淋巴瘤调强放疗后放射性肺损伤分析

发布时间：2018-01-24 10:22

  本文关键词： 淋巴瘤 放射性肺损伤 预测模型　出处：《北京协和医学院》2017年博士论文　论文类型：学位论文

【摘要】：研究背景:淋巴瘤是常见的血液系统肿瘤,而放疗在淋巴瘤的治疗中具有重要地位。在接受膈上放疗的淋巴瘤患者中,放射性肺损伤是胸部主要的剂量限制因素之一。目前淋巴瘤患者的肺部限量主要参考肺癌的标准,但考虑到淋巴瘤患者与肺癌患者的临床和治疗差异,肺部限量或许应有其独特性。研究目的:分析淋巴瘤调强放疗后放射性肺损伤的影响因素,探索各剂量-体积直方图(dose-volume histogram,DVH)参数与放射性肺损伤的关系,建立预测模型观察DVH参数与放射性肺损伤发生概率的相关趋势,寻找淋巴瘤患者肺部剂量限制的临界点。材料与方法:回顾性分析309例调强放疗淋巴瘤患者,其中196例有完整的DVH参数信息。霍奇金淋巴瘤和非霍奇金淋巴瘤患者数分别为151和158例,83.5%的非霍奇金淋巴瘤患者为弥漫大B细胞淋巴瘤。所有患者均采用化疗后受累野放疗的综合治疗方案,5例患者接受了自体干细胞移植。研究的终点事件为所有等级放射性肺损伤和症状性放射性肺损伤,严重程度的评估采用CTCAE 4.0版(肺炎)。建立平滑风险比模型、Lyman-Kutcher-Burman模型(LKB模型)和多因素回归模型观察肺平均剂量(mean lung dose,MLD)和Vx(受照射剂量大于等于x Gy的正常肺组织体积百分比)与放射性肺损伤发生概率的关系,通过ROC曲线寻找潜在限量分界点,通过ROC曲线下面积比较各模型的预测能力。结果:放射性肺损伤的发病率为34.3%,症状性放射性肺损伤为3.6%。对所有等级放射性肺损伤的单因素分析显示,ECOG评分≥2分、乳酸脱氢酶升高、放疗后化疗和所有DVH参数是症状性放射性肺损伤的危险因素,但乳酸脱氢酶升高在多因素分析中丧失了显著性。疗前B症状和V15-V40为症状性放射性肺损伤的危险因素。在目前的淋巴瘤的处方剂量下,MLD和V5-V40与所有等级放射性肺损伤近乎呈线性关系,未发现潜在剂量限制临界点。而对症状性放射性肺损伤而言,低剂量区(V5-V15)与症状性放射性肺损伤相关性差,中等剂量区存在症状性放射性肺损伤的限量临界点(MLD15.1 Gy‘V2028%、V2520%和V3015%),高剂量区(V30-V40)与症状性放射性肺损伤呈线性相关。AUC比较显示,对所有等级症状性放射性肺损伤而言,DVH参数的平滑风险比模型、LKB模型和多因素逻辑回归模型的预测能力无统计学差异。但对症状性放射性肺损伤而言,中-高剂量区的影响更显著,低剂量区的贡献相对较小。结论:要最大程度的降低所有等级放射性肺损伤需根据最佳DVH参数组合最大程度降低肺部受量,但如果只以减少症状为目的,应优先限制中高剂量区的限量问题。
[Abstract]:Background: lymphoma is a common hematologic tumor, and radiotherapy plays an important role in the treatment of lymphoma. Radiation lung injury is one of the main dose limiting factors in the chest. At present, the lung limit of lymphoma patients mainly refers to the standard of lung cancer, but considering the difference of clinical and treatment between lymphoma patients and lung cancer patients. Lung limitation may have to be unique. Objective: to analyze the risk factors of radiation induced lung injury after intensity-modulated radiotherapy for lymphoma. Objective: to explore the relationship between dose-volume histogrammetric volume (DVH) parameters and radiation lung injury. A predictive model was established to observe the correlation between DVH parameters and the probability of radiation-induced lung injury. Materials and methods: 309 patients with IMRT were analyzed retrospectively. Among them, 196 cases had complete DVH parameter information. The number of patients with Hodgkin's lymphoma and non-Hodgkin 's lymphoma were 151 and 158, respectively. 83.5% of patients with non-Hodgkin 's lymphoma were diffuse large B-cell lymphoma. All patients were treated with a combination of post-chemotherapeutic field radiotherapy. Five patients received autologous stem cell transplantation. The endpoint of the study was all grade radiation lung injury and symptomatic radiation lung injury. The severity was assessed using CTCAE version 4.0 (pneumonia). A smooth risk ratio model was established. Lyman-Kutcher-Burman model (LKB model) and multivariate regression model were used to observe mean lung dose. The relationship between MLDs and VX (volume percentage of normal lung tissue with irradiation dose greater than x Gy) and the probability of radiation induced lung injury was found by ROC curve. The predictive ability of each model was compared by the area under the ROC curve. Results: the incidence of radiation lung injury was 34.3%. Symptomatic radiation-induced lung injury was 3.6%. Univariate analysis of all grade radiation-induced lung injury showed that ECOG score 鈮，

本文编号：1459812