|
| 基于新型放疗计划评价工具的放射性直肠炎Nomogram预测模型研究 |
| Study on the Nomogram prediction model for radiation proctitis based on a new radiotherapy plan evaluation toolGE Xudong, CUI Shujun, TIAN Long |
| 投稿时间：2024-11-08 修订日期：2025-02-06 |
| DOI： |
|
 |
| 中文关键词: 前列腺癌 剂量表面积直方图 放射性直肠炎 Nomogram模型 预测 |
| 英文关键词:prostate cancer dose surface histogram radiation proctitis Nomogram model forecast |
| 基金项目:张家口市重点研发计划项目（2421018D） |
|
| 摘要点击次数: 88 |
| 全文下载次数: 0 |
| 中文摘要: |
| 摘 要：[目的] 基于新型放疗计划工具评价结果等临床资料创建前列腺癌（prostate cancer，PCa）放疗致急性放射性直肠炎（radiation proctitis，RP）Nomogram预测模型，评价模型的临床应用价值。[方法] 筛选河北北方学院附属第一医院医学影像部2020年7月至2024年7月PCa患者200例，采用中央随机系统分配法分为建模组（n=140）和验证组（n=60）。基于患者CT影像资料创建新型放疗计划评价工具-剂量表面积直方图（dose and surface histogram，DSH）。基于建模组DSH评价结果等临床资料，采用logistic回归分析急性RP影响因素并创建Nomogram预测模型。通过建模组内部验证和验证组外部验证评价模型的临床应用价值。[结果] Nomogram预测模型显示，肿瘤体积（OR=1.420，95%CI：1.105～2.394）、计划肿瘤体积最大剂量（OR=2.171，95%CI：1.374～3.763）、S70（OR=2.956，95%CI：1.579～5.093）、S78（OR=4.735，95%CI：1.983～8.227）均是影响急性RP发送的独立危险因素（均P＜0.05）。肿瘤体积、计划肿瘤体积最大剂量、S70、S78测急性RP的AUC分别为0.610（95%CI：0.572～0.733）、0.658（95%CI：0.584～0.765）、0.692（95%CI：0.603～0.793）、0.703（95%CI：0.623～0.835），最佳截断值分别为3.02 cm3、73.11 Gy、10.54%、5.22%。建模组内部验证：模型一致性指数为0.832（95%CI：0.781～0.949），校准曲线上观测值同实际值一致性良好，模型能够提供高于各独立危险因素的临床净收益。建模组和验证组外部验证：模型预测两组急性RP的AUC均高于各独立危险因素，两组ROC曲线拟合良好（χ2=1.893，P＞0.05），差异无统计学意义（P＞0.05）。[结论] 基于新型放疗计划工具评价结果等临床资料的急性RP Nomogram预测模型具有一定的临床应用价值，能够为急性RP的防治提供有价值的参考。 |
| 英文摘要: |
| Abstract: [Objective] To create a Nomogram prediction model for acute radiation proctitis (RP) in prostate cancer (PCa) radiotherapy based on the clinical data such as the evaluation result of a new radiotherapy planning tool, and to evaluate the clinical application value of the model. [Methods] 200 PCa patients in the medical imaging department of the First Affiliated Hospital of Hebei North University from July 2020 to July 2024 were selected. They were divided into the modeling group (n=140) and validation group (n=60) by using the central random system allocation method. A new radiotherapy plan evaluation tool based on CT imaging data of the patients was created. It was named as dose surface histogram (DSH). Based on clinical data such as DSH evaluation result in the modeling group, logistic regression analysis was used to identify acute RP influencing factors and the Nomogram prediction model was created. The clinical application value of the model was evaluated through the internal and external validation within the modeling group. [Results] Nomogram prediction model showed that tumor volume (OR=1.420, 95%CI: 1.105-2.394), maximum dose for planning tumor volume (OR=2.171, 95%CI: 1.374-3.763), S70 (OR=2.956, 95%CI: 1.579-5.093), and S78 (OR=4.735, 95%CI: 1.983-8.227) were all the independent risk factors for acute RP (all P<0.05). AUC values of tumor volume, maximum dose of planning tumor volume, S70 and S78 for acute RP prediction were 0.610 (95%CI: 0.572-0.733), 0.658 (95%CI: 0.584-0.765), 0.692 (95%CI: 0.603-0.793), and 0.703 (95%CI: 0.623-0.835), respectively. The optimal cutoff values were 3.02 cm3, 73.11 Gy, 10.54%, and 5.22%, respectively. Internal validation in the modeling group: The model consistency index was 0.832 (95% CI: 0.781-0.949), and the observation values on the calibration curve were in good agreement with the actual values. The model could provide clinical net benefits higher than those of each independent risk factor. External validation in the modeling and validation group: The AUC predicted by the model for acute RP in two groups were higher than those of each independent risk factor. The ROC curves of the two groups were well fitted (χ2=1.893, P>0.05), with no statistically significant difference (P>0.05). [Conclusions] The acute RP Nomogram prediction model based on clinical data such as the evaluation result of the new radiotherapy planning tool has certain clinical application value and can provide valuable references for the prevention and treatment of acute RP. |
|
在线阅读
查看/发表评论 下载PDF阅读器 |
|
|
|