MR-Linac Radiotherapy - The Beam Angle Selection Problem.

Area of Science:

• Medical Physics
• Radiation Oncology
• Radiotherapy Planning

Background:

• Volumetric Modulated Arc Therapy (VMAT) has reduced the need for static-beam IMRT beam angle optimization.
• The unavailability of VMAT in MR-linac systems necessitates re-evaluating optimal beam angle selection for IMRT.
• Rectal cancer serves as a model for investigating automated beam angle optimization strategies.

Purpose of the Study:

• To investigate computer-aided generation of beam angle class solutions (CS) as a replacement for patient-specific beam angle optimization (BAO).
• To assess the plan quality and efficiency of CS compared to BAO and equi-angular setups.
• To evaluate the potential of CS for simplifying clinical workflows in MR-linac treatment planning.

Main Methods:

• 23 rectal cancer patients treated on a Unity MR-linac were analyzed.
• Beam Angle Optimization (BAO) plans with 7-12 beams were generated.
• Class solutions (CS) were derived from BAO plans and compared with BAO and equi-angular (EQUI) plans.

Main Results:

• CS plans demonstrated highly similar quality to BAO plans for >7 beams.
• Both CS and BAO plans were superior to equi-angular setups.
• CS₉ reduced bowel/bladder mean dose by 22% compared to EQUI₉; EQUI required double the beams for comparable quality.

Conclusions:

• Computer-generated beam angle class solutions (CS) can replace individualized BAO without compromising plan quality.
• CS reduces planning complexity, calculation times, and simplifies clinical workflow for MR-linac treatments.
• CS and BAO significantly outperform equi-angular setups, and CS can prevent time-consuming re-optimization in adaptive treatments.