图像引导鼻咽癌调强放疗位置误差导致剂量差异分析

Analysis of dose deviation by set-up error in the image guided intensity modulated radiotherapy fornasopharyngeal carcinoma

目的分析千伏特锥形束CT（CBCT）对鼻咽癌调强放疗靶区中心位置变化及由位置偏差导致的剂量变化。方法对21例鼻咽癌患者调强放疗，全程376次CBCT得到每次靶区中心变化值。在上述变化值中随机抽取10、15次值输入逆向调强放疗计划系统中，照射角度、子野序列、权重等均不变，只对中心发生变化；累积10、15次偏差计划得到新的剂量分布，通过偏差公式得到相对应次数标准计划剂量偏差情况。结果未经校正的376次CBCT在左右、上下、前后方向的系统误差和随机误差分别为0．75mm和1．13mm，0．92mm和2．15mm，0．82mm和1．24mm，利用双参数模型计算CTV外扩4、6，4mm生成胛V，左右、上下、前后方向〈2mm的偏差比例分别为82．8％、76．0％、81．8％，〉3mm偏差分别为6．2％、9．8％、7．1％。在15次累积计划时靶区接受95％处方剂量（D95）偏差为-7．5％～-11．9％，D50的为-5．1％～-8．2％。结论每次较小的摆位误差可能对靶区或正常器官的剂量影响较小，但整个疗程累计30余次的误差结果可能导致较大剂量差异，通过双参数模型计算生成PTV并按PTV实现的计划和图像引导放疗可弥补这些剂量的不足。

Objective To discuss the set-up isocenter error based on kilovolt cone beam computed tomography （KVCBCT） and to investigate dose deviation led to set-up isocenter error. Methods 21 cases of nasopharyngeal carcinoma （NPC） treated with image guided intensity modulated radiotherapy （IG-IMRT） were investigated. The online KVCBCT scan, rigid image registration, set-up error was gained for 376 sets before radiotherapy. We sampled ten and fifteen setup isocenter error in the 376 sets randomly. Without changing beam angle, fields size and leaf sequences and dose weight et al. , we only replaced new isocenter and accumulated the new plan for ten or fifteen plans. We compared the percentage deviation between ten, fifteen times accumulated plans and normal ten , fifteen times plans. Results All 376 sets of KVCBCT image were analyzed for 21 cases. Under the condition of non-correction, the setup isocenter errors are 0. 75 mm±1.13 ram, 0. 92 mm±2. 15 ram,0. 82 mm±1.24 mm in left-right, superior-inferior and anterior- posterior directions respectively. So, we developed the margins which were 4 ram,6 ram,4 mm in three directions respectively from clinical tumor volume to planning tumor volume （PTV） calculated by two parameters model. In the fifteen accumulated plan, the deviation in the dose of 95% PTV （ D95 ） was -7. 5 % - - 11.9% , and the deviation in the Ds0 was -5.1% - -8.2%. Conclusions It is possible of small effects to normal organs and targets because of small error of patient displacement in one fraction. However, many small errors can led to considerable dose difference in targets and normal tissue in thirty fractions of all treatments period. So, according to two parameters model, PTV margin can be designed new olanning and deoended on IG-IMRT technique, which it will be significantly reduced these dose differences.