Session Item

The aim of this study is to determine and validate the local pass rates criteria for pre-treatment patient plan specific QA using commercial PerFRACTION software (Sun Nuclear Corporation, Melbourne, FL) for Halcyon treatment plans.

Halcyon v2.0 (Varian Medical Systems, USA) uses a pre-configured beam model and produces IMRT/VMAT plans. Halcyon treatment plans’ use of dynamic modulation of dose rate, field aperture and gantry speed demands diligent pre-treatment verification. The centre currently verifies all treatment plans with Varian’s Portal Dosimetry (PD), which uses Anisotropic Analytical Algorithm (AAA) to predict the fluence pattern; this is the same algorithm used to generate clinical plans. The main disadvantage of PD is the lack of projecting delivered dose distribution onto a 3D patient dataset. PD uses an integrated measurement of 2D fluence in which the verification is done by field-by-field basis and composite plan.  Dose reconstruction using information from the delivery trajectory log files of Halcyon could be more useful to verify the deliverability of the plan. Recently, Sun Nuclear has released the PerFRACTION software module to analyse trajectory log files, to provide a 3D dose distribution on a 3D patient CT dataset and compare with DICOM RT Dose input of the Halcyon plan. The centre also performs plan-of-the-month phantom-based PSQA on selected plans using ArcCHECK (Sun Nuclear Corporation) to complement PD.

Ten Halcyon patient treatment plans were produced in Eclipse v15.6 (Varian Medical System), using AAA algorithm. The tested plans cover treatment sites including head & neck, lung, prostate, bladder, rectum and breast. When the delivery log file is captured by PerFRACTION, the dose calculation is performed using a Collapsed Cone Convolution Superposition Algorithm. PD and ArcCHECK phantom measurements were performed at the planned gantry and collimator angles. A composite dose image of all treatment fields was created for the PD measurements. Gamma analysis was performed for the PerFRACTION, PD and ArcCHECK measurements. The correlation between the gamma passing rates of composite PD (Gamma 2%-2mm) and ArcCHECK (Gamma 3%-2mm) and PerFRACTION (Gamma 2%-2mm) was evaluated using the Bland-Altman method.

The mean pass rates were: PerFRACTION - 97.2% (SD ± 2.0%); PD - 98.3% (SD ± 1.8%); and ArcCHECK - 98.7% (SD ± 1.4%). Bland-Altman analysis indicates a good correlation between both PD and ArcCHECK versus PerFRACTION. Among the data plotted, only one datum was found to be an outlier in both correlation analysis.  

This study confirms suitability of using PerFRACTION as a log file-based plan verification system for Halcyon with a pass rate criterion of 97% for 2%-2mm Gamma analysis (global). It is independent from the vendor’s PD and will be more efficient than using phantom-based ArcCHECK PSQA (which will be used at a lower frequency).