Impact of Lesion Morphology and Separation On One- Versus Two-Isocenter Frameless Radiosurgery
J Roper*, V Chanyavanich, G Betzel, A Dhabaan, Winship Cancer Institute of Emory University, Atlanta, GASU-E-T-685 Sunday 3:00PM - 6:00PM Room: Exhibit Hall
Purpose: To investigate the dosimetric and localization effects of one- versus two-isocenter SRS.
Methods: A library of 50 retrospective patient plans was compiled for cases of two brain lesions. One- and two-isocenter plans were generated. One-isocenter plans consisted of four non-coplanar-VMAT arcs; isocenter was positioned midway between two lesions. In the two-isocenter approach, each lesion was planned separately using four non-coplanar DCA. DCA plans were summed. All plans were normalized so that 21 Gy covers 100% of the PTVs.
Dosimetry and localization errors were studied for one- and two-isocenter plans. Normal tissue dose, V12, was evaluated as a function of lesion morphology and separation distance. Plans were exported to Velocity, where localization errors were simulated using rotations of 0°, 0.5°, 1°, and 2° about each isocenter. The effect of rotational misalignments on PTV coverage was compared for D95 and minimum dose.
Results: Findings from 10 cases are reported. Analysis of one- versus two-isocenter plans shows that the V12 is comparable at lesion separations of 8 cm or less, while V12 is higher for one-isocenter plans at greater separations. As lesion volume increases, V12 increases more for one-isocenter plans.
The effects of localization errors on target coverage were investigated. For 2° rotational errors on one-isocenter plans, D95 decreases on average by 7%, whereas the decrease for two-isocenter plans is 0.5%. Minimum PTV dose decreases more substantially: 12% versus 1% respectively for the one- and two-isocenter plans. The above values are averaged over all lesion separations. For the largest separation of 15 cm, minimum PTV dose decreases by 37% for the one-isocenter plan.
Conclusions: Localization becomes more important as separation increases for one-isocenter SRS. Findings suggest that lesion morphology and separation should be considered when deciding between one- and two-isocenter SRS because of the effects on normal tissue dose and target coverage.