Applying artificial intelligence and computational modeling to nanomedicine.

Achieving specific and targeted delivery of nanomedicines to diseased tissues is a major challenge. This is because the process of designing, formulating, testing, and selecting a nanoparticle delivery vehicle for a specific disease target is governed by complex multivariate interactions. Computational modeling and artificial intelligence are well-suited for analyzing and modeling large multivariate datasets in short periods of time. Computational approaches can be applied to help design nanomedicine formulations, interpret nanoparticle-biological interactions, and create models from high-throughput screening techniques to improve the selection of the ideal nanoparticle carrier. In the future, many steps in the nanomedicine development process will be done computationally, reducing the number of experiments and time needed to select the ideal nanomedicine formulation.

Toward Predicting Nanoparticle Distribution in Heterogeneous Tumor Tissues.

Nanobio interaction studies have generated a significant amount of data. An important next step is to organize the data and design computational techniques to analyze the nanobio interactions. Here we developed a computational technique to correlate the nanoparticle spatial distribution within heterogeneous solid tumors. This approach led to greater than 88% predictive accuracy of nanoparticle location within a tumor tissue. This proof-of-concept study shows that tumor heterogeneity might be defined computationally by the patterns of biological structures within the tissue, enabling the identification of tumor patterns for nanoparticle accumulation.

Delineating the tumour microenvironment response to a lipid nanoparticle formulation.

Nanoparticles can reduce cytotoxicity, increase circulation time and increase accumulation in tumours compared to free drug. However, the value of using nanoparticles for carrying small molecules to treat tumours at the cellular level has been poorly established. Here we conducted a cytodistribution analysis on Doxorubicin-treated and Doxil-treated tumours to delineate the differences between the small molecule therapeutic Doxorubicin and its packaged liposomal formulation Doxil. We found that Doxil kills more cancer cells, macrophages and neutrophils in the 4T1 breast cancer tumour model, but there is delayed killing compared to its small molecule counterpart Doxorubicin. The cellular interaction with Doxil has slower uptake kinetics and the particles must be degraded to release the drug and kill the cells. We also found that macrophages and neutrophils in Doxil-treated tumours repopulated faster than cancer cells during the relapse phase. While researchers conventionally use tumour volume and animal survival to determine a therapeutic effect, our results show diverse cell killing and a greater amount of cell death in vivo after Doxil liposomes are administered. We conclude that the fate and behaviour of the nanocarrier influences its effectiveness as a cancer therapy. Further investigations on the interactions between different nanoparticle designs and the tumour microenvironment components will lead to more precise engineering of nanocarriers to selectively kill tumour cells and prolong the therapeutic effect.

Macrophages Actively Transport Nanoparticles in Tumors After Extravasation.

Nanoparticles need to navigate a complex microenvironment to target cells in solid tumors after extravasation. Diffusion is currently the accepted primary mechanism for nanoparticle distribution in tumors. However, the extracellular matrix can limit nanoparticle diffusion. Here, we identified tumor-associated macrophages as another key player in transporting and redistributing nanoparticles in the tumor microenvironment. We found tumor-associated macrophages actively migrate toward nanoparticles extravasated from the vessels, engulfing and redistributing them in the tumor stroma. The macrophages can carry the nanoparticles 2-5 times deeper in the tumor than passive diffusion. The amount of nanoparticles transported by the tumor-associated macrophages is size-dependent. Understanding the nanoparticle behavior after extravasation will provide strategies to engineer them to navigate the microenvironment for improved intratumoral targeting and therapeutic effectiveness.

Impact of Tumor Barriers on Nanoparticle Delivery to Macrophages.

The delivery of therapeutic nanoparticles to target cells is critical to their effectiveness. Here we quantified the impact of biological barriers on the delivery of nanoparticles to macrophages in two different tissues. We compared the delivery of gold nanoparticles to macrophages in the liver versus those in the tumor. We found that nanoparticle delivery to macrophages in the tumor was 75% less than to macrophages in the liver due to structural barriers. The tumor-associated macrophages took up more nanoparticles than Kupffer cells in the absence of barriers. Our results highlight the impact of biological barriers on nanoparticle delivery to cellular targets.

Specific Endothelial Cells Govern Nanoparticle Entry into Solid Tumors.

The successful delivery of nanoparticles to solid tumors depends on their ability to pass through blood vessels and into the tumor microenvironment. Here, we discovered a subset of tumor endothelial cells that facilitate nanoparticle transport into solid tumors. We named these cells nanoparticle transport endothelial cells (N-TECs). We show that only 21% of tumor endothelial cells located on a small number of vessels are involved in transporting nanoparticles into the tumor microenvironment. N-TECs have an increased expression of genes related to nanoparticle transport and vessel permeability compared to other tumor endothelial cells. The N-TECs act as gatekeepers that determine the entry point, distribution, cell accessibility, and number of nanoparticles that enter the tumor microenvironment.

A framework for designing delivery systems.

The delivery of medical agents to a specific diseased tissue or cell is critical for diagnosing and treating patients. Nanomaterials are promising vehicles to transport agents that include drugs, contrast agents, immunotherapies and gene editors. They can be engineered to have different physical and chemical properties that influence their interactions with their biological environments and delivery destinations. In this Review Article, we discuss nanoparticle delivery systems and how the biology of disease should inform their design. We propose developing a framework for building optimal delivery systems that uses nanoparticle-biological interaction data and computational analyses to guide future nanomaterial designs and delivery strategies.

The dose threshold for nanoparticle tumour delivery.

Nanoparticle delivery to solid tumours over the past ten years has stagnated at a median of 0.7% of the injected dose. Varying nanoparticle designs and strategies have yielded only minor improvements. Here we discovered a dose threshold for improving nanoparticle tumour delivery: 1 trillion nanoparticles in mice. Doses above this threshold overwhelmed Kupffer cell uptake rates, nonlinearly decreased liver clearance, prolonged circulation and increased nanoparticle tumour delivery. This enabled up to 12% tumour delivery efficiency and delivery to 93% of cells in tumours, and also improved the therapeutic efficacy of Caelyx/Doxil. This threshold was robust across different nanoparticle types, tumour models and studies across ten years of the literature. Our results have implications for human translation and highlight a simple, but powerful, principle for designing nanoparticle cancer treatments.

Liposome Imaging in Optically Cleared Tissues.

Three-dimensional (3D) optical microscopy can be used to understand and improve the delivery of nanomedicine. However, this approach cannot be performed for analyzing liposomes in tissues because the processing step to make tissues transparent for imaging typically removes the lipids. Here, we developed a tag, termed REMNANT, that enables 3D imaging of organic materials in biological tissues. We demonstrated the utility of this tag for the 3D mapping of liposomes in intact tissues. We also showed that the tag is able to monitor the release of entrapped therapeutic agents. We found that liposomes release their cargo >100-fold faster in tissues in vivo than in conventional in vitro assays. This allowed us to design a liposomal formulation with enhanced ability to kill tumor associated macrophages. Our development opens up new opportunities for studying the chemical properties and pharmacodynamics of administered organic materials in an intact biological environment. This approach provides insight into the in vivo behavior of degradable materials, where the newly discovered information can guide the engineering of the next generation of imaging and therapeutic agents.

Endothelialized collagen based pseudo-islets enables tuneable subcutaneous diabetes therapy.

Pancreatic islets are fragile cell clusters and many isolated islets are not suitable for transplantation. Furthermore, following transplantation, islets will experience a state of hypoxia and poor nutrient diffusion before revascularization, which is detrimental to islet survival; this is affected by islet size and health. Here we engineered tuneable size-controlled pseudo-islets created by dispersing de-aggregated islets in an endothelialized collagen scaffold. This supported subcutaneous engraftment, which returned streptozotocin-induced diabetic mice to normoglycemia. Whole-implant imaging after tissue clearing demonstrated pseudo-islets regenerated their vascular architecture and insulin-secreting ß-cells were within 5 µm of a perfusable vessel - a feature unique to this approach. By using an endothelialized collagen scaffold, this work highlights a novel "bottom-up" approach to islet engineering that provides control over the size and composition of the constructs, while enabling the critical ability to revascularize and engraft when transplanted into the clinically useful subcutaneous space.

The entry of nanoparticles into solid tumours.

The concept of nanoparticle transport through gaps between endothelial cells (inter-endothelial gaps) in the tumour blood vessel is a central paradigm in cancer nanomedicine. The size of these gaps was found to be up to 2,000 nm. This justified the development of nanoparticles to treat solid tumours as their size is small enough to extravasate and access the tumour microenvironment. Here we show that these inter-endothelial gaps are not responsible for the transport of nanoparticles into solid tumours. Instead, we found that up to 97% of nanoparticles enter tumours using an active process through endothelial cells. This result is derived from analysis of four different mouse models, three different types of human tumours, mathematical simulation and modelling, and two different types of imaging techniques. These results challenge our current rationale for developing cancer nanomedicine and suggest that understanding these active pathways will unlock strategies to enhance tumour accumulation.

Nanoparticle Size Influences Antigen Retention and Presentation in Lymph Node Follicles for Humoral Immunity.

Lymph node follicles capture and retain antigens to induce germinal centers and long-lived humoral immunity. However, control over antigen retention has been limited. Here we discovered that antigen conjugated to nanoparticle carriers of different sizes impacts the intralymph node transport and specific cell interaction. We found that follicular dendritic cell (FDC) networks determine the intralymph node follicle fate of these nanoparticles by clearing smaller ones (5-15 nm) within 48 h and retaining larger ones (50-100 nm) for over 5 weeks. The 50-100 nm-sized nanoparticles had 175-fold more delivery of antigen at the FDC dendrites, 5-fold enhanced humoral immune responses of germinal center B cell formation, and 5-fold more antigen-specific antibody production over 5-15 nm nanoparticles. Our results show that we can tune humoral immunity by simply manipulating the carrier size design to produce effectiveness of vaccines.

Assessing micrometastases as a target for nanoparticles using 3D microscopy and machine learning.

Metastasis of solid tumors is a key determinant of cancer patient survival. Targeting micrometastases using nanoparticles could offer a way to stop metastatic tumor growth before it causes excessive patient morbidity. However, nanoparticle delivery to micrometastases is difficult to investigate because micrometastases are small in size and lie deep within tissues. Here, we developed an imaging and image analysis workflow to analyze nanoparticle-cell interactions in metastatic tumors. This technique combines tissue clearing and 3D microscopy with machine learning-based image analysis to assess the physiology of micrometastases with single-cell resolution and quantify the delivery of nanoparticles within them. We show that nanoparticles access a higher proportion of cells in micrometastases (50% nanoparticle-positive cells) compared with primary tumors (17% nanoparticle-positive cells) because they reside close to blood vessels and require a small diffusion distance to reach all tumor cells. Furthermore, the high-throughput nature of our image analysis workflow allowed us to profile the physiology and nanoparticle delivery of 1,301 micrometastases. This enabled us to use machine learning-based modeling to predict nanoparticle delivery to individual micrometastases based on their physiology. Our imaging method allows researchers to measure nanoparticle delivery to micrometastases and highlights an opportunity to target micrometastases with nanoparticles. The development of models to predict nanoparticle delivery based on micrometastasis physiology could enable personalized treatments based on the specific physiology of a patient's micrometastases.

Elimination Pathways of Nanoparticles.

Understanding how nanoparticles are eliminated from the body is required for their successful clinical translation. Many promising nanoparticle formulations for in vivo medical applications are large (>5.5 nm) and nonbiodegradable, so they cannot be eliminated renally. A proposed pathway for these nanoparticles is hepatobiliary elimination, but their transport has not been well-studied. Here, we explored the barriers that determined the elimination of nanoparticles through the hepatobiliary route. The route of hepatobiliary elimination is usually through the following pathway: (1) liver sinusoid, (2) space of Disse, (3) hepatocytes, (4) bile ducts, (5) intestines, and (6) out of the body. We discovered that the interaction of nanoparticles with liver nonparenchymal cells ( e. g., Kupffer cells and liver sinusoidal endothelial cells) determines the elimination fate. Each step in the route contains cells that can sequester and chemically or physically alter the nanoparticles, which influences their fecal elimination. We showed that the removal of Kupffer cells increased fecal elimination by >10 times. Combining our results with those of prior studies, we can start to build a systematic view of nanoparticle elimination pathways as it relates to particle size and other design parameters. This is critical to engineering medically useful and translatable nanotechnologies.

Three-Dimensional Imaging of Transparent Tissues via Metal Nanoparticle Labeling.

Chemical probes are key components of the bioimaging toolbox, as they label biomolecules in cells and tissues. The new challenge in bioimaging is to design chemical probes for three-dimensional (3D) tissue imaging. In this work, we discovered that light scattering of metal nanoparticles can provide 3D imaging contrast in intact and transparent tissues. The nanoparticles can act as a template for the chemical growth of a metal layer to further enhance the scattering signal. The use of chemically grown nanoparticles in whole tissues can amplify the scattering to produce a 1.4 million-fold greater photon yield than obtained using common fluorophores. These probes are non-photobleaching and can be used alongside fluorophores without interference. We demonstrated three distinct biomedical applications: (a) molecular imaging of blood vessels, (b) tracking of nanodrug carriers in tumors, and (c) mapping of lesions and immune cells in a multiple sclerosis mouse model. Our strategy establishes a distinct yet complementary set of imaging probes for understanding disease mechanisms in three dimensions.