Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2016 Dec 7.
Published in final edited form as: Phys Med Biol. 2015 Nov 12;60(23):9079–9094. doi: 10.1088/0031-9155/60/23/9079

Acoustic Cavitation-Based Monitoring of the Reversibility and Permeability of Ultrasound-Induced Blood-Brain Barrier Opening

Tao Sun 1,*, Gesthimani Samiotaki 1, Shutao Wang 1, Camilo Acosta 1, Cherry C Chen 1, Elisa E Konofagou 1,2
PMCID: PMC4668271  NIHMSID: NIHMS739617  PMID: 26562661

Abstract

Cavitation events seeded by microbubbles have been previously reported to be associated with MR- or fluorescent-contrast enhancement after focused ultrasound (FUS)-induced blood-brain barrier (BBB) opening. However, it is still unknown whether bubble activity can be correlated with the reversibility (the duration of opening and the likelihood of safe reinstatement) and the permeability of opened BBB, which is critical for the clinical translation of using passive cavitation detection to monitor, predict and control the opening. In this study, the dependence of acoustic cavitation on the BBB opening duration, permeability coefficient and histological damage occurrence were thus investigated. Transcranial pulsed FUS at 1.5 MHz in the presence of systemically circulating microbubbles was applied in the mouse hippocampi (n = 60). The stable and inertial cavitation activities were monitored during sonication. Contrast-enhanced MRI was performed immediately after sonication and every 24 h up to 6 days thereafter, to assess BBB opening, brain tissue permeability and potential edema. Histological evaluations were used to assess the occurrence of neurovascular damages. It was found that stable cavitation was well correlated with: 1) the duration of the BBB opening (r2 = 0.77); 2) the permeability of the opened BBB (r2 = 0.82); 3) the likelihood of safe opening (P < 0.05, safe opening compared to cases of damage; P < 0.0001, no opening compared to safe opening). The inertial cavitation dose was correlated with the resulting BBB permeability (r2 = 0.72). Stable cavitation was found to be more reliable than inertial cavitation at assessing the BBB opening within the pressure range used in this study. This study demonstrates that the stable cavitation response during BBB opening holds promise for predicting and controlling the restoration and pharmacokinetics of FUS-opened BBB. The stable cavitation response therefore showed great promise in predicting the BBB opening duration, enabling thus control of opening according to the drug circulation time. In addition, avoiding adverse effects in the brain and assessing the pharmacokinetics of the compounds delivered can also be achieved by monitoring and controlling the stable cavitation emissions.

1. Introduction

Focused ultrasound (FUS) with systemically administered microbubbles has been shown to transiently induce localized and reversible blood-brain barrier (BBB) opening for drug delivery to the brain (Hynynen and McDannold 2001, Burgess and Hynynen 2013). BBB prevents permeation of molecules greater than 400 Da from the vasculature to the brain parenchyma, rendering many potent drugs ineffective (Pardridge 2005). Induced by FUS, oscillating microbubbles enhance local mechanical effects to the targeted vasculature, facilitating the transient opening of the BBB in the regions of interest (Aryal et al 2014) and the improvements of other brain drug delivery paradigms (Wang et al 2014b, Chen et al 2014). Several therapeutic agents have thus been successfully delivered including antibodies (Kinoshita et al 2006, Raymond et al 2008), chemotherapeutic agents (Park et al 2012, Treat et al 2012, Liu et al 2010), viral vectors (Thévenot et al 2012, Hsu et al 2013, Wang et al 2015), and neurotrophic factors (Wang et al 2012, Samiotaki et al 2015).

One key element to be established prior to clinical translation is to develop a real-time feedback indicator to control and predict the physiological changes after BBB opening. Contrast-enhanced MRI has been employed to detect the opening (Hynynen and McDannold 2001, Tung et al 2011a, McDannold et al 2012), assess any potential damage (McDannold et al 2012, Tung et al 2011b), map the permeability (Vlachos et al 2010, Park et al 2012) and monitor BBB closing (Samiotaki et al 2012). However, MRI-based evaluation of these physiological parameters, especially the reversibility and permeability, cannot be implemented with high temporal efficiency.

On the other hand, it has been well accepted that the microbubble cavitation activity plays a critical role in FUS-induced BBB opening (Tung et al 2011b). Stable volumetric microbubble oscillation (stable cavitation) and/or transient bubble collapse (inertial cavitation) due to associated oscillations have been previously investigated (McDannold et al 2006, Tung et al 2010, 2011a, 2011b, O'Reilly and Hynynen 2012). Transcranial passive cavitation detection (PCD) serves as a powerful tool in detecting microbubble activity in real time by characterizing their strength, mode and location through the scattering signature (Arvanitis et al 2013).

The feasibility of correlating acoustic cavitation emissions with BBB opening assessments has previously been reported (McDannold et al 2006, Tung et al 2011b, O'Reilly and Hynynen 2012, Arvanitis et al 2012, Fan et al 2014). Harmonic (Arvanitis et al 2012) and ultraharmonic (O'Reilly and Hynynen 2012) emissions have been proposed as indicator candidates of opening outcomes. However, most of the previous studies used MRI contrast techniques (Arvanitis et al 2012, O'Reilly and Hynynen 2012, Tung et al 2011b) to assess the opening volume. The reversibility and permeability of the opened BBB may be more vital in clinical applications. The reversibility was depicted by the duration of BBB opening and the likelihood of safe opening. Predicting and controlling the reversibility (the duration of opening and the likelihood of safe reinstatement) are important because 1) the duration of the BBB opening is critical for drug delivery because different drugs have distinct circulation times; 2) developing a real-time indicator to assess the likelihood of safe opening during FUS exposure is vital for to minimize collateral damage. The permeability coefficient Ktrans represents the pharmacokinetics of the MR tracer through the opened BBB, and it was shown to be associated with BBB opening volume and closing timelines (Vlachos et al 2011, Samiotaki et al 2012). This coefficient was also found to be correlated with the payload of a chemotherapy agent (doxorubixin) (Park et al 2012), indicating that the Ktrans has the potential to indicate drug delivery to the brain after BBB opening.

In the current study, we investigate the role of acoustic cavitation in the restoration and pharmacokinetics of FUS-opened BBB, aiming at evaluating the cavitation monitoring in its assessments of the permeability, duration, and safety of BBB opening. Microbubble activity in response to ultrasound are dependent upon their initial dimensions and acoustic pressures (Ferrara et al 2007, Qin et al 2009, Tung et al 2011b). Here, monodisperse microbubbles with three different diameter ranges (1-2, 4-5, and 6-8 μm) in acoustic fields at different pressure amplitudes were thus used to induce variant cavitation activities. Cavitation activity was monitored during FUS treatment and quantitatively correlated with the time for BBB to close, the permeability coefficient and the histological damage occurrence.

2. Materials and Methods

Animals

All animal studies presented herein were approved by the Columbia University Institutional Animal Care and Use Committee. A total of 60 male mice (Weight: 24.21±1.72g; C57BL/6, Harlan Laboratories, Indianapolis, IN) were anesthetized with a mixture of oxygen (0.8 L/min at 1.0 bar, 21°C) and 1.5 - 2.0% vaporized isoflurane (SurgiVet, Smiths Medical PM, Dublin, OH) during treatment. The fur on the scalp was removed by an electric trimmer and depilatory cream in order to minimize acoustic impedance mismatch.

Microbubble preparation

Lipid-shelled monodisperse microbubbles with three different diameter ranges (1-2, 4-5, and 6-8 μm) were prepared in-house and size-isolated using the differential centrifugation method according to a previously published protocol (Feshitan et al 2009). A Multisizer III particle counter (Beckman Coulter, Opa Locka, FL) with a 30-μm aperture was used to measure the size distribution and concentration. The concentration of microbubble solution was diluted using phosphate-buffered saline (PBS) to 8×108 numbers/ml prior to intravenous administration through the tail vein.

FUS

A single-element, spherical-segment FUS transducer (center frequency: 1.5 MHz; focal length: 60 mm, diameter: 60 mm; Imasonic SAS, Voray-sur-l'Ognon, France) was driven by a function generator (33220A, Agilent, Palo Alto, CA). A 50-dB power amplifier (325LA, Electronic Navigation Industries, Rochester, NY) was used to amplify the transmitted waves. Microbubbles in the systematic circulation interacted with the transmitted ultrasound, and the re-radiated waves from microbubbles were received by a single-element FUS transducer (center frequency: 10 MHz; focal length: 60 mm; diameter: 22.4 mm; Olympus NDT, Waltham, MA). This cavitation detector was positioned through a central hole of the FUS transducer so that their foci overlapped. Cavitation signal acquisition was controlled by a pulser-receiver system (Olympus NDT, Waltham, MA) that was connected to a digitizer (Gage Applied Technologies Inc., Lachine, QC, Canada).

A 5-s sonication with a pulse repetition frequency (PRF) of 5 Hz and pulse length (PL) of 100 cycles was carried out prior to microbubble administration in order to obtain an acoustic response baseline, which was used to normalize the cavitation dose quantification. Upon completion of the microbubble injection, 60-s FUS with the same PRF and PL was applied to the targeted area. The peak rarefactional pressures (PRPs) in situ were estimated according to the calibration described in our previous reports (Wang et al 2014a). The right hippocampus was targeted following a grid-guided targeting procedure (Choi et al 2007) while the left served as the control. The FUS focus was placed 3 mm beneath the skull so that the focal region overlapped with the hippocampus (Tung et al 2011b).

Cavitation signal acquisition and analysis

The cavitation dose was calculated based on the integrated area under the curve of temporal power variance of the cavitation signals monitored during FUS exposure. Two cavitation parameters that characterize the cavitation behaviors were calculated: stable cavitation dose (SCD) and inertial cavitation dose (ICD). The SCD was an indicator of the strength of stable oscillation. Each PCD-recorded pulse was transformed into a power spectrum using a fast Fourier transform at a sampling frequency of 50 MHz. Harmonic and ultraharmonic components were identified as the peak value within a 300-kHz bandwidth around each harmonic (nfc, n = 1, 2, 3... fc = 1.5 MHz) and a 100-kHz bandwidth around each ultraharmonic (nfc/2, n = 3, 5, 7...) frequency in the range between 4 and 12 MHz. The SCD was then quantified by integrating the amplitude of the harmonic and ultraharmonic responses over the entire sonication duration. The ICD, which represented the scale of microbubble collapsing, was quantified based on the root mean square (RMS) of the broadband emission amplitude within specific in-harmonic windows (bandwidth: 400 kHz) between 4 and 12 MHz. These RMS amplitudes were then integrated over the same duration. The net cavitation emissions from microbubbles could then be determined by subtracting the baseline measured prior to bubble administration.

It should be noted that the ultraharmonic responses may not be detected due to the high level of broadband noise, the ambient pressure (Sun et al 2012) and/or the presence of the skull. In order to ensure that the ultraharmonic signal can be distinguished from the variation of the background noise, i.e., to increase the signal-to-noise ratio (SNR), the following algorithm was used in the quantification. We first calculated the mean plus three times of the standard deviation of the signals in each in-harmonic range that was used in the ICD quantification. The calculated value was subsequently defined as the noise level of each ultraharmonic frequency bin prior to this in-harmonic range. The ultraharmonic cavitation event was considered observed only if the ultraharmonic value increased above the aforementioned noise level.

Acoustic Parameters

Main Study

In the main study, we chose 0.30 – 0.60 MPa FUS with microbubbles with three different size distributions to induce BBB opening based on previous reports using the same FUS experimental setup (Tung et al 2011b).

ICD-based safety assessments

A previous study (Chen and Konofagou 2014) indicated that delivering large compounds (> 500 kDa) would be associated with the detection of strong inertial cavitation. In order to test the ICD-based safety assessments, we added two more groups with higher-pressure exposures (0.75 and 0.90 MPa) using 4-5 μm microbubbles to induce more inertial cavitation events.

MRI

All MR scans were performed using a 9.4-T MR system (DRX400, Bruker Medical, Billerica, MA). After each sonication and allowing 15 min for animal transport, each mouse was placed inside the vertical bore, while being anesthetized at 30-40 breaths/min with isoflurane gas (1% - 2%). Pre-contrast T2 weighted, Dynamic contrast-enhanced MRI (DCE-MRI) and T1-weighted MR sequences were acquired on Day 0 and every 24 h up to 6 days. T2-weighted MRI was performed to detect edema using T2-RARE sequence (TR/TE: 3300/10.9 ms, spatial resolution: 86 × 86 μm2, slice thickness: 500 μm). DCE-MRI was performed using a 2-D FLASH T1-weighted sequence (TR/TE = 230/2.9 ms, spatial resolution: 130 × 130 μm2, slice thickness: 600 μm). During the third acquisition of the dynamic sequence, the animal was injected with a 0.30-mL bolus of gadodiamide (Omniscan, GE Healthcare, Princeton, NJ) intraperitoneally (IP) as described elsewhere (Vlachos et al 2010). A T1-weighted 2D FLASH acquisition sequence (TR/TE: 230/3.3 ms, spatial resolution: 100 × 100 μm2, slice thickness: 400 μm) was acquired 40 min after IP administration of gadodiamide.

The volume of opening/edema and permeability were analyzed following the approaches reported by our group (Samiotaki et al 2012, Samiotaki and Konofagou 2013, Wang et al 2014a). Briefly, the volume of opening was quantified with volumetric measurements of hyper-intense voxels in post-contrast T1-weighted MR images. Similarly, edema volume was calculated based on hyper-intense voxels in pre-contrast T2-weighted MR images. The permeability was assessed using the general kinetic model (GKM) on the basis of DCE-MR images (Vlachos et al 2010). The arterial input function in GKM was determined by averaging the gadodiamide concentration changes in the internal carotid artery from the entire cohort of wild-type mice. Ktrans, the tracer transfer rate from blood plasma to extracellular extravascular space, served as the quantitative indicator of permeability.

Histology

All mice were euthanized and transcardially perfused with 30 mL PBS followed by 60 mL 4% paraformaldehyde on Day 7. The heads were harvested and then soaked in paraformaldehyde for 24 h, followed by skull removal and fixation again in 4% paraformaldehyde for 6 days. The brains were paraffin embedded and sectioned at 6 μm in thickness. Hematoxylin and eosin (H&E) staining was performed for histological analysis. For each brain, there were 32 sections prepared, with 180 μm spacing between planes.

Statistical Analysis

An unpaired two-tailed Student's t-test was performed to compare the permeability coefficients between the treated and the sham group on each day. The cavitation dependence on the likelihood of no opening, safe opening or opening with damage was examined using one-way ANOVA, followed by a post-hoc Tukey's honest significant difference test. Error bars in this study represent standard deviations and values of P < 0.05 were considered significant. The BBB was considered “closed” when the opening volume fell below mean + three times of standard deviations of the sham group for each day. All statistical analyses were performed using GraphPad Prism 6 (La Jolla, CA).

3. Results

BBB Opening

The opening volume and permeability of the BBB were assessed using MRI, as shown in Figure 1. Three-dimensional T1-weighted MRI confirmed the opening and the corresponding Ktrans maps provided the pharmacokinetic evaluations of different ultrasound parameters.

Figure 1. BBB Opening Evaluation.

Figure 1

Open in a new tab

Typical 3D reconstructed T1-weighted MR images (shown in left for each group) and Ktrans maps (shown in right for each group) of the central slice. The BBB opening assessments were microbubble-size and acoustic pressure dependent.

BBB opening were induced by 0.30 – 0.60 MPa FUS with monodispersed microbubbles within three different diameter ranges (Table 1). As shown in Figure 1 and Table 1, opening volumes and Ktrans coefficients were augmented as the FUS pressure and microbubble size increased. The bubble size effects were evident (Table 1). Three opening outcomes listed in Table 1 using 1-2 μm microbubbles were found to be dramatically different from those of larger bubbles. Interestingly, the opening can be reinstated within a specific BBB opening duration in the groups of ‘0.60 MPa, 1-2 μm’ and ‘0.30 MPa, 4-5 μm’ without any significant inter-animal variability, i.e., openings in Group ‘0.60 MPa, 1-2 μm’ all closed on Day 1 and openings in Group ‘0.30 MPa, 4-5 μm’ all closed on Day 3. It indicated that these two parameter sets could potentially be used to control the BBB recovery time reliably.

Table 1.

Quantified opening volumes, Ktrans coefficients and duration of BBB opening

Bubble Diameter (μm) Pressure
0.30 MPa 0.45 MPa 0.60 MPa
n Opening Volume (mm3) Ktrans (min-1) Duration of Opening (Day#) n Opening Volume (mm3) Ktrans (min-1) Duration of Opening (Day#) n Opening Volume (mm3) Ktrans (min-1) Duration of Opening (Day#)
1-2 μm 5 1.36 ± 2.72 0.004±0.010 0.2 ± 0.40 4 1.67±1.74 0.006±0.002 0.6 ± 0.49 5 5.67±3.75 0.011±0.006 1.0 ± 0
4-5 μm 4 11.12±0.73 0.012±0.002 3.0 ± 0 7 31.10±5.37 0.020±0.004 4.8 ± 0.40 5 45.33±3.22 0.026±0.006 4.4 ± 0.49
6-8 μm 5 15.36±2.61 0.016±0.007 3.2 ± 0.75 6 35.42±4.62 0.025±0.004 3.8 ± 0.75 5 44.65±8.90 0.031±0.005 4.7 ± 0.47
Open in a new tab

Cavitation Dependence of Time Required to Close

To investigate whether the cavitation dose can be a promising indicator to predict the reversibility of BBB, the temporal variation of SCD and ICD was obtained (Figure 2). The SCD showed good agreement (r2 = 0.77) with closing time, suggesting that the SCD could potentially be used in predicting reversibility. However, the ICD-based prediction was not shown as reliable (r2 = 0.27). In addition, the duration of the opened BBB was found to be bubble-size dependent (Figure 2). Smaller microbubbles (1-2 μm) induced shorter openings within 1 day, while larger bubbles (4-5 and 6-8 μm) produced longer openings up to 2-6 days. Consistent with the bubble-size effects in the opening assessments, stable cavitation activities of 1-2 μm microbubbles can be differentiated from those of larger bubbles (P < 0.0001) with a threshold level at around 50 dB (Figure 2a). The ICD of 1-2 μm microbubbles were also found to be statistically different (Figure 2b; P < 0.05) compared to that of larger bubbles.

Figure 2. Measured (a) stable and (b) inertial cavitation doses on the sonication day as a function of the duration of BBB opening.

Figure 2

Open in a new tab

Number of mice: 41 (excluding the cases with LO, the sham group and one off-target case); 1-2, 4-5 and 6-8 μm groups were labeled in red, blue and black, respectively; 0.30, 0.45 and 0.60 MPa groups were marked by circle, square and upper triangle, respectively.

Permeability and Its Cavitation Relevance

The permeability closing timeline was characterized both qualitatively and quantitatively in Figure 3. Four mice were found to have 6-day-long openings based on T1-weighted MRI assessment on Day 6. We define those four animals as Long-term Opening (LO) cases, while others that closed within Day 1-6 are Reversible Opening (RO) cases. LO cases showed higher permeability coefficients than that of RO cases for each day of the experiment. In addition, Ktrans of LO was found to be significantly higher compared to that of the sham till Day 4. However, the Ktrans of the RO decreased to the same level as that of sham after Day 1, indicating that the “open” state of LO lasted longer than that of RO. Interestingly, the Ktrans closing curves of RO and LO (Figure 3) underwent different temporal variations. Ktrans in the LO cases reached a peak on Day 1 and then gradually declined while Ktrans in the RO cases decreased monotonically.

Figure 3. Permeability (Ktrans) closing timeline.

Figure 3

Open in a new tab

Typical permeability maps and quantitative Ktrans of (a) 4-5 μm and (b) 6-8 μm groups. Number of mice: 48 (excluding one off-target case); Sham, 0.30, 0.45 and 0.60 MPa groups were labeled in grey, red, blue and black, respectively; RO: reversible opening; LO: long-term opening. All statistical analyses were t-tests compared with the sham group for each day. *: P < .05; **: P < .01. Error bar represents mean ± std.

Having established the relationship between cavitation responses and reversibility of the opened BBB, we further investigated the relationship between cavitation dose and permeability coefficient. Both SCD (r2 = 0.82) and ICD (r2 = 0.73) were found to be linearly correlated with the Ktrans on the sonication day (Figure 4). Ktrans of LO cases and their corresponding SCDs held the largest values among others.

Figure 4. (a) Stable and (b) inertial cavitation dose dependencies on the permeability coefficients on the sonication day.

Figure 4

Open in a new tab

Number of mice: 45 (excluding the sham group and one off-target case); 1-2, 4-5 and 6-8 μm groups were labeled in red, blue and black, respectively; 0.30, 0.45 and 0.60 MPa groups were marked by circle, square and upper triangle, respectively. LO subjects were labeled in hollow circles.

Safety Analysis

To identify edema and neurovascular damage after ultrasound exposure, we performed pre-contrast T2-weighted MRI and H&E histological analyses, respectively. Damage, including red blood cell extravasations and dark neurons, was found in four mice (excluding the one under off-targeting exposure), which were also the four LO cases. Figure 5 compared the opening volumes, T2 enhancements and H&E staining results of the RO and LO cases under the same FUS exposure. Contrary to the RO subjects, the BBB of the LO subjects remained open with edema on Day 5.

Figure 5. Safety assessments.

Figure 5

Open in a new tab

T1-weighted MR images on Day 0 (T1-Day0), T1-weighted and T2-weighted MR images on Day 5 (T1-Day 5 and T2-Day 5, respectively), histologic evaluations (H&E) of the control and sonicated sides were compared between RO and LO cases under the same ultrasound parameters (0.60 MPa, 6-8 μm microbubbles). Scale bars = 200 μm; scale bar in the inset = 50 μm.

All animals in the main study excluding the sham group were then categorized into three groups according to the H&E evaluations (Figure 6). Cavitation responses among the three groups were compared using one-way ANOVA. The ANOVA showed significance and the post-hoc multi-comparison revealed the significantly different SCD (P < 0.05) compared across ‘no opening’, ‘safe opening’ and ‘opening with damage’ cases. Note that the four cases with histological damage are the same cases with edema on Day 5 so that we did not differentiate the edema and histological damage. These findings indicate that the SCD may also be used as a measure of likelihood for safe opening, which further demonstrated the reliability of SCD-based prediction of opening outcomes.

Figure 6. (a) Stable and (b) inertial cavitation doses compared across no opening, safe opening and opening with damage cases.

Figure 6

Open in a new tab

Number of mice: 46 (pressure range: 0.30 – 0.60 MPa); *: P < .05; ***: P < .001; ****: P < .0001; ns: no significance. Error bar represents mean ± std.

Finally, in order to further investigate the ICD-based prediction of the safety assessments, we did an extensive study including two more groups with higher-pressure exposures (0.75 and 0.90 MPa) using 4-5 μm microbubbles. Figure 7 showed the cavitation comparisons across ‘safe openings’, ‘opening with edema’ and ‘opening with histological damage’ (Note that all the subjects were successfully opened by 0.30 – 0.90 MPa FUS with 4-5 μm microbubbles). These results revealed the significantly different SCD (P < 0.01) and ICD (P < 0.01) compared across safe opening, opening with edema and opening with histological damage subjects. The ICD-based prediction of safety could thus be accomplished after FUS exceeds certain pressure (i.e., stronger inertial cavitation activities are induced).

Figure 7. (a) Stable and (b) inertial cavitation doses compared across safe opening, opening with edema and opening with histological damage cases.

Figure 7

Open in a new tab

Number of mice: 29 (4-5 μm microbubbles, pressure range: 0.30 – 0.90 MPa); **: P < .01; ***: P < .001; ns: no significance. Error bar represents mean ± std.

4. Discussion

In this study, the feasibility of assessing the restoration and pharmacokinetics of FUS-opened BBB based on monitoring cavitation signals was investigated. Our results revealed that stable cavitation signals from oscillating microbubbles can be utilized to potentially predict and control the duration of BBB opening, the likelihood of safe opening and the permeability of the opened BBB. A real-time feedback method based on monitoring SCD could then be used to control and predict the physiological changes during FUS exposure.

Cavitation dependence and mechanisms

Our results indicated that, at low to moderate pressure levels (0.30 – 0.60 MPa) in vivo, SCD showed good correlations with the reinstatement time, likelihood of safe opening and Ktrans; while ICD was only well correlated with Ktrans. Possible causes may be attributed to cavitational mechanisms. Sustained stable cavitation could thus be the dominant physical mechanism in opening the BBB at low to moderate pressure levels, at which FUS exposures are normally kept to avoid damage to the brain. The inertial cavitation threshold is higher than the occurrence of sustained stable cavitation (Bader and Holland 2013). Within the range of 0.30 – 0.60 MPa, stable cavitating microbubbles cause physiological effects, while only a small number of nuclei undergo inertial cavitation (Tung et al 2011b). Stable cavitation may thus dominate the cavitation activities to induce BBB opening.

Additionally, SCD may also be used to gauge the likelihood for safe opening. This finding is in agreement with the numerical investigation by Bader et al. (Bader and Holland 2013). The threshold of inertial cavitation predicted by the Mechanical Index (MI,PRP∕fc), which is the standard indicator for adverse bio-effects, is MI = 0.4 (Holland C K, O'Brien Jr W D, Crum L A 2000, Bader and Holland 2013). The cavitation index (ICAV, PRP/fc) was developed by Bader et al. to predict the occurrence of stable cavitation. Stable cavitation components are likely to occur when 0.09 ≤ ICAV ≤ 0.45 (Bader and Holland 2013). The parameters used in the present study (0.30 – 0.60 MPa at 1.5MHz) appeared in the range of CAV from 0.2 to 0.4, and MI from 0.245 to 0.49, suggesting that the likelihood of stable cavitation was evident and that of the inertial cavitation was lower (i.e., smaller number of nuclei would undergo inertial cavitation).

Pharmacodynamic Evaluations

The good correlation between cavitation responses with BBB permeability (Figure 4) indicated that monitoring cavitation activities would aid in monitoring the pharmacokinetics of the drugs to be delivered. Interestingly, Figure 3 showed that the Ktrans peak in the LO cases was reached 24 h after BBB opening, whereas the permeability of RO decreased as time progressed. This may be attributed to the edema that was detected until Day 5 (Figure 5). T2-weighted MR images of LO subjects indicated larger hyper-intense areas on the sonicated side on Day 1 than that on Day 0. The existence of edema may affect the contrast in DCE-MRI as well as the quantification of Ktrans.

Control the opening

To monitor and control the opening using real-time feedback during FUS is the ultimate goal for clinical translation. O'Reilly et al. (O'Reilly and Hynynen 2012) designed a semi-close-loop controlling system for BBB opening. They detected ultraharmonic components in the loop, and decreased the pressure to different target levels (% pressure to achieve ultraharmonics) once ultraharmonic emission detected. Target levels were found to be associated with MRI enhancement and damage occurrence. Albeit the potential of this controlling paradigm, one limitation is that the controlling relied on the successful detection of ultraharmonics, and then it became open-loop for the remainder of the sonications. However, it has been reported that ultraharmonic emissions were only occasionally observed in some PCD systems (Arvanitis et al 2012), which might affect the performance of ultraharmonic-based controller. Moreover, many other important matrixes, such as the cavitation doses of harmonics, ultraharmonics and broadband noise, had not been implemented into the feed-back control. Based on our findings, monitoring of these cavitation doses is associated with not only the direct opening assessments (such as MRI and/or fluorescence contrast enhancement), but also the other opening's characteristics (such as the opening duration, safety and permeability). To achieve the desired opening, FUS pressure, PL and/or PRF can be programmed based on the cavitation monitoring feedback. In addition, the cumulative cavitation doses may serve as the end or critical turning point of FUS exposure. For example, one can stop or decrease the level of FUS exposure once a certain amount of SCD and/or ICD has been reached. However, it should be noted that results of the present study were in agreement with O'Reilly et al.'s that the damage occurrence did not correlate with wideband responses well while it was associated with strong stable cavitation response. It suggested that the window to achieve safe opening was limited and we may have to be conservative in setting the parameters in the controller. For instance, one may want to avoid opening in 5 days due to the safety concern.

Prior to FUS exposure, selecting more suitable parameter sets also becomes important in tailoring the opening. In this study, cavitation activities were induced by microbubbles in FUS fields with various acoustic pressures. In order to reliably tailor the opening, we have uncovered the optimal type of bubble and FUS pulse trains. For instance, the BBB can be reinstated with 100% probability within the same amount of time when sonicated under the following parameter sets: [0.60 MPa, 1-2 μm] (closed on Day 1) and [0.30 MPa, 4-5 μm] (closed on Day 3). These two parameter sets may be used to control the BBB reversibility more reliably than other parameter sets.

Safety assessments

Typically attributed to adverse bio-effects, inertial cavitation is normally intended to be avoided by lowering FUS pressure in opening the BBB (Tung et al 2011b) and other microbubble-mediated ultrasound therapy (Bader and Holland 2013). However, previous reports (Chen and Konofagou 2014) have shown that inertial cavitation has been associated with the successful delivery of agents larger than 500 kDa when Definity® is used. This finding suggests that delivering larger-size agents may require inertial cavitation when smaller bubbles are used. Our results (Figure 7) indicated that ICD-based safety control could be accomplished when stronger inertial cavitation activities are induced, which would ultimately help clinicians to monitor the likelihood of safe opening during treatment using ICD.

Another interesting finding is that the cases with “opening with edema” and cases with “opening with histological damage” were identical under 0.60 MPa, while they did not fully coincide at 0.75 and 0.90 MPa. Three different situations were then found: 1) with edema but without histological damage; 2) without edema but with histological damage; and 3) with edema and with histological damage. The first situation can be attributed to the sensitivity of H&E histological evaluations. The second may result from physiological reinstatement of BBB. Future studies will investigate the detailed physiological mechanisms and their correlation with cavitation activities.

Limitations

There are several limitations to this study. First, MRI assessment was performed on a daily basis, so that the detection of closing is not precise in terms of hours. In other words, the time required for BBB reinstatement may be overestimated by no more than 24 hrs. Moreover, it should be noted that the opening findings are tied to the MR tracer used in the assessments. Based on our previous results assessed by fluorescently tagged dextrans, the position and areas of opening were agent-size relevant (Choi et al 2010). Thus, MR contrast agents at different molecular weights would result in distinct reversibility and permeability outcomes. Lastly, the global cavitation matrixes (SCD and ICD) were found to be bubble-size dependent. It would be interesting to normalize the SCD and ICD by an appropriate scaling factor, such as the scattering cross section of microbubbles, so that more detailed cavitation mechanisms may be uncovered. The cross section of larger bubbles would result in more scattering, and thus more cavitation emissions (Ainslie and Leighton 2011). Unfortunately, it is hard to calculate the scattering cross section at this time, since an elaborate simulation study will need to be performed. We will assess this effect in our future work.

5. Conclusion

In summary, we have shown that monitoring of the cavitation behavior during FUS can reliably predict the duration of opening, the permeability of the induced BBB opening and the likelihood of safe opening. The stable cavitation dose may therefore provide a real-time predictor of the properties of the induced reversible disruption. Finally, the dependence of the BBB reversibility on the bubble diameter and FUS pressure allows for the control of the safety profile of this technique.

Acknowledgments

This study was supported in part by National Institute of Health grants R01 EB009041 and R01 AG038961. The authors wish to thank Dr. Yao-Sheng Tung, Oluyemi Olumolade, Anushree Srivastava, Dr. Hong Chen, Shih-Ying Wu, and Dr. Fabrice Marquet from Biomedical Engineering Department of Columbia University, for their important assistance and suggestions.

References

  1. Ainslie M a, Leighton TG. Review of theory for scattering and extinction cross-sections, damping factors and resonance frequencies of spherical gas bubbles. J. Acoust. Soc. Am. 2011;130:3184–208. doi: 10.1121/1.3628321. Online: http://scitation.aip.org/content/asa/journal/jasa/130/5/10.1121/1.3628321. [DOI] [PubMed] [Google Scholar]
  2. Arvanitis CD, Livingstone MS, McDannold N. Combined ultrasound and MR imaging to guide focused ultrasound therapies in the brain. Phys. Med. Biol. 2013;58:4749–61. doi: 10.1088/0031-9155/58/14/4749. Online: http://www.ncbi.nlm.nih.gov/pubmed/23788054. [DOI] [PMC free article] [PubMed] [Google Scholar]
  3. Arvanitis CD, Livingstone MS, Vykhodtseva N, McDannold N. Controlled ultrasound-induced blood-brain barrier disruption using passive acoustic emissions monitoring. PLoS One. 2012;7:e45783. doi: 10.1371/journal.pone.0045783. Online: http://www.pubmedcentral.nih.gov/articlerender.fcgi?artid=3454363&tool=pmcentrez&rendertype=abstract. [DOI] [PMC free article] [PubMed] [Google Scholar]
  4. Aryal M, Arvanitis CD, Alexander PM, McDannold N. Ultrasound-mediated blood-brain barrier disruption for targeted drug delivery in the central nervous system. Adv. Drug Deliv. Rev. 2014;72:94–109. doi: 10.1016/j.addr.2014.01.008. Online: http://dx.doi.org/10.1016/j.addr.2014.01.008. [DOI] [PMC free article] [PubMed] [Google Scholar]
  5. Bader KB, Holland CK. Gauging the likelihood of stable cavitation from ultrasound contrast agents. Phys. Med. Biol. 2013;58:127–44. doi: 10.1088/0031-9155/58/1/127. Online: http://www.ncbi.nlm.nih.gov/pubmed/23221109. [DOI] [PMC free article] [PubMed] [Google Scholar]
  6. Burgess A, Hynynen K. Noninvasive and targeted drug delivery to the brain using focused ultrasound. ACS Chem. Neurosci. 2013;4:519–26. doi: 10.1021/cn300191b. Online: http://www.ncbi.nlm.nih.gov/pubmed/23379618. [DOI] [PMC free article] [PubMed] [Google Scholar]
  7. Chen H, Chen CC, Acosta C, Wu S-Y, Sun T, Konofagou EE. A New Brain Drug Delivery Strategy: Focused Ultrasound-Enhanced Intranasal Drug Delivery. PLoS One. 2014;9:e108880. doi: 10.1371/journal.pone.0108880. Online: http://dx.plos.org/10.1371/journal.pone.0108880. [DOI] [PMC free article] [PubMed] [Google Scholar]
  8. Chen H, Konofagou EE. The size of blood-brain barrier opening induced by focused ultrasound is dictated by the acoustic pressure. J. Cereb. Blood Flow Metab. 2014;34:1197–204. doi: 10.1038/jcbfm.2014.71. Online: http://www.ncbi.nlm.nih.gov/pubmed/24780905. [DOI] [PMC free article] [PubMed] [Google Scholar]
  9. Choi JJ, Pernot M, Small S a, Konofagou EE. Noninvasive, transcranial and localized opening of the blood-brain barrier using focused ultrasound in mice. Ultrasound Med. Biol. 2007;33:95–104. doi: 10.1016/j.ultrasmedbio.2006.07.018. Online: http://www.ncbi.nlm.nih.gov/pubmed/17189051. [DOI] [PubMed] [Google Scholar]
  10. Choi JJ, Wang S, Tung Y-S, Morrison B, Konofagou EE. Molecules of various pharmacologically-relevant sizes can cross the ultrasound-induced blood-brain barrier opening in vivo. Ultrasound Med. Biol. 2010;36:58–67. doi: 10.1016/j.ultrasmedbio.2009.08.006. Online: http://www.pubmedcentral.nih.gov/articlerender.fcgi?artid=2997717&tool=pmcentrez&rendertype=abstract. [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Fan CH, Liu HL, Ting CY, Lee YH, Huang CY, Yan-Jung M, Wei KC, Yen TC, Yeh CK. Submicron-bubble-enhanced focused ultrasound for blood-brain barrier disruption and improved CNS drug delivery. PLoS One. 2014;9 doi: 10.1371/journal.pone.0096327. [DOI] [PMC free article] [PubMed] [Google Scholar]
  12. Ferrara K, Pollard R, Borden M. Ultrasound microbubble contrast agents: fundamentals and application to gene and drug delivery. Annu. Rev. Biomed. Eng. 2007;9:415–47. doi: 10.1146/annurev.bioeng.8.061505.095852. Online: http://www.ncbi.nlm.nih.gov/pubmed/17651012. [DOI] [PubMed] [Google Scholar]
  13. Feshitan JA, Chen CC, Kwan JJ, Borden MA. Microbubble size isolation by differential centrifugation. J. Colloid Interface Sci. 2009;329:316–24. doi: 10.1016/j.jcis.2008.09.066. Online: http://www.ncbi.nlm.nih.gov/pubmed/18950786. [DOI] [PubMed] [Google Scholar]
  14. Holland CK, O'Brien WD, Jr, Crum LAFPL, T A F Bioeffects in tissues with gas bodies. J. Ultrasound Med. 2000;19:97–108. doi: 10.7863/jum.2000.19.2.97. Online: http://www.jultrasoundmed.org/content/19/2/97.abstract. [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. Hsu P-H, Wei K-C, Huang C-Y, Wen C-J, Yen T-C, Liu C-L, Lin Y-T, Chen J-C, Shen C-R, Liu H-L. Noninvasive and targeted gene delivery into the brain using microbubble-facilitated focused ultrasound. PLoS One. 2013;8:e57682. doi: 10.1371/journal.pone.0057682. Online: http://www.pubmedcentral.nih.gov/articlerender.fcgi?artid=3584045&tool=pmcentrez&rendertype=abstract. [DOI] [PMC free article] [PubMed] [Google Scholar]
  16. Hynynen K, McDannold N. Noninvasive MR Imaging–guided Focal Opening of the Blood-Brain Barrier in Rabbits1. Radiology. 2001 doi: 10.1148/radiol.2202001804. Online: http://radiology.rsna.org/content/220/3/640.short. [DOI] [PubMed]
  17. Kinoshita M, McDannold N, Jolesz FA, Hynynen K. Targeted delivery of antibodies through the blood-brain barrier by MRI-guided focused ultrasound. Biochem. Biophys. Res. Commun. 2006;340:1085–90. doi: 10.1016/j.bbrc.2005.12.112. Online: http://www.ncbi.nlm.nih.gov/pubmed/16403441. [DOI] [PubMed] [Google Scholar]
  18. Liu H-L, Hua M-Y, Chen P-Y, Chu P-C, Pan C-H, Yang H-W, Huang C-Y, Wang J-J, Yen T-C, Wei K-C. Blood-brain barrier disruption with focused ultrasound enhances delivery of chemotherapeutic drugs for glioblastoma treatment. Radiology. 2010;255:415–25. doi: 10.1148/radiol.10090699. [DOI] [PubMed] [Google Scholar]
  19. McDannold N, Arvanitis CD, Vykhodtseva N, Livingstone MS. Temporary disruption of the blood-brain barrier by use of ultrasound and microbubbles: safety and efficacy evaluation in rhesus macaques. Cancer Res. 2012;72:3652–63. doi: 10.1158/0008-5472.CAN-12-0128. Online: http://www.pubmedcentral.nih.gov/articlerender.fcgi?artid=3533365&tool=pmcentrez&rendertype=abstract. [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. McDannold N, Vykhodtseva N, Hynynen K. Targeted disruption of the blood-brain barrier with focused ultrasound: association with cavitation activity. Phys. Med. Biol. 2006;51:793–807. doi: 10.1088/0031-9155/51/4/003. Online: http://www.ncbi.nlm.nih.gov/pubmed/16467579. [DOI] [PubMed] [Google Scholar]
  21. O'Reilly M, Hynynen K. Blood-Brain Barrier: Real-time Feedback-controlled Focused Ultrasound Disruption by Using an Acoustic Emissions–based Controller. Radiology. 2012;263 doi: 10.1148/radiol.11111417. Online: http://radiology.rsna.org/content/263/1/96.short. [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. Pardridge WM. The blood-brain barrier: bottleneck in brain drug development. NeuroRx. 2005;2:3–14. doi: 10.1602/neurorx.2.1.3. Online: http://www.pubmedcentral.nih.gov/articlerender.fcgi?artid=539316&tool=pmcentrez&rendertype=abstract. [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Park J, Zhang Y, Vykhodtseva N, Jolesz FA, McDannold NJ. The kinetics of blood brain barrier permeability and targeted doxorubicin delivery into brain induced by focused ultrasound. J. Control. Release. 2012;162:134–42. doi: 10.1016/j.jconrel.2012.06.012. Online: http://www.pubmedcentral.nih.gov/articlerender.fcgi?artid=3520430&tool=pmcentrez&rendertype=abstract. [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. Qin S, Caskey CF, Ferrara KW. Ultrasound contrast microbubbles in imaging and therapy: physical principles and engineering. Phys. Med. Biol. 2009;54:R27–57. doi: 10.1088/0031-9155/54/6/R01. Online: http://www.pubmedcentral.nih.gov/articlerender.fcgi?artid=2818980&tool=pmcentrez&rendertype=abstract. [DOI] [PMC free article] [PubMed] [Google Scholar]
  25. Raymond SB, Treat LH, Dewey JD, McDannold NJ, Hynynen K, Bacskai BJ. Ultrasound enhanced delivery of molecular imaging and therapeutic agents in Alzheimer’s disease mouse models. PLoS One. 2008;3:e2175. doi: 10.1371/journal.pone.0002175. Online: http://www.pubmedcentral.nih.gov/articlerender.fcgi?artid=2364662&tool=pmcentrez&rendertype=abstract. [DOI] [PMC free article] [PubMed] [Google Scholar]
  26. Samiotaki G, Acosta C, Wang S, Konofagou EE. Enhanced delivery and bioactivity of the neurturin neurotrophic factor through focused ultrasound—mediated blood–brain barrier opening in vivo. J. Cereb. Blood Flow Metab. 2015;35:611–22. doi: 10.1038/jcbfm.2014.236. Online: http://www.nature.com/doifinder/10.1038/jcbfm.2014.236. [DOI] [PMC free article] [PubMed] [Google Scholar]
  27. Samiotaki G, Konofagou E. Dependence of the Reversibility of Focused-Ultrasound-Induced Blood–Brain Barrier Opening on Pressure and Pulse Length In Vivo. IEEE Trans. Ultrason. 2013;60:2257–65. doi: 10.1109/TUFFC.2013.6644731. Online: http://www.ncbi.nlm.nih.gov/pmc/articles/PMC3968797/ [DOI] [PMC free article] [PubMed] [Google Scholar]
  28. Samiotaki G, Vlachos F, Tung Y-S, Konofagou EE. A quantitative pressure and microbubble-size dependence study of focused ultrasound-induced blood-brain barrier opening reversibility in vivo using MRI. Magn. Reson. Med. 2012;67:769–77. doi: 10.1002/mrm.23063. Online: http://www.pubmedcentral.nih.gov/articlerender.fcgi?artid=3658823&tool=pmcentrez&rendertype=abstract. [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. Sun T, Jia N, Zhang D, Xu D. Ambient pressure dependence of the ultra-harmonic response from contrast microbubbles. J. Acoust. Soc. Am. 2012;131:4358–64. doi: 10.1121/1.4707512. Online: http://www.ncbi.nlm.nih.gov/pubmed/22712910. [DOI] [PubMed] [Google Scholar]
  30. Thévenot E, Jordão JF, O'Reilly M a, Markham K, Weng Y-Q, Foust KD, Kaspar BK, Hynynen K, Aubert I. Targeted delivery of self-complementary adeno-associated virus serotype 9 to the brain, using magnetic resonance imaging-guided focused ultrasound. Hum. Gene Ther. 2012;23:1144–55. doi: 10.1089/hum.2012.013. Online: http://www.ncbi.nlm.nih.gov/pubmed/22838844. [DOI] [PMC free article] [PubMed] [Google Scholar]
  31. Treat LH, McDannold N, Zhang Y, Vykhodtseva N, Hynynen K. Improved anti-tumor effect of liposomal doxorubicin after targeted blood-brain barrier disruption by MRI-guided focused ultrasound in rat glioma. Ultrasound Med. Biol. 2012;38:1716–25. doi: 10.1016/j.ultrasmedbio.2012.04.015. Online: http://www.ncbi.nlm.nih.gov/pubmed/22818878. [DOI] [PMC free article] [PubMed] [Google Scholar]
  32. Tung Y-S, Marquet F, Teichert T, Ferrera V, Konofagou EE. Feasibility of noninvasive cavitation-guided blood-brain barrier opening using focused ultrasound and microbubbles in nonhuman primates. Appl. Phys. Lett. 2011a;98:163704. doi: 10.1063/1.3580763. Online: http://www.pubmedcentral.nih.gov/articlerender.fcgi?artid=3094460&tool=pmcentrez&rendertype=abstract. [DOI] [PMC free article] [PubMed] [Google Scholar]
  33. Tung Y-S, Vlachos F, Choi JJ, Deffieux T, Selert K, Konofagou EE. In vivo transcranial cavitation threshold detection during ultrasound-induced blood-brain barrier opening in mice. Phys. Med. Biol. 2010;55:6141–55. doi: 10.1088/0031-9155/55/20/007. Online: http://www.ncbi.nlm.nih.gov/pubmed/20876972. [DOI] [PMC free article] [PubMed] [Google Scholar]
  34. Tung Y-S, Vlachos F, Feshitan JA, Borden MA, Konofagou EE. The mechanism of interaction between focused ultrasound and microbubbles in blood-brain barrier opening in mice. J. Acoust. Soc. Am. 2011b;130:3059–67. doi: 10.1121/1.3646905. Online: http://www.pubmedcentral.nih.gov/articlerender.fcgi?artid=3248062&tool=pmcentrez&rendertype=abstract. [DOI] [PMC free article] [PubMed] [Google Scholar]
  35. Vlachos F, Tung Y-S, Konofagou E. Permeability dependence study of the focused ultrasound-induced blood-brain barrier opening at distinct pressures and microbubble diameters using DCE-MRI. Magn. Reson. Med. 2011;66:821–30. doi: 10.1002/mrm.22848. Online: http://www.ncbi.nlm.nih.gov/pubmed/21465543. [DOI] [PMC free article] [PubMed] [Google Scholar]
  36. Vlachos F, Tung Y-S, Konofagou EE. Permeability assessment of the focused ultrasound-induced blood-brain barrier opening using dynamic contrast-enhanced MRI. Phys. Med. Biol. 2010;55:5451–66. doi: 10.1088/0031-9155/55/18/012. Online: http://www.ncbi.nlm.nih.gov/pubmed/20736501. [DOI] [PMC free article] [PubMed] [Google Scholar]
  37. Wang F, Shi Y, Lu L, Liu L, Cai Y, Zheng H, Liu X, Yan F, Zou C, Sun C, Shi J, Lu S, Chen Y. Targeted delivery of GDNF through the blood-brain barrier by MRI-guided focused ultrasound. PLoS One. 2012;7:e52925. doi: 10.1371/journal.pone.0052925. Online: http://www.pubmedcentral.nih.gov/articlerender.fcgi?artid=3531370&tool=pmcentrez&rendertype=abstract. [DOI] [PMC free article] [PubMed] [Google Scholar]
  38. Wang S, Olumolade OO, Sun T, Samiotaki G, Konofagou EE. Noninvasive, neuron-specific gene therapy can be facilitated by focused ultrasound and recombinant adeno-associated virus. Gene Ther. 2015;22:104–10. doi: 10.1038/gt.2014.91. Online: http://www.nature.com/doifinder/10.1038/gt.2014.91. [DOI] [PMC free article] [PubMed] [Google Scholar]
  39. Wang S, Samiotaki G, Olumolade O, Feshitan JA, Konofagou EE. Microbubble type and distribution dependence of focused ultrasound-induced blood-brain barrier opening. Ultrasound Med. Biol. 2014a;40:130–7. doi: 10.1016/j.ultrasmedbio.2013.09.015. Online: http://www.ncbi.nlm.nih.gov/pubmed/24239362. [DOI] [PMC free article] [PubMed] [Google Scholar]
  40. Wang S, Sun T, Acosta C, Karakatsani ME, Olumolade O, Samiotaki G. Enhancement of Direct Brain Infusion with Focused Ultrasound and Microbubbles. 2014b:2169–72. [Google Scholar]

RESOURCES