young men), when symptomatic, is treated with CT-guided RFA with reported success rates of 89–95%, as reviewed by Motamedi et al. [64]. Two recent studies from Italy examined MRgHIFU as an alternative therapy for non-spinal osteoid osteoma. Napoli et al. [65] reported a small series (n = 6) with good efficacy and without major complications. Geiger et al. [66] performed a prospective, multicentre clinical trial (n = 29) and demonstrated complete clinical success (pain score of 0) in 90% of patients over 1 year follow-up, without adverse events. These preliminary results support the possibility of comparable clinical efficacy between CT-guided RFA and MRgHIFU. The latter is an appealing alternative given its non-invasive nature and lack of associated ionising radiation exposure for a predominantly young patient population. Additional randomised clinical studies will further help to define the role of MRgHIFU in treatment of osteoid osteoma.

HIFU in connective tissue/extra-abdominal desmoid tumours

Extra-abdominal desmoid tumours lack the capacity for metastasis, but can be locally aggressive, and are prone to recur even after complete resection [67]. Symptomatic, recurrent tumours after primary surgical resection are often treated with radiation therapy, with or without neoadjuvant chemotherapy, to avoid the morbidity of repeated surgical resections [68]. Both the non-invasive nature and the repeatability of HIFU therapy make it an appealing modality in this context. Wang et al. [69] reported results of USgHIFU ablation for eight patients with multiple recurrent extra-abdominal desmoid tumours, with overall >50% decrease in tumour volume over mean follow-up of 30 months. Preliminary data reported by Ghanouni et al. [70] using MRgHIFU to treat five patients with extra-abdominal desmoid tumours showed no serious adverse events. Further studies are needed to evaluate long-term outcomes and success in local disease control vs. conventional radiotherapy for non-surgical or recurrent disease.

HIFU in thyroid and parathyroid tumours

The cost and potential morbidity of surgery for patients with benign, but enlarging thyroid nodules or symptomatic parathyroid adenomas has inspired a number of minimally invasive approaches to therapy – including percutaneous ethanol injections, laser ablation, or RFA [71,72]. In Europe HIFU has also recently been explored as a non-invasive means to treat these conditions. Esnault et al. [73] reported a feasibility study of 25 patients with multinodular goitre who underwent USgHIFU ablation therapy prior to surgery. Successful ablation was seen in 95% of treated nodules demonstrated by either coagulative necrosis or cavitation on histopathological analysis. HIFU treatments were well tolerated with good patient safety – transient local pain and minor skin burns were the most common adverse events. Kovatcheva et al. [74] described USgHIFU ablation therapy for four patients with symptomatic primary hyperparathyroidism who declined surgery. Serum parathyroid hormone level normalised in two of the four patients post-treatment. Mild to moderate transient subcutaneous oedema was the most common adverse effect observed. These initial studies are promising, and further studies, including randomised trials examining long-term outcomes vs. standard of care, will help to define the role of HIFU in thyroid and parathyroid disease.

HIFU in kidney tumours

Compared to the 1990s, renal cancers are currently diagnosed at a significantly earlier stage and smaller overall size in the USA [75,76]. Approximately 38% of these tumours are less than 3 cm at time of diagnosis [76] and 60% are asymptomatic [77]. For peripheral, radiologically suspicious lesions 1–4 cm in size, partial nephrectomy is currently the treatment of choice [78]. However, reported surgical complication rates range from 9–33%, due in large part to the many medical co-morbidities commonly encountered in this patient population [79], and thus minimally invasive alternatives are desirable. Percutaneous and laparoscopic thermal ablation (e.g. radiofrequency, microwave, or cryoablation) have been employed for these small tumours in recent clinical trials demonstrating 93–95% recurrence free survival at 5 years [80,81]. In Europe, pilot clinical trials of laparoscopic HIFU devices (trochar-driven, ultrasound-guided HIFU transducers directly approximated to the renal tumour) have demonstrated successful tumour ablation, with the potential benefit of no required tumour puncture vs. other minimally invasive ablative techniques [82,83]. Further studies are needed to clarify the oncological success and reliability of laparoscopic HIFU.

Attempts at extracorporeal HIFU ablation of kidney tumours have shown unsatisfactory tumour destruction with a high rate of incomplete ablation due to respiratory motion and acoustic impedance from overlying ribs and perinephric fat [21,84–86]. As mentioned above, there is ongoing research in the development of motion gating for MRgHIFU therapy [25] and MR thermography [26] that will theoretically allow for more effective and accurate targeting of organs prone to respiratory movement – including the liver and kidneys.

HIFU in brain tumours

Trans-cranial HIFU is an appealing, non-invasive treatment modality for neurosurgical interventions. Efficient ultrasound transduction through the skull, however, has posed a significant obstacle that has prevented this therapy from entering clinical practice since early experiments in the 1950s [87]. In the last 10 years, the advent of MR thermography and phased-array HIFU transducers paved the way for a pilot clinical trial in the USA that examined ablative MRgHIFU for treatment of glioblastoma multiforme [88]. McDannold et al. [88] attempted MRgHIFU therapy in three patients with glioblastoma using a hemispherical phased-array HIFU transducer. They successfully focused the transcranial ultrasound beam within the brain parenchyma and generated a heat signature detectable with MR thermography, but unfortunately were limited by the device power available at the time, and were unable to achieve thermal coagulation at the targeted lesions. Subsequent studies have successfully utilised MRgHIFU with hemispheric phased-array transducers and computed tomographic correction algorithms to generate targeted intracranial foci of thermal coagulation for treatment of chronic neuropathic pain [89,90], essential tremor [91] and Parkinson’s disease [92]. These results are promising in anticipation of forthcoming oncological feasibility studies.

In addition to ablation, HIFU combined with systemically administered microbubbles has been shown to locally and transiently disrupt the blood–brain barrier (BBB) at discrete targets in pre-clinical studies [93]. Over the last 10 years, pre-clinical studies in murine cancer models of both primary and metastatic brain tumours have shown successful delivery of multiple systemic therapies to the brain with FUS-mediated BBB opening. Successful immunotherapies include Herceptin [94], and targeted delivery of natural killer cells to breast tumours metastasised to the brain [95]. Chemotherapy agents, including 1,3-bis(2-chloroethyl)-1-nitrosourea (BCNU) [96,97], epirubicin [98] and doxorubicin [99–101] have also been successfully transported across the BBB following FUS exposure, with associated decrease in tumour growth and improvement in animal survival. These results are encouraging, and future human trials of these techniques will help to elucidate their role in clinical treatment of brain cancer.

Conclusion and future applications

High intensity focused ultrasound is an effective, non-invasive technique to achieve tumour ablation and to facilitate targeted medical therapy to tumours. Although HIFU technology has been in clinical use since the 1950s, limited complimentary technologies have hindered its advancement into clinical oncology beyond a small spectrum of disease. Today, HIFU has reached an exciting threshold of both transducer and guidance-system development. The advent of MRI guidance systems with real-time thermal mapping, phased-array transducers with computed tomographic correction algorithms, endoluminal transducers, motion gating for both MRgHIFU therapy and MR thermometry, temperature mapping for diagnostic ultrasound, and numerous innovations in medical nanotechnology (such as microbubble-assisted BBB disruption, and encapsulated chemotherapy and gene therapy agents capable of targeted, thermally induced delivery) have paved the way for oncological applications that were previously impractical in western medicine. HIFU is poised to bridge the promising data from preliminary trials to more robust, randomised clinical trials for treatment of liver, breast, pancreatic, bone, desmoid, thyroid, parathyroid, kidney and brain tumours. Although completion of such trials is challenging and time-consuming for relatively uncommon diseases, the results will help to better define the emergent role of HIFU in management of a broad spectrum of oncology.

Declaration of interest

The authors report no conflicts of interest. The authors alone are responsible for the content and writing of the paper.