Moving toward Noninvasive, Focused Ultrasound Therapeutic Delivery of Drugs in the Brain: Prolonged Opening of Blood-Brain Barrier May Not Be Needed

See also the article by Morse et al in this issue.

Introduction

Delivery of therapeutic agents across the blood-brain barrier (BBB), especially with spatial selection of the target, has not yet reached mainstream medical care. It requires complex and invasive pharmacologic interventions, such as intra-arterial administration of hypertonic mannitol. The importance of trans-BBB delivery cannot be underestimated. The ability to allow drug delivery into the brain (including nucleic acid and gene delivery vehicles), preferably to particular areas, will have a profound effect on the practices of modern medicine. It will aid in treatment of brain tumors, Alzheimer disease, Parkinson disease, and many other conditions and diseases. Image-guided, focused ultrasound is on the verge of providing transient BBB opening, and the study by Morse et al presented in this issue of Radiology (1) proposes an unexpected yet very simple approach modification that may aid in achieving the desired therapeutic goals.

Focused ultrasound ablative treatments in the brain have already reached the clinic. High-power, localized ultrasound hyperthermia under MRI guidance and temperature control is approved for treatment of essential tremor (2) and tremor-dominant Parkinson disease (and is being investigated for several other applications). Image-guided therapy systems capable of focusing ultrasound through the intact skull are now in practical use in many image-guided therapy centers worldwide. In addition to thermal ablation, these focused ultrasound systems may provide localized drug delivery using ultrasound power levels that are orders of magnitude lower than those used in ablative hyperthermia.

As a tool for transient BBB opening, focused ultrasound typically results in increased vascular permeability that persists for up to several hours following insonations. Pioneered by Hynynen et al, this procedure includes intravascular administration of gas-filled biocompatible microbubbles, routinely used as blood pool contrast ultrasound imaging agents in the clinic (3). These microbubbles compress and expand upon the action of ultrasound pressure waves. Injected intravenously, the microbubbles, which are smaller than blood cells, pass through all the vasculature, including the vessels and capillaries in the area of the brain where ultrasound is focused on the target tissue. Vascular endothelium is mechanically massaged by the expanding and compressing microbubbles that pass through the vessel lumen. In response to this localized mechanical action, tight junctions between endothelial cells are altered so the BBB stops being an obstacle to drug transport across the vessel wall and therapeutic agents can be delivered to the brain. This mechanical stimulation also results in active transport across the BBB and a reduction in drug efflux.

In animal models, success in drug delivery has been exciting. In a rodent model of glioma, 40% of animals receiving BBB opening in combination with intravascular microbubbles and long-circulating doxorubicin liposomes were essentially cured (ie, disease-free for 140 days) (4). The drug (including nanoparticle drug) was likely delivered to the areas where glioma cells were present and active, yet the vasculature there was not sufficiently leaky to allow drug delivery without ultrasound. Delivery of therapeutics to treat Alzheimer disease in a murine model demonstrated reduction of amyloid plaque (5), and clinical testing of the BBB opening has also been initiated (6). In these trials, focused ultrasound was mostly performed under MRI guidance and clearly demonstrated transfer of MRI contrast agent to the brain. Delivery of antitumor agents has also reached clinical trial stage. Early results look promising (7).

There is concern that while the BBB is open, other materials besides the therapeutic agent from the bloodstream may enter the brain. These materials (eg, highly abundant serum albumin) may exert some levels of neurotoxicity. Therefore, it may be beneficial to assure effective trans-BBB delivery yet minimize the exposure of the brain to undesirable blood components. In the traditional process of scientific invention, researchers try to find a compromise in which positive results are not overshadowed by undesirable side effects (eg, lower acoustic power and longer pulse length). An alternative, disruptive, and nonlogical methodology from decades ago (8) applies nontraditional designs (known in the USA as Technical Inventive Problem Solving), including a suggestion to “design the system in an opposite direction.” In the study by Morse et al (1), they have done just that, proposing an alternative version of focused ultrasound application to microbubbles that allows trans-BBB delivery while reducing delivery of undesirable materials (eg, serum albumin).

Instead of using prolonged acoustic pulses at moderate intensity, as others do, Morse and colleagues proposed using sequences of short pulses in rapid succession. Instead of traditional long 10 000-cycle ultrasound pulses delivered at 0.5-Hz intervals, rapid short-pulse treatment consists of short five-cycle ultrasound pulses with 1.25-KHz pulse repetition frequency and 10-msec burst length. In combination with intravascular microbubbles, rapid short-pulse ultrasound allows the transfer of materials across the BBB without opening the barrier for more than just a few minutes in a murine model. Morse et al (1) reported that a model drug (3 kDa fluorescent dextran) is successfully transferred to the brain of experimental animals. Microscopy-assessed delivery pattern is more uniform for rapid short-pulse ultrasound than for the prolonged-pulse traditional ultrasound. Most importantly, by using rapid short-pulse ultrasound, albumin levels in the brain assessed by histology are 3.4-fold lower than with the traditional long-pulse ultrasound. As long-pulse ultrasound delivers two orders of magnitude more energy into the tissue, it is not surprising to observe some level of tissue damage. This damage is not detected in brain tissues treated with rapid short-pulse ultrasound.

The exact mechanisms of the events observed with rapid short-pulse ultrasound–based BBB opening are not yet fully understood. More experiments will be necessary to understand them. The kinetic curve for dye accumulation in the brain with rapid short-pulse ultrasound would definitely benefit from more data points between 0–10 minutes, which is a limitation of the study by Morse et al (1). Depending on the results of such a study, one might even find that the BBB remains open for less than 10 minutes following ultrasound treatment. Quantitative assessment of the amount of the delivered drug, beyond just the fluorescence microscopy of histology samples, would also be very helpful to assess if the therapeutic index would be improved with rapid short-pulse ultrasound. The size and chemical nature of molecules or particles capable of crossing the BBB following rapid short-pulse insonation should also be investigated to see if results differ from those observed in the traditional insonation schemes. Other next steps should include assessment of rapid short-pulse ultrasound in influencing edema levels (BBB opening may lead to enhanced water transfer into the tissue) as well as the levels of sterile inflammation (9). The nature and specific parameters (particle concentration and size distribution, gas and shell type) of the microbubbles used in this intervention might also be important.

As all the devices that deliver therapeutic ultrasound should be capable of adjusting pulse length and pulse repetition frequency from the control software, translation of rapid short-pulse ultrasound into the clinical trial domain should not be too complicated. The overall usefulness of the proposed rapid short-pulse ultrasound treatment strategy must be carefully considered in relation to the nature of the provided therapy (eg, neuromodulation vs Alzheimer plaque inhibitors vs brain tumor, with large or small treated volume) as well as to the importance and comparative levels of neurotoxicity due to albumin extravasation. The desired type of drug is also important: is it long circulating or rapidly excreted? Critical-toxicity organs for the drug, in addition to the final goal of curative therapeutic efficacy and improvement of therapeutic index, should also be considered. The short-pulse ultrasound approach is already used in drug delivery research. We find that a single 10-cycle, microsecond-long ultrasound pulse is sufficient to convert to a microbubble a superheated perfluorocarbon nanodroplet attached to a loaded red blood cell. Microbubble expansion leads to membrane puncture and immediate localized payload release (10). A combination of this technique with rapid short-pulse ultrasound may further improve drug delivery efficacy.

Design of the proper acoustic pulses, including their intensity, polarity, frequency, pulse length, repetition frequency, and duty cycle will continue to play a decisive role in the study design of image-guided ultrasound intervention and implementation of this technique for therapeutic delivery across the blood-brain barrier.