切换至 "中华医学电子期刊资源库"

中华医学超声杂志(电子版) ›› 2026, Vol. 23 ›› Issue (04) : 292 -299. doi: 10.3877/cma.j.issn.1672-6448.2026.04.005

浅表器官超声影像学

超声特征联合Ki-67预测乳腺癌腋窝淋巴结转移的价值
张玲玲1, 肖沁妤2, 方晓政1, 周利杰2,()   
  1. 1 310053 杭州,浙江中医药大学
    2 314000 浙江嘉兴,嘉兴大学附属第一医院超声医学科
  • 收稿日期:2025-08-26 出版日期:2026-04-01
  • 通信作者: 周利杰

Ultrasound features combined with Ki-67 for predicting axillary lymph node metastasis in breast cancer

Lingling Zhang1, Qinyu Xiao2, Xiaozheng Fang1, Lijie Zhou2,()   

  1. 1 Zhejiang Chinese Medical University, Hangzhou 310053, China
    2 Department of Ultrasound, Affiliated Hospital of Jiaxing University, the First Hospital of Jiaxing, Jiaxing 314000, China
  • Received:2025-08-26 Published:2026-04-01
  • Corresponding author: Lijie Zhou
引用本文:

张玲玲, 肖沁妤, 方晓政, 周利杰. 超声特征联合Ki-67预测乳腺癌腋窝淋巴结转移的价值[J/OL]. 中华医学超声杂志(电子版), 2026, 23(04): 292-299.

Lingling Zhang, Qinyu Xiao, Xiaozheng Fang, Lijie Zhou. Ultrasound features combined with Ki-67 for predicting axillary lymph node metastasis in breast cancer[J/OL]. Chinese Journal of Medical Ultrasound (Electronic Edition), 2026, 23(04): 292-299.

目的

探讨基于超声特征及临床病理参数Ki-67构建的列线图模型对乳腺癌患者腋窝淋巴结转移(ALNM)的预测价值。

方法

选取2020年10月至2023年12月230例来自嘉兴大学附属第一医院经手术确诊的乳腺癌患者作为研究对象,根据是否发生ALNM将患者分为转移组(149例)和无转移组(81例)。按7∶3比例随机将患者分为训练集(161例)和验证集(69例)。比较训练集中ALNM者与无ALNM者超声特征及临床病理参数的差异,包括原发病灶最大直径、位置、边界、形态、边缘、纵横比、血流信息、微钙化、年龄、体质量指数、肿瘤指标(糖类抗原125、糖类抗原153)、癌胚抗原、病理分型、组织学分级、雌激素受体、孕激素受体及Ki-67增殖指数。使用Logistic回归分析探究ALNM的独立危险因素,基于上述因素构建列线图预测模型,绘制受试者操作特征曲线、校准曲线、决策曲线分析评估该模型的诊断效能及临床适用性。

结果

多因素Logistic回归分析显示:Ki-67(OR=2.27,95%CI:1.07~4.81)、肿瘤最大直径(OR=2.34,95%CI:1.12~4.89)、肿瘤边缘(OR=3.26,95%CI:1.57~6.79)及微钙化(OR=2.31,95%CI:1.11~4.81)均为ALNM的独立预测因素(P<0.05)。基于以上变量构建预测乳腺癌ALNM的列线图模型,该模型训练集曲线下面积(AUC)为0.75(95%CI:0.67~0.83),验证集AUC为0.77(95%CI:0.65~0.89),校准曲线显示模型预测概率与腋窝淋巴结实际概率基本一致。

结论

基于超声特征肿瘤最大直径、肿瘤边缘、微钙化及病理参数Ki-67建立的列线图模型对预测ALNM具有一定价值。

Objective

To explore the utility of a nomogram model incorporating ultrasound features and the clinicopathological parameter Ki-67 in predicting axillary lymph node metastasis (ALNM) in patients with breast cancer.

Methods

A total of 230 patients with surgically confirmed breast cancer from the First Affiliated Hospital of Jiaxing University were included. The patients were categorized based on the presence or presence of ALNM into either a metastasis group (n=149) or a non-metastasis group (n=81). Using a 7:3 random allocation ratio, the cohort was divided into a training set (n=161) and a validation set (n=69). In the training set, ultrasound characteristics and clinicopathological parameters were compared between patients with and without ALNM. Assessed variables included maximum tumor diameter, lesion location, boundary, morphology, margin, aspect ratio, blood flow, presence of microcalcifications, age, body mass index (BMI), serum tumor markers (CA125, CA153, and carcinoembryonic antigen), pathological type, histological grade, estrogen receptor status, progesterone receptor status, and the Ki-67 proliferation index. Logistic regression analysis was used to identify independent risk factors for ALNM. A nomogram prediction model was constructed based on these factors. The model's performance was evaluated using receiver operating characteristic (ROC) curves (with area under the curve [AUC]), calibration curves, and decision curve analysis (DCA) to assess its diagnostic accuracy and clinical applicability.

Results

Multivariable logistic regression analysis identified four independent predictors of ALNM: Ki-67 expression level (odds ratio [OR]=2.27, 95% confidence interval [CI]: 1.07–4.81), maximum tumor diameter (OR=2.34, 95%CI: 1.12–4.89), tumor margin status (OR=3.26, 95%CI: 1.57–6.79), and presence of microcalcifications (OR=2.31, 95%CI: 1.11–4.81) (all P<0.05). A nomogram was developed using these variables. The model had an AUC of 0.75 (95%CI: 0.67–0.83) in the training set and 0.77 (95%CI: 0.65–0.89) in the validation set. The calibration curve indicated good agreement between predicted probabilities and actual outcomes. Decision curve analysis supported the model's clinical utility across a range of threshold probabilities.

Conclusion

The nomogram model constructed based on ultrasound features (maximum tumor diameter, tumor margin, and microcalcifications) and the pathological parameter Ki-67 contributes valuable predictive information for ALNM in breast cancer.

表1 训练集与验证集乳腺癌患者临床病理及超声特征差异比较
变量 训练集(161例) 验证集(69例)
未转移组(56例) 转移组(105例) 统计值 P 未转移组(25例) 转移组(44例) 统计值 P
年龄(岁,
±s
57.25±11.07 57.64±11.80 t=-0.20 0.839 53.32±9.76 57.98±10.07 t=-1.87 0.066
BMI(kg/m2
±s
24.06±4.23 23.57±2.19 t=0.81 0.421
23.52±2.30 24.17±3.03 t=-0.94 0.351
CA125[例(%)] χ2=0.00 0.959 χ2=0.00 1.000
≤22 U/ml 43(76.79) 81(77.14) 21(84.00) 36(81.82)
>22 U/ml 13(23.21) 24(22.86) 4(16.00) 8(18.18)
CA153[例(%)] χ2=0.58 0.448 χ2=3.52 0.061
≤20 U/ml 48(85.71) 85(80.95) 25(100) 36(81.82)
>20 U/ml 8(14.29) 20(19.05) 0(0) 8(18.18)
CEA[例(%)] χ2=1.28 0.258 χ2=1.61 0.205
≤5 ng/ml 51(91.07) 89(84.76) 25(100) 39(88.64)
>5 ng/ml 5(8.93) 16(15.24) 0(0) 5(11.36)
病理分型[例(%)] χ2=0.35 0.552 χ2=0.85 0.357
其他类型 6(10.71) 7(6.67) 5(20.00) 4(9.09)
浸润性导管癌 50(89.29) 98(93.33) 20(80.00) 40(90.91)
组织学分级[例(%)] χ2=1.26 0.262 χ2=2.35 0.125
1~2级 27(48.21) 41(39.05) 12(48.00) 13(29.55)
3级 29(51.79) 64(60.95) 13(52.00) 31(70.45)
雌激素受体[例(%)] χ2=0.02 0.889 χ2=0.50 0.479
阴性 17(30.36) 33(31.43) 7(28.00) 16(36.36)
阳性 39(69.64) 72(68.57) 18(72.00) 28(63.64)
孕激素受体[例(%)] χ2=0.01 0.941 χ2=1.56 0.211
阴性 21(37.50) 40(38.10) 7(28.00) 19(43.18)
阳性 35(62.50) 65(61.90) 18(72.00) 25(56.82)
Ki-67[例(%)] χ2=10.03 0.002 χ2=4.68 0.030
≤14% 32(57.14) 33(31.43) 14(56.00) 13(29.55)
>14% 24(42.86) 72(68.57) 11(44.00) 31(70.45)
象限[例(%)] - 0.841 - 0.567
内上 12(21.43) 19(18.10) 6(24.00) 12(27.27)
外上 32(57.14) 62(59.05) 11(44.00) 17(38.64)
中央区 1(1.79) 5(4.76) 0(0) 4(9.09)
外下 9(16.07) 17(16.19) 8(32.00) 10(22.73)
内下 2(3.57) 2(1.90) 0(0) 1(2.27)
最大直径[例(%)] χ2=9.45 0.002 χ2=4.68 0.030
≤2 cm 35(62.50) 39(37.14) 14(56.00) 13(29.55)
>2 cm 21(37.50) 66(62.86) 11(44.00) 31(70.45)
边界[例(%)] χ2=2.18 0.139 χ2=1.60 0.206
不清晰 33(58.93) 74(70.48) 12(48.00) 28(63.64)
清晰 23(41.07) 31(29.52) 13(52.00) 16(36.36)
形态[例(%)] χ2=0.14 0.711 χ2=0.71 0.399
不规则 50(89.29) 97(92.38) 17(68.00) 34(77.27)
规则 6(10.71) 8(7.62) 8(32.00) 10(22.73)
边缘毛刺征[例(%)] χ2=10.38 0.001 χ2=7.10 0.008
34(60.71) 36(34.29) 18(72.00) 17(38.64)
22(39.29) 69(65.71) 7(28.00) 27(61.36)
纵横比[例(%)] χ2=2.91 0.088 χ2=0.64 0.424
>1 6(10.71) 3(2.86) 4(16.00) 3(6.82)
≤1 50(89.29) 102(97.14) 21(84.00) 41(93.18)
血流信息[例(%)] χ2=0.13 0.718 χ2=1.12 0.291
12(21.43) 20(19.05) 5(20.00) 14(31.82)
44(78.57) 85(80.95) 20(80.00) 30(68.18)
微钙化[例(%)] χ2=10.20 0.001 χ2=6.39 0.011
35(62.50) 38(36.19) 17(68.00) 16(36.36)
21(37.50) 67(63.81) 8(32.00) 28(63.64)
表2 乳腺癌腋窝淋巴结转移的多因素Logistic回归分析
图1 预测乳腺癌腋窝淋巴结转移的列线图模型
图2 预测乳腺癌腋窝淋巴结转移(ALNM)列线图模型超声应用案例。二维超声图像提示右乳外上象限探及大小约1.8 cm×1.1 cm低回声区,边界欠清晰,形态不规则,边缘呈毛刺征(图a箭头所示),其内见成簇点状强回声(图b箭头所示);病理提示ALNM(图c所示),免疫组化提示Ki-67增殖指数约为20%。经列线图模型计算得分为0+100+70+70=240分,对应的ALNM概率为80%
图3 预测乳腺癌腋窝淋巴结转移列线图模型的受试者操作特征曲线。图a为训练集预测效果,图b为验证集预测效果 注:AUC为曲线下面积,CI为可信区间
图4 预测乳腺癌腋窝淋巴结转移列线图模型的校准曲线。图a为训练集校准曲线,图b为验证集校准曲线
图5 预测乳腺癌腋窝淋巴结转移列线图模型临床决策分析曲线。图a为训练集临床决策分析曲线,图b为验证集临床决策分析曲线
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
Chang JM, Leung JWT, Moy L, et al. Axillary nodal evaluation in breast cancer: state of the art [J]. Radiology, 2020, 295(3): 500-515.
3
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.
4
Thalji SZ, Cortina CS, Guo MS, et al. Postoperative complications from breast and axillary surgery [J]. Surg Clin North Am, 2023, 103(1): 121-139.
5
Gradishar WJ, Moran MS, Abraham J, et al. Breast Cancer, Version 3.2024, NCCN Clinical Practice Guidelines in Oncology [J]. J Natl Compr Canc Netw, 2024, 22(5): 331-357.
6
徐国栋, 李冰, 黄绮国. 早期乳腺癌非前哨淋巴结转移的相关因素分析及转移风险评估 [J]. 中国普通外科杂志, 2024, 33(7): 1192-1198.
7
Laohawiriyakamol S, Mahattanobon S, Puttawibul P. Intraoperative molecular analysis of total tumor load in sentinel lymph node: a predictor of axillary status in early breast cancer [J]. Asian Pac J Cancer Prev, 2022, 23(1): 349-354.
8
Qiu C, Wei Y, Li J. Can negative axillary ultrasound reliably predict pathologically negative axillary lymph node status in breast cancer patients with cT ≤3 cm, cN0, and HER2-positive?-a retrospective, single-institution study [J]. Gland Surgery, 2024, 13(8): 1511-1521.
9
Jung JG, Ahn SH, Lee S, et al. No axillary surgical treatment for lymph node-negative patients after ultra-sonography [NAUTILUS]: protocol of a prospective randomized clinical trial [J]. BMC Cancer, 2022, 22(1): 189.
10
党燕, 韦艳艳, 朱来敏, 等. 多模态磁共振技术对乳腺癌腋窝淋巴结转移的预测价值 [J]. 医学影像学杂志, 2024, 34(9): 49-54.
11
刘琪, 常才, 李佳伟. 乳腺导管原位癌的影像学表现与分子分型、组织病理学分级及预后相关性的研究进展 [J]. 中国癌症杂志, 2024, 34(2): 201-209.
12
蔡林利, 宋宏萍, 巨艳, 等. 超声结合临床病理指标模型对T1-2期乳腺癌腋窝淋巴结转移的预测价值 [J/OL]. 中华医学超声杂志(电子版), 2024, 21(2): 143-150.
13
孙帼, 谢迎东, 徐超丽, 等. 超声联合临床特征的列线图模型预测甲状腺乳头状癌淋巴结转移的价值 [J/OL]. 中华医学超声杂志(电子版), 2023, 20(7): 734-742.
14
孙舒涵, 陈雅静, 宗晴晴, 等. 基于超声的深度学习列线图预测乳腺癌新辅助化疗后腋窝淋巴结状态的研究 [J/OL]. 中华医学超声杂志(电子版), 2025, 22(2): 97-105.
15
石丽楠, 曹春莉, 桑田, 等. 基于乳腺癌超声特征及临床病理指标的列线图预测腋窝淋巴结转移风险 [J]. 中国医学影像学杂志, 2024, 32(4): 332-338.
16
周婉, 施静, 周锋盛, 等. 超声造影联合临床指标预测乳腺癌患者腋窝淋巴结转移的效能 [J]. 中国超声医学杂志, 2024, 40(12): 1349-1353.
17
Zheng B, Chen Q. Novel model based on ultrasound predict axillary lymph node metastasis in breast cancer [J]. BMC Med Imaging, 2023, 23(1): 135.
18
Chen Y, Xie Y, Li B, et al. Automated breast ultrasound (ABUS)-based radiomics nomogram: an individualized tool for predicting axillary lymph node tumor burden in patients with early breast cancer [J]. BMC Cancer, 2023, 23(1): 340.
19
Chen Y, Wang L, Dong X, et al. Deep learning radiomics of preoperative breast MRI for prediction of axillary lymph node metastasis in breast cancer [J]. J Digit Imaging, 2023, 36(4): 1323-1331.
20
Chintapally N, Englander K, Gallagher J, et al. Tumor characteristics associated with axillary nodal positivity in triple negative breast cancer [J]. Diseases, 2023, 11(3): 118.
21
朱阳阳, 聂芳, 吕文豪, 等. 乳腺癌常规超声及超声造影特征与HER-2表达的相关性研究 [J]. 中国超声医学杂志, 2020, 36(7): 606-609.
22
Xu S, Wang Q, Hong Z. The correlation between multi-mode ultrasonographic features of breast cancer and axillary lymph node metastasis [J]. Front Oncol, 2024, 14: 1433872.
23
石红, 杨彦辉, 张亚萍, 等. Nomogram预测乳腺浸润性导管癌Ki-67表达水平 [J]. 中国超声医学杂志, 2023, 39(1): 17-21.
24
Bai X, Wang Y, Song R, et al. Ultrasound and clinicopathological characteristics of breast cancer for predicting axillary lymph node metastasis [J]. Clin Hemorheol Microcirc, 2023, 85(2): 147-162.
25
Huang ZB, Mo SJ, Li GT, et al. Prognosticating axillary lymph node metastasis in breast cancer through integrated photoacoustic imaging, ultrasound, and clinical parameters [J]. Breast Cancer Res, 2025, 27(1): 123.
26
Sun P, Han J, Li M, et al. Spectral ultrasound combined with clinical pathological parameters in prediction of axillary lymph node metastasis in breast cancer [J]. J Ultrasound Med, 2024, 43(12): 2311-2324.
27
Xiong J, Zuo W, Wu Y, et al. Ultrasonography and clinicopathological features of breast cancer in predicting axillary lymph node metastases [J]. BMC Cancer, 2022, 22(1): 1155.
28
高卓然, 赵诚, 李晶晶, 等. 基于多模态超声成像特征构建的乳腺癌雄激素受体表达水平预测模型的临床价值 [J]. 中国超声医学杂志, 2024, 40(6): 649-653.
29
杨兴益, 董瑞生, 郭浩. 多模态超声特征与乳腺癌YEATS4、SOX12表达状态的相关性研究 [J]. 放射学实践, 2021, 36(5): 663-668.
30
Jiang M, Li CL, Luo XM, et al. Radiomics model based on shear-wave elastography in the assessment of axillary lymph node status in early-stage breast cancer [J]. Eur Radiol, 2022, 32(4): 2313-2325.
31
郭旭, 余卫峰, 王丽玲, 等. 剪切波弹性成像联合常规超声预测乳腺癌腋窝淋巴结转移的临床价值 [J]. 临床超声医学杂志, 2024, 26(12): 1009-1015.
32
Ferroni G, Sabeti S, Abdus-Shakur T, et al. Noninvasive prediction of axillary lymph node breast cancer metastasis using morphometric analysis of nodal tumor microvessels in a contrast-free ultrasound approach [J]. Breast Cancer Res, 2023, 25(1): 65.
[1] 崔蕊, 杨文利, 李栋军, 王子杨, 陈伟, 李逸丰, 沈琳, 刘倩, 李逸凡, 刘宸琛. L33-8s高频线阵超声探头在眼部睫状体离断诊断中的应用价值[J/OL]. 中华医学超声杂志(电子版), 2026, 23(04): 271-275.
[2] 张崇霄, 胡紫玥, 崔立坤, 王璐, 卢漫. 甲状腺乳头状癌术后淋巴结转移的超声特征[J/OL]. 中华医学超声杂志(电子版), 2026, 23(04): 283-291.
[3] 莫遵玉, 赖莉萍, 李水平, 葛肖艳. 基于多种超声参数构建乳腺癌高负荷腋窝淋巴结转移的预测模型[J/OL]. 中华乳腺病杂志(电子版), 2026, 20(02): 82-90.
[4] 赵珂欣, 王蓉, 张钧, 杨哲, 石钰环, 姬雅楠, 刘敏丽, 张生军. 雷公藤提取物在三阴性乳腺癌中的作用机制[J/OL]. 中华乳腺病杂志(电子版), 2026, 20(01): 55-59.
[5] 仲洋杨, 邓舒瑶, 李永杰, 李庄. 男性隐匿性乳腺癌一例[J/OL]. 中华乳腺病杂志(电子版), 2026, 20(01): 60-63.
[6] 黎钰敏, 朱晓旭, 陈子昂, 谢晓娜, 张金丽, 张志. 高频超声联合剪切波弹性成像在增生性瘢痕厚度与硬度评估中的应用价值[J/OL]. 中华损伤与修复杂志(电子版), 2026, 21(02): 90-96.
[7] 杨国涛, 毛华杰, 陈志军, 陈金亮, 车军, 江起庭. 超声辅助定位经皮通道松解对慢性腕管综合征的临床疗效分析[J/OL]. 中华损伤与修复杂志(电子版), 2026, 21(02): 97-101.
[8] 常人元, 雷涛, 王彤, 陈宇, 李志梅, 高欣, 周楠, 曹磊, 董艳芳, 刘静静. rt-PA剂量对中高危急性肺栓塞患者超声辅助置管溶栓期间肺动脉压和出血风险的影响[J/OL]. 中华肺部疾病杂志(电子版), 2026, 19(01): 42-48.
[9] 东小鸽, 樊海宁, 侯立朝, 杜凯豪, 刘海刚, 汪占金, 薛伟伟, 石亚超, 魏六木, 王展. 三维可视化技术、ICG荧光示踪和术中实时超声在肝癌个性化精准诊疗中应用[J/OL]. 中华肝脏外科手术学电子杂志, 2026, 15(02): 181-189.
[10] 李鹏, 张维桢, 武国帅, 马伊凡, 张灵强. 超声造影在胰腺疾病中的应用研究进展[J/OL]. 中华肝脏外科手术学电子杂志, 2026, 15(01): 119-123.
[11] 顾雨佳, 王晓军, 曹海波, 周林强, 石磊, 于涛, 任峰, 蒋才奇, 卞杰勇. 多模态辅助定位在合并颅骨缺损脑积水脑室-腹腔分流术中的临床应用[J/OL]. 中华神经创伤外科电子杂志, 2026, 12(01): 24-29.
[12] 韩静, 李西羊, 张桂芝. 不同浓度罗哌卡因超声引导下腹横肌平面阻滞在结直肠癌术后镇痛中的效果对比[J/OL]. 中华消化病与影像杂志(电子版), 2026, 16(01): 79-85.
[13] 王悦扬, 蔡雁. 孕前子宫肌瘤患者高强度超声聚焦治疗后的妊娠结局[J/OL]. 中华产科急救电子杂志, 2026, 15(01): 38-42.
[14] 杨斯博, 孙娅慧, 刘淑萍, 张海涛, 孙津津, 李利. 经J-VALVE TF系统治疗重度主动脉瓣反流围手术期超声评估一例[J/OL]. 中华诊断学电子杂志, 2026, 14(01): 44-48.
[15] 周健伟, 顾辰昊, 申雪纯, 杨菲, 何胜虎, 徐冰. 表现为重度二尖瓣反流的低流量低压差重度主动脉瓣狭窄一例[J/OL]. 中华诊断学电子杂志, 2026, 14(01): 49-52.
阅读次数
全文


摘要


AI


AI小编
你好!我是《中华医学电子期刊资源库》AI小编,有什么可以帮您的吗?