基于影像组学不同插值预测脑转移瘤EGFR和HER2表达状态：一项双中心研究

doi:10.13418/j.issn.1001-165x.2023.5.19

中国临床解剖学杂志 ›› 2023, Vol. 41 ›› Issue (5): 608-613.doi: 10.13418/j.issn.1001-165x.2023.5.19

基于影像组学不同插值预测脑转移瘤EGFR和HER2表达状态：一项双中心研究

李嫣然¹，王俭¹，徐彩霞²，靳勇³*

1.新疆医科大学第一附属医院影像中心，乌鲁木齐 830011； 2.长治医学院药学系，山西长治 046000；
3.长治市人民医院影像科，山西长治 046000

收稿日期:2022-06-10 出版日期:2023-09-25 发布日期:2023-10-17
通讯作者: 靳勇，副主任医师，E-mail：350252832@qq.com
作者简介:李嫣然（1996-），女，博士，研究方向：神经影像学，E-mail：13632325045@163.com
基金资助:
山西省2020年度“四个一批”科技兴医创新计划重点专项（2020XM38）

Prediction of EGFR and HER2 expression in brain metastases based on different radiomics interpolator: a two-center study

Li Yanran ¹, Wang Jian ¹, Xu Caixia ², Jin Yong ³*

1. Department of Radiology, The First Affiliated Hospital of Xinjiang Medical University, Urumqi 830011, Xinjiang Province, China；2. Department of Pharmacy, Changzhi Medical College, Changzhi 046000, Shanxi Province, China；3. Department of Radiology, Changzhi People’s Hospital, Changzhi 046000, Shanxi Province, China

Received:2022-06-10 Online:2023-09-25 Published:2023-10-17

摘要/Abstract

摘要： 目的　基于影像组学的不同插值预测脑转移瘤表皮生长因子受体（epidermal growth factor receptor，EGFR）和表皮生长因子受体2（human epidermal growth factor receptor 2，HER2）表达状态，并探索预测效果最佳的插值方法。方法　回顾来自两个机构共100例腺癌脑转移患者资料（56例患者基因表达为突变型EGFR或HER2阳性，44例患者基因表达为野生型EGFR或HER2阴性），在T1WI增强序列选择sitkNearestNeighbor，sitkLinear和sitkBSplines 3种插值分别提取1409个特征，将患者按7：3随机分为训练集和测试集，使用最小绝对收缩选择算子（LASSO）选择信息性特征，支持向量机（support vector machine，SVM）构建诊断模型，训练集用于模型训练，独立测试集评估模型的预测性能，ROC曲线计算模型的准确率、敏感度和特异度，ROC曲线下面积（area under curve，AUC）评估模型的预测性能。结果　基于sitkBSplines插值选定的19个影像组学特征建立的模型显示良好的预测能力，在训练集中，模型的AUC为0.99，分类准确率为0.95，敏感度为0.92，特异度为0.97；在独立测试集中，AUC为0.86，灵敏度0.82，特异度0.78，准确率0.8。sitkBSplines在中心1及中心2的建模中均表现出高于其他插值的鉴别性能。而基于sitkLinear选定的9个影像组学特征建立的模型在训练集和独立测试集的AUC分别为0.74和0.53。结论　MRI影像组学模型对预测腺癌脑转移瘤中EGFR突变/HER2状态具有一定应用价值，其中基于sitkBSplines插值提取的影像组学特征预测效能最优。

关键词: 脑转移瘤, 　影像组学, 　EGFR, 　HER2

Abstract: Objective To predict epidermal growth factor receptor (EGFR) mutation status and human epidermal growth factor receptor 2 (HER2) expression status of brain metastasis based on different radiomics interpolator and to explore the optimal interpolator for prediction. Methods Data from 100 patients with brain metastasis from adenocarcinoma (56 with mutant EGFR/HER2+, 44 with wild-type EGFR/HER2-) from 2 institutions were retrospectively reviewed and analyzed. Contrast-enhanced T1-weighted imaging (T1-CE) sequence was selected for radiomics features extraction by using 3 different interpolators (sitkNearestNeighbor, sitkLinear and sitkBSplines). A total of 1409 radiomics features were extracted from each MR interpolator. The patients were randomly divided into training coherent and independent testing coherent according to 7:3. The least absolute shrinkage selection operator (LASSO) was used to select informative features, a radiomics signature was built with the support vector machine (SVM) model of the training cohort, and the radiomics signature performance was evaluated by using an independent testing data set. The accuracy, sensitivity and specificity of the model was calculated by the ROC curve. The predictive performance of the model was assessed by the ROC area under curve (AUC). Results Nineteen selected radiomics features based on sitkBSplines interpolator showed good discrimination in both the training and independent test cohorts. The radiomics signature yielded an AUC of 0.99, a classification accuracy of 0.95, sensitivity of 0.92, and specificity of 0.97 in the training cohort, and an AUC of 0.86, a classification accuracy of 0.8, sensitivity of 0.82, and specificity of 0.78 in the independent testing data set. SitkBSplines showed better discrimination performance than other interpolations in both Center 1 and Center 2 modeling. Nine selected radiomics signature features based on sitkLinear interpolator yielded an AUC of 0.74 in training cohort and an AUC of 0.53 in the independent testing cohort. Conclusions Radiomics signature model has certain application value in predicting EGFR mutation /HER2 status in adenocarcinoma brain metastases, among which radiomics features based on sitkBSplines interpolation is the most effective in discrimination performance.

Key words: Brain metastasis; , , Radiomics; , , EGFR; , , HER2

中图分类号:

R739.41
R445

李嫣然, 王俭, 徐彩霞, 靳勇. 基于影像组学不同插值预测脑转移瘤EGFR和HER2表达状态：一项双中心研究[J]. 中国临床解剖学杂志, 2023, 41(5): 608-613.

Li Yanran, Wang Jian, Xu Caixia, Jin Yong. Prediction of EGFR and HER2 expression in brain metastases based on different radiomics interpolator: a two-center study[J]. Chinese Journal of Clinical Anatomy, 2023, 41(5): 608-613.

参考文献 14

[1]	Lemmon MA, Schlessinger J. Cell signaling by receptor tyrosine kinases[J]. Cell,2010, 141(7): 1117-1134. DOI: 10.1016/j.cell. 2010. 06.011.
[2]	Avraham R, Yarden Y. Feedback regulation of EGFR signalling: decision making by early and delayed loops[J]. Nat Rev Mol Cell Biol, 2011, 12(2): 104-117. DOI: 10.1038/nrm3048.
[3]	Borrero-García LD, Del Mar Maldonado M, Medina-Velázquez J, et al. Rac inhibition as a novel therapeutic strategy for EGFR/HER2 targeted therapy resistant breast cancer[J]. BMC Cancer, 2021, 21(1): 652. DOI: 10.1186/s12885-021-08366-7.
[4]	Karami E, Soliman H, Ruschin M, et al. Quantitative MRI biomarkers of stereotactic radiotherapy outcome in brain metastasis[J]. Sci Rep, 2019, 9(1): 19830. DOI: 10.1038/s41598-019-56185-5.
[5]	Galldiks N, Kocher M, Ceccon G, et al. Imaging challenges of immunotherapy and targeted therapy in patients with brain metastases: response, progression, and pseudoprogression[J]. Neuro Oncol, 2020, 22(1): 17-30. DOI: 10.1093/neuonc/noz147.
[6]	Koay EJ, Lee Y, Cristini V, et al. A visually apparent and quantifiable CT imaging feature identifies biophysical subtypes of pancreatic ductal adenocarcinoma[J]. Clin Cancer Res, 2018, 24(23): 5883-5894. DOI: 10.1158/1078-0432.Ccr-17-3668.
[7]	Grossmann P, Stringfield O, EL-Hachem N, et al. Defining the biological basis of radiomic phenotypes in lung cancer[J]. Elife, 2017, 6: e23421. DOI: 10.7554/eLife.23421.
[8]	Tu W, Sun G, Fan L, et al. Radiomics signature: a potential and incremental predictor for EGFR mutation status in NSCLC patients, comparison with CT morphology[J]. Lung Cancer, 2019, 132: 28-35. DOI: 10.1016/j.lungcan.2019.03.025.
[9]	Mei D, Luo Y, Wang Y, et al. CT texture analysis of lung adenocarcinoma: can Radiomic features be surrogate biomarkers for EGFR mutation statuses[J]. Cancer Imaging, 2018, 18(1): 52. DOI: 10.1186/s40644-018-0184-2.
[10]	Zhou J, Tan H, Li W, et al. Radiomics signatures based on multiparametric MRI for the preoperative prediction of the HER2 status of patients with breast cancer[J]. Acad Radiol, 2021, 28(10): 1352-1360. DOI: 10.1016/j.acra.2020.05.040.
[11]	An Z, Aksoy O, Zheng T, et al. Epidermal growth factor receptor and EGFRvIII in glioblastoma: signaling pathways and targeted therapies[J]. Oncogene, 2018, 37(12): 1561-1575. DOI: 10.1038/s41388-017-0045-7.
[12]	Sabbah DA, Hajjo R, Sweidan K. Review on epidermal growth factor receptor (EGFR) structure, signaling pathways, interactions, and recent updates of EGFR inhibitors[J]. Curr Top Med Chem, 2020, 20(10): 815-834. DOI: 10.2174/1568026620666200303123102.
[13]	Cruz RGB, Madden SF, Brennan K, et al. A transcriptional link between HER2, JAM-A and FOXA1 in breast cancer[J]. Cells, 2022, 11(4): 735. DOI: 10.3390/cells11040735.
[14]	Roviello G, Catalano M, Iannone LF, et al. Current status and future perspectives in HER2 positive advanced gastric cancer[J]. Clin Transl Oncol, 2022, 24(6): 981-996. DOI: 10.1007/s12094-021-02760-0.

基于影像组学不同插值预测脑转移瘤EGFR和HER2表达状态：一项双中心研究

Prediction of EGFR and HER2 expression in brain metastases based on different radiomics interpolator: a two-center study

PDF (PC)

可视化

摘要/Abstract

引用本文

使用本文

参考文献 14

相关文章 4

Metrics

本文评价

推荐阅读 0

[1]	蔡梅钦, 罗伦, 何海勇, 郑文汉, 张保豫, 李文胜, 郭英. 神经内镜扩大经鼻入路在颅咽管瘤手术的应用[J]. 中国临床解剖学杂志, 2021, 39(4): 469-472.
[2]	邓昭健, 蒋其俊, 甘正凯. 眉弓入路显微手术治疗颅咽管瘤的临床应用及疗效[J]. 中国临床解剖学杂志, 2019, 37(6): 704-707.
[3]	王守森, 丁陈禹, 秦勇, 魏梁锋, 魏淋, 肖德勇, 王如密. 垂体腺瘤患者蝶鞍形态的MRI研究[J]. 中国临床解剖学杂志, 2015, 33(2): 160-165.
[4]	胡晓芳, 王永魁, 邱小忠, 王国保. Nix和miR-520h对神经胶质瘤发生的影响及相互作用[J]. 中国临床解剖学杂志, 2014, 32(5): 590-594.