Corines Marina J, Coffey Kristen, Dou Eda, Lobaugh Stephanie, Zheng Junting, Hwang Sinchun, Feigin Kimberly
Memorial Sloan Kettering Cancer Center, Department of Radiology, New York, NY, USA.
University of California San Francisco, Department of Radiology and Biomedical Imagery, San Francisco, CA, USA.
J Breast Imaging. 2022 Nov-Dec;4(6):600-611. doi: 10.1093/jbi/wbac053. Epub 2022 Aug 29.
To determine prevalence and frequency of malignancy among bone lesions detected on breast MRI and to identify clinical and imaging features associated with bone metastases from breast cancer (BC), as bone lesions are suboptimally evaluated on breast imaging protocols and can present a diagnostic challenge.
This IRB-approved retrospective review of breast MRIs performed from June 2009 to June 2018 identified patients with bone lesions. Demographic, clinical, and MRI features were reviewed. Clinical outcome of bone lesions was determined based on pathology and/or additional diagnostic imaging. All benign lesions had ≥ 2 years of imaging follow-up. Statistics were computed with Fisher's exact and Wilcoxon rank sum tests.
Among all patients with breast MRI, 1.2% (340/29 461) had bone lesions. Of these, 224 were confirmed benign or metastatic BC by pathology or imaging follow-up, with 70.1% (157/224) be- nign and 29.9% (67/224) metastatic. Bone metastases were associated with BC history (P < 0.001), with metastases occurring in 58.2% (53/91) of patients with current BC, 17.9% (14/78) patients with prior BC, and 0.0% (0/55) without BC. Bone metastases were associated with invasive and ad- vanced stage BC and, on MRI, with location in sternum, ribs, or clavicles, larger size, multiplicity, andT1 hypointensity (all P < 0.01 in tests of overall association).
Of clinically confirmed breast MRI-detected bone lesions, 30% were bone metastases; all were detected in patients with current or prior BC. Metastases were associated with advanced stage, invasive carcinoma, larger lesion size, multiplicity, low T1 signal, and non-spine location.
确定乳腺MRI检测到的骨病变中恶性肿瘤的患病率和频率，并识别与乳腺癌（BC）骨转移相关的临床和影像学特征，因为在乳腺成像方案中对骨病变的评估欠佳，且可能带来诊断挑战。
这项经机构审查委员会（IRB）批准的对2009年6月至2018年6月期间进行的乳腺MRI的回顾性研究，确定了有骨病变的患者。回顾了人口统计学、临床和MRI特征。根据病理和/或其他诊断成像确定骨病变的临床结果。所有良性病变均有≥2年的影像学随访。采用Fisher精确检验和Wilcoxon秩和检验进行统计计算。
在所有接受乳腺MRI检查的患者中，1.2%（340/29461）有骨病变。其中，224例经病理或影像学随访确诊为良性或转移性BC，70.1%（157/224）为良性，29.9%（67/224）为转移性。骨转移与BC病史相关（P<0.001），58.2%（53/91）的现患BC患者、17.9%（14/78）的既往BC患者和0.0%（0/55）的无BC患者发生转移。骨转移与浸润性和晚期BC相关，在MRI上，与胸骨、肋骨或锁骨部位、更大尺寸、多发性和T1低信号相关（在总体关联测试中，所有P<0.01）。
在临床确诊的乳腺MRI检测到骨病变中，30%为骨转移；所有骨转移均在现患或既往有BC的患者中检测到。转移与晚期、浸润性癌、更大的病变尺寸、多发性、低T1信号和非脊柱部位相关。
