Room: AAPM ePoster Library
No Action Level (NAL) protocol to manage physician presence effectively during MRI guided online adaptive radiation therapy for head and neck cancer.
MRI guided online adaptive radiotherapy on the Elekta Unity platform presents two scenarios during clinical workflow that require physician verification/approval: Image Fusion between the reference (CT/MR) and the daily MR image and subsequent approval of adapted treatment plan. We reviewed daily adaptation results from 9 head and neck cancer patient treatments, (total of 241 fractions, while aspiring to establish an action level protocol to manage physician presence daily at the MR Linac effectively. Typical treatment prescription for head and neck cancer patients included doses as high as 2.2 Gy per fraction up to a total of 33 fractions. Adaptation results included (a) daily patient shift information based on image fusion and (b) dosimetric criteria for adaptive plan generation.
For the Adapt to Position (ATP) workflow: overall mean error in patient shift based on online image fusion was -0.03, 0.07 and 0.01 cm (RL, SI, AP). SD of systematic error was 0.33. 0.26 and 0.14 cm (RL, SI, AP). SD of random error was 0.17, 0.29 and 0.2 cm (RL, SI, AP). All dosimetric criteria were usually always met (within the 2 % underdose or overdose levels for target and OARs respectively) during online plan adaptation for shifts in the ranges specified above.
An action level threshold of 0.5 cm patient shift (along any direction) was deemed to be acceptable as long as all dosimetric criteria were satisfied following plan adaptation. Physician presence at the MR Linac was only required if either shifts exceeded 0.5 cm or if any of the dosimetric criteria exceeded 2% tolerance levels on more than three subsequent fractions.
Funding Support, Disclosures, and Conflict of Interest: Some of the co-authors in this work receive corporate funding support from Elekta Inc.
MR, Radiation Therapy, Image-guided Therapy
IM/TH- MRI in Radiation Therapy: MRI/Linear accelerator combined- IGRT and tracking