Construction of a model for the improved planning of MCO-informed VMAT in RayStation using a knowledge base of clinical IMRT-MCO treatment plans

Abstract

Intensity Modulated Radiation Therapy (IMRT) is a type of external beam radiation therapy that has proven effective at treating many cancers. A related therapy type, Volumetric Modulated Arc Therapy (VMAT), has the potential to provide comparable dose coverage to tumor sites while better sparing nearby organs at risk (OARs). Multi-criteria Optimization (MCO) is an algorithm that is used to optimize a patient's personalized IMRT treatment plan. VMAT treatment plans cannot be optimized using early versions of the MCO algorithm. The purpose of this study was to construct a model for the automated generation of VMAT treatment plans for prostate cancers using a knowledge base of previously implemented IMRT-MCO treatment plans. An initial model configuration was iteratively refined to produce VMAT plans that represent a quality 'first pass" that can be further optimized by trained treatment planners. The clinical implementation of a model like this one could significantly improve the timeliness of standard non-MCO VMAT optimization methods.