Abstract: Objective　To study the accuracy of the placement of cervical cancer with intensity-modulated radiation therapy (IMRT) adjusted by cone beam CT (CBCT) and six-dimensional 6D treatment bed and the boundary of the target area. Methods　Thirty-six patients with cervical cancer who were 35-70 years old treated at Sun Yat-sen University Cancer Center from October 2020 to April 2021 were randomly selected. Their Karnofsky score was over 80. After volumetric intensity modulated radiation therapy (VMAT), each patient underwent CBCT scans a total of 714 times before each treatment to obtain a volume CT image, which was then registered with the localization CT image, and performed 6D registration again, and the 6D setup error value was recorded. The body mass index (BMI) of each patient was recorded. The influences of BMI on the coronal plane Rtn, sagittal plane Pitch, transverse plane Roll, and other directions in the 6D registration was analyzed. According to the Van Herk formula M_PTV=2.5∑+0.7σ, the corresponding extrapolation boundary value (MPTV) in the translation direction was calculated. Independent sample t test and paired t test were performed on all the data. Results　The absolute values of the placement errors of 3D registration in the left and right (LR), head-to-foot (SI), and ventral dorsal (AP) directions were (2.1±0.9), (3.2±1.2), and (2.3±1.2) mm, and those of 6D registration were (1.8±0.6), (3.2±1.0), and (2.2±0.8) mm. The paired t test was used for the original data, and there was no statistical difference in the SI direction (P>0.05), and were in the LR and AP directions (all P<0.05). Under the influence of BMI, there were statistical differences in the the rotation and positioning errors in the directions of yaw angle (Rtn), pitch angle (Pitch), and roll angle (Roll) between the normal and obese patients (all P<0.05). The 6D registration was reduced by 1.0, 2.0, and 1.6 mm in LR, SI, and AP as compared with the 3D registration by calculating the extrapolation boundary. Conclusions　For CBCT combined with 6D treatment table online correction, 6D image registration is better than 3D image registration. According to the formula M_PTV=2.5∑+0.7σ, the 6D extension boundary is smaller than the 3D extension boundary; in the clinical radiotherapy for cervical cancer, the use of 6D image registration combined with a six-dimensional bed has more advantages than a single three-dimensional bed, which can reduce the dose of normal tissues around the target area and improve the accuracy of tumor radiotherapy.

Key words: Cervical cancer, CBCT, 6D bed, Set-up error, BMI, Outer boundary

摘要： 目的　锥形束CT（CBCT）联合六维（6D）治疗床纠正子宫颈癌调强放疗摆位精度及靶区外放边界的研究。方法　随机选取2020年10月至2021年4月中山大学肿瘤防治中心36例子宫颈癌患者，年龄35～70岁；卡氏评分>80分。接受容积调强放射治疗（VMAT），每例患者每次治疗前进行CBCT扫描共714次，得到容积CT图像，重建后和定位CT图像进行配准，先进行3D配准，记录摆位误差数值，再次进行6D配准，记录6D摆位误差数值；记录各患者的体质量指数（BMI），并分析BMI在6D配准中冠状面偏转角（Rtn）、矢状面俯仰角（Pitch）、横断面翻滚角（Roll）方向的影响；按Van Herk公式M_PTV=2.5∑+0.7σ计算平移方向的相应外放边界值（M_PTV）。对所有数据分别进行独立样本t检验和配对t检验。结果　3D配准和6D配准在左右（LR）、头脚（SI）、腹背（AP）方向的摆位误差取绝对值分别为（2.1±0.9）、（3.2±1.2）、（2.3±1.2）mm；（1.8±0.6）、（3.2±1.0）、（2.2±0.8）mm；对原始数据进行配对样本t检验中，SI方向比较差异无统计学意义（P>0.05），在LR和AP方向比较差异均有统计学意义（均P<0.05）；在BMI影响下，正常与肥胖患者在Rtn、Pitch、Roll方向的旋转摆位误差，差异均有统计学意义（均P<0.05）；通过计算外放边界6D配准比3D配准在LR、SI和AP分别减少1.0、2.0、1.6 mm。结论　CBCT联合6D治疗床在线纠正，6D图像配准优于3D图像配准；通过外放边界公式M_PTV=2.5∑+0.7σ计算出6D的外放边界比3D的外放边界小，这对于在临床上子宫颈癌的放疗使用6D图像配准联合6D床比单使用3D床更具有优势，能够减少靶区周围正常组织受量，提高肿瘤放疗的精确性。

关键词: 子宫颈癌, CBCT, 6D床, 摆位误差, BMI, 外扩边界