Department of Radiation Oncology, Shandong Cancer Hospital and Institute, Jinan, China.
J Appl Clin Med Phys. 2012 Nov 8;13(6):4017. doi: 10.1120/jacmp.v13i6.4017.
The purpose of this study was to compare positional and volumetric differences of planning target volumes (PTVs) defined on axial three dimensional CT (3D CT) and four dimensional CT (4D CT) for liver cancer. Fourteen patients with liver cancer underwent 3D CT and 4D CT simulation scans during free breathing. The tumor motion was measured by 4D CT. Three internal target volumes (ITVs) were produced based on the clinical target volume from 3DCT (CTV3D): i) A conventional ITV (ITVconv) was produced by adding 10 mm in CC direction and 5 mm in LR and and AP directions to CTV3D; ii) A specific ITV (ITVspec) was created using a specific margin in transaxial direction; iii) ITVvector was produced by adding an isotropic margin derived from the individual tumor motion vector. ITV4D was defined on the fusion of CTVs on all phases of 4D CT. PTVs were generated by adding a 5 mm setup margin to ITVs. The average centroid shifts between PTVs derived from 3DCT and PTV4D in left-right (LR), anterior-posterior (AP), and cranial-caudal (CC) directions were close to zero. Comparing PTV4D to PTVconv, PTVspec, and PTVvector resulted in a decrease in volume size by 33.18% ± 12.39%, 24.95% ± 13.01%, 48.08% ± 15.32%, respectively. The mean degree of inclusions (DI) of PTV4D in PTVconv, and PTV4D in PTVspec, and PTV4D in PTVvector was 0.98, 0.97, and 0.99, which showed no significant correlation to tumor motion vector (r = -0.470, 0.259, and 0.244; p = 0.090, 0.371, and 0.401). The mean DIs of PTVconv in PTV4D, PTVspec in PTV4D, and PTVvector in PTV4D was 0.66, 0.73, and 0.52. The size of individual PTV from 4D CT is significantly less than that of PTVs from 3DCT. The position of targets derived from axial 3DCT images scatters around the center of 4D targets randomly. Compared to conventional PTV, the use of 3D CT-based PTVs with individual margins cannot significantly reduce normal tissues being unnecessarily irradiated, but may contribute to reducing the risk of missing targets for tumors with large motion.
本研究旨在比较肝癌患者在自由呼吸状态下基于轴向三维 CT（3D CT）和四维 CT（4D CT）勾画的计划靶区（PTV）的位置和体积差异。14 例肝癌患者行 3D CT 和 4D CT 模拟扫描。4D CT 测量肿瘤运动。基于 3D CT（CTV3D）生成 3 个内部靶区（ITV）：i）CTV3D 外扩 10 mm 在 CC 方向，5 mm 在 LR 和 AP 方向，得到常规 ITV（ITVconv）；ii）利用横轴面特定边界生成特定 ITV（ITVspec）；iii）通过个体肿瘤运动矢量的各向同性边界生成 ITVvector。4D CT 所有时相融合定义 ITV4D。将 5mm 摆位误差加入 ITVs 生成 PTV。PTV 由 3D CT 生成的 PTV 与 PTV4D 之间在左右（LR）、前后（AP）和头脚（CC）方向的中心位置偏移接近零。与 PTVconv、PTVspec 和 PTVvector 相比，PTV4D 的体积分别减少了 33.18%±12.39%、24.95%±13.01%和 48.08%±15.32%。PTV4D 包含 PTVconv、PTVspec 和 PTVvector 的平均包容度（DI）分别为 0.98、0.97 和 0.99，与肿瘤运动矢量无显著相关性（r=-0.470、0.259 和 0.244；p=0.090、0.371 和 0.401）。PTVconv 包含 PTV4D、PTVspec 包含 PTV4D 和 PTVvector 包含 PTV4D 的平均 DI 分别为 0.66、0.73 和 0.52。4D CT 生成的单个 PTV 体积明显小于 3D CT 生成的 PTV。从轴向 3D CT 图像获得的靶区位置随机散布在 4D 靶区中心周围。与常规 PTV 相比，使用基于 3D CT 的个体化 PTV 并不能显著减少不必要照射的正常组织，但可能有助于降低肿瘤运动较大时靶区漏照的风险。
