| 朱志鹏,蓝茂英,欧雪,等.解剖结构对左侧乳腺癌根治术后不同拉弧角度放疗计划选择研究[J].中华放射医学与防护杂志,2025,45(9):884-891.Zhu Zhipeng,Lan Maoying,Ou Xue,et al.Selection of radiotherapy plans with different arc angles for left-sided breast cancer based on anatomical structures[J].Chin J Radiol Med Prot,2025,45(9):884-891 |
| 解剖结构对左侧乳腺癌根治术后不同拉弧角度放疗计划选择研究 |
| Selection of radiotherapy plans with different arc angles for left-sided breast cancer based on anatomical structures |
| 投稿时间：2025-06-25 |
| DOI：10.3760/cma.j.cn112271-20250625-00214 |
| 中文关键词: 深吸气屏气 解剖结构 左侧乳腺癌根治术后 分段弧 |
| 英文关键词:Deep inspiration breast hold (DIBH) Anatomical structure Left-sided breast cancer Arc |
| 基金项目:广西壮族自治区卫健委自筹课题(Z-A20230603,Z-A20230606);广东省医学科学技术研究基金(A2023250) |
|
| 摘要点击次数: 7175 |
| 全文下载次数: 883 |
| 中文摘要: |
| 目的 比较基于深吸气屏气下(DIBH)左侧乳腺癌根治术后4种不同拉弧角度放疗计划差异,并探索是否能利用患者解剖结构参数选择合适拉弧方式。方法 选取51例2020年至2022年在广西医科大学第二附属医院和广州医科大学附属第一医院行DIBH下左侧乳腺癌根治术后放疗患者。51例患者中采用简单随机抽取40例患者行SV120、SV100、SV80和SV60 4种拉弧计划,剩余的11例患者仅进行SV60计划优化。比较上述计划的靶区和危及器官(心脏、肺、右乳)剂量参数,单个分段弧出束时间。提取上述51例患者解剖结构(心脏、肺和右乳)分别为PHeart、PLungl和PBreast并作为自变量,以SV60计划是否为良好作为因变量,按照训练集(45例)和测试集(6例)划分,使用多因素回归模型[P(SV60)]将上述自变量和因变量进行拟合,分类阈值为0.5。结果 随着拉弧角度增加,其对靶区和左肺的调制能力也会增强,但右肺的V5和单个分段弧平均出束时间也随之增大。SV120计划的计划靶区V107%(6.84%)、均匀性(0.13)和适形度(0.81)以及左肺平均量(1 330.97 cGy)都优于其他3种计划且差异有统计学意义(W = 0~99,P < 0.001)。SV60的右乳平均量(198.97 cGy)、心脏(440.35 cGy)、右肺V5(0.27%)均小于其他3种计划,其中右肺V5和右乳平均量差异有统计学意义(W = 0~157, P < 0.001),而其他3种计划之间(除右肺V5)差异均无统计学意义(P > 0.05)。4种拉弧方式的单个分段弧平均出束时间分别为22.0、19.1、16.1和14.4 s,差异有统计学意义(χ2 = 93.0, P < 0.001)。多因素回归模型发现3个解剖结构PLungl、PHeart和PBreast与SV60计划为良好成负相关性(t = -64.84、-28.20、-24.45, P < 0.001),模型在训练集上的准确率为 93.33%,精确率为 92.86%;在测试集中,模型准确率和精确率均达到了 100%。结论 对于深吸气屏气左侧乳腺癌根治术后放疗患者,结合患者的解剖结构选择合适拉弧角度是一种可行方式。 |
| 英文摘要: |
| Objective To compare four postmastectomy radiotherapy plans with different arc angles for left-sided breast cancer prepared using deep inspiration breath hold (DIBH) technique, in order to explore the feasibility of selecting optimal arc angles based on the parameters of patients' anatomical structures. Methods A total of 51 patients who underwent postmastectomy radiotherapy for left-sided breast cancer using DIBH at the Second Affiliated Hospital of Guangxi Medical University and the First Affiliated Hospital of Guangzhou Medical University were selected. Among these, 40 patients selected using simple random sampling were treated with four radiotherapy plans with different arc angles, labeled as SV120, SV100, SV80, and SV60, while the remaining 11 patients were treated with only the SV60 plans. The dose parameters of the target volumes (TVs) and organs at risk (the heart, lungs, and contralateral breast), as well as the beam-on times of individual arcs, in these plans were compared. The parameters of the anatomical structures (i.e., the heart, lung, and breast) of the 51 patients, including PHeart, PLungl, and PBreast, were extracted. Taking these parameters as independent variables and the quality of the SV60 plans as the dependent variable, fitting was conducted using P(SV60)—a multivariate logistic regression model—based on a training set (45 patients) and a testing set (six patients). The classification threshold of the plans was set at 0.5. Results As the arc angle increased, the plans exhibited improved modulation capabilities for TVs and the left lung. However, the V5 of the right lung and the average beam-on time of a single arc also increased. The SV120 plans demonstrated significantly better V107% of planning target volume (PTV; 6.84%), homogeneity index (HI; 0.13), conformity index (CI; 0.81), and mean dose to the left lung (1330.97 cGy) compared to the other three types of plans, with statistically significant differences (W = 0-99,P < 0.001). The SV60 plans displayed lower mean doses to the contralateral breast (198.97 cGy) and the heart (440.35 cGy) and lower V5 of the right lung (0.27%) than the other three types of plans, with significant differences in V5 of the right lung and the mean dose to the contralateral breast (W = 0-157,P < 0.001). In contrast, no significant difference was observed in these parameters (except for V5 of the right lung) among the other three types of plans. The four types of plans exhibited average beam-on times of individual arcs of 22.0, 19.1, 16.1, and 14.4 s, respectively, with statistically significant differences (χ2 = 93.0,P < 0.001). A multivariate logistic regression model revealed that PLungl, PHeart, and PBreast were negatively correlated with the quality of the SV60 plans (t = -64.84, -28.20, -24.45, P < 0.001). This model yielded an accuracy of 93.33% and a precision of 92.86% in the training set, while its accuracy and precision reached 100% in the testing set. Conclusions For patients treated with postmastectomy radiotherapy for left-sided breast cancer using DIBH, it is feasible to select appropriate arc angles based on patients’ anatomical structures. |
| HTML 查看全文 查看/发表评论 下载PDF阅读器 |
| 关闭 |
|
|
|