Ultrasound as a treatment modality for neurological diseases

Neurological disorders are a particular challenge for therapeutic intervention — ultrasound has emerged as a novel approach with a broad range of applications

With an ageing population, neurological disorders present an increasing challenge to our health care systems. Although antibodies are increasingly being explored for therapeutic intervention,1 the inefficiency of their uptake by the brain means that the estimated cost of a vaccine to treat neurodegenerative disorders such as Alzheimer disease (AD) will exceed US$25 000 per patient per year.2 Not only is this expected to challenge the health care systems of many countries, it also raises ethical issues associated with making these vaccines available to every patient.

Potential ways of overcoming these challenges include:

• Rendering the need for antibodies obsolete through the use of cheaper, non-biologic drugs that are currently in development.

• Improving antibodies by increasing their affinity or modifying their specificity.

• Facilitating antibody uptake by the brain through modification, such as by making them either bi-specific (one arm targeting transporters to get the antibody across the brain) or smaller (by generating single-chain fragment variables).3,4

• Proceeding with the available antibodies, but increasing their uptake by the brain and the relevant target cells.

Achieving the last of these approaches would not only significantly reduce costs but also overcome the major challenge of the blood–brain barrier (BBB), which characterises brain diseases in general. Although the BBB serves an important protective role, it precludes the use of all large-molecule neurotherapeutics and more than 98% of small-molecule drugs from entering the brain and engaging with their target.5 In this article, we argue that an “old” modality — focused ultrasound — can specifically address the challenges presented by the high cost of vaccines for AD, and brain disorders more generally, by transiently opening the BBB.6

Ultrasound is a mechanical pressure wave with a frequency above the range of human hearing. It is used in diagnostic imaging for various applications such as obstetrics and cardiology, and in the examination of the abdomen and musculoskeletal system. Its use is dictated by the fact that although ultrasound may be readily propagated through air (eg, parking sensors) or liquid–solid composites (eg, tissues), it is reflected at air–liquid/solid interfaces, which explains why it does not travel through the lungs. Similarly, because cancellous bone is extremely attenuating to ultrasound propagation, current neurological diagnostic imaging is performed over transcranial windows, where the skull is thin enough to allow adequate signal transduction. However, to date, such windows are only suitable for the imaging of structures in limited areas, such as the blood vessels basal to the brain. Although, broadly, the instrumentation for therapeutic ultrasound is similar to that of diagnostic applications, the former uses longer pulses of higher power ultrasound. Established applications include lithotripsy and physiotherapy.

In the past, ultrasound treatment through the skull was thought not to be feasible. However, it has recently been used successfully by titrating acoustic energy to temperatures high enough to ablate small brain volumes.7 In a study of 76 patients with moderate to severe essential tremor, focused ultrasound using a magnetic resonance imaging (MRI)-compatible transducer system was found to significantly improve tremor and total disability scores.7

Ultrasound may also be applied through its interaction with intravenously injected microbubbles to transiently open the BBB (obicodilation), which can be used to facilitate drug delivery to the brain.6 Microbubbles can be generated in the laboratory,8 although proprietary agents are available. Ultrasound causes microbubbles to undergo sustained oscillations of enlargement and contraction with the ultrasound cycle. This exerts pressure on the epithelial cells lining the cerebral blood vessels, which causes a disintegration of tight junction complexes, thereby facilitating transport of molecules in both directions across the BBB.9

Achieving obicodilation in humans presents a challenge, as the human skull distorts and absorbs up to 90% of the ultrasound energy. This currently necessitates the use of a large phased array device and requires a lengthy procedure in an MRI machine, which is not tolerated by all patients.10 Whereas extracranial ultrasound devices may be optimal for a single session of obicodilation in precise, deep-seated diseases of the basal ganglia, they are not suitable for repeatedly targeting diffuse and superficial pathologies such as those that occur in AD or tumours close to the skull.

Carpentier and colleagues reported the interim results of a phase 1/2a clinical trial in patients with recurrent glioblastomas, using an implantable transducer for repeated obicodilation before receiving systemic chemotherapy with carboplatin.11 The authors reported BBB disruption in the absence of any tissue damage. Although the ultrasound emissions were directed to areas with roles in motor function and sensory processes, patients did not present clinical symptoms or unusual sensations. Similarly, the 11 patients with epilepsy in the study remained stable. Interestingly, although the primary objective was not to assess efficacy, the region encompassed by the ultrasound field showed no detectable tumour progression on MRI in most of the patients who had confirmed BBB disruption. Although still at an early stage, the study also suggests potential applications for ultrasound beyond treating tumour tissue, such as in the treatment of AD.11

AD is characterised by deposition of the peptide amyloid β (Aβ) as amyloid plaques, and of tau protein as neurofibrillary tangles.12 Current therapeutic strategies aim to either prevent or slow down the build-up of toxic forms of Aβ and tau, or to facilitate their clearance.13 AD and related disorders such as frontotemporal dementia have been modelled in mice by expressing pathogenic forms of relevant genes found in familial cases of these diseases.12 Strategies that are currently being clinically evaluated, including vaccinations, have their foundation in animal work using transgenic mouse models such as the APP23 strain. By developing a novel scanning ultrasound (SUS) method that uses a motor-driven scanning mode to achieve obicodilation repeatedly over several weeks, in the absence of any therapeutic agents, a massive reduction of Aβ pathology was achieved in APP23 mice, and memory functions were restored to wild-type levels.8 This improvement occurred in the absence of any damage.

Aβ pathology is extracellular and SUS has been shown to activate microglia to digest this peptide.8 In contrast, the tau pathology that characterises AD and other tauopathies (such as frontotemporal dementia) is intracellular. In a follow-up study targeting tau pathology, it was demonstrated that this pathology could also be reduced and cognitive impairment ameliorated by applying repeated SUS treatments,14 although what causes these improvements mechanistically remains to be determined. This study also showed that SUS facilitated BBB passage of a therapeutic anti-tau antibody and its efficient intracellular uptake by neurons that harboured pathological tau.14 Because insoluble protein aggregates are a unifying feature of neurodegenerative diseases, SUS may therefore, by extension, also be a potential strategy for treating other conditions such as Parkinson disease.

In exploring the potential of ultrasound technology for therapeutic applications, safety is an obvious concern. Fortunately, several studies indicate that a therapeutic effect can be achieved in the absence of overt damage.6,15 A reason for this is that ultrasound is highly tunable and, when parameters are chosen carefully, BBB opening can be safely achieved, as shown not only in transgenic AD mouse models and wild-type mice8,16 but also in larger animals17 and humans.11 Importantly, ultrasound treatment for up to 20 months in non-human primates did not alter neurological functions, including visual, cognitive, motivational and motor processes.17

Drawing on the findings of Carpentier and colleagues in relation to brain tumours,11 the use of implantable devices could be an option for the clinical ultrasound treatment of patients with AD or other neurodegenerative diseases. However, being invasive, this may not be accepted by such patients, especially at an early stage of disease. Another possibility is to develop ways of safely and non-invasively delivering ultrasound at sufficient energies through the human skull.6 Obicodilation also greatly facilitates therapeutic uptake of, for example, an antibody targeting tau,14 or the anti-angiogenic antibody bevacizumab targeting gliomas, with brain uptake being increased up to 57-fold.18 For a disease of pandemic proportions such as AD, with a projected one million cases by 2050,19 obicodilation allows lower antibody doses, which would significantly reduce costs to the health care system.

We believe that there is a future for ultrasound-mediated treatment of neurological disorders, provided that we can solve the problem of the attenuation of ultrasound in the human skull and implement a safety monitoring protocol.