PROJECT SUMMARY Stereotactic Body Radiotherapy (SBRT) has been established as an effective, safe, and feasible first-line option in lung cancer. The quest for dose escalation to increase tumor control and survival benefit, and simultaneously to minimize normal tissue complication, requires precise radiation delivery to a region including the cancerous target and the smallest possible ring region around the target (called margin) to ensure sufficient dose to the target in the presence of positioning uncertainty, mainly caused by respiratory motion. This is especially critical for SBRT, as it is vulnerable to geometry errors due to high dose per fraction, a steep dose gradient, and the long delivery time that increases chances of changes in motion. Pre-treatment imaging technologies have been used to aid the position of tumor against the treatment beam. Yet, a margin size of 6- 10 mm is still routinely used to accommodate systematic and random changes in respiratory motion magnitude, baseline, and period. Consequently, high dose (close to prescription level) is delivered to a large volume of normal tissue, yielding toxicity concern. There is a strong, but unmet need for margin reduction in lung SBRT. The ultimate form of margin reduction is 4D radiotherapy (RT) that modifies multi-leaf collimator positions in real time in response to tumor motion to always direct radiation to the tumor. Knowing tumor position in real time is the prerequisite for 4D RT. Nonetheless, to date, no approach can provide accurate and reliable 3D tumor tracking. To address this problem, our group proposed in 2016 a tumor tracking scheme by measuring scattered x-ray photons and lately demonstrated its potential feasibility using state-of-the-art photon counting detection technology. The overall goal of this proposal is to develop and translate a prototype system with appropriate capability for 3D tumor tracking, to evaluate its performance in phantom and patient studies, and to demonstrate its clinical impacts, by jointing the strong and complementary expertise of teams at Johns Hopkins University, Massachusetts General Hospital, and Varex Imaging. We will pursue four specific aims (SAs). SA1. Develop a prototype hardware system to image tumor region with scattered kV photons. SA2. Develop a software system for data processing pipeline and tumor tracking. SA3. Integrate the developed systems to form a prototype and perform phantom experiments to characterize its performance. SA4. Implement the system in clinic, perform patient studies to demonstrate feasibility and advantages of real-time 3D tumor tracking, and translate it to end users. The innovation of this project is a novel system measuring scattered kV x-ray photons to image tumor regions and using it to track tumor motion. Deliverability is ensured by preliminary studies and the team with complementary expertise. Our project will overcome a major hurdle in 4D RT, enabling margin reduction in lung cancer SBRT and henc...