A comprehensive formulation for volumetric modulated arc therapy planning.

Area of Science:

• Radiation Oncology
• Medical Physics
• Computational Biology

Background:

• Volumetric Modulated Arc Therapy (VMAT) is a standard radiation therapy technique.
• Current VMAT optimization uses greedy heuristics, potentially compromising plan consistency and quality.
• A novel direct aperture optimization method is proposed to address these limitations.

Purpose of the Study:

• Introduce and evaluate a novel direct aperture optimization method for VMAT (comVMAT).
• Overcome limitations of current greedy VMAT optimization heuristics.
• Enhance plan consistency and quality in radiation therapy.

Main Methods:

• Formulated VMAT planning as an optimization problem with L2-norm fidelity and anisotropic total variation terms.
• Utilized a level set function for aperture shapes and penalized MLC motion range.
• Employed a proximal-class optimization solver and alternating optimization strategy for simultaneous fluence and aperture optimization.
• Generated single-arc comVMAT plans and compared them to clinical VMAT (clnVMAT) plans.

Main Results:

• comVMAT optimization converged within 600 iterations.
• comVMAT plans consistently reduced OAR doses (average 6.59% max, 7.45% mean reduction).
• Achieved high PTV coverage (within 0.25% of prescription dose).
• comVMAT dose distribution resembled static IMRT plans with beam orientation optimization.

Conclusions:

• The novel non-greedy VMAT approach optimizes all beams in an arc and directly generates deliverable apertures.
• Single-arc comVMAT fully utilizes digital linear accelerator capabilities.
• The new single VMAT algorithm generates superior plans compared to existing two-arc VMAT algorithms.