Primary content


L Holloway (Lead)

S Arumugam

A Xing

T Young

D Thwaites


To investigate the clinical significance of different Volumetric Modulated Arc Therapy (VMAT) delivery errors and the sensitivity of the ArcCHECK detector to those errors.


Two Prostate and two Head and Neck radiotherapy VMAT plans were considered. Three types of errors in Multi Leaf Collimator (MLC) position and dose delivery were simulated. i. Gantry Independent ii. Gantry dependent and iii. Control point (CP) specific errors. The impact of simulated errors was evaluated using clinical dosimetric end points.  The plans with simulated errors were measured using ArcCHECK and compared with the error free plan.


A minimum 3 mm MLC error and 5% output error was detectable for the gantry angle independent errors and a minimum 4 mm MLC error and 4% dose error was identifiable for gantry dependent errors. For the prostate plan all these errors resulted in a greater than 5% reduction in mean Planning Target Volume (PTV) dose and greater than 5% increase in rectum equivalent uniform dose (EUD), and thus would be expected to have a clinical impact. Errors introduced to the CPs did not have a clinically significant impact.


ArcCHECK is able to detect VMAT delivery errors that are clinically significant. This work will help establish more specific quality assurance procedures for radiotherapy.