基于乳腺X线摄影及DCE-MRI机器学习模型预测乳腺癌新辅助治疗后病理完全缓解:双中心研究

doi:10.6040/j.issn.1671-7554.0.2024.1296

摘要/Abstract

摘要： 目的探讨基于乳腺X线摄影及动态增强磁共振成像(dynamic contrast-enhanced magnetic resonance imaging, DCE-MRI)的机器学习模型预测乳腺癌患者新辅助治疗(neoadjuvant therapy, NAT)后病理完全缓解(pathological complete response, pCR)的价值。方法回顾性分析2016年8月至2023年7月于青岛大学附属医院(机构1)及烟台毓璜顶医院(机构2)接受NAT后行手术的396例乳腺癌患者资料,来自机构1的320例患者按7∶3比例随机分为训练集和验证集,来自机构2的76例患者作为独立的外部验证集。对患者NAT前乳腺X线摄影及DCE-MRI图像进行感兴趣区域(region of interest, ROI)勾画、特征提取、特征筛选,使用支持向量机(support vector machine, SVM)机器学习算法构建影像组学模型。对临床特征进行单因素-多因素逻辑回归分析,保留具有统计学意义的临床独立预测因子并构建临床模型。将联合影像组学模型与临床独立预测因子使用SVM机器学习算法联合构建综合模型。采用受试者工作特征(receiver operating characteristic, ROC)曲线的曲线下面积(area under the curves, AUC)、准确性、敏感性、特异性和F1-score评价模型的性能,绘制校准曲线评价模型的拟合优度,采用决策曲线分析(decision curve analysis, DCA)评估模型的临床应用价值。结果联合影像组学模型预测性能高于临床模型、乳腺X线摄影影像组学模型和MRI影像组学模型,其在训练集、验证集和外部验证集AUC分别为0.899、0.850及0.765。综合模型预测性能最佳,其在训练集、验证集和外部验证集AUC分别为0.918、0.856、0.795,且该模型具有良好的校准能力和临床收益。Delong检验示临床模型与综合模型的AUC的差异有统计学意义(P<0.05)。结论基于乳腺X线摄影及DCE-MRI的机器学习模型可以预测乳腺癌患者NAT后pCR,且具有较高的预测性能。

关键词: 乳腺癌, 影像组学, 新辅助治疗, 乳腺X线摄影, 磁共振成像

Abstract: Objective To investigate the value of machine learning models based on mammography and dynamic contrast-enhanced magnetic resonance imaging(DCE-MRI)in predicting pathological complete response(pCR)in breast cancer patients after neoadjuvant therapy(NAT). Methods A retrospective analysis of 396 breast cancer patients who underwent NAT followed by surgery from August 2016 to July 2023 at the Affiliated Hospital of Qingdao University(Institution 1)and Yantai Yuhuangding Hospital(Institution 2)was performed. A total of 320 patients from Institution 1 were randomly divided into a training set and a validation set in a ratio of 7∶3, and 76 patients from Institution 2 served as an independent external validation set. Regions of interest(ROI)were delineated on pre-NAT mammography and DCE-MRI images, followed by feature extraction and feature selection. The radiomics model was constructed using the support vector machine(SVM)algorithm. Clinical features underwent univariate and multivariate analyses, and statistically significant independent predictors were used to construct the clinical model. The comprehensive model integrating the radiomics signature and clinical predictors was constructed using the SVM algorithm. The performance of the models was evaluated using the area under the receiver operating characteristic(AUC)curve, accuracy, sensitivity, specificity, and F1-score. The calibration efficiency of the predictive models was evaluated by drawing calibration curves and decision curve analysis(DCA)was performed to evaluate the clinical utility of the predictive models. Results The combined radiomics model demonstrated better predictive performance than the clinical model, mammography radiomics model and MRI radiomics model, with AUCs of 0.899, 0.850, and 0.765 in the training, validation, and external validation sets, respectively. The comprehensive model showed the best predictive performance, with AUCs of 0.918, 0.856, and 0.795 in the training, validation, and external validation sets, respectively. The comprehensive model exhibited good calibration ability and clinical benefit. The Delong test showed statistically significant difference between the clinical model and the comprehensive model(P<0.05). Conclusion Machine learning models based on mammography and DCE-MRI effectively predict pCR in breast cancer patients after NAT, and demonstrate preferable predictive performance.

Key words: Breast cancer, Radiomics, Neoadjuvant therapy, Mammography, Magnetic resonance imaging

中图分类号:

R737.9

刘晶晶,庞婧,赵晓丹,林昕,付敏,陈静静. 基于乳腺X线摄影及DCE-MRI机器学习模型预测乳腺癌新辅助治疗后病理完全缓解:双中心研究[J]. 山东大学学报 (医学版), 2025, 63(1): 60-72.

LIU Jingjing, PANG Jing, ZHAO Xiaodan, LIN Xin, FU Min, CHEN Jingjing. Machine learning model based on mammography and DCE-MRI to predict pathological complete response after neoadjuvant therapy in breast cancer: a dual center research[J]. Journal of Shandong University (Health Sciences), 2025, 63(1): 60-72.

参考文献

[1] Bray F, Laversanne M, Sung H, et al. Global cancer statistics 2022: GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185 countries[J]. CA Cancer J Clin, 2024, 74(3): 229-263.
[2] 中国抗癌协会乳腺癌专业委员会,中华医学会肿瘤学分会乳腺肿瘤学组.中国抗癌协会乳腺癌诊治指南与规范(2024年版)[J]. 中国癌症杂志, 2023, 33(12): 1092-1187. The Society of Breast Cancer China Anti-Cancer Association, Breast Oncology Group of the Oncology Branch of the Chinese Medical Association. Guidelines for breast cancer diagnosis and treatment by China Anti-cancer Association(2024 edition)[J]. China Oncology, 2023, 33(12): 1092-1187.
[3] 初竹秀,赵文静,李小燕,等.218例女性乳腺癌患者行新辅助化疗及伴随分子标志物改变的临床价值[J].山东大学学报(医学版), 2021, 59(9): 130-139. CHU Zhuxiu, ZHAO Wenjing, LI Xiaoyan, et al. Significance of neoadjuvant chemotherapy and molecular marker changes in 218 women with breast cancer[J]. Journal of Shandong University(Health Sciences), 2021, 59(9): 130-139.
[4] Reig B, Heacock L, Lewin A, et al. Role of MRI to assess response to neoadjuvant therapy for breast cancer[J]. J Magn Reson Imaging, 2020, 52(6): 1587-1606.
[5] 陈鑫,陈小波,梁演婷,等.基于MRI影像组学的可解释性机器学习模型预测乳腺癌新辅助治疗病理完全缓解[J]. 临床放射学杂志, 2024, 43(10): 1666-1672. CHEN Xin, CHEN Xiaobo, LIANG Yanting, et al. Interpretable machine learning model based on MRI radiomics for predicting pathological complete response to neoadjuvant therapy in breast cancer[J]. Journal of Clinical Radiology, 2024, 43(10): 1666-1672.
[6] Yu YM, Wang ZB, Wang Q, et al. Radiomic model based on magnetic resonance imaging for predicting pathological complete response after neoadjuvant chemotherapy in breast cancer patients[J]. Front Oncol, 2023, 13: 1249339. doi: 10.3389/fonc.2023.1249339.
[7] 李小苑,杨志企,陈湘光,等.联合ADC组学标签和激素受体预测乳腺癌新辅助化疗病理完全化解[J].中国CT和MRI杂志, 2024, 22(1): 88-90. LI Xiaoyuan, YANG Zhiqi, CHEN Xiangguang, et al. Combined apparent diffusion coefficient maps radiomics signature and hormone receptors predict pathological complete response of neoadjuvant chemotherapy in breast cancer[J]. Chinese Journal of CT and MRI, 2024, 22(1): 88-90.
[8] 冯晓丹,刘菲菲,杜瑶,等.基于超声影像组学及临床特征的可解释性机器学习模型预测乳腺癌新辅助化疗疗效[J].中国超声医学杂志, 2024, 40(3): 278-282. FENG Xiaodan, LIU Feifei, DU Yao, et al. Predicting the effect of interpretable machine learning model based on ultrasound radiomics and clinical features on neoadjuvant chemotherapy for breast cancer[J]. Chinese Journal of Ultrasound in Medicine, 2024, 40(3): 278-282.
[9] Yu YF, He ZF, Ouyang J, et al. Magnetic resonance imaging radiomics predicts preoperative axillary lymph node metastasis to support surgical decisions and is associated with tumor microenvironment in invasive breast cancer: a machine learning, multicenter study[J]. EBioMedicine, 2021, 69: 103460. doi: 10.1016/j.ebiom.2021.103460.
[10] Mao N, Yin P, Wang QL, et al. Added value of radiomics on mammography for breast cancer diagnosis: a feasibility study[J]. J Am Coll Radiol, 2019, 16(4 Pt A): 485-491.
[11] 南帅明,邓雪英. MRI评价乳腺癌新辅助治疗疗效的研究进展[J].临床放射学杂志, 2021, 40(11): 2219-2223.
[12] Hayward JH, Linden OE, Lewin AA, et al. ACR appropriateness criteria^® monitoring response to neoadjuvant systemic therapy for breast cancer: 2022 update[J]. J Am Coll Radiol, 2023, 20(5s): S125-s145.
[13] Zeng Q, Xiong F, Liu L, et al. Radiomics based on DCE-MRI for predicting response to neoadjuvant therapy in breast cancer[J]. Acad Radiol, 2023, 30(Suppl 2): S38-S49.
[14] Mayerhoefer ME, Materka A, Langs G, et al. Introduction to radiomics[J]. J Nucl Med, 2020, 61(4): 488-495.
[15] Zheng D, He XJ, Jing J. Overview of artificial intelligence in breast cancer medical imaging[J]. Clin Med, 2023, 12(2): 419. doi: 10.3390/jcm12020419.
[16] Liu Y, Li X, Zhu LN, et al. Preoperative prediction of axillary lymph node metastasis in breast cancer based on intratumoral and peritumoral DCE-MRI radiomics nomogram[J]. Contrast Media Mol Imaging, 2022, 2022: 6729473. doi: 10.1155/2022/6729473.
[17] Hua YC, Peng QQ, Han JQ, et al. A two-center study of a combined nomogram based on mammography and MRI to predict ALN metastasis in breast cancer[J]. Magn Reson Imaging, 2024, 110: 128-137. doi: 10.1016/j.mri.2024.04.019.
[18] 美国放射学院. 王殊,洪楠,译. 乳腺影像报告与数据系统图谱(2013版)[M]. 北京:北京大学医学出版社,2016.
[19] Tian YQ, Hua H, Peng QQ, et al. Preoperative evaluation of Gd-EOB-DTPA-enhanced MRI radiomics-based nomogram in small solitary hepatocellular carcinoma(≤3 cm)with microvascular invasion: a two-center study[J]. J Magn Reson Imaging, 2022, 56(5): 1459-1472.
[20] Wang XL, Hua H, Han JQ, et al. Evaluation of multiparametric MRI radiomics-based nomogram in prediction of response to neoadjuvant chemotherapy in breast cancer: a two-center study[J]. Clin Breast Cancer, 2023, 23(6): e331-e344.
[21] Di Cosimo S, Depretto C, Miceli R, et al. Mammographic density to predict response to neoadjuvant systemic breast cancer therapy[J]. J Cancer Res Clin Oncol, 2022, 148(4): 775-781.
[22] Zhao R, Lu H, Li YB, et al. Nomogram for early prediction of pathological complete response to neoadjuvant chemotherapy in breast cancer using dynamic contrast-enhanced and diffusion-weighted MRI[J]. Acad Radiol, 2022, 29(Suppl 1): S155-S163.
[23] Paakkola NM, Karakatsanis A, Mauri D, et, al. The prognostic and predictive impact of low estrogen receptor expression in early breast cancer: a systematic review and meta-analysis[J]. ESMO Open, 2021, 6(6): 100289. doi: 10.1016/j.esmoop.2021.100289.
[24] Zheng GY, Hou J, Shu ZY, et al. Prediction of neoadjuvant chemotherapy pathological complete response for breast cancer based on radiomics nomogram of intratumoral and derived tissue[J]. BMC Med Imaging, 2024, 24(1): 22. doi: 10.1186/s12880-024-01198-4.
[25] Jiang T, Song JD, Wang XY, et al. Intratumoral and peritumoral analysis of mammography, tomosynthesis, and multiparametric MRI for predicting Ki-67 level in breast cancer: a radiomics-based study[J]. Mol Imaging Biol, 2022, 24(4): 550-559.
[26] 张玉姣,宋德领,王燕飞,等.基于乳腺X线图像影像组学列线图对乳腺癌腋窝淋巴结转移的预测价值[J].放射学实践, 2022, 37(1): 48-54. ZHANG Yujiao, SONG Deling, WANG Yanfei, et al. Predictive value of breast canceraxillary lymph node metastasis based on the radiomics nomogram of mammography[J]. Radiologic Practice, 2022, 37(1): 48-54.
[27] 万宏燕,徐井旭,杨瑜,等.基于X线摄影影像组学特征鉴别乳腺良恶性肿块的价值[J].医学影像学杂志, 2023, 33(5): 773-776. WAN Hongyan, XU Jingxu, YANG Yu, et al. Radiomics features of mammography in distinguishing benign and malignant breast masses[J]. Journal of Medical Imaging, 2023, 33(5): 773-776.
[28] Skarping I, Larsson M, Förnvik D. Analysis of mammograms using artificial intelligence to predict response to neoadjuvant chemotherapy in breast cancer patients: proof of concept[J]. Eur Radiol, 2022, 32(5): 3131-3141.
[29] Zhang K, Lin J, Lin F, et al. Radiomics of contrast-enhanced spectral mammography for prediction of pathological complete response to neoadjuvant chemotherapy in breast cancer[J]. J Xray Sci Technol, 2023, 31(4): 669-683.
[30] 刘凯文,金莹莹,王守巨.人工智能辅助的磁共振成像在评估乳腺癌新辅助化疗中的应用综述[J].数据采集与处理, 2024, 39(4): 794-812. LIU Kaiwen, JIN Yingying, WANG Shouju. Artificial intelligence-assisted magnetic resonance imaging in assessment of neo adjuvant chemotherapy for breast cancer: a review[J]. Journal of Data Acquisition and Processing, 2024, 39(4): 794-812.
[31] Caballo M, Sanderink WBG, Han L, et al. Four-dimensional machine learning radiomics for the pretreatment assessment of breast cancer pathologic complete response to neoadjuvant chemotherapy in dynamic contrast-enhanced MRI[J]. J Magn Reson Imaging, 2023, 57(1): 97-110.
[32] 岳文怡,张洪涛,高珅,等.基于机器学习的组合模型在预测乳腺癌新辅助化疗疗效中的价值[J].磁共振成像, 2024, 15(3): 93-99. YUE Wenyi, ZHANG Hongtao, GAO Shen, et al. Radiomics based on combined machine learning models for prediction of the response to neoadjuvant chemotherapy in mass enhancement breast cancer using magnetic resonance imaging[J]. Chinese Journal of Magnetic Resonance Imaging, 2024, 15(3): 93-99.
[33] Sutton EJ, Onishi N, Fehr DA, et al. A machine learning model that classifies breast cancer pathologic complete response on MRI post-neoadjuvant chemotherapy[J]. Breast Cancer Res, 2020, 22(1): 57. doi: 10.1186/s13058-020-01291-w.
[34] Periyasamy S, Prakasarao A, Menaka M, et al. Support vector machine based methodology for classification of thermal images pertaining to breast cancer[J]. J Therm Biol, 2022, 110: 103337. doi: 10.1016/j.jtherbio.2022.103337.
[35] Kuramoto Y, Wada N, Uchiyama Y. Prediction of pathological complete response using radiomics on MRI in patients with breast cancer undergoing neoadjuvant pharmacotherapy[J]. Int J Comput Assist Radiol Surg, 2022, 17(4): 619-625.

多维度评价

Viewed

Full text

Abstract

Cited

Shared

Discussed