MENU

Click here to

×

Are you sure ?

Yes, do it No, cancel

Effect of Eye Torsion On Setup Accuracy and Planning Dosimetry in the Ocular Melanoma Proton Treatment

M Mamalui1*, M Rutenberg1, J Bolling2, R Dagan1, (1) University of Florida/Radiation Oncology, Jacksonville, FL, (2) Mayo Jacksonville, FL

Presentations

(Sunday, 7/12/2020)   [Eastern Time (GMT-4)]

Room: AAPM ePoster Library

Purpose: Torsional eye movement is an eye rotation about the line of sight (i.e.,line connecting macula/fovea with fixation target).When eye makes any horizontal/vertical gaze shift, line of sight shifts as well. We find 6-10 degrees torsional differences between simulation and treatment in 20% of our cases.
Our aim is twofold: a)arrive at a torsional tolerance number that ensures robust GTV coverage based on our existent treatment margins;b)recommend patient registration method in cases where unaccounted torsion exists.
Methods: analysis: I.15 anonymized plans were selected to perform simulated torsion ±[3,5,8,10,12]degrees sensitivity analysis based on tumor location: tumor around optic disk/macula area (5); tumor around equator(5); tumor anterior to equator(5).TumorDVHtorsion was evaluated vs TumorDVHnominal. Torsion tolerance was extracted based on BooleanAND of 2conditions:1)|TumorDVHtorsion-TumorDVHnominal|<5% at 95% prescription Dose;2) not more than5%difference in dose covering 100%of TargetGTV surface area, both for one fraction out of 4(note we plan to target GTV).
II.To recommend setup method when torsion is present, we analyzed 15x4daily image pairs for 15selected plans: a)created 'perfect setup' plan (reflects true eye position during treatment);b)create 2types of registrations for the treated plan:1)based on modeling patient's daily torsion; 2)identifying&favoring clips closest to the tumor; c)evaluate both registrations vs 'perfect setup'plan; d)evaluate registrations vs each other.
Results: analysis shows 8-10degrees torsion tolerance can be used for most posterior/anterior tumors (irregular/strongly asymmetric shapes not included). 5 degree tolerance is required for targets around eye equator. Registering planned clip geometry while identifying&favoring clips closest to the tumor shows superior TumorDVH difference vs positioning using torted geometry.
Conclusions:To the best of our knowledge, this is the first systematic investigation of the torsional sensitivity/robustness of the ocular proton planning. Torsional tolerance of 5 degrees to ensure robust GTV coverage was derived. Preferable patient registration method based on identifying&favoring nearest clips to tumor(when unaccounted acquired torsion exists)was established.

Keywords

Not Applicable / None Entered.

Taxonomy

Not Applicable / None Entered.

Contact Email