ESTRO 2022

Session Item

Implementation of new technology and techniques

Session Code: 7002

Session Type: Poster (digital)

Track: Physics

Mechanical and dosimetric accuracy of Dynamic Trajectory Radiotherapy delivery on a C-arm linac

Jenny Bertholet, Switzerland

Presentation Number: PO-1683

Abstract

Abstract Title:

Mechanical and dosimetric accuracy of Dynamic Trajectory Radiotherapy delivery on a C-arm linac

Authors:

Jenny Bertholet¹, Paul-Henry Mackeprang¹, Hannes L. Loebner¹, Silvan Mueller¹, Gian Guyer¹, Yanick Wyss¹, Daniel Frei¹, Werner Volken¹, Olgun Elicin¹, Daniel M Aebersold¹, Michael K Fix¹, Peter Manser¹

¹Inselspital, Bern University Hospital and University of Bern, Division of Medical Radiation Physics and Department of Radiation Oncology, Bern, Switzerland

Show Affiliations

Purpose or Objective

Dynamic Trajectory Radiotherapy (DTRT) extends volumetric modulated arc therapy (VMAT) with table and collimator rotation during beam-on. This technique has shown improved healthy tissue sparing for equivalent target coverage compared to VMAT in head and neck (HN) radiotherapy. This study demonstrates the deliverability of DTRT on a C-arm linac, reporting on the mechanical and dosimetric accuracy.

Material and Methods

DTRT plans were created for different HN cases on an anthropomorphic phantom. Six cases were had sequential boost resulting in 13 plans with 2 Gy/fraction and one case was a single vocal cord irradiation (SVCI) with 3.63 Gy/fraction resulting in a total of 27 trajectories (1 trajectory corresponds to 1 full gantry rotation). Gantry-table-collimator paths were determined by an A* path-searching algorithm minimizing beam’s eye view OAR-target overlap within the collision-free space determined using a case-specific collision model. Maximum gradients of 3° table or collimator rotation per degree gantry rotation were allowed and paths were smoothed using a 10-points (20°) moving average to avoid abrupt table motion.

All trajectories were delivered on the phantom using developer mode on a TrueBeam linac (Varian Medical Systems) equipped with a 120-leaf MLC and a PerfectPitch 6-degree-of-freedom table. All machine log-files were recorded in order to assess the mechanical deviations for all dynamic axes, calculated as the difference between expected and actual values. Correlation between speed and deviation was evaluated for gantry, table and collimator angles.

Dosimetric validation was carried out with transversal plane film measurements on the phantom for one case (2 plans, 4 trajectories).

Results

All plans were delivered successfully without interlock or collision.

The mean delivery time was 2.4 minutes (range: 1.9-2.9 minutes) per trajectory. The root-mean-square (RMS) deviation were 0.02°, 0.12° and 0.03° for gantry, table and collimator angles, respectively. Maximum deviations were 0.13°, 0.16° and 0.17° for gantry, table and collimator angles, respectively.

The Pearson’s correlation coefficient between speed and deviation was high (negative) for table and collimator angles (<-0.99, p<<0.01) but low for gantry angle (0.16, p<<0.01). Figure 1 shows example trajectories and deviations for gantry, table and collimator angles for one plan. Although gantry angle deviations appear higher when there is a change in direction for table or collimator rotation, correlation between deviation and speed or acceleration of any component was low.

The mean RMS deviation for all moving MLC leaves was 0.17 mm (maximum RMS deviation: 0.27 mm).

The passing rates between measured dose on film and calculated dose were 93.9% and 95.8% (global gamma, 2%/2mm, 10% dose threshold).

Conclusion

DTRT plans for HN radiotherapy are deliverable with high mechanical and dosimetric accuracy on a TrueBeam linac and with clinically acceptable delivery times.

This work was supported by Varian Medical Systems.