Although cancer is driven by mutations in individual cells, tumors are complex environments, resembling organs more than simple clusters of cells. Recent studies have shown that the microenvironment within tumors is wildly heterogeneous, owing to tumors with diverse genetic profiles as well as interactions between the tumor parenchymal cells and the surrounding stromal cells. This heterogeneity is linked to the malignancy of tumors and their resistance to therapy. Despite the importance of the heterogeneous tumor microenvironment, there are no tools that can adequately visualize its dynamics in vivo. This proposal seeks to develop a new imaging technology - super-localization ultrasound imaging - which relies on highly dynamic phase-change nanoparticles to create a high resolution map of molecular expressions within tumors. The phase-change contrast agents, termed laser-activated nanodetectors (LANDs), consist of a lipid monolayer stabilizing a liquid perfluorohexane core with an encapsulated dye. Upon irradiation with a pulsed laser, the LANDs undergo a rapid vaporization and a transient microbubble is formed, resulting in an increased signal in ultrasound imaging. As the perfluorohexane cools below its boiling point (56 C) the LANDs recondense into their stable droplet state. The time of recondensation is a stochastic process, and it depends on the droplet size, incident interrogating ultrasound energy, and the local tissue environment. Therefore, high frame rate ultrasound imaging is able to capture single recondensation events. The difference between successive ultrasound images reveals the response of the imaging system to only a single activated LAND. After fitting this response to the point spread function of the imaging system, the exact position of the LAND can be localized with much greater precision than the system's diffraction/bandwidth limited resolution. After conjugating the LANDs with molecular-specific antibodies, a high-resolution map of the molecular expressions in tissue can be obtained. This application focuses on the development and optimization of the LANDs as well as the associated ultrasound imaging techniques. Molecular targeting to the epidermal growth factor receptor will be achieved via a directional conjugation to a monoclonal antibody. The particles will be synthesized and fully characterized with dynamic light scattering, ultraviolet-visible spectrophotometry, and transmission electron microscopy. Simultaneously, custom ultrasound imaging sequences will be developed and optimized to detect the recondensation of individual LANDs with high sensitivity. The overall improvement in resolution will be fully characterized in tissue-mimicking phantoms. Finally, in vivo studies with a xenograft mouse model of cancer will be used to study the safety of the technique, the delivery of LANDs to the tumor, the molecular specificity, and the ability of super-localization ultrasound imaging to detect small spatial variations in a heterogeneous tumor. The end result will be a versatile framework that can be used to study a wide variety of molecular expressions in tumors in real-time and in vivo. This has broad implications ranging from furthering our understanding of the progression of cancer to personalized prediction of treatment response in the clinic.
Cancer remains a leading cause of death worldwide owing, in part, to resistance to treatment by small pockets of cells. New tools that can help researchers understand the complex tumor microenvironment could aid in the development of new therapeutics and predict the success of therapeutic strategies.
