Abstract
Ultra-high dose rate (UHDR) radiation delivery, termed FLASH radiotherapy (FLASH-RT), has the potential to
reduce normal tissue damage, with significant clinical cancer treatment ramifications. Current evidence suggests
that FLASH-RT reduces functional damage to normal brain, colon, lung, and skin, at the same dose values.
Although controversial, some studies show this reduction in tissue damage as great as 40%, despite tissue type
alpha/beta ratio and total dose remain significant unknown factors. Most surprisingly tumor tissue appears
unaffected by FLASH radiation sparing, suggesting that the therapeutic ratio achieved through FLASH-RT is
higher than conventional RT. The potential to achieve higher therapeutic ratios make FLASH-RT very attractive
for future human use to address radioresistant tumors that would otherwise result in excessive radiotoxicity
during conventional treatment. Despite clinical translation of FLASH-RT to large animal and human studies being
urgently warranted, the majority of the preclinical work for FLASH-RT remains in small animals, where concerns
regarding dose and dose rate inhomogeneities cannot be adequately assessed. A major barrier towards
translational FLASH-RT is that optimization of treatment plans for large animals and humans with larger/deeper
tissues and complicated geometries remains challenging. In order to demonstrate the hypothesized superiority
of FLASH-RT, it is essential to conduct the studies FLASH-RT studies under controlled conditions and in clinically
realistic workflows. This begins with a treatment planning system {TPS) capable of generating and delivering
comparable plans for FLASH/CONV-RT within a minimally modified clinical setting. In the current proposal, we
are committed to developing such a TPS for the first time and making our software available to the scientific
community. We will achieve this by first generating beam models for FLASH-capable linacs. These beam models
will then be used to implement advanced planning and delivery technology utilizing passive intensity modulation.
Next, we will develop tumor control probability and normal tissue complication probability models deployable on
the TPS. Lastly, our work will culminate in demonstrating successful deliveries of optimized plans with
comprehensive characterization of plan delivery using quality assurance systems uniquely available at
Dartmouth. Due to the urgency to transfer this paradigm-shifting therapy, high-risk of the first initiative, the
exploratory, and developmental nature of the proposal, the R21 grant mechanism is leveraged and justified. The
successful completion of the proposed work will advance the state-of-the-art in FLASH-RT by bridging the
extensive gap between the basic science of FLASH and the clinical needs of FLASH for translation to human
trials. The team has leading expertise in radiation physics, informatics, and radiation oncology, and are supported
by existing FLASH-capable linacs, TPS prototypes, unique UHDR dosimetry technologies invented at
Dartmouth, canine pilot trials, future human trials in planning and leading industrial partners. Taken together, the
team is ideally poised to conduct the proposed work of exploratory and developmental nature.