investigator_user investigator user funding collaborators pending menu bell message arrow_up arrow_down filter layers globe marker add arrow close download edit facebook info linkedin minus plus save share search sort twitter remove user-plus user-minus
  • Project leads
  • Collaborators

Personalized Motion Management for truly 4D Lung Stereotactic Body Radiotherapy

Amit Sawant

1 Collaborator(s)

Funding source

National Institutes of Health (NIH)
Respiratory motion causes significant geometric and dosimetric uncertainties in lung cancer radiotherapy (RT). The impact of such uncertainties is amplified in hypofractionated regimens such as stereotactic body radiotherapy (SBRT), where very high, potent doses are delivered in relatively few fractions. Lung SBRT achieves excellent local control (>80%) but also shows significant collateral toxicity (10 - 28%). Several clinical studies have reported a strong correlation between toxicity and radiation dose. Thoracic anatomy changes continuously in all four dimensions (4D=3D+time) from cycle-to-cycle and day-to- day. A common limitation of current motion management techniques is that they discard large amounts of this 4D information and do not capture nor adequately account for cycle-to-cycle variations. We hypothesize that completely accounting for all four dimensions at each RT step will significantly improve dose-sparing and, consequently, lead to reduced toxicity. In response to PAR-10-169, we form a multidisciplinary academic-industrial collaboration between UT Southwestern Medical Center (UTSW), University of Utah (Utah), University of Maryland (UMD), Varian Medical Systems and VisionRT. Our goal is to create a comprehensive 4DRT motion management solution that achieves e50% dose-sparing of serial structures and 30-50% more sparing of normal lung compared to current clinical lung SBRT. Towards this goal, we present a systematic, hypothesis-driven research plan. In Aim 1, we will investigate a novel binning-free maximum a posteriori (MAP) 4DCT reconstruction. The 4DCT will be parameterized by real-time surface photogrammetry (VisionRT) to create a high-spatiotemporal- resolution 4D motion model that describes the internal volume as a function of external surface over several respiratory cycles. The VisionRT system is installed in the CT-simulation room as well as the treatment room, thus serving as a common link between the CT-sim and the dose delivery stages. In Aim 2, we will investigate 4D optimization to create deliverable treatment plans that account for motion over multiple respiratory cycles. We will also investigate the novel concept of using motion as an additional degree of freedom rather than a constraint. In Aim 3, we will investigate real-time beam adaptation using multileaf collimator (MLC) tracking. This technique will reshape the beam so as to follow all of the complex changes (translation, rotation and deformation) of the tumor and surrounding organs. We will investigate closed-loop RT via a voxel-level dosimetric reconstruction of each delivered fraction; to be used for verification and, f necessary, for daily replanning. Our industrial partners will incorporate our research findings int two research 4DRT prototypes which will be deployed at UTSW and UMD for end-user validation. Validation will be performed using a deformable lung motion phantom and data from lung cancer patients. The latter will consist of 4DCT, and surface tracking data and in-room kV x-ray fluoroscopy. Finally, we will form physician-physicist teams to develop practice guidelines, quality assurance and education frameworks to facilitate clinical translation.

Related projects