Bursting microbubbles: How nanobubble contrast agents can enable the future of medical ultrasound molecular imaging and image-guided therapy

Abstract

The field of medical ultrasound has undergone a significant evolution since the development of microbubbles as contrast agents. However, because of their size, microbubbles remain in the vasculature and therefore have limited clinical applications. Building a better—and smaller—bubble can expand the applications of contrast-enhanced ultrasound by allowing bubbles to extravasate from blood vessels—creating new opportunities. In this review, we summarize recent research on the formulation and use of nanobubbles (NBs) as imaging agents and as therapeutic vehicles. We discuss the ongoing debates in the field and reluctance to accepting NBs as an acoustically active construct and a potentially impactful clinical tool that can help shape the future of medical ultrasound. We hope that the overview of key experimental and theoretical findings in the NB field presented in this article provides a fundamental framework that will help clarify NB–ultrasound interactions and inspire engagement in the field.

Introduction

Biomedical ultrasound (US) imaging is a well-established clinical tool used for the diagnosis and management of a broad range of diseases. With applications ranging from fetal imaging to echocardiography, its diagnostic use is exceeded only by two-dimensional X-ray imaging [1]. US imaging uses sound waves above 20 kHz to construct images based on the interaction of the sound waves with the surrounding environment. The US waves, generated by a transducer (for imaging typically in the 1–20 MHz range), will be reflected, scattered, or absorbed by tissue and boundaries between tissues, such as skin and bone. The timing and strength of the returned echoes can then be used to determine the tissue location (assuming the speed of sound is known) and scattering characteristics with high spatial and temporal resolution. US images can be collected at over 20 frames per second, with a spatial resolution approaching 200 microns for higher frequency transducers. For a detailed overview of the physics behind US, please refer to the book by Richard Cobbold [2] (Fundamentals of Biomedical Ultrasound).

In addition to high temporal and spatial resolution of image acquisition, the popularity of US has been driven by notable advantages over X-ray computed tomography or magnetic resonance imaging (MRI), such as its well-established safety profile, low cost, portability, and broad accessibility. These benefits tend to outweigh some of the biggest challenges with US such as operator dependence and poor soft tissue contrast. The contrast and anatomical detail in US images has not been considered favorable compared with other modalities (e.g. MRI). One technology that has significantly improved US image quality and expanded US applications worldwide is the development of bubble-based contrast agents.

The advent of microbubble (MB) contrast agents has stimulated innovative strategies for cancer detection, therapy, and post-therapy monitoring, especially in the past two decades [3, 4∗∗, 5]. The use of gas as a contrast agent for US was first noted over 50 years ago [6]. Since then, several generations of this concept evolved from using unstabilized carbon dioxide (which was rapidly dissipated in the blood) to shell-stabilized constructs of more hydrophobic gasses. Current clinical MB formulations comprised a hydrophobic gas core (such as perfluoropropane—C₃F₈ or sulfur hexafluoride—SF₆) stabilized by a lipid, polymer, or protein shell. Bubbles generate contrast because of their interactions with the US waves. When placed in an acoustic field, bubbles oscillate in response to positive and negative pressure changes; these oscillations are distinct from the surrounding medium because of differences in the compressibility and density between the gas and blood and the viscoelastic properties of the shell, as described in the following [7,8]. MBs have United States Food and Drug Administration approval for enhancement of echocardiographs and detection of liver lesions [4] in adult and pediatric [9] populations and have been investigated widely in numerous other clinical diagnostic applications. An excellent review of clinical diagnostic use of MBs and their safety profile compared with other contrast media was recently published by Erlichman et al. [5]. MBs have also been examined in numerous therapeutic applications ranging from neuromodulation to sensitization of tumors to radiation therapy, and several approaches are currently in clinical trials [7,10∗, 11, 12]. Although a detailed summary of the current state of the art with MBs is beyond the scope of this review, we offer some additional thoughts about this field in the second part of this review.

MBs have had a significant impact in the field; however, MBs have some inherent challenges which may limit their utility in applications such as molecular imaging and image-guided therapy. The development of MB applications has been constrained by a relatively large particle diameter (∼1–8 μm), which confines the MB to the vasculature, limiting their targets to upregulated biomarkers only within the intravascular space [13, 14, 15, 16∗∗]. This reduces the potential for MBs to target biomarkers outside of the endothelium. MBs are typically produced at lower concentrations (∼10⁶ to 10⁸ bubbles/ml), which reduce their longevity both in ambient conditions and in vivo. Increasing the concentration is feasible, but leads to significant attenuation and shadowing, thus may not be practical to implement [17]. For clinical use, MBs are produced via on-site ‘activation’, which ranges from the hydration of a freeze-dried powder (Lumason®) to mechanical agitation of sealed vials containing lipid solutions and gas (Definity ®). This process yields highly heterogeneous bubble populations, which range from submicron to ∼2- to 10-micron bubble diameters. It is worth noting that, in some cases, the polydispersity of MBs could be an advantage because it allows the use of many different frequency transducers and enables imaging of many anatomical targets. However, the general recent community consensus is that monodisperse bubbles could provide increased sensitivity and specificity to detect lower MB concentrations, critical for molecular imaging applications. Likewise, a reduction in bubble diameter to the nanoscale can complement MB capabilities and provide a robust platform for expanding the biomedical applications of US. Thus, in this review, we will focus on work associated with stable, echogenic, uniform NBs.

Nanobubbles (also frequently referred to as sub-micron or nanoscale bubbles, abbreviated here as NBs) have been known to exist in commercially available MB formulations such as Definity® for nearly 20 years [18,19]. As mentioned previously, the MB formation, or “activation”, occurs via self-assembly of solubilized amphiphiles around a hydrophobic gas core. This relatively uncontrolled process results in a highly heterogeneous bubble population with a significant submicron component. However, as explained in detail in the following, the conventional theory would dictate that with decreasing bubble radius the backscatter from coated bubbles decreases significantly, and the resonant frequency increases rapidly. Thus, the submicron component of the acoustic activity was thought to be insignificant compared with that of the larger MBs (∼1–5 microns in diameter, which, by coincidence, has a resonant frequency close to the operating range of many commercial US imaging devices, about 3–5 MHz) and was largely dismissed. It was not until the intentional formulation of echogenic NBs by Oeffinger and Wheatleyet al in 2004 and the first demonstration of their activity in vivo in 2006 by the same group [20,21] that the idea of NBs and their application as contrast agents in biomedical US became recognized. Concurrently, reports of circulating NBs coalescing in extravascular space into imageable MBs were published by Rapoport et al. [22]. However, as a field, echogenic NBs did not see rapid adoption until approximately 2010, when a burst of publications reported on their use in cancer imaging [23∗∗, 24, 25, 26, 27, 28, 29, 30, 31]. In the last decade, research into the application of NBs in diagnostic US and US-mediated drug delivery has seen rapid growth as shown in Figure 1A. The growth can also be correlated with the more widespread adoption of bulk NBs by the physics and colloid communities [32].

Early experimental reports of echogenic NBs garnered little attention from the contrast-enhanced US (CEUS) and/or MB and biophysics communities because of a lack of concrete, overwhelming evidence demonstrating that the measured or visualized acoustic activity was indeed stemming from the submicron bubbles and not from larger (near 1 micron or above) bubbles which may have been contaminating the total bubble population.

Theoretically, the backscatter generated from the interaction of US with a particle is highly dependent on radius, with a 1 μm particle scattering 10⁶ times more than a 0.1 μm particle at the clinical frequency range (3–15 MHz, ignoring potential resonance). However, this assumes Rayleigh scattering independent of the effect of shell composition of the bubble oscillator such as shell viscosity, shell elasticity, and surface tension, as well as the acoustic pressure. At higher pressures and/or reduced surface tension of the bubble shell, the nonlinear activity of coated bubbles significantly contributes to its scattering strength. As described in detail later in this review, the current hypothesis is that the nonlinear activity due to the lipid shell coupled with the high particle concentration per imaging voxel is thought to be the major driving force behind the acoustic activity of NBs.

Recent progress in research using NBs for medical US imaging and drug delivery applications has been aided by the rapid developments in theoretical physics and experimental biophysics which provided strong evidence for the existence of bulk NBs as a viable, long-lasting construct [32]. Concurrently, several key experimental developments in imaging applications provided evidence that acoustically active NBs may be a viable contrast agent for US imaging. These were (1) the development of NB formulation methodologies with rigorous size control to reduce the size of the NBs to 200–400 nm and to remove all bubbles larger than 1 micron, which could be contaminating the solution and reducing control of acoustic response [28,30,33], (2) availability of instrumentation for quantitative, precise measurement of particle diameter, concentration, and, most importantly, buoyancy and total gas volume [34∗∗, 35, 36], (3) manipulation of the NB shell to reduce interfacial tension and increase deformability using surfactants such as Tween 80 and Pluronic and edge activators such as propylene glycol, (4) the theoretical modeling of bubble acoustic response at lower surface tension and higher acoustic pressures [37,38], and (5) rigorously controlled in vitro and in vivo demonstrations of NB acoustic activity, extravasation, and imaging characteristics compared with MBs [29,39, 40, 41]. In the following sections, we detail these developments in the NB formulation and characterization. We then briefly discuss a theoretical framework for understanding the acoustic activity of NBs, provide an overview of the in vivo imaging and therapeutic applications, and finally discuss emerging future directions in this exciting, still-developing field.