Early-Stage Alzheimer Disease Image-guided Therapy Clinical Trial Serendipity: Glymphatic Efflux and Prolonged Meningeal Venous Permeability Enhancement

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

Alexander L. Klibanov, PhD, is an associate professor with a primary appointment in the Cardiovascular Division, Department of Medicine, at University of Virginia School of Medicine. He also has joint appointments in the biomedical engineering and radiology departments at that institution. His research interests are in the areas of molecular imaging (mostly with ultrasound) and ultrasound-triggered drug and gene delivery. He is a chemist by initial training, and a graduate of Moscow University Chemistry Department.

Alzheimer disease (AD) has been in the crosshairs of the radiology community for years, from diagnostic imaging (1) to assessing the efficacy of therapeutic interventions (2). The idea to enhance the dissolution of amyloid-β plaque deposits in the brain of a person with AD is gaining momentum and recently reached the clinical trial stage. The proposed approach uses the transient opening of the blood-brain barrier (BBB) with a combination of therapeutic US and intravascular microbubbles known to facilitate material exchange between blood plasma and the interstitial space in the brain. On the basis of results from preclinical animal testing, the BBB opening may enhance fluid traffic in the brain, known as glymphatic efflux, which carries away poorly soluble AD plaque components. Initial clinical testing of this approach reported no adverse events and no worsening of cognitive scores at 3-month follow-up (3).

The focused ultrasound treatment begins after placing the patient in the transcranial focused ultrasound system, which is located inside the MRI scanner. Then, intravenous administration of a gadolinium-containing MRI contrast agent enables confirmation that the BBB is intact by showing no contrast material retention in the brain. Then, intravenous administration of gas-filled microbubbles occurs; these particles are known to radiologists and cardiologists as blood-pool US contrast agents for clinical use in diagnostic imaging. As soon as microbubbles enter the vascular bed, the transcranial focused ultrasound treatment begins under MRI guidance. The action of ultrasound pressure waves on the US contrast agent particles within the insonated tissue vasculature results in the cyclic compression and expansion of bubble gas microphase, with a severalfold excursion in volume. The use of US with low frequency (eg, 220 kHz) ensures minimal attenuation of US by the skull; the use of the US phase-array transducer with 1024 elements allows tight focusing of acoustic energy through the skull. Many of the vibrating bubbles are in physical contact with the endothelial lining of the vessel wall or are surrounded by capillary walls. This cyclic mechanical vibration results in transient loosening of tight junctions of the BBB endothelial lining. The focused ultrasound procedure takes seconds to complete for each of the treatment spots (focal zones of several millimeters). Subsequent intravenous injection of gadolinium-based MRI contrast material shows that the BBB in the US-treated areas of the brain has opened quickly, and MRI contrast material accumulates in the brain parenchyma. The intensity and duration of the acoustic pulses allows control of the BBB opening time: it may last as long as several hours (3), or it may be as short as several minutes (4). Repetition of the MRI procedure (eg, 24 hours later), at first, without contrast material, shows a lack of long-term retention of gadolinium from the prior contrast agent administration. At that time, intravenous administration of the fresh dose of gadolinium-containing contrast material shows no accumulation of incoming MRI contrast material in the insonated areas, as the BBB there has already closed.

An intriguing observation during the course of a larger ongoing BBB opening study in individuals with AD (clinical trial no. NCT03671889) (5) was made in several patients: Mehta et al (6) report an additional delayed and unexpected effect of the focused ultrasound and microbubble BBB opening. After focused ultrasound treatment of the hippocampal target site, as described previously, the perivenous space showed MRI contrast enhancement. The enhancement was visible not just immediately after the BBB opening procedure but also 24 hours later when the original insonated area had BBB closure already, so the brain parenchyma was contrast-negative. At that time point, the freshly administered gadolinium-containing contrast agent revealed perivenous MRI signal enhancement, which implied the lasting period of permeability of the blood-meningeal barrier (BMB) in the affected regions of postcapillary venules and meningeal veins. This conclusion was made from the observation that gadolinium-containing contrast material could exit the veins and reach the perivenous meningeal space in the areas away from sites of the original BBB opening insonation foci. This extended opening of the BMB remained for 24–48 hours. In all study participants, it spontaneously resolved within 1 week. A variety of MRI techniques showed the safety of the study and enabled confirmation of a lack of hemorrhage, infarct, edema, demyelination, or gliosis. There were no findings to suggest long-term parenchymal or extraparenchymal retention of gadolinium. Neurologic and neuropsychologic examinations of all participants did not show deterioration 12–19 months after completion of the study.

The observation of perivenous BMB remaining open at the 24-hour time point, or even longer, may direct us to some important phenomena that take place as a consequence of the prior off-site transient BBB opening. One of the questions that arises pertains to the potential mechanism of the prolonged BMB opening. Another is the respective set of consequences for the rate of clearance and the potential pathways of interstitial fluid drainage. The article by Mehta and colleagues offers an important insight into these processes in humans (6). This follows the earliest clinical trial assessment of glymphatic events after BBB opening in patients with AD, as first observed by Meng et al (7).

Mehta and colleagues suggest that during the BBB opening phase, the infiltration of serum components into the interstitial space takes place, inducing a secondary neuroimmunologic response, which leads to an extended period of the observed perivenous gadolinium-containing contrast agent accumulation (6).The arrival of these serum components into the interstitium, as they drain through the interstitial and glymphatic spaces, may be the reason for the observed extended BMB opening. The prolonged period of BMB opening might also relate to the presence of AD plaque amyloid-β or be linked to the need for clearance of amyloid-β.

In response to this report of perivenous BMB permeability enhancement, further steps are necessary to achieve a better understanding of the underlying mechanisms. First, a larger patient cohort is necessary. Second, Mehta and colleagues (6) mention macrophage activation and leukocyte trafficking to the affected veins (see Figure 6 in their article). Thus, the level of inflammation in those tissues also needs to be assessed with any of the available clinical imaging tools, such as PET/CT (with fluorodeoxyglucose enhancement), SPECT (indium 111 white blood cells), or MRI. Third, researchers may have to perform more animal testing to confirm that the extended period of perivenous BMB opening is observed in the preclinical setting. Preclinical animal studies will enable histologic confirmation of vessel permeability and leukocyte accumulation. Finally, systemic use of anti-inflammatory agents might also influence the level of venous inflammation and BMB opening. If gadolinium-containing MRI contrast material can enter a perivenous space, then other molecules may be able to do the same. The direct inflammatory response to focused ultrasound treatment is well known in the literature but has not yet been studied at the off-site perivascular venous area and BMB.

Acoustic parameters are another matter of interest. Current ultrasound wave pattern as used in the study by Mehta et al (6) have a rather short duration. They are literally seconds per spot altogether, with relatively short ultrasound pulses (2.6 msec each, spaced by 30.4 msec, with 10 cycles total). They also have peak negative acoustic pressure between 500 KPa and 1000 KPa, quite large amplitude values. Potential modification of the insonation parameters might include a variation of the pulse length, pulse repetition frequency, and overall treatment time, as in the study by Morse et al (4). This may allow control of the level of BBB opening and influx of the serum components into interstitial space, as suggested by Mehta and colleagues to be responsible for the postulated venous inflammation and prolonged BMB opening. Counterintuitively, use of a long interval between the pulses (eg, intermittent at 1 Hz or less) might even increase the level of bioeffects and of the BBB opening level because the bubbles might have sufficient time between US pulses to enter into the capillaries before US action is initiated again (8). The microbubble dose and the exact nature of the available US contrast material (the gas used, the shell, and particle size) might also be used to control the efficacy of BBB opening.

Another important issue is the brain lymphatic system. Since its discovery by Louveau et al in 2015 (9), the existence of a lymphatic system in the brain has been confirmed in numerous species, including imaging studies in nonhuman primates and humans (10). The relationship of glymphatic flow and the lymphatic system might play an important role in the phenomena observed by Mehta et al (6) and may require more detailed consideration.

Overall, the article by Mehta et al (6) opens up the set of interesting phenomena and tools to augment US and microbubble blood-brain barrier opening, with longer-term back-end blood-meningeal barrier permeability, which might become useful for Alzheimer disease therapy and beyond. For instance, it might be considered for the enhancement of blood levels of circulating tumor-related extracellular vesicles and tumor-specific nucleic acid, to be used for liquid biopsy diagnostics.