Knowledge-Based RapidPlan Volumetric Modulated Arc Therapy Model in Nasopharyngeal Carcinoma

Area of Science:

• Radiation Oncology
• Medical Physics
• Cancer Treatment Planning

Background:

• Knowledge-based planning software like RapidPlan (RP) aims to enhance radiation therapy planning efficiency and quality.
• Volumetric modulated arc therapy (VMAT) is a common technique for delivering radiation doses.

Purpose of the Study:

• To develop and evaluate a RapidPlan (RP) model for nasopharyngeal carcinoma (NPC) VMAT.
• To compare the quality and efficiency of RP-generated plans against manually created plans.

Main Methods:

• A cohort of 160 NPC patients' existing plans were used for RP model training.
• A separate cohort of 33 patients was used to compare RP and manual plans using dose-volume histograms, isodose curves, and physician evaluations.
• Plan quality was assessed based on target coverage, organ-at-risk (OAR) doses, and physician scores.

Main Results:

• RP plans were completed within 1 hour, significantly improving efficiency.
• RP plans showed superior conformity for planning target volume 70 and better OAR sparing compared to manual plans.
• Physician evaluations favored RP plans for their dose-volume histograms, OAR sparing, conformity, and overall quality, leading to a 100% clinical acceptance rate.

Conclusions:

• The developed RP Eclipse model efficiently generates high-quality VMAT plans for NPC.
• RP Eclipse shows promise for integration into clinical workflows and future research, including multi-institutional studies.