MRI及临床病理特征对乳腺癌人表皮生长因子受体2表达状态的鉴别诊断价值

分享到微信朋友圈

• 特别关注 •

沈怡媛尤超蔺璐奕周嘉音顾雅佳

Cite this article as: SHEN Y Y, YOU C, LIN L Y, et al. Diagnostic value of MRI features combined with clinicopathologic features in predicting the expression of human epidermal growth factor receptor 2 in breast cancer[J]. Chin J Magn Reson Imaging, 2024, 15(1): 6-13.本文引用格式：沈怡媛, 尤超, 蔺璐奕, 等. MRI及临床病理特征对乳腺癌人表皮生长因子受体2表达状态的鉴别诊断价值[J]. 磁共振成像, 2024, 15(1): 6-13. DOI:10.12015/issn.1674-8034.2024.01.002.

摘要

[摘要] 目的探讨MRI联合临床病理特征在乳腺癌人表皮生长因子受体2（human epidermal growth factor receptor 2, HER-2）表达状态中的鉴别诊断价值，尤其是在HER-2低表达乳腺癌中的鉴别诊断价值。材料与方法 回顾性分析2018年1月至2019年12月在复旦大学附属肿瘤医院经病理证实为乳腺癌的患者治疗前乳腺MRI图像，205例患者均行双侧乳腺平扫及增强MRI检查。根据免疫组织化学和荧光原位杂交结果将HER-2状态分为HER-2阴性（包括零、低表达）和阳性（过表达）。分析各组临床病理特征及MRI特征，临床病理特征包括年龄、月经状态、雌激素受体（estrogen receptor, ER）、孕激素受体（progesterone receptor, PR）、激素受体（hormone receptor, HR）、分子分型和Ki-67水平。MRI特征包括纤维腺体类型、背景实质强化、多灶或多中心、瘤内T2WI高信号、瘤周水肿、病灶类型、病灶大小、肿块形状、边缘、内部强化模式、非肿块强化分布及内部强化模式。单因素分析中，对于HER-2阴、阳性组间比较，年龄采用独立样本t检验，病灶大小采用Mann-Whitney U检验，其余临床病理特征及MRI特征采用χ²检验；对于HER-2零、低和过表达组的比较，年龄采用单因素方差分析，病灶大小采用Kruskal-Wallis H检验；其余临床病理特征及MRI特征采用χ²检验。多因素分析采用二元logistic回归分析，用受试者工作特征曲线下面积（area under the curve, AUC）、敏感度和特异度评价模型的诊断效能。结果 HER-2阴性中零表达59例、低表达79例，HER-2阳性（过表达）67例。HER-2阴性与阳性组临床病理特征中，ER、PR、HR和分子分型差异有统计学意义（P均<0.001），MRI特征中肿块边缘差异有统计学意义（P=0.020）。进一步比较HER-2低表达组与零表达组、HER-2低表达组与过表达组，临床病理特征中，ER、PR、HR、分子分型和Ki-67水平（以中位数40%为截断值）组间差异具有统计学意义（ER、PR、HR、分子分型：P均<0.001；Ki-67：P<0.001，P=0.037）；MRI特征中，瘤内T2WI高信号与肿块形状组间差异具有统计学意义（瘤内T2WI高信号：P=0.031，P=0.011；肿块形状：P=0.012，P=0.025），且肿块边缘在HER-2低表达与零表达组间差异有统计学意义（P=0.036）。联合临床病理和MRI特征的多因素分析提示，PR状态、Ki-67水平及肿块形状是鉴别乳腺癌HER-2低表达与零表达的独立预测因素，AUC、敏感度和特异度分别为0.772、79.7%和70.9%；PR状态及瘤内T2高信号是鉴别HER-2低表达与过表达的独立预测因素，AUC、敏感度、特异度分别为0.793、69.8%和76.1%。结论 MRI影像特征对乳腺癌HER-2表达状态具有鉴别诊断价值，尤其在HER-2低表达与零表达或过表达乳腺癌鉴别诊断中。联合临床病理特征，PR阳性、Ki-67低于40%、肿块形状不规则和瘤内T2WI高信号可提示HER-2低表达乳腺癌。

[Abstract] Objective To explore the value of magnetic resonance imaging (MRI) features combined with clinicopathologic features in distinguishing human epidermal growth factor receptor 2 (HER-2) expression status, especially in HER-2-low breast cancer.Materials and Methods The pre-treatment breast MRI images of 205 patients with pathologically confirmed breast cancer from January 2018 to December 2019 at Fudan University Shanghai Cancer Center were retrospectively analyzed. All patients underwent a bilateral breast scan and a dynamic contrast enhancement MRI. HER-2 status was categorized into HER-2 negative (including HER-2-zero and HER-2-low) and HER-2 positive based on immunohistochemistry and fluorescence in situ hybridization results. Clinicopathologic features and MRI features were analyzed in each group. Clinicopathologic features included age, menstrual status, estrogen receptor (ER), progesterone receptor (PR), hormone receptor (HR), molecular subtypes and Ki-67 level. MRI features included fibroglandular tissue, background parenchymal enhancement, multifocal or multicentric, intratumoral T2WI high signal, peritumoral edema, lesion type, lesion size, shape, margin and internal enhancement pattern of the mass, and distribution and internal enhancement pattern of non-mass enhancement. In the univariate analysis, for the comparison between HER-2 negative and positive groups, an independent sample t-test was used for age, a Mann-Whitney U test was used for lesion size, and a χ² test was used for the remaining clinicopathologic and MRI features. For the comparison of HER-2 zero, low, and overexpression groups, a one-way analysis of variance was used for age, a Kruskal-Wallis H test was used for lesion size, and a χ2 test was used for the remaining clinicopathologic and MRI features. Multifactorial analysis was performed by binary logistic regression analysis, and the diagnostic efficacy of the model was evaluated by area under the curve (AUC), sensitivity and specificity of the receiver operating characteristic.Results There were 67 HER-2-positive (HER-2-overexpression), 59 HER-2-zero and 79 HER-2-low cases. Between the HER-2-negative and positive group, the difference in clinicopathologic features of ER, PR, HR, and molecular typing were statistically significant (all P<0.001), and the differences in the margin of the mass in MRI features were statistically significant (P=0.020). Further comparing the HER-2 low with HER-2-zero group or HER-2-overexpression group, the differences in clinicopathologic features were statistically significant in ER, PR, HR, molecular subtypes, and Ki-67 levels (with a cutoff value of 40% of the median) between HER-2-low and HER-2-zero or HER-2-overexpression group (ER, PR, HR, and molecular subtypes: all P<0.001; Ki-67: P<0.001, P= 0.037); among the MRI features, the differences in the intratumoral T2WI hyperintensity and mass shape were statistically significant between HER-2-low and HER-2-zero or HER-2-overexpression group (intratumoral T2WI hyperintensity: P=0.031, P=0.011; mass shape: P=0.012, P=0.025), and the difference in the mass margin was statistically significant between HER-2-zero and HER-2-low group (P=0.036). In the multifactorial analysis combining clinicopathologic and MRI features, PR status, Ki-67 and mass shape were independent predictors to distinguish HER-2-low and -zero expression, with an AUC, sensitivity, and specificity of 0.772, 79.7%, and 70.9%, respectively; and PR status and intratumoral T2WI hyperintensity were independent predictors to distinguish HER-2-low versus -overexpression, with an AUC, sensitivity, and specificity of 0.793, 69.8% and 76.1%, respectively.Conclusions MRI features have a differential diagnostic value for HER-2 expression status in breast cancer, especially in distinguishing HER-2 low-expression and HER-2-zero or -overexpression status. Combining clinicopathologic features and MRI features, PR positivity, Ki-67 lower than 40%, irregular mass shape, and intratumoral T2WI hyperintensity can indicate HER-2 low-expression breast cancer.

[关键词] 乳腺癌；人表皮生长因子受体2；磁共振成像；鉴别诊断；低表达；靶向治疗

[Keywords] breast cancer；human epidermal growth factor receptor 2；magnetic resonance imaging；differential diagnosis；low expression；targeted therapy

作者信息

沈怡媛 ^{1,
2}   尤超 ^{1,
2}   蔺璐奕 ^{1,
2}   周嘉音 ^{1,
2}   顾雅佳 ^{1,
2*}

¹ 复旦大学附属肿瘤医院放射诊断科，上海　200032

² 复旦大学上海医学院肿瘤学系，上海　200032

通信作者：顾雅佳，E-mail：guyajia@126.com

作者贡献声明：：顾雅佳设计本研究方案，对稿件重要内容进行了修改，并获得了国家自然科学基金项目、吴阶平医学基金会临床科研专项资助基金资助；沈怡媛起草和撰写稿件，获取、分析及解释本研究数据；尤超设计本研究方案，解释数据，对稿件重要内容进行了修改，获得了上海市卫生健康委员会面上项目基金资助；蔺璐奕采集、分析本研究数据，对稿件知识性内容进行批评性审阅并对稿件内容进行修改；周嘉音采集研究数据，对稿件知识性内容作批评性审阅及支持性贡献并对稿件内容进行修改。全体作者都同意发表最后的修改稿，同意对本研究的所有方面负责，确保本研究的准确性和诚信。

出版信息

基金项目： 国家自然科学基金项目 82071878 上海市卫生健康委员会面上项目 202240241 吴阶平医学基金会临床科研专项资助基金 320.6750.2022-11-24

收稿日期：2023-10-07

接受日期：2023-12-29

中图分类号：R445.2 R737.9

文献标识码：A

DOI: 10.12015/issn.1674-8034.2024.01.002

本文引用格式：沈怡媛, 尤超, 蔺璐奕, 等. MRI及临床病理特征对乳腺癌人表皮生长因子受体2表达状态的鉴别诊断价值[J]. 磁共振成像, 2024, 15(1): 6-13. DOI:10.12015/issn.1674-8034.2024.01.002.

正文

0 引言

乳腺癌是全球女性最常见的恶性肿瘤，占全年新增病例数的11.7%^[¹^]。人表皮生长因子受体2（human epidermal growth factor receptor 2, HER-2）是影响乳腺癌患者预后的主要标志物之一，其表达会异常激活酪氨酸激酶下游信号，导致细胞生长不受调节和控制^[²^,³^]。抗HER-2靶向药物已被证明可以显著改善HER-2阳性乳腺癌患者的生存和预后，而HER-2阴性患者却缺乏有效靶向药物^[⁴^,⁵^]。但近期的DESTINY-Breast04 3期试验表明，HER-2阴性患者中的低表达状态人群可获益于新的靶向药物^[⁵^,⁶^]。此外，研究发现HER-2低表达人群的疗效、预后均不同于HER-2零表达人群，具有独特的生物学行为和临床特征^[⁶^,⁷^]，这表明沿用原有的阴、阳二分类来区分HER-2状态不足以满足临床诊治需求，对HER-2零、低及过表达状态进一步细分具有重要意义。

目前临床上HER-2状态检测使用免疫组化（immunohistochemistry, IHC）和荧光原位杂交（fluorescence in situ hybridization, FISH）。HER-2低表达定义为IHC 1+或2+且FISH阴性，高达50%的乳腺癌患者呈HER-2低表达状态^[⁸^]。但组织病理结果需要通过侵入性手术或空芯针活检取材，且对肿瘤整体的评估有限^[⁹^,¹⁰^]。MRI具有良好的软组织分辨率，被认为是评估乳腺癌最准确的手段^[¹¹^,¹²^,¹³^]，在乳腺癌鉴别诊断中起到重要作用。既往有关HER-2表达状态的MRI研究多集中在阴性与阳性的鉴别^[¹⁴^,¹⁵^,¹⁶^,¹⁷^,¹⁸^]，仅有少量研究关注HER-2低表达乳腺癌。有研究使用扩散模型鉴别HER-2状态，但仅能较好区分HER-2低表达和过表达^[¹⁹^]；亦有研究使用影像组学模型鉴别HER-2低表达和零表达，但图像处理和建模过程较为繁琐^[²⁰^,²¹^,²²^]。本研究采用的MRI特征可在常规阅片中直接获得，可直观、简便地提示HER-2低表达乳腺癌的影像特征。此外，研究发现HER-2低表达乳腺癌具有独特的临床病理特征：与HER-2零表达和过表达组相比，HER-2低表达组雌激素受体（estrogen receptor, ER）及孕激素受体（progesterone receptor, PR）阳性比例明显较高，而Ki-67水平较低^[⁶^]。因此，本研究的目的是探讨MRI联合临床病理特征在鉴别乳腺癌HER-2表达状态，尤其在低表达乳腺癌中的价值，以期提供HER-2低表达乳腺癌特征的影像学视角，为临床医师对HER-2低表达患者制定合理的治疗方案提供帮助，从而有助于实现精准诊疗。

1 材料与方法

1.1 研究对象

回顾性分析2018年1月至2019年12月在复旦大学附属肿瘤医院就诊的205例乳腺癌患者的临床、病理及MRI资料。纳入标准：（1）经病理证实为乳腺癌；（2）具有治疗前MRI影像；（3）具有完整HER-2评估结果，IHC为2+者须具备FISH结果。排除标准：图像质量差或具有明显穿刺痕迹。病例纳排流程见图1。本研究遵守《赫尔辛基宣言》，并经复旦大学附属肿瘤医院伦理委员会批准，免除受试者知情同意，批准文号：NCT04461990。

点击查看大图

图1 乳腺癌患者纳入和排除流程图。HER-2：人表皮生长因子受体2。
Fig. 1 A flowchart of patients inclusion and exclusion. HER-2: human epidermal growth factor receptor 2.

1.2 仪器与方法

采用西门子公司MAGNETOM Skyra 3.0 T MR扫描仪（爱尔兰根，德国）。患者取俯卧位，双侧乳腺置于16通道乳腺专用相控阵线圈内。乳腺MRI扫描方案包括：三平面定位（冠状面、矢状面、横断面），横断面梯度回波T1WI序列，横断面快速自旋回波T2WI压脂序列，动态对比增强（dynamic contrast-enhanced, DCE）序列。DCE采用3D T1压脂梯度回波序列，平扫一期后，使用高压注射器将钆喷酸葡胺（商品名称：马根维显，Magnevist，拜耳公司，韦恩，美国）以2 mL/s的速度注入手背静脉，剂量为0.1 mmol/kg，对比剂注射后，立刻团注15 mL的生理盐水，注射完成后开始连续5期的动态增强扫描。各序列详细扫描参数见表1。

点击查看表格

表1 MRI扫描参数
Tab. 1 Technical parameters of MRI scanning

1.3 图像分析

由两名从事乳腺影像诊断的放射科医师（分别为工作年限2年的住院医师和15年的副主任医师）对MRI图像进行独立分析，意见不一致时，由第3名具有30年乳腺影像诊断经验的放射科主任医师进行评定，达成一致意见。根据美国放射学会乳腺影像报告和数据系统（breast imaging reporting and data system, BI-RADS）标准对MRI特征进行评估^[¹²^]。评估内容包括：（1）纤维腺体类型；（2）背景实质强化；（3）多灶或多中心；（4）瘤内T2WI高信号；（5）瘤周水肿；（6）病灶类型；（7）病灶大小；（8）肿块形状；（9）肿块边缘；（10）肿块内部强化模式；（11）非肿块强化分布；（12）非肿块强化内部强化模式。

1.4 组织病理学评估

HER-2表达结果根据2018年美国临床肿瘤学会/美国病理学家学会检测指南在空芯针活检中确定^[¹⁰^]。HER-2阴性定义为IHC评分为0、1+或2+且FISH阴性，阳性定义为IHC评分为3+或2+且FISH阳性。其中，HER-2阴性细分为HER-2零表达（IHC评分为0）和低表达（IHC评分为1+或2+且FISH阴性），HER-2阳性即为HER-2过表达。此外，ER及PR阳性定义为IHC表达水平≥1%。激素受体（hormone receptor, HR）阳性状态为ER或PR阳性。根据IHC确定的ER、PR、HER-2状态和肿瘤增殖指数Ki-67，将乳腺癌分为以下几个分子亚型^13]：（1）腔面A型；（2）腔面B/HER-2阴性型；（3）腔面B/HER-2阳性型；（4）HER-2过表达型；（5）三阴性型。

1.5 统计学分析

所有数据采用SPSS 20.0软件进行统计学分析。对计量资料首先采用Levene方差齐性检验，符合正态分布且方差齐的计量资料以均数±标准差表示，组间比较采用单因素方差分析或独立样本t检验。非正态分布的计量资料以M（Q1, Q3）表示，组间比较采用Kruskal-Wallis H检验或Mann-Whitney U检验。计数资料以例数（百分比）表示，组间比较采用χ²检验。P<0.05为差异有统计学意义。选择单因素分析结果中P<0.1的因素进行多因素分析，建立鉴别HER-2状态的二元logistic回归模型，并将分析所得的独立预测因素用R语言（4.2.3版本，http://cran.r-project.org）软件生成诺模图，利用受试者工作特征（receiver operator characteristic, ROC）曲线下面积（area under the curve, AUC）、敏感度和特异度评估模型的诊断效能。

2 结果

2.1 一般资料

共205例乳腺癌患者纳入本研究，均为女性，年龄21~90岁。HER-2阴性138例，其中零表达者59例、低表达者79例，过表达者即HER-2阳性者，为67例。

2.1.1 HER-2阴性与阳性患者的临床病理特征

HER-2阴、阳性患者的临床病理特征见表2。两组间年龄、月经状态及Ki-67水平差异无统计学意义（P均>0.05）。HER-2阴性患者ER、PR阳性者比例显著高于阳性者（P均<0.05）。

点击查看表格

表2 HER-2阴性和阳性组临床病理特征比较
Tab. 2 Comparison of clinicopathologic features of HER-2-negative and HER-2-positive groups

2.1.2 HER-2零表达、低表达及过表达患者的临床病理特征

HER-2零、低、过表达患者的临床病理特征见表3。HER-2低表达组与零表达组及与过表达组相比，年龄和月经状态差异无统计学意义（P均>0.05）；ER、PR阳性者比例显著高于零表达或低表达者，Ki-67水平则显著低于其他两组（P均<0.05）。此外，在分子分型中，三阴性型者多见于HER-2零表达患者，腔面型则多见于HER-2低表达者，差异具有统计学意义（P均<0.001）。

点击查看表格

表3 三组HER-2表达水平患者的临床病理特征比较
Tab. 3 Comparison of clinicopathologic features of patients with three HER-2 status groups

2.2 MRI特征在鉴别HER-2表达状态中的价值

2.2.1 HER-2阴性与阳性患者MRI特征比较

HER-2阴、阳性组在纤维腺体类型、背景实质强化、是否多灶或多中心、是否存在瘤内T2WI高信号、瘤周水肿、病灶类型、大小、肿块形状和内部强化模式及非肿块强化分布和内部强化模式的差异无统计学意义（P>均0.05），仅肿块边缘在组间具有显著差异（P=0.020）。

2.2.2 HER-2低表达与零表达及低表达与过表达患者MRI特征比较

HER-2低表达与零表达及低表达与过表达患者MRI特征比较见表4。HER-2低表达者在瘤内T2WI高信号及肿块形状与其他两组具有显著差异（P均<0.05），更常表现为存在瘤内T2WI高信号和不规则形肿块。此外，与HER-2零表达者相比，低表达者更常表现出不清晰的肿块边缘，差异具有统计学意义（P=0.036）（图2）。

点击查看大图

图2 不同HER-2表达水平病例。2A~2B：女，41岁，HER-2-零表达病例。左乳外侧圆形病灶，于DCE-MRI呈边缘强化模式（2A），T2WI未见瘤内低信号（2B）。IHC示ER（-），PR（-），Ki-67为80%。2C~2D：女，36岁，HER-2-低表达病例。右乳外侧不规则形病灶，于DCE-MRI呈不均匀强化，边缘不清晰（2C），T2WI见瘤内高信号（2D）。IHC示ER（+），PR（+），HER-2（2+），FISH（-），Ki-67为30%。2E~2F：女，54岁，HER-2-过表达病例。右乳外侧类圆形病灶，于DCE-MRI呈不均匀强化（2E），T2WI未见瘤内高信号（2F）。IHC示ER（-），PR（-），HER-2（3+），Ki-67为60%。HER-2：人表皮生长因子受体2；DCE-MRI：动态对比增强MRI；IHC：免疫组织化学；FISH：荧光原位杂交；ER：雌激素受体；PR：孕激素受体；DCE-1：增强第1期。
Fig. 2 Cases of different HER-2 expression level. 2A-2B: Female, 41 years old, HER-2-zero expression case. The lesion is located in the left breast, with round shape and rim enhancement pattern on DCE-MRI (2A). There is no intratumoral hyperintensity on T2WI (2B). IHC shows ER (-), PR (-), and Ki-67 is 80%. 2C-2D: Female, 36 years old, HER-2-low expression case. The lesion is located in the right breast, with irregular shape, non-circumscribed margin and heterogeneous enhancement pattern on DCE-MRI (2C). Intratumoral hyperintensity can be seen on T2WI (2D). IHC shows ER (+), PR (+), HER-2 (2+), FISH (-), and Ki-67 is 30%. 2E-2F: Female, 54 years old, HER-2-overexpression case. The lesion is located in the right breast, with oval shape and heterogeneous enhancement on DCE-MRI (2E). There is no intratumoral hyperintensity on T2WI (2F). IHC shows ER (-), PR (-), HER-2 (3+), Ki-67 is 60%. HER-2: human epidermal growth factor receptor 2; DCE-MRI: dynamic contrast-enhancement MRI; IHC: immunohistochemistry; FISH: fluorescence in situ hybridization; ER: estrogen receptor; PR: progesterone receptor; DCE-1: enhancement of first phase.

点击查看表格

表4 三组HER-2表达水平患者的MRI特征比较
Tab. 4 Comparison of MRI features of patients with three HER-2 status groups

2.2.3 临床病理特征及MRI特征联合模型

结合临床病理特征行二元logistic回归分析后，可得到PR状态、Ki-67水平及肿块形状是HER-2低与零表达鉴别中的独立预测因素（P均<0.05）。PR状态及瘤内T2WI高信号则是HER-2低与过表达鉴别中的独立预测因素（P均<0.05）（表5）。联合上述特征分别建立的鉴别HER-2低表达与零或过表达的logistic回归模型，并用诺模图展示（图3）。在诺模图中，每个独立预测因素的值对应一个具体分数，将图中所有变量得分之和投影到总分量表上，以获得每个病灶HER-2低表达的概率（图3）。模型鉴别HER-2低表达和零表达的AUC、敏感度及特异度分别为0.772 [95% 置信区间（confidence interval, CI）：0.686~0.858]、79.7%、70.9%；鉴别HER-2低表达和过表达的AUC、敏感度及特异度分别为0.793（95% CI：0.719~0.866）、69.8%、76.1%（图4）。

点击查看大图

图3 联合临床病理及MRI特征建立预测HER-2低表达诺模图。3A：联合PR状态、Ki-67和肿块形状鉴别HER-2低表达和零表达；3B：联合PR状态和瘤内T2WI高信号鉴别HER-2低表达和过表达。HER-2：人表皮生长因子受体2；PR：孕激素受体。
Fig. 3 Nomograms by combining clinicopathological and MRI features for predicting HER-2-low. 3A: Combining PR status, Ki-67 and mass shape to differentiate HER-2-low and -zero expression groups. 3B: Combining PR status and intratumoral T2WI hyperintensity to differentiate HER-2-low and -over expression groups. HER-2: human epidermal growth factor receptor 2; PR: progesterone receptor.

点击查看大图

图4 联合临床病理及MRI特征建立logistic回归模型。4A：联合PR状态、Ki-67和肿块形状鉴别HER-2低表达和零表达；4B：联合PR状态和瘤内T2WI高信号鉴别HER-2低表达和过表达。AUC：曲线下面积；CI：置信区间；PR：孕激素受体；HER-2：人表皮生长因子受体2。
Fig. 4 Logistic models by combining clinicopathological and MRI features. 4A: Combining PR status, Ki-67 and mass shape to differentiate HER-2-low and -zero expression groups. 4B: Combining PR status and intratumoral T2WI hyperintensity to differentiate HER-2-low and -over expression groups. AUC: area under the curve; CI: confidence interval; PR: progesterone receptor; HER-2: human epidermal growth factor receptor 2.

点击查看表格

表5 鉴别HER-2低表达与零表达及低表达与过表达组logistic回归分析结果
Tab. 5 Results of logistic regression analysis to differentiate HER-2-low versus HER-2-zero and HER-2-low versus HER-2-overexpression groups

3 讨论

本研究探讨了MRI特征在鉴别乳腺癌HER-2状态中的价值，并联合临床病理特征和MRI特征建立了鉴别HER-2低表达乳腺癌的logistic回归模型。研究表明，在HER-2阴性及阳性的二分类鉴别时，MRI特征中仅肿块边缘具有价值；但在HER-2零、低和过表达的三分类鉴别时，MRI特征中的瘤内T2WI高信号、肿块形状及边缘具有较大价值。当联合MRI特征和临床病理特征，构建的logistic回归模型在鉴别乳腺癌HER-2低表达和零或低表达时具有较好的诊断效能。

3.1 MRI特征鉴别HER-2阴、阳性状态的价值

本研究结果表明，在临床病理特征中，HER-2阴性与阳性乳腺癌患者年龄、Ki-67水平差异无统计学意义，与既往研究结果一致^[¹⁵^,²³^]。在MRI特征，ELIAS等^[²⁴^]的系统综述分析了HER-2阳性乳腺癌DCE-MRI特征，发现仅快速初始动力学和洗脱型曲线与HER-2阳性乳腺癌显著相关；王婷婷等^[²⁵^]的研究发现HER-2阳性者较常发生淋巴结转移；李周丽等^[¹⁶^]的研究发现时间信号曲线在阴、阳性组间存在显著差异。本研究角度与既往研究不同，更聚焦于阅片时可直接观察到的MRI影像特征，结果显示仅肿块边缘在HER-2阴性和阳性乳腺癌患者中存在显著差异，而其他特征均未观察到显著差异，这可能是HER-2阴性者中低表达与零表达影像存在异质性所致。因此，本研究进一步探讨了HER-2低表达与其他两组的影像差异。

3.2 MRI特征鉴别HER-2低表达与零表达及与过表达状态的价值

本研究结果显示，尽管MRI特征在鉴别HER-2阴、阳性时仅有少量有意义的特征，但其在HER-2零、低和过表达的三分类鉴别中更具优势。无论是与HER-2零表达还是过表达者相比，低表达者都更常表现为瘤内T2WI高信号和不规则肿块。瘤内T2WI高信号提示丰富的细胞质、基质水肿和坏死^[²⁶^,²⁷^]。坏死通常是由于肿瘤生长迅速，肿瘤内部缺血缺氧所致^[²⁸^]。本研究中，HER-2低表达病灶最大径的中位数高于零表达，与以上分析相符。在肿块形态方面，既往研究发现其与HER-2状态有关。三阴性乳腺癌与圆或椭圆形态具有强关联^[²⁹^,³⁰^]，HER-2过表达乳腺癌则多报道为不规则肿块^[¹⁴^,¹⁷^]，但尚无关注HER-2低表达的研究。本研究中不规则肿块以HER-2低表达最多，达82.8%。对于肿块边缘，CHEN等^[³¹^]认为其与HR状态有关，HR阳性者可表现出不清晰的边缘。本研究中，HER-2低表达组HR阳性比例显著高于其他两组，低表达者亦较零表达更常表现出边缘不清晰，与其结果一致。

此外，病灶类型和肿块强化模式可提示HER-2低和零表达，但未达到显著性水平，这可能与样本量有限有关。关于病灶类型，本研究结果显示，HER-2零表达者有74.6%表现为孤立性肿块；而HER-2低表达者仅55.7%呈孤立性肿块，其他患者则常以非肿块强化或肿块伴非肿块强化的形式出现。既往研究中，三阴性乳腺癌多表现为孤立性肿块^[³²^,³³^,³⁴^]，而HER-2阳性乳腺癌因导管内成分较多^[³⁵^,³⁶^]，出现非肿块强化的比例较其他分子亚型高^[³⁷^,³⁸^]。在本研究中，HER-2零表达组三阴性乳腺癌占比最高，孤立性肿块占比亦最高，而当HER-2表达水平上升为低表达时，肿块性强化比例减低，非肿块强化比例升高，与以上分析一致。既往研究表明关于肿块强化模式，既往研究表明边缘强化通常与血管生成和血管内皮生长因子表达及HR表达缺乏有关^[³⁹^,⁴⁰^]。本研究中边缘强化以零表达最多，低表达次之，与两组HR表达情况相符。

3.3 临床病理及MRI特征联合模型鉴别HER-2低表达水平

此前一些研究基于MRI定量指标或影像组学鉴别了HER-2表达水平。在鉴别HER-2阴、阳性方面，张成孟等^[¹⁸^]构建了基于DCE-MRI瘤内及瘤周影像组学模型，在验证集中AUC可达0.799。在鉴别HER-2低表达方面，MAO等^[¹⁹^]使用4种扩散模型的指标区分HER-2状态，发现αCTRW是最佳鉴别指标，AUC可达0.802，但仅能较好区分HER-2低表达和过表达。亦有一些影像组学模型进一步提升了对HER-2低表达乳腺癌的鉴别能力^[²⁰^,²¹^,²²^]，但图像处理与建模步骤繁琐，限制其在临床广泛应用。本研究中的logistic回归模型显示，联合临床病理特征和影像特征可获得类似效果，但更为直观、便捷，可在常规阅片中起到提示作用。

3.4 优势与局限性

本研究具有以下优势：（1）从乳腺癌HER-2表达状态及低表达的临床热点问题出发，探讨了MRI特征在鉴别HER-2表达状态，尤其是低表达中的应用价值；（2）常规MRI特征便捷、易得、一目了然，可以在临床工作中作为辅助诊断的简易方式。

本研究存在一定局限性：（1）为回顾性单中心研究，且样本量较小，存在一定选择偏倚。我们拟在今后的工作中扩大样本量，并寻求多中心合作以进行外部验证，增强结果的可信度。（2）常规MRI特征易受阅片者主观因素及MRI信噪比对图像的干扰，后续需开展定量研究验证结论和优化上述问题。

4 结论

综上所述，MRI特征对乳腺癌HER-2表达状态具有鉴别诊断价值，尤其在识别HER-2低表达乳腺癌中更具优势。结合MRI和临床病理特征分析发现，PR阳性、Ki-67低于40%、不规则的肿块形状和瘤内T2WI高信号能够提示HER-2低表达乳腺癌，这将有助于临床医师对潜在获益于新型ADC药物的HER-2低表达患者制定合理的治疗方案。

参考文献

[1]

SUNG H, FERLAY J, SIEGEL R L, et al. Global cancer statistics 2020: GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185 countries[J]. CA Cancer J Clin, 2021, 71(3): 209-249. DOI: 10.3322/caac.21660.

[2]

WULFKUHLE J D, BERG D, WOLFF C, et al. Molecular analysis of HER2 signaling in human breast cancer by functional protein pathway activation mapping[J]. Clin Cancer Res, 2012, 18(23): 6426-6435. DOI: 10.1158/1078-0432.CCR-12-0452.

[3]

MARCHIO C, ANNARATONE L, MARQUES A, et al. Evolving concepts in HER2 evaluation in breast cancer: Heterogeneity, HER2-low carcinomas and beyond[J/OL]. Semin Cancer Biol. 2021, 72: 123-135 [2023-02-16]. https://pubmed.ncbi.nlm.nih.gov/32112814/. DOI: 10.1016/j.semcancer.2020.02.016.

[4]

HAMILTON E, SHASTRY M, SHILLER S M, et al. Targeting HER2 heterogeneity in breast cancer[J/OL]. Cancer Treat Rev, 2021, 100: 102286 [2023-09-20]. https://www.sciencedirect.com/science/article/pii/S0305737221001341?via%3Dihub. DOI: 10.1016/j.ctrv.2021.102286.

[5]

DIECI M V, MIGLIETTA F. HER2: a never ending story[J]. Lancet Oncol, 2021, 22(8): 1051-1052. DOI: 10.1016/S1470-2045(21)00349-1.

[6]

ZHANG G, REN C, LI C, et al. Distinct clinical and somatic mutational features of breast tumors with high-, low-, or non-expressing human epidermal growth factor receptor 2 status[J/OL]. BMC Med, 2022, 20 (1): 142 [2023-04-29]. https://www.ncbi.nlm.nih.gov/pubmed/35484593. DOI: 10.1186/s12916-022-02346-9.

[7]

DAI L J, MA D, XU Y Z, et al. Molecular features and clinical implications of the heterogeneity in Chinese patients with HER2-low breast cancer [J/OL]. Nat Commun, 2023, 14 (1): 5112 [2023-08-22]. https://www.nature.com/articles/s41467-023-40715-x. DOI: 10.1038/s41467-023-40715-x.

[8]

ZHU S J, LU Y J, FEI X C, et al. Pathological complete response, category change, and prognostic significance of HER2-low breast cancer receiving neoadjuvant treatment: a multicenter analysis of 2489 cases[J]. Br J Cancer, 2023, 129(8): 1274-1283. DOI: 10.1038/s41416-023-02403-x.

[9]

CHEN R X, QI Y N, HUANG Y, et al. Diagnostic value of core needle biopsy for determining HER2 status in breast cancer, especially in the HER2-low population[J]. Breast Cancer Res Treat, 2023, 197(1): 189-200. DOI: 10.1007/s10549-022-06781-3.

[10]

WOLFF A C, HAMMOND M E H, ALLISON K H, et al. Human epidermal growth factor receptor 2 testing in breast cancer: American society of clinical oncology/college of American pathologists clinical practice guideline focused update[J]. J Clin Oncol, 2018, 36(20): 2105-2122. DOI: 10.1200/JCO.2018.77.8738.

[11]

MANN R M, KUHL C K, KINKEL K, et al. Breast MRI: guidelines from the European Society of Breast Imaging[J]. Eur Radiol, 2008, 18(7): 1307-1318. DOI: 10.1007/s00330-008-0863-7.

[12]

MALHAIRE C, SELHANE F, SAINT-MARTIN M J, et al. Exploring the added value of pretherapeutic MR descriptors in predicting breast cancer pathologic complete response to neoadjuvant chemotherapy[J/OL]. Eur Radiol, 2023 [2023-09-20]. https://link.springer.com/article/10.1007/s00330-023-09797-5. DOI: 10.1007/s00330-023-09797-5.

[13]

MANN R M, CHO N, MOY L. Breast MRI: state of the art[J]. Radiology, 2019, 292(3): 520-536. DOI: 10.1148/radiol.2019182947.

[14]

代婷, 苏桐, 王锐, 等. DCE-MRI影像特征对乳腺癌激素受体、HER-2及三阴性乳腺癌的预测价值[J]. 磁共振成像, 2023, 14(4): 57-67. DOI: 10.12015/issn.1674-8034.2023.04.011.

DAI T, SU T, WANG R, et al. Predictive value of DCE-MRI features of breast cancer on hormone receptor, HER-2 and triple negative breast cancer[J]. Chin J Magn Reson Imag, 2023, 14(4): 57-67. DOI: 10.12015/issn.1674-8034.2023.04.011.

[15]

李芹, 黄琰, 杨孟, 等. 合成MRI参数直方图特征预测乳腺浸润性导管癌人表皮生长因子受体2表达状态的研究[J]. 中华放射学杂志, 2021, 55(12): 1294-1300. DOI: 10.3760/cma.j.cn112149-20210620-00584.

LI Q, HUANG Y, YANG M, et al. The histogram features of quantitative parameters from synthetic MRI in predicting the expression of human epithelial growth factor receptor 2 in breast invasive ductual carcinoma[J]. Chin J Radiol, 2021, 55(12): 1294-1300. DOI: 10.3760/cma.j.cn112149-20210620-00584.

[16]

李周丽, 陈基明, 高静, 等. MRI影像组学模型术前预测乳腺癌人表皮生长因子受体2表达状态的价值[J]. 磁共振成像, 2023, 14(4): 82-88. DOI: 10.12015/issn.1674-8034.2023.04.014.

LI Z L, CHEN J M, GAO J, et al. Value of MRI image omics model in preoperative prediction of human epidermal growth factor receptor 2 expression in breast cancer[J]. Chin J Magn Reson Imag, 2023, 14(4): 82-88. DOI: 10.12015/issn.1674-8034.2023.04.014.

[17]

张超鑫, 何建鑫, 于敏, 等. 乳腺浸润性导管癌DCE-MRI特征与HER2表达状态相关性研究[J]. 中国CT和MRI杂志, 2023, 21(1): 92-94, 183. DOI: 10.3969/j.issn.1672-5131.2023.01.031.

ZHANG C X, HE J X, YU M, et al. Correlation study of imaging features of DCE-MRI and HER2 status in patient with breast carcinoma[J]. Chin J CT MRI, 2023, 21(1): 92-94, 183. DOI: 10.3969/j.issn.1672-5131.2023.01.031.

[18]

张成孟, 丁治民, 陈鹏, 等. 基于DCE-MRI瘤内及瘤周影像组学联合TIC分型及Ki-67预测乳腺癌患者HER-2表达[J]. 磁共振成像, 2023, 14(4): 68-75. DOI: 10.12015/issn.1674-8034.2023.04.012.

ZHANG C M, DING Z M, CHEN P, et al. Prediction of HER-2 expression in breast cancer patients based on DCE-MRI intratumor and peritumoral imaging combined with TIC typing and Ki-67[J]. Chin J Magn Reson Imag, 2023, 14(4): 68-75. DOI: 10.12015/issn.1674-8034.2023.04.012.

[19]

MAO C, HU L, JIANG W, et al. Discrimination between human epidermal growth factor receptor 2 (HER2)-low-expressing and HER2-overexpressing breast cancers: a comparative study of four MRI diffusion models[J/OL]. Eur Radiol, 2023 [2023-09-20]. https://link.springer.com/article/10.1007/s00330-023-10198-x. DOI: 10.1007/s00330-023-10198-x.

[20]

BIAN X Q, DU S Y, YUE Z B, et al. Potential antihuman epidermal growth factor receptor 2 target therapy beneficiaries: the role of MRI-based radiomics in distinguishing human epidermal growth factor receptor 2-low status of breast cancer[J]. J Magn Reson Imaging, 2023, 58(5): 1603-1614. DOI: 10.1002/jmri.28628.

[21]

GUO Y, XIE X, TANG W, et al. Noninvasive identification of HER2-low-positive status by MRI-based deep learning radiomics predicts the disease-free survival of patients with breast cancer[J/OL]. Eur Radiol, 2023 [2023-09-20]. https://link.springer.com/article/10.1007/s00330-023-09990-6. DOI: 10.1007/s00330-023-09990-6.

[22]

RAMTOHUL T, DJERROUDI L, LISSAVALID E, et al. Multiparametric MRI and radiomics for the prediction of HER2-zero, -low, and-positive breast cancers[J/OL]. Radiology, 2023, 308(2): e222646 [2023-08-01]. https://pubmed.ncbi.nlm.nih.gov/37526540/. DOI: 10.1148/radiol.222646.

[23]

YOU C, LI J, ZHI W, et al. The volumetric-tumour histogram-based analysis of intravoxel incoherent motion and non-Gaussian diffusion MRI: association with prognostic factors in HER2-positive breast cancer[J/OL]. Journal of translational medicine, 2019, 17 (1): 182 [2023-07-02]. https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6604303/. DOI: 10.1186/s12967-019-1911-6.

[24]

ELIAS S G, ADAMS A, WISNER D J, et al. Imaging features of HER2 overexpression in breast cancer: a systematic review and meta-analysis[J]. Cancer Epidemiol Biomarkers Prev, 2014, 23(8): 1464-1483. DOI: 10.1158/1055-9965.EPI-13-1170.

[25]

王婷婷, 周娟, 田宁, 等. 乳腺癌钼靶、MRI征象与分子分型的相关性分析[J]. 磁共振成像, 2019, 10(6): 420-424. DOI: 10.12015/issn.1674-8034.2019.06.005.

WANG T T, ZHOU J, TIAN N, et al. Correlation of mammography, MRI features and molecular typing of breast cancer[J]. Chin J Magn Reson Imag, 2019, 10(6): 420-424. DOI: 10.12015/issn.1674-8034.2019.06.005.

[26]

YOON G Y, CHAE E Y, CHA J H, et al. Imaging and clinicopathologic features associated with pathologic complete response in HER2-positive breast cancer receiving neoadjuvant chemotherapy with dual HER2 blockade[J]. Clin Breast Cancer, 2020, 20(1): 25-32. DOI: 10.1016/j.clbc.2019.06.015.

[27]

MUMIN N A, RAMLI HAMID M T, WONG J H D, et al. Magnetic resonance imaging phenotypes of breast cancer molecular subtypes: a systematic review[J/OL]. Acad Radiol, 2022, 29(Suppl 1): S89-S106 [2023-09-02]. https://pubmed.ncbi.nlm.nih.gov/34481705/. DOI: 10.1016/j.acra.2021.07.017.

[28]

BAE, SHIN S U, RYU H S, et al. Pretreatment MR imaging features of triple-negative breast cancer: association with response to neoadjuvant chemotherapy and recurrence-free survival[J]. Radiology, 2016, 281(2): 392-400. DOI: 10.1148/radiol.2016152331.

[29]

SZEP M, PINTICAN R, BOCA B, et al. Multiparametric MRI Features of Breast Cancer Molecular Subtypes[J/OL]. Medicina (Kaunas), 2022, 58 (12): 1716 [2022-11-23]. https://www.ncbi.nlm.nih.gov/pmc/articles/PMC9785392/. DOI: 10.3390/medicina58121716.

[30]

MOFFA G, GALATI F, COLLALUNGA E, et al. Can MRI Biomarkers Predict Triple-Negative Breast Cancer?[J/OL]. 2020, 10 (12): 1090 [2022-12-15]. https://www.mdpi.com/2075-4418/10/12/1090 DOI: 10.3390/diagnostics10121090.

[31]

CHEN P P, ZHAO S H, GUO W H, et al. Dynamic contrast-enhanced magnetic resonance imaging features and apparent diffusion coefficient value of HER2-positive/HR-negative breast carcinoma[J]. Quant Imaging Med Surg, 2023, 13(8): 4816-4825. DOI: 10.21037/qims-22-1318.

[32]

UEMATSU T, KASAMI M, YUEN S. Triple-negative breast cancer: correlation between MR imaging and pathologic findings[J]. Radiology, 2009, 250(3): 638-647. DOI: 10.1148/radiol.2503081054.

[33]

HUANG J L, LIN Q, CUI C X, et al. Correlation between imaging features and molecular subtypes of breast cancer in young women (≤30 years old)[J]. Jpn J Radiol, 2020, 38(11): 1062-1074. DOI: 10.1007/s11604-020-01001-8.

[34]

ANGELINI G, MARINI C, IACCONI C, et al. Magnetic resonance (MR) features in triple negative breast cancer (TNBC) vs receptor positive cancer (nTNBC)[J/OL]. Clin Imaging, 2018, 49: 12-16 [2023-10-06]. https://pubmed.ncbi.nlm.nih.gov/29120811/. DOI: 10.1016/j.clinimag.2017.10.016.

[35]

SEYFETTIN A, DEDE I, HAKVERDI S, et al. MR imaging properties of breast cancer molecular subtypes[J]. Eur Rev Med Pharmacol Sci, 2022, 26(11): 3840-3848. DOI: 10.26355/eurrev_202206_28951.

[36]

BOISSERIE-LACROIX M, HURTEVENT-LABROT G, FERRON S, et al. Correlation between imaging and molecular classification of breast cancers[J]. Diagn Interv Imaging, 2013, 94(11): 1069-1080. DOI: 10.1016/j.diii.2013.04.010.

[37]

GWEON H M, JEONG J, SON E J, et al. The clinical significance of accompanying NME on preoperative MR imaging in breast cancer patients[J/OL]. PLoS One, 2017, 12(5): e0178445 [2023-05-30]. https://journals.plos.org/plosone/article?id=10.1371/journal.pone.0178445. DOI: 10.1371/journal.pone.0178445.

[38]

KOH J, PARK A Y, KO K H, et al. Can enhancement types on preoperative MRI reflect prognostic factors and surgical outcomes in invasive breast cancer?[J]. Eur Radiol, 2019, 29(12): 7000-7008. DOI: 10.1007/s00330-019-06236-2.

[39]

GIGLI S, AMABILE M I, DAVID E, et al. Morphological and semiquantitative kinetic analysis on dynamic contrast enhanced MRI in triple negative breast cancer patients[J]. Acad Radiol, 2019, 26(5): 620-625. DOI: 10.1016/j.acra.2018.06.014.

[40]

GALATI F, RIZZO V, MOFFA G, et al. Radiologic-pathologic correlation in breast cancer: do MRI biomarkers correlate with pathologic features and molecular subtypes?[J/OL]. Eur Radiol Exp, 2022, 6 (1): 39 [2023-08-08]. https://www.ncbi.nlm.nih.gov/pmc/articles/PMC9357588/. DOI: 10.1186/s41747-022-00289-7.

上一篇 乳腺癌精准诊疗的重要工具——MRI定量分析技术

下一篇 IDEAL-IQ序列在乳腺肿块良恶性鉴别诊断中的应用价值探究