Room: Exhibit Hall | Forum 3
Purpose: To present a method to select the treatment isocenter that minimizes the volume of normal tissue treated during multi-target single-isocenter stereotactic radiosurgery (SRS).
Methods: A statistical framework was developed to quantify the impact of uncertainties on planning target volumes (PTV) for the treatment of multiple metastatic brain tumors with a single treatment isocenter. Shifts in target position due to translational and rotational errors were modeled with independent, zero mean, normal distributions (0.5 mm, 0.5°) in three dimensions added in quadrature. Simulations were performed to determine the volume of normal tissue treated due to margin expansions required to maintain a given probability (95% at 2σ) of target coverage. Tumors were modeled as spherical objects with diameters ranging from 4-40 mm. The separation distance between lesions varied from 40-100 mm. For the two-lesion case, the percent increase in PTV was determined relative to an isocenter placed midway between each lesion for three scenarios including the optimal isocenter that minimized normal tissue volume treated, as well as isocenters at the target’s center-of-mass (COM) and center-of-surface-area (CSA). The three-lesion case was also investigated.
Results: Isocenter placement midway between targets of different sizes is not optimal for multi-target SRS. For two targets, optimal placement of the isocenter reduced the margin volume by up to 17.8% for the range of scenarios considered. Placing the isocenter at the COM or CSA of two lesions resulted in maximum differences of 3.4% and 1.1%, respectively, versus the optimal solution. CSA closely approximated the optimal isocenter location for the three-target case.
Conclusion: Center-of-mass and center-of-surface-area provide useful approximations to select the optimal treatment isocenter for multi-target single-isocenter SRS. When the ratio of the maximum dimension of a large lesion to a small lesion is less than 3, CSA provides a more accurate approximation.