宫颈癌术后调强放疗摆位误差对靶区累积剂量偏差的影响

目的:千伏级锥形束 CT（CBCT）获取分次间宫颈癌术后调强放射治疗(IMRT)摆位误差，分析分次间摆位误差对靶区累积剂量偏差的影响。方法:选取61例宫颈癌术后行调强放疗的患者，全程916次CBCT获取摆位误差，将误差值输入治疗计划系统中，由分次间摆位误差剂量叠加得到累积摆位误差剂量，通过偏差公式与标准计划剂量计算偏差百分比。结果:摆位误差x、y、z方向的偏差和偏移等中心距离分别为0.04（-0.16, 0.25）、-0.05（-0.37, 0.28）、0.10（-0.06, 0.24）和0.55（0.38, 0.78） cm。临床靶区除CTV的HIsum-HIplan和CTV1 的Dsum_D50-Dplan_D50与HIsum-HIplan无统计学差异外，其他临床靶区的配对检验均有统计学差异。计划靶区除PGTVnd的Dsum_median-Dplan_median、Dsum_mean-Dplan_mean、Dsum_D50-Dplan_D50无统计学差异外，其他计划靶区均有统计学差异。累积摆位误差剂量与标准计划剂量分布对比呈现负偏态分布，峰度降低。GTVnd、CTV、CTV1、CTVn、CTV_all与PGTVnd、PTV、PTV1、PTVn、PTV_all剂量偏差均呈降低，计划靶区的累积剂量偏差比临床靶区偏差明显增大。Dmin偏差、D98偏差、D95偏差偏离最大，Dmax偏差、D5偏差、D2偏差变化次之，Dmedian偏差、Dmean偏差、D50偏差变化最小，反S型DVH曲线向左偏移，斜率增大。临床靶区HI偏差均上升。结论:宫颈癌术后调强放疗摆位误差对靶区累积剂量影响存在统计差异性，靶区累积剂量降低、均匀性变劣。宫颈癌术后调强放疗在每次治疗前需进行CBCT位置校准以保证靶区各结构剂量准确性。在放疗计划设计时考虑增加CBCT次数带来额外剂量的风险。

Effects of setup errors in postoperative intensity-modulated radiotherapy of cervical cancer on cumulative target dose deviation

Abstract: Objective To obtain the inter-fractional setup errors in postoperative intensity-modulated radiotherapy (IMRT) of cervical cancer by kilovolt cone beam computed tomography (CBCT), and to analyze the variation of cumulative target dose deviation caused by inter-fractional setup errors. Methods A total of 61 patients treated with postoperative IMRT for cervical cancer were enrolled in the study, and their setup errors were obtained by 916 CBCT scans. The setup errors were inputted into treatment planning system, and the cumulative setup error dose was obtained by superposing the inter-fractional setup error doses. The deviation percentage was computed according to deviation formula and the standard planned dose. Results The setup errors in the x, y, z directions and offset isocentric distances were 0.04 (-0.16, 0.25)、-0.05(-0.37, 0.28)、 0.10(-0.06, 0.24), and 0.55 (0.38, 0.78) cm, respectively. Except for the HIsum-HIplan of CTV as well as the Dsum_D50-Dplan_D50 and HIsum-HIplan of CTV1, there were statistical differences in the paired test on the other clinical target areas, and there was statistical significance in the paired test on planning target areas, except for the Dsum_median-Dplan_median, Dsum?mean-Dplan_mean, Dsum_D50-Dplan_D50 of PGTVnd. The comparison of the cumulative setup error dose with the standard planned dose showed a negative skewed distribution, with a decreased kurtosis. The dose deviations of GTVnd, CTV, CTV1, CTVn, CTV_all and PGTVnd, PTV, PTV1, PTVn, PTV_all were all decreased, and the cumulative dose deviations in the planning target areas were significantly larger than those in the clinical target areas. The deviations of Dmin_deviation, D98_ deviation and D95_deviation were the largest, followed by Dmax_deviation, D5_deviation, D2_deviation, while Dmedian_deviation, Dmean_deviation and D50_deviation had the smallest deviations. The reverse S-type DVH curve was skewed to the left, with an increased slope. The HIdeviation of clinical target areas was increased. Conclusion The setup errors in the postoperative IMRT of cervical cancer which has various effects on the cumulative target dose makes the cumulative target dose reduced and the target dose homogeneity worse. Postoperative IMRT for cervical cancer requires CBCT position calibration before each treatment to ensure the accuracy of the dose of each structure in the target area. The risk of additional dose from increasing CBCT frequency should be considered in radiotherapy planning.