Dogan Nesrin, Leybovich Leonid B, Sethi Anil, Emami Bahman
Loyola University Medical Center, Radiation Oncology Department, 2160 South First Avenue, Maywood, IL 60153, USA.
Phys Med Biol. 2003 May 7;48(9):1133-40. doi: 10.1088/0031-9155/48/9/304.
Due to leaf travel range limitations of the Varian Dynamic Multileaf Collimator (DMLC) system, an IMRT field width exceeding 14.5 cm is split into two or more adjacent abutting sub-fields. The abutting sub-fields are then delivered as separate treatment fields. The accuracy of the delivery is very sensitive to multileaf positioning accuracy. The uncertainties in leaf and carriage positions cause errors in the delivered dose (e.g., hot or cold spots) along the match line of abutting sub-fields. The dose errors are proportional to the penumbra slope at the edge of each sub-field. To alleviate this problem, we developed techniques that feather the split line of IMRT fields. Feathering of the split line was achieved by dividing IMRT fields into several sub-groups with different split line positions. A Varian 21EX accelerator with an 80-leaf DLMC was used for IMRT delivery. Cylindrical targets with varying widths (>14.5 cm) were created to study the split line positions. Seven coplanar 6 MV fields were selected for planning using the NOMOS-CORVUS system. The isocentre of the fields was positioned at the centre of the target volume. Verification was done in a 30 x 30 x 30 cm3 polystyrene phantom using film dosimetry. We investigated two techniques to move the split line from its original position or cause feathering of them: (1) varying the isocentre position along the target width and (2) introduction of a 'pseudo target' outside of the patient (phantom). The position of the 'pseudo target' was determined by analysing the divergence of IMRT fields. For target widths of 14-28 cm, IMRT fields were automatically split into two sub-fields, and the split line was positioned along the centre of the target by CORVUS. Measured dose distributions demonstrated that the dose to the critical structure was 10% higher than planned when the split line crossed through the centre of the target. Both methods of modifying the split line positions resulted in maximum shifts of approximately 1 cm from the original. Therefore, it was concluded that the feathering of the split line may be used for reducing the magnitude of hot/cold spots. This method was tested for an oesophageal cancer case. For a six-field arrangement, it was possible to create three field sub-groups with different split lines. The feathering technique developed in this work does not require any modifications of the radiation fields during the course of treatment because only one treatment plan is used to deliver the entire course of radiation treatments. In addition, this method may be more biologically effective because the split line feathering is achieved for every fraction of radiation.
由于瓦里安动态多叶准直器（DMLC）系统的叶片移动范围有限，超过14.5厘米的调强放疗射野宽度会被分割成两个或更多相邻的对接子野。然后将这些对接子野作为单独的治疗射野进行照射。照射的准确性对多叶定位精度非常敏感。叶片和托架位置的不确定性会在对接子野的匹配线上导致所输送剂量出现误差（例如热点或冷点）。剂量误差与每个子野边缘的半值层斜率成正比。为缓解这一问题，我们开发了对调强放疗射野分割线进行羽化处理的技术。通过将调强放疗射野划分为具有不同分割线位置的几个子组来实现分割线的羽化。使用配备80叶DLMC的瓦里安21EX加速器进行调强放疗照射。创建了宽度不同（>14.5厘米）的圆柱形靶区来研究分割线位置。使用NOMOS - CORVUS系统选择七个共面6兆伏射野进行计划。射野的等中心位于靶区体积的中心。在一个30×30×30立方厘米的聚苯乙烯模体中使用胶片剂量测定法进行验证。我们研究了两种将分割线从其原始位置移动或使其羽化的技术：（1）沿靶区宽度改变等中心位置，以及（2）在患者体外（模体）引入一个“虚拟靶区”。“虚拟靶区”的位置通过分析调强放疗射野的发散度来确定。对于宽度为14 - 28厘米的靶区，调强放疗射野会自动分割成两个子野，并且CORVUS会将分割线沿靶区中心定位。测量的剂量分布表明，当分割线穿过靶区中心时，关键结构所接受的剂量比计划高10%。两种修改分割线位置的方法都导致最大偏移量相对于原始位置约为1厘米。因此，得出的结论是分割线的羽化可用于减小热点/冷点的幅度。该方法在一例食管癌病例中进行了测试。对于六野布置，可以创建具有不同分割线的三个野子组。本研究中开发的羽化技术在治疗过程中不需要对辐射野进行任何修改，因为整个放射治疗疗程仅使用一个治疗计划。此外，这种方法可能在生物学上更有效，因为每次放疗分割都实现了分割线羽化。
