SUBJECT: Ph.D. Proposal Presentation
BY: Keith Griffin
TIME: Wednesday, December 2, 2020, 9:00 a.m.
PLACE: Online, Webex
TITLE: Methodology for Rapid Reconstruction of Neutron Dose for Pediatric Patients Undergoing Passive Scattering Proton Therapy Treatment
COMMITTEE: Dr. Nolan E. Hertel, Co-Chair (NRE)
Dr. Choonsik Lee, Co-Chair (NRE (National Cancer Institute))
Dr. Eric S. Elder (NRE/MP)
Dr. Chris K. Wang (NRE/MP)
Dr. Harald Paganetti (Massachusetts General Hospital)


The National Cancer Institute, in collaboration with Massachusetts General Hospital, is currently establishing the first pediatric proton therapy cohort study to quantify the risk of subsequent malignancies (second primary cancers) following proton therapy as compared to photon therapy. 10,000 pediatric patients treated using proton therapy will be compared to 10,000 patients with similar diagnoses treated using photon therapy. The study will leverage the use of electronic medical records, which hold treatment information, such as patient CT images and treatment plans, to connect the risks of second cancers and mortality to site-specific radiation doses received by the patient. However, current proton therapy treatment planning systems only calculate in-field and near-field proton dose and not the out-of-field dose from secondary neutrons, which may be generated within the treatment nozzle or within the patient's own body. The purpose of my dissertation will be to address underlying out-of-field dosimetry challenges currently faced by our working group, especially for the passive scattering patients. The Monte Carlo simulation package most commonly used in particle therapy, TOPAS, will be compared against other Monte Carlo codes, such as MCNP6 and PHITS, to determine the impact of high-energy physics modeling choices on estimated out-of-field dose. Incomplete (partial anatomy) CT image sets will be extended to full anatomy by developing methods to estimate out-of-field patient anatomy when unavailable for a cohort member. Finally, methodology will be developed that simulates the full passive scattering beamline using treatment planning information retrieved for specific patients to estimate their out-of-field doses. It is currently estimated that full Monte Carlo calculations will not be feasible for each cohort member due to long computation times; therefore, we will also investigate potential methods to accelerate the out-of-field dose estimation and compare to the full simulation results from before. The generated dosimetry data from NCI and MGH will become a rich future resource for multiple outside groups to study the late effects related to radiotherapy exposure in pediatric cancer patients.

Webex link: