Dosimetric influences of rotational setup errors on head and neck carcinoma intensity-modulated radiation therapy treatments.

The purpose of this work is to investigate the dosimetric influence of the residual rotational setup errors on head and neck carcinoma (HNC) intensity-modulated radiation therapy (IMRT) with routine 3 translational setup corrections and the adequacy of this routine correction. A total of 66 kV cone beam computed tomography (CBCT) image sets were acquired on the first day of treatment and weekly thereafter for 10 patients with HNC and were registered with the corresponding planning CT images, using 2 3-dimensional (3D) rigid registration methods. Method 1 determines the translational setup errors only, and method 2 determines 6-degree (6D) setup errors, i.e., both rotational and translational setup errors. The 6D setup errors determined by method 2 were simulated in the treatment planning system and were then corrected using the corresponding translational data determined by method 1. For each patient, dose distributions for 6 to 7 fractions with various setup uncertainties were generated, and a plan sum was created to determine the total dose distribution through an entire course and was compared with the original treatment plan. The average rotational setup errors were 0.7°± 1.0°, 0.1°±1.9°, and 0.3°±0.7° around left-right (LR), anterior-posterior (AP), and superior-inferior (SI) axes, respectively. With translational corrections determined by method 1 alone, the dose deviation could be large from fraction to fraction. For a certain fraction, the decrease in prescription dose coverage (Vp) and the dose that covers 95% of target volume (D95) could be up to 15.8% and 13.2% for planning target volume (PTV), and the decrease in Vp and the dose that covers 98% of target volume (D98) could be up to 9.8% and 5.5% for the clinical target volume (CTV). However, for the entire treatment course, for PTV, the plan sum showed that the average Vp was decreased by 4.2% and D95 was decreased by 1.2Gy for the first phase of IMRT with a prescription dose of 50Gy. For CTV, the plan sum showed that the average Vp was decreased by 0.8% and D98, relative to prescription dose, was not decreased. Among these 10 patients, the plan sum showed that the dose to 1-cm3 spinal cord (D1cm3) increased no more than 1Gy for 7 patients and more than 2Gy for 2 patients. The average increase in D1cm3 was 1.2Gy. The study shows that, with translational setup error correction, the overall CTV Vp has a minor decrease with a 5-mm margin from CTV to PTV. For the spinal cord, a noticeable dose increase was observed for some patients. So to decide whether the routine clinical translational setup error correction is adequate for this HNC IMRT technique, the dosimetric influence of rotational setup errors should be evaluated carefully from case to case when organs at risk are in close proximity to the target.