Guan Xiyin, Peng Jiyou, Sun Jiayao, Xing Xing, Hu Chaosu
Department of Radiation Oncology, Shanghai Proton and Heavy Ion Center, Fudan University Cancer Hospital, Shanghai, China.
Shanghai Key Laboratory of Radiation Oncology, Shanghai, China.
Front Oncol. 2024 May 30;14:1394111. doi: 10.3389/fonc.2024.1394111. eCollection 2024.
We tried to establish the normal tissue complication probability (NTCP) model of temporal lobe injury of recurrent nasopharyngeal carcinoma (NPC) patients after two courses of intensity modulated radiotherapy (IMRT) to provide more reliable dose-volume data reference to set the temporal lobe tolerance dose for recurrent NPC patients in the future.
Recurrent NPC patients were randomly divided into training data set and validation data set in a ratio of 2:1, All the temporal lobes (TLs) were re-contoured as R/L structures and named separately in the MIM system. The dose distribution of the initial IMRT plan was deformed into the second course planning CT via MIM software to get the deformed dose. Equivalent dose of TLs in 2Gy fractions was calculated via linear quadratic model, using an α/β=3 for temporal lobes. NTCP model that correlated the irradiated volume of the temporal lobe and? the clinical variables were evaluated in a multivariate prediction model using AUC analysis.
From Jan. 2010 to Dec. 2020, 78 patients were enrolled into our study. Among which 26 (33.3%) developed TLI. The most important factors affecting TLI was the sum-dose d1.5cc of TL, while the possible clinical factors did not reach statistically significant differences in multivariate analysis. According to NTCP model, the TD5 and TD50 EQD2 dose of sum-dose d1.5cc were 65.26Gy (46.72-80.69Gy) and 125.25Gy (89.51-152.18Gy), respectively. For the accumulated EQD2 dose, the area under ROC shadow was 0.8702 (0.7577-0.9828) in model validation, p<0.001.
In this study, a NTCP model of temporal lobe injury after a second course of IMRT for recurrent nasopharyngeal carcinoma was established. TD5 and TD50 doses of temporal lobe injury after re-RT were obtained according to the model, and the model was verified by validation set data.
尝试建立复发性鼻咽癌（NPC）患者在两程调强放疗（IMRT）后颞叶损伤的正常组织并发症概率（NTCP）模型，为今后设定复发性NPC患者颞叶耐受剂量提供更可靠的剂量体积数据参考。
将复发性NPC患者按2:1的比例随机分为训练数据集和验证数据集，所有颞叶（TLs）在MIM系统中重新勾勒为R/L结构并分别命名。通过MIM软件将初始IMRT计划的剂量分布变形到第二程计划CT上以获得变形剂量。使用线性二次模型计算颞叶在2Gy分次下的等效剂量，颞叶的α/β=3。在多变量预测模型中使用AUC分析评估将颞叶照射体积与临床变量相关联的NTCP模型。
2010年1月至2020年12月，78例患者纳入本研究。其中26例（33.3%）发生颞叶损伤（TLI）。影响TLI的最重要因素是颞叶的d1.5cc总剂量，而可能的临床因素在多变量分析中未达到统计学显著差异。根据NTCP模型，d1.5cc总剂量的TD5和TD50 EQD2剂量分别为65.26Gy（46.72 - 80.69Gy）和125.25Gy（89.51 - 152.18Gy）。对于累积EQD2剂量，模型验证中ROC曲线下面积为0.8702（0.7577 - 0.9828），p<0.001。
本研究建立了复发性鼻咽癌第二程IMRT后颞叶损伤的NTCP模型。根据该模型获得了再程放疗后颞叶损伤的TD5和TD50剂量，并通过验证集数据对模型进行了验证。
