Ultrasound-mediated transport of nanoparticles in tissue: A predictive model

A prerequisite for successful cancer therapy is that all cancer cells are killed. Chemotherapy alone or in combination with other treatments is commonly used in cancer therapy. A problem in chemotherapy is that less than 1 % of the drug is taken up in tumours. Toxic effects towards healthy tissue restrict the doses that can be applied, which limit the therapeutic response. A promising strategy for increasing the tumour uptake of drugs, is to encapsulate drugs into nanoparticles, and take the advantage of the leaky blood vessels in tumours which permit nanoparticles to cross these capillary walls, but not capillaries in normal tissue. However, a challenge using nanoparticles in cancer therapy is to achieve sufficient dosage and homogenous distribution of the nanoparticles to all cancer cells, as the extracellular matrix surrounding the cells limits the penetration of drugs and nanoparticles to the cancer cell. Delivery of nanoparticles to brain tissue is even more challenging due to the blood-brain barrier. Tight junctions between cells lining the microvessels in the brain form a physical barrier. Thus, to be able to treat diseases in the brain, the blood-brain barrier has to be opened safely. Focused ultrasound especially in the presence of gas filled microbubbles has been shown to improve the delivery of nanoparticles and drugs into tumors and across the blood-brain barrier in mice. Microbubbles are gas filled bubbles and when the microbubbles are exposed to ultrasound, they will vibrate causing mechanical stress on the blood vessel wall and improve transport into the tissue. To optimize this promising therapy which now also is tested in cancer patients, we need to understand the underlying mechanisms better. The overall aim of the project was to obtain new knowledge on the transport mechanisms for nanoparticles and drugs in tissues. This was achieved by theoretically describing transport in tissue exposed to ultrasound, and by combining modelling/computer simulations and experimental research. We studied in particular whether ultrasound enhanced diffusion of nanoparticles and lead to a flow called acoustic streaming. A major finding was that with the ultrasound parameters applied, which are typical parameters used both in mice and in patients, acoustic streaming contributed only to a small degree to transport of nanoparticles, whereas nanoparticle diffusion increased. Next, we showed using simulations how ultrasound enhanced diffusion. We found that the particles are “jumping” on and off the fibre network in the extracellular matrix and this “jumping” was increased by ultrasound. The microenvironment in tumours varies, i.e. the vascular density and blood flow, and the amount of proteins and sugars in the extracellular matrix between the cells. Amount of protein in particular the network of collagen fibres is important for the stiffness of the tissue. We studied how differences in the microenvironment could influence the effect of ultrasound and microbubbles on the tumour uptake of nanoparticles. We found that tumours with little collagen, which were rather soft, responded better than stiffer tumours. Using a higher ultrasound pressure might be beneficial for the stiffer tumours, but this needs to be tested further. Increasing the temperature might heat up the tissue. Using simulations, we studied how much the temperature in the tissue was increased when the acoustic pressure increased. The properties of the microbubbles can be important for their efficiency in drug delivery. We have studied two types of microbubbles, namely the microbubbles designed as contrast agents for ultrasound imaging which have a diameter of 3-5 µm, and larger microbubbles with a dimeter of approximately 20-30 µm. The advantage of the large microbubbles is that they will be in contact with a larger part of the vessel wall and the stress on the vessel wall when the microbubbles are vibrating, might be more efficient for drug delivery. In accordance with this hypothesis, simulations indicated that the large microbubbles were indeed more efficient in drug delivery. To optimize delivery of nanoparticles to tumors, it is important to know whether ultrasound and microbubbles have any effect on the blood vessels. Ultrasound and the small microbubbles designed for imaging, reduced the number of functional vessels immediately after treatment, and the vascularization was restored within 1 hour. However, the larger microbubbles had no effect on the vasculature, probably because low acoustic pressures were applied. Altogether, the project has contributed to new knowledge on the mechanisms for ultrasound-induced transport processes through tissue, and the impact of the tumor microenvironment on ultrasound-induced delivery of nanoparticles. The results can be used to design new clinical studies in patients and to optimize the treatment, for instance avoid heating the tissue too much.

Outcomes • A volume-averaged model was established and used to estimate the ultrasound-induced acoustic streaming in an agarose hydrogel. The model showed that with the ultrasound parameters used (1 MHz and 0.5MPa), acoustic streaming was negligible. The main mechanism for the ultrasound-enhanced movement of nanoparticles was diffusion. This was consistent with experimental results measuring the mean square displacement of nanoparticles by single particle tracking. • Molecular dynamics simulations showed that the ultrasound-enhanced diffusion probably was caused by oscillatory diffusion, i.e. the nanoparticles oscillate between a state in contact/no contact with polymer fibers. • The tumor microenvironment of three commonly used murine tumor model were characterized. Vascular parameters, collagen, glycosaminoglycan in the extracellular matrix and stiffness were measured. Such characterization is important for the scientists using these tumor models. • All reported clinical studies and most preclinical studies in mice use microbubbles designed for ultrasound imaging. Such microbubbles were compared with microbubbles with 100 times larger volume. These large microbubbles are formed by the concept called Acoustic Cluster Therapy (ACT). The ACT-bubble had a slightly enhanced impact on nanoparticle transport in extracellular matrix compared to the microbubbles designed for ultrasound imaging, probably because the ACT-bubbles are in contact with a larger part of the blood vessel wall. • Ultrasound and ACT bubbles, monodisperse microbubbles, or microbubbles for imaging opened the blood-brain barrier. Impact • The results in mice demonstrated that tumor uptake of nanoparticles differed between tumor models, indicating that the ultrasound-parameters applied should be different in soft and stiff tumors. This indication should be further tested in other tumor models using various acoustic pressures. • Using the typical US parameters used for ultrasound-mediated drug delivery, a major finding was that ultrasound enhanced diffusion contributed to nanoparticle displacement. Acoustic streaming was negligible. A high duty cycle or higher acoustic pressure, not causing significant tissue damage, should be used to induce acoustic streaming. It is previously reported that high ultrasound frequency is needed to induce acoustic streaming. Our results indicate at a dual frequency transducer can be used to both enhance diffusion at low frequency and cause acoustic streaming at high frequency. • When designing new clinical studies combining chemotherapy with ultrasound and microbubbles the findings reported in this project are of importance. • The results obtained comparing the large ACT bubbles and the microbubbles designed for ultrasound imaging are important new knowledge for the company EXACT Therapeutics.

A prerequisite for successful cancer chemotherapy is that the drugs reach all tumor cells. However, less than 1% of injected drugs accumulate in tumors. Encapsulating drugs into nanoparticles (NPs) improves the accumulation of drugs in tumors and reduces the toxic effect, but the tumor uptake of NPs is low and heterogeneous. Delivery of NPs to brain tissue is even more challenging due to the blood-brain barrier (BBB). Focused ultrasound (FUS) has been found to improve the delivery of NPs to tumor tissue either by acoustic radiation force or cavitation caused by microbubbles (MBs), and temporarily open the BBB. MBs will oscillate in the acoustic field which can generate mechanical effects on the vessel wall, enhancing the vascular permeability. The next step is the penetration throughout the interstitium. This transport process is poorly understood. We hypothesize that MBs in the blood vessels oscillating in the acoustic field cause the capillary wall to vibrate and generate interstitial flow and this flow transports NPs and drugs through the interstitium. To prove our hypothesis, we will establish a theoretical framework describing transport in porous media and combine modelling/computer simulations and experiments. The theoretical work and simulations will be carried out at the Center of Excellence Porelab at NTNU which has extensive competence on porous media transport. The results of the simulations will be continuously challenged and validated through experiments and advanced imaging of NPs in tissues. The interstitium will first be regarded as a network of flow pores, followed by a continuum model. This novel approach will bring new knowledge on FUS-mediated transport through tissue, optimizing FUS-mediated delivery of NPs and drugs and understanding the impact of interstitial flow generated by acoustic radiation force or the vibrating capillary wall. The new knowledge will lead to a model predicting the distribution of NPs and drugs in tumor and brain tissue.