目的 基于磁共振成像（magnetic resonance imaging,MRI）影像组学和临床特征构建鼻咽癌（nasopharyngeal cancer,NPC）患者放疗后鼻咽坏死（nasopharyngeal necrosis after radiotherapy,PRNN）的nomogram预测模型。方法 收集梅州市人民医院2017年1月至2023年10月243例NPC患者作为研究对象,按照8∶2比例分为训练集（n=194）、测试集（n=49）。所有NPC患者均接受放疗,训练集根据患者是否出现PRNN,分为PRNN组（n=57）、无PRNN组（n=137）。比较训练集中两组患者临床资料,LASSO回归筛选MRI影像组学特征,构建影像组学评分,多因素Logistic回归分析NPC患者发生PRNN的影响因素,构建训练集、测试集nomogram预测模型,采用受试者工作特征曲线（receiver operating characteristic curve,ROC）、校准曲线和决策曲线（decision curve analysis,DCA）曲线评估预测模型效能。结果 无PRNN组与PRNN组患者年龄、体质量指数（body mass index,BMI）、AJCC-T分期、二程放疗比较,差异具有统计学意义;LASSO回归筛选获取35个影像组学特征,计算影像组学评分;多因素logistic回归显示,年龄、BMI、AJCC-T分期、二程放疗及影像组学评分是NPC患者发生PRNN的影响因素;将年龄、BMI、AJCC-T分期、二程放疗作为临床模型,影像组学评分作为影像组学模型,年龄、BMI、AJCC-T分期、二程放疗及影像组评分作为组合模型,用于构建nomogram预测模型;ROC曲线、校准曲线及DCA曲线显示,测试集和训练集中组合预测模型具有明显临床净收益和良好预测效能。结论 年龄、BMI、AJCC-T分期、二程放疗及影像组学评分是NPC患者PRNN的影响因素,根据上述因素构建的nomogram预测模型具有良好预测效能和临床适用性。
Abstract
Objective To construct a nomogram prediction model for nasopharyngeal necrosis after radiotherapy (PRNN) in patients with nasopharyngeal cancer (NPC) based on MRI radiomics and clinical features. Methods 243 cases of NPC from Meizhou People's Hospital from January 2017 to October 2023 were selected as the research subjects. They were divided into the training set (n=194) and the test set (n=49) in a ratio of 8∶2. All NPC patients received radiotherapy. The training set was divided into the PRNN group (n=57) and the non-PRNN group (n=137) based on whether the patients presented with PRNN. By comparing the clinical data of the two groups of patients in the training set, LASSO regression was used to screen the MRI imaging features, and an imaging feature score was constructed. Nomogram prediction models for the training set and test set were constructed. Multivariate logistic regression analysis was conducted to identify the influencing factors of PRNN occurrence in NPC patients. Nomogram prediction models for the training set and test set were constructed. Evaluate the performance of the prediction model using receiver operating characteristic curve (ROC), calibration curve, and decision curve analysis (DCA) curve. Result There were statistically significant differences in age, body mass index (BMI), AJCC-T stage, and the second course of radiotherapy between the non-PRNN group and the PRNN group. LASSO regression was used to screen for 35 radiomics features, and the radiomics score was calculated. Multivariate Logistic regression showed that age, BMI, AJCC-T stage, second course of radiotherapy, and radiomics score were the influencing factors for PRNN in NPC patients. Age, BMI, AJCC-T stage, second course of radiotherapy, and radiomics score were used as the clinical model, and the radiomics score as the radiomics model. Age, BMI, AJCC-T stage, second course of radiotherapy, and radiomics score were used as the combined model to construct a nomogram prediction model. ROC curve, calibration curve, and DCA curve showed that the combined prediction model in the test set and training set had significant clinical net benefits and good predictive efficacy. Conclusion Age, BMI, AJCC-T stage, second course of radiotherapy, and radiomics score were the influencing factors for PRNN in NPC patients. The nomogram prediction model constructed based on the above factors had good predictive efficacy and clinical applicability.
关键词
磁共振成像 /
影像组学 /
临床特征 /
鼻咽癌 /
放疗后鼻咽坏死 /
nomogram预测模型
Key words
magnetic resonance imaging /
radiomics /
clinical features /
nasopharyngeal carcinoma /
post-radiotherapy nasopharyngeal necrosis /
nomogram prediction model
{{custom_sec.title}}
{{custom_sec.title}}
{{custom_sec.content}}
参考文献
[1] JAINAR CJE, JACOMINA LE, DEE EC, et al. Global disparities in current clinical trials in nasopharyngeal carcinoma[J]. Int J Radiat Oncol Biol Phys, 2025, 123(4): 1041-1049.
[2] HUANG H, YAO Y, DENG X, et al.Immunotherapy for nasopharyngeal carcinoma: Current status and prospects (Review)[J]. Int J Oncol, 2023, 63(2): 97.
[3] CHOW JCH, CHEUNG KM, AU KH, et al.Radiation-induced hypoglossal nerve palsy after definitive radiotherapy for nasopharyngeal carcinoma: Clinical predictors and dose-toxicity relationship[J]. Radiother Oncol, 2019, 138: 93-98.
[4] FEI ZD, CHEN TJ, QIU XF, et al.Effect of relevant factors on radiation-induced nasopharyngeal ulcer in patients with primary nasopharyngeal carcinoma treated with intensity-modulated radiation therapy[J]. Laryngoscope Investig Otolaryngol, 2020, 5(2): 228-234.
[5] XU M, LIU C, MI J L, et al.Nomogram for prognosis of nasopharyngeal carcinoma based on tumor residue detected by MR imaging at the end of intensity-modulated radiotherapy[J]. Cancer Manag Res, 2020, 12: 4523-4532.
[6] WANG CK, WANG TW, YANG YX, et al.Deep learning for nasopharyngeal carcinoma segmentation in magnetic resonance imaging: a systematic review and meta-analysis[J]. Bioengineering (Basel), 2024, 11(5): 504.
[7] QI YJ, SU GH, YOU C, et al.Radiomics in breast cancer: Current advances and future directions[J]. Cell Rep Med, 2024, 5(9): 101719.
[8] 郗玉珍, 华鹏, 姜锋, 等. Delta影像组学在预测鼻咽癌诱导化疗联合同步放化疗疗效的价值[J]. 临床放射学杂志, 2023, 42(02): 216-221.
[9] 邓智毅, 叶祎菁, 李定波, 等. 基于MRI影像组学特征因素预测Ⅱ~Ⅳa期鼻咽癌患者复发转移风险及辅助化疗受益的临床意义[J]. 中国耳鼻咽喉头颈外科, 2024, 31(08): 477-484.
[10] TANG LL, CHEN YP, CHEN CB, et al.The Chinese society of clinical oncology (CSCO) clinical guidelines for the diagnosis and treatment of nasopharyngeal carcinoma[J]. Cancer Commun (Lond), 2021, 41(11): 1195-1227.
[11] 欧阳希. 初治M0期鼻咽癌VMAT放疗后鼻咽坏死的危险因素分析[D]. 南昌: 南昌大学医学部, 2023.
[12] LI KY, KWOK HM, CHEUK W, et al.Demystifying the challenging diagnosis of post-radiation nasopharyngeal necrosis on multimodality imaging[J]. J Med Imaging Radiat Oncol, 2024, 68(7): 805-807.
[13] YANG XL, LIN L, HE SS, et al.Nasopharyngeal necrosis following intensity-modulated radiation therapy of primary nasopharyngeal carcinoma-incidence rate and predictors of risk[J]. BMC Cancer, 2025, 25(1): 802.
[14] 宗丹, 黄文轩, 郭业松, 等. 鼻咽癌首程根治性放疗后鼻咽坏死治疗策略及预后分析[J]. 中华放射肿瘤学杂志, 2024, 33(9): 797-803.
[15] LI XY, SUN XS, LIU SL, et al.The development of a nomogram to predict post-radiation necrosis in nasopharyngeal carcinoma patients: a large-scale cohort study[J]. Cancer Manag Res, 2019, 11: 6253-6263.
[16] YANG K, AHN YC, NAM H, et al.Clinical features of post-radiation nasopharyngeal necrosis and their outcomes following surgical intervention in nasopharyngeal cancer patients[J]. Oral Oncol, 2021, 114: 105180.
[17] YUAN J, WU M, QIU L, et al.Tumor habitat-based MRI features assessing early response in locally advanced nasopharyngeal carcinoma[J]. Oral Oncol, 2024, 158: 106980.
[18] 贡瑞敏, 赵月琳, 万桂伶, 等. MRI纹理分析+DWI预测鼻咽癌放化疗应答性的临床研究[J]. 河北医科大学学报, 2025, 46(06): 649-655.
[19] HOU J, HE Y, LI H, et al.MRI-based radiomics models predict cystic brain radionecrosis of nasopharyngeal carcinoma after intensity modulated radiotherapy[J]. Front Neurol, 2024, 15: 1344324.
[20] 王卓, 张少茹, 周云舒, 等. 基于多参数MRI影像组学的列线图预测鼻咽癌诱导化疗效果[J]. 中国医学影像学杂志, 2023, 31(05): 459-466.
[21] HUANG L, YANG Z, ZENG Z, et al.MRI-based radiomics models for the early prediction of radiation-induced temporal lobe injury in nasopharyngeal carcinoma[J]. Front Neurol, 2023, 14: 1135978.
基金
梅州市社会发展科技计划项目 “基于MRI放射组学和临床特征的机器学习模型预测鼻咽癌治疗后发生鼻咽坏死”（2024C0301045）
PDF(4210 KB)
