Dosimetrically motivated beam-angle optimization for non-coplanar arc radiotherapy with and without dynamic collimator rotation.

Bertholet, Jenny; Zhu, Chengchen; Guyer, Gian; Mueller, Silvan; Volken, Werner; Mackeprang, Paul-Henry; Loebner, Hannes A; Stampanoni, Marco F M; Aebersold, Daniel M; Fix, Michael K; Manser, Peter (2024). Dosimetrically motivated beam-angle optimization for non-coplanar arc radiotherapy with and without dynamic collimator rotation. Medical physics, 51(2), pp. 1326-1339. American Association of Physicists in Medicine AAPM 10.1002/mp.16899

[img]
Preview
Text
Medical_Physics_-_2023_-_Bertholet_-_Dosimetrically_motivated_beam_angle_optimization_for_non_coplanar_arc_radiotherapy.pdf - Published Version
Available under License Creative Commons: Attribution-Noncommercial-No Derivative Works (CC-BY-NC-ND).

Download (6MB) | Preview

BACKGROUND

Non-coplanar techniques have shown to improve the achievable dose distribution compared to standard coplanar techniques for multiple treatment sites but finding optimal beam directions is challenging. Dynamic collimator trajectory radiotherapy (colli-DTRT) is a new intensity modulated radiotherapy technique that uses non-coplanar partial arcs and dynamic collimator rotation.

PURPOSE

To solve the beam angle optimization (BAO) problem for colli-DTRT and non-coplanar VMAT (NC-VMAT) by determining the table-angle and the gantry-angle ranges of the partial arcs through iterative 4π fluence map optimization (FMO) and beam direction elimination.

METHODS

BAO considers all available beam directions sampled on a gantry-table map with the collimator angle aligned to the superior-inferior axis (colli-DTRT) or static (NC-VMAT). First, FMO is performed, and beam directions are scored based on their contributions to the objective function. The map is thresholded to remove the least contributing beam directions, and arc candidates are formed by adjacent beam directions with the same table angle. Next, FMO and arc candidate trimming, based on objective function penalty score, is performed iteratively until a desired total gantry angle range is reached. Direct aperture optimization on the final set of colli-DTRT or NC-VMAT arcs generates deliverable plans. colli-DTRT and NC-VMAT plans were created for seven clinically-motivated cases with targets in the head and neck (two cases), brain, esophagus, lung, breast, and prostate. colli-DTRT and NC-VMAT were compared to coplanar VMAT plans as well as to class-solution non-coplanar VMAT plans for the brain and head and neck cases. Dosimetric validation was performed for one colli-DTRT (head and neck) and one NC-VMAT (breast) plan using film measurements.

RESULTS

Target coverage and conformity was similar for all techniques. colli-DTRT and NC-VMAT plans had improved dosimetric performance compared to coplanar VMAT for all treatment sites except prostate where all techniques were equivalent. For the head and neck and brain cases, mean dose reduction-in percentage of the prescription dose-to parallel organs was on average 0.7% (colli-DTRT), 0.8% (NC-VMAT) and 0.4% (class-solution) compared to VMAT. The reduction in D2% for the serial organs was on average 1.7% (colli-DTRT), 2.0% (NC-VMAT) and 0.9% (class-solution). For the esophagus, lung, and breast cases, mean dose reduction to parallel organs was on average 0.2% (colli-DTRT) and 0.3% (NC-VMAT) compared to VMAT. The reduction in D2% for the serial organs was on average 1.3% (colli-DTRT) and 0.9% (NC-VMAT). Estimated delivery times for colli-DTRT and NC-VMAT were below 4 min for a full gantry angle range of 720°, including transitions between arcs, except for the brain case where multiple arcs covered the whole table angle range. These times are in the same order as the class-solution for the head and neck and brain cases. Total optimization times were 25%-107% longer for colli-DTRT, including BAO, compared to VMAT.

CONCLUSIONS

We successfully developed dosimetrically motivated BAO for colli-DTRT and NC-VMAT treatment planning. colli-DTRT and NC-VMAT are applicable to multiple treatment sites, including body sites, with beneficial or equivalent dosimetric performances compared to coplanar VMAT and reasonable delivery times.

Item Type:

Journal Article (Original Article)

Division/Institute:

04 Faculty of Medicine > Department of Haematology, Oncology, Infectious Diseases, Laboratory Medicine and Hospital Pharmacy (DOLS) > Clinic of Radiation Oncology > Medical Radiation Physics
04 Faculty of Medicine > Department of Haematology, Oncology, Infectious Diseases, Laboratory Medicine and Hospital Pharmacy (DOLS) > Clinic of Radiation Oncology

UniBE Contributor:

Zhu, Chengchen, Guyer, Gian Mauro Carlo, Müller, Silvan Andreas, Volken, Werner, Mackeprang, Paul-Henry, Löbner, Hannes Anton, Aebersold, Daniel Matthias, Fix, Michael, Manser, Peter

Subjects:

500 Science > 530 Physics
600 Technology > 610 Medicine & health

ISSN:

0094-2405

Publisher:

American Association of Physicists in Medicine AAPM

Language:

English

Submitter:

Pubmed Import

Date Deposited:

27 Dec 2023 11:01

Last Modified:

10 Feb 2024 00:15

Publisher DOI:

10.1002/mp.16899

PubMed ID:

38131614

Uncontrolled Keywords:

beam angle optimization dynamic trajectory radiotherapy non-coplanar radiotherapy

BORIS DOI:

10.48350/190732

URI:

https://boris.unibe.ch/id/eprint/190732

Actions (login required)

Edit item Edit item
Provide Feedback