Dosimetric Analysis of Intra-Fraction Motion Detected by Surface Guided Radiation Therapy During Linac Stereotactic Radiosurgery

Abstract

Stereotactic radiosurgery (SRS) requires very sharp dose gradients, high conformality and typically uses minimal expansion margins. Immobilization with an open face mask is more comfortable and less invasive than frame-based immobilization but concerns about intra-fraction motion must be addressed. Surface guided radiation therapy (SGRT) is an attractive option for intra-fraction patient monitoring because it is continuous, has sub-millimeter accuracy and uses no ionizing radiation. The purpose of this study was to investigate the dosimetric consequences of uncorrected intra-fraction patient motion detected during frameless Linac-based SRS.All SRS patients (pts) were immobilized in an SRS-specific open face thermoplastic mask with bite plate and were monitored during treatment using SGRT with tolerances of 1 mm translations and 1° rotations; the treatment beam automatically turned off if these tolerances were exceeded. SRS pts received 18 or 21 Gy in 1 fraction or 27 Gy in 3 fractions based on tumor size and location. GTV to PTV margins ranged from 0 to 3 mm. Twelve pts were treated with circular cones and 13 were treated with MLC-based plans with the technique chosen based on PTV size. If the SGRT system detected motion greater than tolerance during treatment, conebeam CT(CBCT) was repeated and the necessary shifts were made prior to continuing treatment. For the 25 pts with intra-fraction 3D vector shifts of at least 1 mm, we offset the isocenter in the treatment planning system using the translational shifts from the repeat CBCT and re-calculated the plans with the planned MU to determine the dosimetric effect the shift would have caused had it not been detected. The GTV and PTV coverage and normal brain V12 for the shifted plans were compared to the planned values. Wilcoxon signed rank tests were used to compare planned and shifted dosimetric indices and median 2 sample tests were used to investigate these differences between cone and MLC plans.From the recalculated plans, the resulting median V12 increased by 0.24 cc (P = 0.006) and the median GTV and PTV coverage decreased 9% (P < 0.001) and 10.2 % (P < 0.001), respectively. PTV coverage decreased more for shifted cone plans than for shifted MLC plans (11.6% vs 4.7%, P = 0.011) but the median V12 differences were found to be significantly larger for MLC plans (-0.34 cc vs -0.01 cc, P = 0.011), due to the sharper dose gradients and higher conformality for the cone plans compared to the MLC plans. Differences in GTV coverage between cone and MLC plans were not statistically significant.SGRT detected clinically meaningful intra-fraction motion during frameless SRS, which could potentially lead to large underdoses and increased normal brain dose if not corrected. Dosimetric differences between planned and shifted plans were statistically significant.R.D. Foster: None. B.J. Moeller: Independent Contractor; Novant Health. Partner; SERO. Stock; Moderna. M. Robinson: None. M. Bright: SEAAPM.J.L. Ruiz: Stock; Obseva USA, Inc. C.J. Hampton: None. J.H. Heinzerling: Research Grant; Vision RT, Inc, AstraZeneca. Honoraria; Vision RT, Inc, AstraZeneca. Consultant; Vision RT, Inc.