术前MR成像中计算机辅助诊断生成的乳腺癌动力学特征与原发性可手术浸润性乳腺癌患者无病生存期的关联
这项回顾性研究纳入329例新诊断浸润性乳腺癌女性患者,利用计算机辅助诊断(CAD)系统分析术前MR成像的动力学参数。中位随访50个月期间36例复发。CAD测量的峰值强化和流出成分较高与无病生存期较差显著相关,淋巴血管侵犯也与较差的无病生存期相关。
要点
- CAD测量的术前MR峰值强化可区分有无复发患者,AUC为0.728。
- 峰值强化越高、流出成分越高,无病生存期越差。
- 病理证实的淋巴血管侵犯与较差无病生存期相关(HR=3.011)。
核心结论
- 术前MR成像中CAD测量的峰值强化和流出成分较高与原发性可手术乳腺癌患者较差的无病生存期显著相关。
研究设计与样本
- 回顾性研究,纳入329例连续女性患者(平均年龄52.9岁,范围32-88岁),均为新诊断浸润性乳腺癌,在2012年1月至2013年2月间接受术前MR成像和手术;使用商用CAD系统回顾性分析所有MR图像;参考标准为无病生存期随访(中位50个月,范围15-55个月)。
关键影像表现/征象
- CAD测量的峰值强化(注射对比剂后第一序列中最高像素信号强度)可区分复发与无复发患者,AUC=0.728(95%CI:0.676,0.775;P<.001)。
- CAD参数包括峰值强化、血管容积(强化病灶总体积)和延迟强化模式(肿瘤内流出、平台和持续强化成分的比例)。
诊断标准与关键数值
- 多因素Cox分析显示,较高的峰值强化与较差无病生存期相关(HR=1.001;95%CI:1.000,1.002;P=.004)。
- 较高的流出成分与较差无病生存期相关(HR=1.029;95%CI:1.005,1.054;P=.017)。
- 组织病理学检查示淋巴血管侵犯与较差无病生存期相关(HR=3.011;95%CI:1.302,6.962;P=.010)。
临床应用与阅片要点
- CAD生成的峰值强化和流出成分可在术前MR阅片时量化评估,并与无病生存期风险关联。
- 淋巴血管侵犯是病理指标,但结合CAD动力学参数可提供额外预后信息。
局限性
- 本文为回顾性研究。
- 需进一步研究以验证CAD动力学特征的预后价值。
以上解读依据论文摘要(全文未开放获取或许可不允许翻译转载)。
英文原文(摘要)· 点击展开
Purpose To retrospectively investigate the relationship between the kinetic features of breast cancer assessed with computer-aided diagnosis (CAD) at preoperative magnetic resonance (MR) imaging and disease-free survival in patients with primary operable invasive breast cancer. Materials and Methods This retrospective study was approved by the institutional review board. The requirement to obtain informed consent was waived. The authors identified 329 consecutive women (mean age, 52.9 years; age range, 32-88 years) with newly diagnosed invasive breast cancer who had undergone preoperative MR imaging and surgery between January 2012 and February 2013. All MR images were retrospectively reviewed by using a commercially available CAD system, and the following kinetic parameters were noted for each lesion: peak enhancement (highest pixel signal intensity in the first series obtained after administration of contrast material), angio-volume (total volume of the enhancing lesion), and delayed enhancement profiles (the proportions of washout, plateau, and persistently enhancing component within a tumor). Cox proportional hazards modeling was used to identify the relationship between CAD-generated kinetics and disease-free survival after adjusting for clinical-pathologic variables. Results A total of 36 recurrences developed at a median follow-up of 50 months (range, 15-55 months). CAD-measured peak enhancement at preoperative MR imaging enabled differentiation between patients with and patients without recurrence (area under the receiver operating characteristic curve = 0.728; 95% confidence interval [CI]: 0.676, 0.775; P < .001). Multivariate Cox analysis showed that a higher peak enhancement (hazard ratio [HR] = 1.001; 95% CI: 1.000, 1.002; P = .004), a higher washout component (HR = 1.029; 95% CI: 1.005, 1.054; P = .017), and lymphovascular invasion at histopathologic examination (HR = 3.011; 95% CI: 1.302, 6.962; P = .010) were associated with poorer disease-free survival. Conclusion Higher values of CAD-measured peak enhancement and washout component at preoperative MR imaging were significantly associated with poorer disease-free survival of patients with primary operable breast cancer. © RSNA, 2017.
引用
- Kim JJ, Kim JY, Kang HJ, et al. Computer-aided Diagnosis-generated Kinetic Features of Breast Cancer at Preoperative MR Imaging: Association with Disease-free Survival of Patients with Primary Operable Invasive Breast Cancer. Radiology. 2017. PMID: 28253106. doi:10.1148/radiol.2017162079
供执业医师工作参考,不替代临床判断;数值与分级以指南原文为准。