Ultrasound contrast microbubbles to predict the microsphere distribution during transarterial radioembolization with holmium microspheres, an in vitro proof of concept study

Abstract

Transarterial radioembolization (TARE) is an established treatment method for non-resectable liver tumors. One of the challenges of the approach is the accurate prediction of the microsphere biodistribution in the liver. We propose to use ultrasound contrast microbubbles as holmium microsphere precursors, which allows real-time prediction of the microsphere trajectories and biodistribution using dynamic contrast-enhanced ultrasound (DCE-US). The immediate goal in this in vitro study was to investigate the predictive capabilities of microbubbles as microsphere precursors. The study was conducted in an experimental in vitro model which represents the bifurcating right branch of the hepatic artery. A controlled injection of experimental BR-14 ultrasound contrast microbubbles and non-radioactive holmium-165 microspheres was performed in separate consecutive experiments in an arterial flow phantom. The microbubbles and microspheres were collected separately at the outlets of the phantom and counted using a Coulter counter to determine their distribution over the different outlets. The flow profile, the injection velocity, and the catheter position were monitored during the measurements to ensure stability. The results showed a good correlation between the microbubble and the microsphere distributions (p = 0.0038, r = 0.88) measured at the outlets. Differences in the distributions could be attributed to the characteristics of microbubbles and microspheres alone (e.g. particle size and concentration), since critical parameters were kept stable between the two experiments. The current in vitro study provides confidence that the microsphere biodistribution can be predicted using contrast microbubbles. The comparison provided by this study forms a foundation for the development of a DCE-US guided TARE treatment.

Keywords: In-vitro model; biodistribution; holmium microspheres; particle flow behavior; transarterial radioembolization; ultrasound contrast microbubbles.

Figure 1.

(A) The 3D CAD model representing a simplified right branch of the hepatic artery in SolidWorks (Dassault Systèmes, Wyman St Waltham, Massachusetts, USA) and (B) the RHA phantom made from PDMS including the capillary tubes, where the cavities represent the arterial lumen, visualized by filling the lumen with a mix of blue dye (Ecoline 533 indigo, Royal Talens, Apeldoorn, the Netherlands) and blood mimicking fluid.

Figure 2.

Schematic overview of the experimental in vitro setup used for the comparison between microbubble and microsphere trajectories.

Figure 3.

(A) A schematic overview and (B) a photo of the right branch of the hepatic artery phantom, including the cameras and LED light sources for analysis of the catheter position and movement. Note that Figure 3A is flipped with respect to Figure 2.

Figure 4.

Analysis of the catheter position, with (A) the desired 9 o’clock position of the catheter tip, showing the catheter opening (light grey) and catheter wall (dark grey) in the cross-section of the right branch of the hepatic artery phantom, and (B, C) the measurements of the catheter tip location on the first frame for both the top view (B) as well as the side view (C). Reference lines were used for measuring the catheter tip location, where the blue vertical lines indicate the diameter of the catheter, the red lines indicate the lumen diameter, and the green line (B) indicates the distance between the catheter and the start of the bifurcation fixed at a distance of 11.1 mm.

Figure 5.

Mean waveforms and standard deviations of (A) the BR-14 microbubble measurements and (B) the ¹⁶⁵Ho microsphere measurements.

Figure 6.

Heat maps of the average catheter location and movement of (A) the BR-14 microbubble measurements and (B) the ¹⁶⁵Ho microsphere measurements. The position of the catheter tip is expressed as a percentage of its presence at a location over the total duration of the injection, in the injection plane.

Figure 7.

(A) BR-14 microbubble and (B) ¹⁶⁵Ho microsphere distributions for all measurements, expressed relative to the total number of particles that have been measured over all outlets per measurement. (C) Average distributions of the microbubbles (MB) and microspheres (MS) including standard deviations (2σ), expressed as a percentage of the total number of particles in all outlets.