Effects of Acoustic Radiation Force on the Binding Efficiency of BR55, a VEGFR2-Specific Ultrasound Contrast Agent

Abstract

This work describes an in vivo study analyzing the effect of acoustic radiation force (ARF) on the binding of BR55 VEGFR2-specific contrast-agent microbubbles in a model of prostatic adenocarcinoma in rat. A commercial ultrasound system was modified by implementing high duty-cycle 3.5-MHz center frequency ARF bursts in a scanning configuration. This enabled comparing the effects of ARF on binding in tumor and healthy tissue effectively in the same field of view. Bubble binding was established by measuring late-phase enhancement in amplitude modulation (AM) contrast-specific imaging mode (4 MHz, 150 kPa) 10 min after agent injection when the unbound bubbles were cleared from the circulation. Optimal experimental conditions, such as agent concentration (0.4 × 10⁸–1.6 × 10⁸ bubbles/kg), acoustic pressure amplitude (26–51 kPa) and duty-cycle (20%–95%) of the ARF bursts, were evaluated in their ability to enhance binding in tumor without significantly increasing binding in healthy tissue. Using the optimal conditions (38 kPa peak-negative pressure, 95% duty cycle), ARF-assisted binding of BR55 improved significantly in tumor (by a factor of 7) at a lower agent dose compared with binding without ARF, and it had an insignificant effect on binding in healthy tissue. Thus, the high binding specificity of BR55 microbubbles for targeting VEGFR2 present at sites of active angiogenesis was confirmed by this study. Therefore, it is believed that based on the results obtained in this work, ultrasound molecular imaging using target-specific contrast-agent microbubbles should preferably be performed in combination with ARF.

Introduction

Contrast enhanced ultrasound (CEUS) imaging has seen tremendous improvements over the last two decades thanks to the advent of so-called second generation blood-pool agents, viz. SonoVue^® (Bracco Imaging S.p.A., Milan, Italy), Optison™ (GE Healthcare, Chalfont St Giles, UK), Definity^® (Lantheus Medical Imaging, North Billerica, MA, USA) and Sonazoid™ (GE Healthcare) (de Jong et al. 2009). In parallel, continuous improvements in microbubble detection sensitivity and contrast-to-tissue ratio were made possible thanks to the development of sophisticated low mechanical index (MI) contrast-specific imaging techniques, such as pulse inversion (PI), amplitude modulation (AM) and contrast pulse sequence (CPS) imaging (Rafter et al. 2004). Because of the high contrast sensitivity of the imaging techniques, in combination with the nonlinear characteristics of these agents (Frinking et al. 2000), individual bubbles can now be tracked in real-time while flowing through the smallest branches of the vascular network. This makes contrast ultrasound a valuable clinical tool for identifying, for example, vascularization within tumors and, thereby, improving the detection and diagnosis of cancer.

A new and perhaps even more exciting area for contrast ultrasound is in molecular imaging. Real-time imaging, high spatial resolution and extraordinary sensitivity for microbubble detection make ultrasound, in combination with target-specific contrast agents, a powerful modality for molecular imaging. Ligand-bearing microbubbles can be targeted to specific receptors, delineating pathologies that would otherwise be difficult to detect. A common strategy for preparing such targeted contrast agents is to couple specific biotinylated antibodies to streptavidin-functionalized microbubbles (Klibanov 2009). These products are very convenient for proof-of-concept studies in animals but they are not suitable for use in humans because of the risk of provoking an immune response. Recently, a target-specific contrast agent for human use, BR55 (Pochon et al. 2010), has been developed without potentially immunogenic proteins such as streptavidin or antibodies. BR55 microbubbles are functionalized with a binding peptide specific for the human vascular endothelium growth factor receptor 2 (VEGFR2), which is one of the important molecular markers of angiogenesis, the process of growing new capillary blood vessels, and has been linked to the progression and aggressiveness of many tumor types (Folkman 1992; Carmeliet 2005; Hicklin and Ellis 2005).

Target-specific agents show great potential in ultrasound molecular imaging applications, their binding efficiency being determined by an effective interaction between ligands present on the bubbles and target receptors expressed on the vascular endothelium. This interaction depends, among others, on local agent concentration and physiologic factors such as receptor density, vessel diameter and wall shear rate. Indeed, for given physiologic factors, a linear relationship has been observed in vitro between agent concentration and bubble adhesion on P-selectin-coated microcapillaries (Rychak et al. 2005) and in vivo between agent dose and late-phase enhancement in rat prostate adenocarcinoma (Tardy et al. 2010). However, only small fractions of the total number of injected bubbles interact and bind to the target site, even at high agent concentrations. In the in vivo study just mentioned, for example, a high tumor detection specificity was obtained (a tumor-to-healthy tissue signal ratio of 20 was observed) but at an agent dose corresponding to approximately 10 times the typical imaging dose used with a nontargeted agent such as SonoVue™ (Schneider 1999). For establishing bubble binding at this high dose, late-phase enhancement was measured 10 min after injection when the unbound bubbles were cleared from the circulation. Moreover, it was demonstrated in the same in vivo study that the late-phase signal in healthy prostate tissue, although significantly lower than in tumor, also increased with agent dose. Therefore, alternative solutions have been explored that enhance binding efficiency at lower agent dose, resulting in improved tumor detection specificity and, possibly, earlier detection of bound bubbles in tumor.

Acoustic radiation force (ARF) (Fowlkes et al. 1993; Leighton 1994; Dayton et al. 1997) has been described as a means to manipulate bubbles in a flow by pushing them away from the center of the vessel toward the vessel wall (Dayton et al. 1999a, 1999b; Zhao et al. 2004; Rychak et al. 2005, 2007). Consequently, ARF promotes ligand-receptor interaction and, thus, may enhance binding of target-specific microbubbles in tumor. However, it is unknown if ARF may also increase binding in healthy tissue, which would deteriorate tumor detection specificity. In this work, ARF-assisted binding of BR55 microbubbles was investigated in a model of prostatic adenocarcinoma in rat. Optimal experimental conditions, such as agent concentration, acoustic pressure amplitude and duty-cycle of the ARF bursts, were evaluated in their ability to enhance binding in tumor without significantly increasing binding in healthy tissue. Additionally, the effect of ARF on the binding efficiency of BR55 in tumor and healthy tissue was compared with that of a nontargeted blood-pool agent using the optimized conditions.