Introduction

Glioblastoma is the most common malignant primary brain tumor with an annual incidence rate of approximately 3.21 per 100,000 population [1]. Glioblastoma is a dismal disease with overall survival time of approximately 15 months [2, 3]. First-line treatment includes gross total resection and adjuvant combined chemotherapy with temozolomide and fractionated radiotherapy, followed by maintenance temozolomide [2, 4]. Unfortunately, the progression of glioblastomas is inevitable, and efficacy of the second-line treatment options is limited [5, 6]. Diffuse astrocytomas typically have slower growth rate and a more indolent course than high grade gliomas. However, low-grade tumors remain incurable and can progress to become high-grade tumors causing rapid disease progression and clinical deterioration [7, 8]. There remains an urgent need for new treatment strategies that could help to optimize prognosis of glioma patients [9, 10].

Sonodynamic therapy (SDT) involves systemic delivery of non-toxic chemical agents (sonosensitizers) that accumulate in tumor cells or environment. These agents are activated and become cytotoxic by exposure to low-intensity targeted ultrasound. Essentially, both the sensitization and ultrasound exposure are not tumoricidal by themselves; instead the cytotoxic events occur when both are combined [11]. SDT represents an emerging approach that offers the possibility of incisionless eradication of solid tumors in a site-directed manner, and this approach is becoming increasingly studied for treatment of gliomas [12, 13].

Focused ultrasound (FUS) is an emerging technology that allows controlled, spatially and temporally precise delivery of ultrasound energy to intracranial targets [14, 15]. High-intensity FUS can be used for thermal ablation and is an effective method for thalamotomy and subthalamtomy for essential tremor [16] and tremor predominant Parkinson’s disease [17]. Application of high-intensity FUS for treatment of brain tumors is limited by typically a narrow treatment envelope and long sonication time required to ablate significant tumor volume [18, 19]. At lower intensities, FUS can be applied without causing tissue damage, can be used to sonicate larger volumes and is being actively investigated for local and temporary disruption of the blood brain barrier to enhance delivery of chemotherapeutics into brain tumors [14, 20] with encouraging results in animal models and initial clinical trials [14, 18]. Therefore, the FUS can be used for SDT of intracranial gliomas with excellent spatial precision and accuracy.

In this article, we will provide comprehensive review of the concept and biological actions of SDT, review published pre-clinical studies of SDT for gliomas, and discuss potential clinical applications and future directions of SDT for gliomas.

The concept of sonodynamic therapy

Activation of a non-toxic compound by an external physical stimulus was first discovered back in 1900 s [21], by interaction of certain dyes with light, and was termed photodynamic therapy (PDT). It is worth reviewing some of the well-established concept of PDT, as some of these are shared with SDT.

PDT relies on the activation of photosensitizers by absorption of a photon of light with the appropriate wavelength. In the presence of oxygen, the excited photosensitizer forms reactive oxygen species (ROS) that can directly induce cellular damage by rapidly oxidizing cellular components. Most photosensitizers used for anti-cancer PDT operate more via creation the type II ROS rather than type I. Several compounds have received clinical approval for PDT, including tetrapyrrole structures, such as porphyrins, chlorins, bacteriochlorins and phthalocyanines. Synthetic dyes and natural products, such as hypericin, riboflavin and curcumin, have been investigated [22]. The selective nature of tumor targeting in PDT is thought to occur through tumoral accumulation via the EPR (enhanced permeability and retention) effect in tumors with leaky vasculature and defective lymphatic drainage [23].The most effective photosensitizers are typically hydrophobic compounds that accumulate in tumor cells and intercalate into membrane structures. The three main cell killing mechanisms of PDT are apoptotic, necrotic and autophagy-associated cell death that are related to photosensitizer localization in the different organelles.

Sonodynamic therapy (SDT) relies on the activation of sonosensitizers (non-toxic compounds) that upon ultrasound activation become cytotoxic by generation of ROS. Other, broader, definition of SDT have been proposed, to include non-chemical-based forms of non-thermal ultrasound therapy as well [24], with one of the most important biological effects being drug delivery [25] and direct destabilization of the plasma membrane, a mechanism known as sonoporation, that can enhance compound transport across the cell membrane [26].

In this review, we restrict SDT to the ultrasound activation of photochemical sensitizers.

Similar to PDT, SDT requires the combination of interaction of a chemical, ultrasound, and oxygen. The major advantage of SDT is the ability to provide more than ten of centimeters of penetration of ultrasound energy into soft tissues depending on the ultrasound frequency, and the possibility of delivering a tightly focused ultrasound beam for focal treatment.

SDT started by evaluating tumor localizing porphyrins in ultrasound-induced reactions [27]. These early studies suggested that cell damage enhancement was probably mediated via single oxygen generated by activation of hematoporfirin by sonoluminescence, the generation of light by collapsing cavitation bubbles [28]. Since then, many different sonosensitizers have proven to be effective sonosensitizers, including (i) porphyrin derivatives, such as hematoporphyrin monomethyl ether (HMME), protoporphyrin IX disodium salt (PpIX), pheophorbide A, Photofrin, Photofrin II, ATX-70, ATX-S10, Ce6 and DCPH-P-Na(I); (ii) xanthene derivates, including erythosien B, rose Bengal; (iii) and inorganic sonosensitizers, such as Ti02 [29]. Several mechanisms of action of the ultrasound have since been identified (Fig. 1). In addition to direct activation of sonosensitizers by sonoluminescence and the subsequent generation of free radicals, mechanisms also include direct generation of free radicals by pyrolysis-mediated processes also taking place in the close vicinity of hot collapsing cavitation bubbles [27, 30]. Free radicals are formed via direct pyrolysis of the sonosensitizers, breaking apart the sensitizer generating free radicals that can react with other endogenous substrates to generate ROS, or by interaction with hydroxyls and hydrogen radicals formed by pyrolysis of water. Another possible mechanism of action of SDT relies on physical destabilization of the plasma membrane by the sonosensitizer, due to cell susceptibility to the mechanical action of the ultrasound, such as local shear force. While the activation process by ultrasound during SDT is generally considered to be induced by cavitation, and potentially via sonoluminescence, cell apoptosis, and pyrolysis, the exact mechanism of SDT is probably not governed by a universal phenomenon, and still needs to be analyzed based on the ultrasound treatment conditions and the formulation of a sonosensitizer [31,32,33]. Two major sonochemical products arise from ultrasonically-induced cavitation: free radicals and singlet oxygen, although the role of this later has been controversial in SDT [31]. Free radicals can induce a chain reaction of lipid peroxidation and cell damage, while singlet oxygen, once entering excited singlet state, is capable of oxidizing cellular contents. Available evidence strongly suggests a pivotal role for ROS in SDT [33] and the remaining questions mostly are concerned with the mechanism of ROS generation and which ROS are mainly responsible for mediating cytotoxic effects of SDT. Immune-modulation anti-tumor effect of SDT was also suggested [34].

Fig. 1
figure 1

Mechanisms of SDT (illustrated with 5-ALA as sonosensitizer). ROS can be produced by (1) pyrolysis of the compound (in this example ppIX) or water, (2) activation of the sonosensitizer by light produced during sonoluminescence. (3) Alternatively, compound affinity for the plasma membrane could make the cells more susceptible to mechanical stresses caused by the ultrasound. All three routes will lead to downstream signaling leading mainly to apoptosis

Full size image

In Vitro studies of SDT

In vitro studies using a variety of sonosensitizers and different human and rat glioma cell lines generated evidence for the suggesting effectiveness of SDT for gliomas. SDT has been evaluated in in vitro studies using a variety of sonosensitizers, including porphyrin derivatives, such as hematoporphyrin monomethyl ether (HMME) [35, 36], and Photofrin [37,38,39], aluminum phthalocyanine disulfonate [40] and 5-aminolevulinic acid (5-ALA) [41,42,43]. Glioma cell lines used in several published studies include the C6 [35, 36, 41] and the F98 rat glioma cell lines [40], the U251 [37,38,39, 43], U105 [37], U87 [41, 43] malignant glioma cell lines and the U251Oct − 3/4 glioma stem-like cells [43]. Variable FUS parameters were utilized across studies (Table 1).

Table 1 In vitro studies evaluating sonodynamic therapy of glioma cell lines
Full size table
Table 2 Studies exploring sonodynamic therapy in animal models
Full size table

Hao et al. reported an improved apoptotic rate of HMME-SDT treated C6 rat glioma cells as compared to HMME or ultrasound alone. Ca+ 2 overload played a primary role in the apoptotic process which was associated with an increased production of ROS, decreased mitochondrial membrane potential (MMP), and increase in cyt-c [35]. In the study by Dai et al., the apoptotic effect of hematoporphyrin monomethyl ether (HMME)-SDT on C6 glioma cells was greater than HHME or ultrasound treatment alone. HMME-SDT induced apoptosis of C6 glioma cells due to increased ROS production and decreased mitochondrial membrane potential (MMP). Upregulation of caspase-9, caspase-3, and Bax expression and downregulation of Bcl-2 expression suggested a pivotal role of the mitochondrial signal pathway in the apoptotic process [36].

Hayashi et al., reported sensitivity to low-level ultrasound of both U251 and U105 human glioma cells. Photofrin enhanced ultrasound-induced cell death of the U251 cells expressing LRP/α2MR, but not in U105 cells not expressing LRP/α2MR [37]. Xu et al. reported decreased susceptibility of glioma stem-like cells to SDT than the U251 glioma cell line due to ABCG2 protein over expression causing efflux of the sonosensitizer [39]. Addition of fumitremorgin C, an ABCG2 inhibitor, resulted in a significant increase in the Photofrin-SDT mediated relative production of ROS suggesting the usefulness of fumitremorgin C in the SDT treatment of ABCG2-expressing malignant glioma cells [38].

Suehiro et al. and Sheehan et al. demonstrated greater tumoricidal effect of 5-ALA-SDT than FUS and 5-ALA alone on the U251 [43], C6 [41], U87 [41, 43] glioma cell lines and U251Oct − 3/4 glioma stem-like cells [43]. Bilmin et al. reported significant cytotoxic effects of 5-ALA-SDT on RG2 rat glioma cells [44]. Ju et al. demonstrated increased apoptosis, increased production of ROS, and loss of MMP with the addition of hyperthermia to 5-ALA-SDT in vitro. Higher levels of proteins Bax and cleaved caspase-3, 8, and 9 and lower level of bcl-2 were noted in the SDT-hyperthermia group than in the SDT alone group, the hyperthermia alone group and the control group [42]. In the study by Gonzales et al. compared to FUS alone or bleomycin alone, aluminum phthalocyanine disulfonate -SDT and bleomycin significantly inhibited growth of F98 glioma cells as three-dimensional tumor spheroids [40].

Animal models of SDT

SDT has been studied in small animal intracranial and subcutaneous glioma xenograft models (Table 2). An immunodeficient murine intracranial glioma model was used in one study [43]. The majority of studies used C6 glioma cells. Other cell lines included human glioblastoma U87 MG-Red-FLuc [42, 43, 45], U-118 MG [46] and SNB19 cells [42] ,and F98 rat malignant glioma cells [47]. The most commonly used radio-sensitizers were 5-ALA followed by sinoporphyrin. Fluorescein, hematoporphyrin monomethyl ether, Rose Bengal and iRGD modified DVDMS liposome were also tested. FUS parameters (intensity, duty cycle etc.) varied across studies and were reported inconsistently thus making challenging to make reliable comparisons between studies.

In all published studies, treatment with SDT was associated with inhibition of glioma growth and/or tumor proliferation of intracranial and subcutaneous glioma models across multiple tumor cell lines, cell lines, tested sonosentizers and FUS parameters. SDT was associated with decreased tumor growth and with longer survival of animals treated with SDT when compared to control animals [43, 45, 48,49,50]. Upregulation of apoptotic cell death mechanisms after SDT was the most widely studied mechanisms of action of SDT in in vivo glioma models. Overexpression of Terminal deoxynucleotidyl transferase dUTP-mediated nick-end labeling (TUNEL) and upregulation of caspase-3, 8, 9, Bax and Cyto-C and downregulated expression of Bcl-2 were demonstrated with SDT [1, 3, 6, 8]. SDT was also shown to downregulate glioma angiogenesis as evident by reduction of micro-vessel density and expression of vascular endothelial growth factor (VEGF) [50].

SDT and FUS sonication of normal and peritumoral brain tissue was proven to be generally safe [47, 51, 52]. Ohmura with colleagues explored long term-effects of FUS applying 10 and 15 W/cm2, 1.04 MHz, 5-minute exposure, and found that at sonication at 15 W/cm2 caused brain lesions in all treated animals while sonication at 10 W/cm2 did not cause any lesions at 4-weeks post-sonication [52]. On the other hand, Nonanka with colleagues reported that sonication of rat brain caudoputamen region at 25 W/cm2 at 1 MHz for 5 min did not cause brain damage on histological examination independently from sonosensitizer (Rose Bengal) use. However, sonication at 110 W/cm2 at 1 MHz for 3 min induced sharply delineated coagulation necrosis in the majority of animals with greater dose of sonosensiter being associated with greater risk and size of a lesion [51].

Clinical evidence

Notwithstanding the great number of pre-clinical studies that yielded very promising results, efforts to translate this therapeutic approach into clinical practice has been limited to date. The first evidence of the effect of SDT in clinical practice has been described by Inui et al. in a case-report in which SDT was combined with GcMAF-based immunotherapy to treat a patient with terminal breast cancer (invasive ductal carcinoma, grade 3, ER+, PR+, HER2+, right axillary tumor, spinal metastases, intrapleural nodular tumor and right pleural effusion) [53]. SDT was performed using chlorin e6 and 5-ALA as sonosensitizers, and a total of 19 treatments of SDT were conducted in a three-month time span. This treatment protocol gave surprising results–the axillary tumor and intra-pleural nodular tumor disappeared completely; tumor markers were dramatically reduced; and no appreciable side effects were reported. The mechanism that is proposed to be behind the efficacy of this combined approach is the initiation of direct inflammatory necrosis inside tumors, coupled with the production of antitumor immunity via antigen-presenting cells to prevent immune escape [53]. A similar approach was used by the same research group to treat a patient with NSCLC (lung adenocarcinoma, stage 3B), using sonodynamic therapy coupled with GcMAF-based immunotherapy, TTF therapy and ozone therapy. The average survival of such patients with standard treatment protocols is only 8 months, while using this new approach no tumor growth was reported at 15 months without side effects [54].

A recent phase II clinical trial investigated the effect of DVDMS-SDT for peripheral artery disease (PAD) and demonstrated reduced plaque inflammation and improved walking performance of patients treated with SDT when compared to placebo [55]. Inflammation plays a central role in the development of atherosclerosis, and SDT is well-known for its immunomodulatory action, such as induction macrophage apoptosis, promotion of cholesterol efflux, and stabilization of atherosclerotic plaques.

As of 01/02/2021 there is one ongoing phase 0 single center, first in human, open-label study that uses ascending energy doses of SDT delivered via MRgFUS combined with intravenous 5-ALA that aims to assess safety and efficacy this approach in up to 30 patients with recurrent high-grade gliomas (NCT04559685). Eligible participants who are scheduled for tumor resection will be administered intravenous 5-ALA approximately 6–7 h prior to receiving sonodynamic therapy (SDT). Another phase 1 clinical trial has been recently approved (IRB - IRCCS C.Besta – 70/2020; 9/16/2020; 9/75) to assess the safety and feasibility of 5-ALA mediated SDT in patients with glioblastoma: 10 patients will receive SDT and the undergo a clinical and radiological follow-up for 3 weeks prior to tumor resection. In another Phase 0 single center trial (NCT04559685) ascending energy doses of SDT utilizing the MRgFUS combined with intravenous ALA (administered 6–7 h before SDT) will be tested in up to 30 patients diagnosed with recurrent high-grade gliomas who have measurable disease at recurrence defined as at least one contrast-enhancing lesion with a volume of at least 6 cm3 and ≤ 20cm3 of targeted treatment area. The authors will perform dose-escalation and time-escalation of SDT. Two other studies are focused on applying SDT for PAD (NCT03967730) and carotid atherosclerotic plaque (NCT03871725). Considering the substantial body of promising pre-clinical evidence documenting efficacy of SDT for gliomas and the potential translatability of this approach to other malignant brain tumors, clinical trials exploring safety and efficacy of SDT in glioma and other brain tumor patients are anticipated in the near future.

Future directions

To capitalize on the sonoluminescence mechanisms, novel strategies of SDT have recently been proposed, coupling sonosensitizer on the membrane of ultrasound contrast agents, shelled gas bubbles, lipid stabilized microbubble (MB), under the rationale that placing sensitizers in close proximity to MBs undergoing inertial cavitation could enhance their efficiency at generating ROS through sonoluminescence or pyrolysis-mediated processes [56]. Another potentially favorable permutation of this approach also involves attachment of a sonosensitizer to the surface oxygen-carrying MB to improve the sonodynamic effects under hypoxic conditions [57].

Because some sensitizers appear to be sensitive to both ultrasound and light, the combination of SDT with PDT has been proposed to increase therapeutic efficacy [58] and/or reduce required dose of chemicals, and benefit from deeper penetration depth and superior focusing capability in tissues compared to laser irradiation. Several preclinical studies have reported benefit of sonophotodynamic therapy (SPDT), showing stronger therapeutic effect and reduction in the required dose of chemicals, that could help protect peripheral tissue from collateral damage [29].

FUS technology is progressing rapidly. The Exablate Neuro (INSIGHTEC, Israel) platform allows spatially precise delivery of ultrasound energy to intracranial targets under MRI guidance. The NaviFUS System (NAVIFUS, Taiwan) uses pre-treatment CT/MRI images and neuro-navigation tracking system to target ultrasound energy. However, these platforms are used mostly used in the setting of BBB opening, and their value for SDT remains to be determined.

Further pre-clinical studies exploring therapeutic and immunomodulation actions of SDT are warranted to harness the full therapeutic potential of this promising approach.

Conclusions

Abundant pre-clinical studies in glioma in vitro and animal models strongly suggest that SDT is a potent and promising therapeutic approach for gliomas that can induce anti-tumor effects via activation of apoptosis, anti-tumor immune response, and via other biological mechanisms. Safety of SDT for surrounding normal brain tissues has been reliably documented in pre-clinical studies of using small animal glioma models. A fuller understanding of optimal sonication parameters as well as doses and types of sonosensitizers for gliomas is imperative. Clinical experience with SDT for gliomas remains limited, but several clinical studies of SDT for brain tumor patients are now underway.