Dosimetric impact of hysteresis in gantry rotation on clinical VMAT plans.

Abstract

Gantry angle inaccuracies during treatment delivery affect the dose distribution received by the patient. These inaccuracies may arise from stretching of the gantry belt, leading to hysteresis in the gantry angle. The purpose of this work is to determine the impact of gantry hysteresis on the dose distribution of representative volumetric modulated arc therapy plans and to evaluate whether current quality assurance guidelines for gantry angle tolerances are adequate in these cases. Simulations were performed in Eclipse by encoding gantry errors with various amounts of hysteresis. The impact on the dose constraints and the gamma index passing rates computed relative to the as-planned dose distribution was examined in five clinical plans: brain stereotactic radiosurgery, prostate with nodes, lung stereotactic body radiation therapy (SBRT) and spine SBRT. The hysteresis thresholds, defined as the amount of gantry angle error causing a 1.0% dose-constraint change or a gamma reduction to 90%, were identified for gross tumor volumes, planning target volumes (PTVs) and selected organs at risks (OARs). D99%, D98% and gamma (1.0%, 0.01 mm) in the targets decreased with increasing hysteresis. The hysteresis threshold for the D98% or D99% ratios to the planned values ranged from 1.0° to 11.0°, with the smallest value (indicating the highest sensitivity) observed in the multi-met brain plan. Thresholds for the gamma analysis ranged from 0.5° to 7.7°, with the smallest value in the double-met lung SBRT plan. All dose metrics in the OARs decreased with increasing hysteresis, except for the D25% ratio in the bladder of the prostate/pelvis plan. The corresponding gamma passing rates decreased as hysteresis increased. Maintaining gantry hysteresis within 1.0° allowed dose-volume histogram -based metrics to remain within 1.0% of the planned values for all targets and OARs. The corresponding gamma passing rates remained above 90% for all OARs and for all but two PTVs. Limiting hysteresis to 0.5° ensured that all metrics stayed within tolerance.