Abstract
Focused ultrasound (FUS), applied in conjunction with systemically administered gas-filled microbubbles (MBs), locally and transiently disrupts the blood–brain and blood–tumor barriers (BBB/BTB). These disruptions enable the delivery of blood-borne therapeutics whose passage into tissue would otherwise be heavily restricted or completely obstructed under normal physiological conditions. In addition to facilitating delivery, FUS-mediated BBB/BTB disruption can induce secondary biological effects, including increased inflammation, altered blood flow, and the ability, in the context of cerebral cavernous malformations (CCMs), to alter the progression of neurovascular lesions. Yet, even as FUS BBB/BTB disruption rapidly advances toward clinical adoption for these indications, important questions about therapeutic efficacy and response to treatment still remain unresolved. This is in part due to limitations of currently deployed imaging approaches. Quantitative MRI methods have the potential to address these questions, but they have not been well-leveraged for this purpose before. The content of this thesis is segmented into four projects that leverage quantitative MRI across multiple disease contexts to investigate the therapeutic potential of FUS-mediated BBB/BTB disruption.
In the first project, we evaluated whether FUS-mediated BBB disruption enhances delivery of small and moderate-sized agents to cerebral cavernous malformations (CCMs), hemorrhagic vascular lesions of the central nervous system that lack effective pharmacologic therapies. Using T1-mapping MRI, we quantified delivery of ~1 kDa (MultiHance) and ~17 kDa (GadoSpin D) contrast agents with and without FUS. FUS more than doubled MultiHance delivery in the lesion core and tripled it in the perilesional space, while GadoSpin D showed even greater relative gains, with 21.7-fold and 3.8-fold increases, respectively. These findings provide a quantitative framework for how FUS augments delivery to CCMs and inform therapeutic selection to potentially provide patients with new treatment options.
In the second project, we evaluated the ability of FUS+MB-mediated BTB disruption to augment the delivery of small and moderately sized contrast agents in a peripheral tumor model of triple negative breast cancer (TNBC), both at baseline and following anti-angiogenic therapy designed to induce vascular normalization (VN). VN aims to rebalance tumor vasculature by restoring pro- and anti-angiogenic signaling, resulting in reduced vessel diameter, increased pericyte coverage, and decreased permeability and interstitial fluid pressure. Using T1-mapping MRI, we quantified delivery of a 605 Da (Gadovist) and ~17 kDa (GadoSpin D) contrast agent to assess how VN modulates FUS-enhanced transport. VN+FUS increased both exposure (~64%) and delivery (1.89-fold) of Gadovist relative to controls, with trends toward increased exposure from VN alone. For GadoSpin D, neither VN nor FUS alone improved delivery; however, VN priming enabled FUS to increase exposure by ~45% relative to controls and ~74% relative to FUS alone. These findings demonstrate that VN can reshape the tumor microenvironment to enhance FUS-mediated delivery in a size-dependent manner.
In the third project, we addressed a critical and unresolved question: how therapeutic size influences delivery following FUS-mediated BBB/BTB disruption. While FUS has been shown to increase the delivery of a broad range of therapeutics, clinically relevant agents span a wide range of hydrodynamic diameters, from small molecules to antibodies and nanoparticles, and the extent to which delivery efficiency depends on size remains poorly understood. Accurately quantifying this relationship remains challenging with conventional MRI techniques, which are limited by sensitivity and contrast agent availability. To overcome these limitations, we introduced quantitative susceptibility mapping (QSM) as a novel MRI-based approach to quantify the delivery of superparamagnetic iron oxide nanoparticles (IONPs) following FUS-mediated BBB/BTB disruption. Using QSM, we evaluated size-dependent delivery in both naïve brain and glioma-bearing mice modeling glioblastoma and suggests that an optimum size for FUS-mediated delivery to both naïve brain and to brain tumors exists at approximately 23nm.
Fourth, we investigated how FUS-mediated BBB disruption influences the progression of CCMs. While our prior work has shown that FUS BBB disruption reduces lesion growth and formation, the underlying changes in lesion composition remain poorly understood. Given that CCM progression is closely linked to iron accumulation due to their hemorrhagic nature, we leveraged QSM, a technique highly sensitive to paramagnetic blood products and clinically used to evaluate CCMs to quantify lesion susceptibility in a murine model. We found that FUS did not alter average lesion susceptibility but significantly impacted the trajectory of total susceptibility change relative to pre-FUS values. Specifically, the change in total susceptibility remained relatively stable in FUS-treated lesions, whereas untreated lesions demonstrated a consistent increase over time. These findings provide new insight into how FUS interacts with CCM biology and suggest a potential therapeutic benefit beyond volume control.
