摘要
目的探讨基于^(18)F-FDG PET/CT影像组学特征建立的模型对人表皮生长因子受体2(HER2)阳性乳腺癌患者新辅助靶向联合化疗预后评估的价值。方法回顾性分析2016年1月至2022年8月间在天津医科大学肿瘤医院行治疗前^(18)F-FDG PET/CT检查的132例HER2阳性乳腺癌女性患者[(50±11)岁]的资料, 按8∶2分层抽样为训练组(105例)和验证组(27例), 记录临床病理资料及无进展生存(PFS)。在PET和CT图像上进行病灶勾画及影像组学特征提取, 用最小绝对收缩和选择算子(LASSO)算法筛选训练组影像组学特征, 计算影像组学评分(Rad-score)。采用Cox比例风险回归分析筛选新辅助靶向联合化疗后PFS的危险因素, 构建列线图模型, 计算一致性指数(C-index)以评估预测效能。结果单因素Cox回归分析显示, N分期和Rad-score是HER2阳性乳腺癌新辅助靶向联合化疗后PFS的相关危险因素[风险比(HR):2.36(95%CI:1.04~5.37)、14.50(95%CI:3.39~62.13), P值:0.040、<0.001]。多因素Cox回归分析显示, Rad-score是PFS的独立危险因素[HR=13.32(95%CI:3.10~57.20), P<0.001]。联合N分期和Rad-score得到的列线图模型比单独的Rad-score模型能更准确地预测PFS:训练组的C-index分别为0.80和0.74, 验证组的C-index分别为0.77和0.71。结论基于治疗前^(18)F-FDG PET/CT的影像组学能预测HER2阳性乳腺癌患者新辅助靶向联合化疗后PFS;结合影像组学特征与临床危险因素的列线图模型能更有效地预测预后。
Objective:To explore the value of a model based on ^(18)F-FDG PET/CT radiomics features in assessing the prognosis of patients with human epidermal growth factor receptor 2(HER2)-positive breast cancer undergoing neoadjuvant targeted chemotherapy.Methods:This retrospective analysis included 132 female patients(age(50±11)years)diagnosed with HER2-positive breast cancer who underwent ^(18)F-FDG PET/CT prior to treatment between January 2016 and August 2022 in Tianjin Medical University Cancer Institute and Hospital.Data were split into training(105 cases)and validation(27 cases)cohorts using stratified sampling(8∶2).Clinical pathological data and progression-free survival(PFS)were recorded.PET and CT images were annotated for lesion delineation and radiomics features extraction.The least absolute shrinkage and selection operator(LASSO)algorithm was used to select features in the training cohort,and the radiomics score(Rad-score)was calculated.Cox proportional hazards regression analysis was performed to identify risk factors for PFS.A nomogram model was constructed,and the concordance index(C-index)was calculated to assess predictive performance.Results:Univariate Cox regression showed that N stage(hazard ratio(HR)=2.36,95%CI:1.04-5.37,P=0.040)and Rad-score(HR=14.50,95%CI:3.39-62.13,P<0.001)were related to PFS in patients with HER2-positive breast cancer after neoadjuvant therapy.Multivariate analysis indicated the Rad-score as an independent risk factor for PFS(HR=13.32,95%CI:3.10-57.20,P<0.001).The nomogram model combining N stage and Rad-score predicted PFS more accurately than the Rad-score model alone,with C-indexes of 0.80 vs 0.74 in the training cohort,and 0.77 vs 0.71 in the validation cohort.Conclusions:Radiomics based on pre-treatment ^(18)F-FDG PET/CT can predict PFS in patients with HER2-positive breast cancer undergoing neoadjuvant targeted chemotherapy.The nomogram model combining radiomics features and clinical risk factor improves prognostic prediction.
作者
万星
朱磊
张利卜
朱湘
徐文贵
Wan Xing;Zhu Lei;Zhang Libo;Zhu Xiang;Xu Wengui(Department of Molecular Imaging and Nuclear Medicine,Tianjin Medical University Cancer Institute and Hospital,National Clinical Research Center for Cancer,Tianjin's Clinical Research Center for Cancer,Key Laboratory of Breast Cancer Prevention and Therapy,Ministry of Education,Key Laboratory of Cancer Prevention and Therapy,Tianjin 300060,China)
出处
《中华核医学与分子影像杂志》
北大核心
2025年第9期537-542,共6页
Chinese Journal of Nuclear Medicine and Molecular Imaging
基金
天津市医学重点学科(专科)建设项目(TJYXZDXK-009A)。
