Clinical Analysis For Contouring、Setup Variation And Parotid Glands Variation In Intensity-modulated Radiotherapy For Nasopharyngeal Carcinoma Patients

Objective: During intensity-modulated radiotherapy for nasopharyngeal carcinomapatients, there exist many uncertainties, many studies only concentrate on one of them, thisstudy investigate the differences in target delineation, set-up and organ motion in order toguide the planning target volume margin and planning organs at risk margin and to guidethe best time for altering the radiation planning.Methods: Choose10pathology proven nasopharyngeal carcinoma patients in ourinstitute between2011-2012years, we acquire their planning image and weekly repeatedimages in weeks1-5.3experts contour parotid glands, submandibular glands and spinalcord respectively, and then we compare their delineation differences. Infuse the planningCT scan and weekly repeated scans, construct spatial three-dimensional coordinate systemin treatment planning system, choose reference points and detect weekly set-up variations,count setup systematic error and setup random error, and further, we can count planningtarget volume. Contoured both of parotid glands in planning CT image and weeklyrepeated CT images and calculated weekly volume, displacement and dosimetric changesof both parotid glands and detected the most distinguished time of their changes.According to the displacements of centers of both parotid glands in spatial three-dimensional directions, we can determine the planning organs at risk volume for bothparotid glands.3months after completion of radiation, we follow up the xerostomia ofpatients and detect the relevance between xerostomia and changes of volume anddosimetry for parotid glands.Results: The delineation difference for spinal cord is between2.7%and4.3%which is the smallest, the parotid glands are respectively-2.3%～5.6%and-1.4%～3.8%and for submandibular glands are5.5%～7.6%and-4.1%～6.2%, which is the biggest.During radiotherapy, the setup variation in lateral、anterior-inferior and anterior-posteriorare respectively（0.93±0.71）mm、（1.12±0.84）mm、（1.21±1.12）mm, The setup systematicerror and setup random error are both <2mm in3-dimensional directions and4.5mm-5mmPTV margin is enough. After5weeks of radiation, the volume of left and right parotidglands decreased by29%-44.9%and28%-57.1%, the dosimetry increased to148.3%-259.4%and198.1%-274.7%respectively. The displacements for centers of bothparotid glands in spatial3-dimensional directions are （3.86±3.39）、（4.15±2.17）、（2.89±2.84）；（3.56±2.18）、（3.83±2.96）、（3.25±2.96）respectively. We can detected thatthe most distinguished changes for volume, displacement and dosimetry for the left parotidgland was in the third, and the most distinguished changes for right parotid gland were inthe third week, second week and fourth week respectively and the change range for themall was larger in the first3weeks than the later2weeks. The distinguished systematic errorfor ipsilateral parotid gland is2.19mm, however,2.23mm for the controlateral parotidgland. Theoretically, the planning organ at risk margin for parotid glands is6.5mm,especially, for the controlateral parotid gland. The xerostomia for most patients are all mildor moderate, and none of them suffer from severe xerostomia, and there exist relevancebetween changes of volume, dosimetry and xerostomia.Conclusion: There are many uncertainties in intensity-modulated radiotherapy, weshould detect them, decrease the contouring differences for targets and organs at risk,drawing planning target volume and planning organ at risk volume will beneficial in clinical practice, we suggest that replanning in the third week will further decreased theradiation dosimetry of parotid glands.