文献速递乳腺USMR

多模态超声和MRI预测三阴性和HER2阳性乳腺癌肿瘤浸润淋巴细胞:单中心回顾性研究

Prediction of tumour-infiltrating lymphocytes in triple-negative and HER2-positive breast cancer using multi-modality ultrasound and MRI: a single-institution retrospective study.
PubMed 文献 · Br J Radiol · 2025-01-01

本研究回顾性纳入2018年4月至2023年11月432例接受超声和MRI的三阴性及HER2阳性乳腺癌患者,以20%为阈值区分高/低肿瘤浸润淋巴细胞(TILs),228例(52.7%)为高TILs。基于超声和MRI特征构建的列线图在训练集AUC为0.884(95%CI 0.846-0.921),测试集AUC为0.883(95%CI 0.825-0.940),校准曲线显示良好校准能力。

要点

  • 超声椭圆形/圆形、边界清晰、回声不均匀及MRI肿块样病变、均匀强化与高TILs相关。
  • 基于超声和MRI特征的列线图预测高TILs的AUC在训练集为0.884,测试集为0.883。
  • 治疗前超声和MRI特征可能无创预测三阴性和HER2阳性乳腺癌TILs水平。

核心结论

  • 基于超声和MRI影像特征的列线图对三阴性和HER2阳性乳腺癌高TILs水平具有预测价值,可能有助于治疗决策和预后评估。

研究设计与样本

  • 单中心回顾性研究,纳入2018年4月至2023年11月432例接受超声和MRI的三阴性及HER2阳性乳腺癌患者,以组织学淋巴细胞浸润比例20%为阈值区分高/低TILs。

关键影像表现/征象

  • 超声表现:椭圆形/圆形(OR 0.148,P=0.001)和边界清晰(OR 0.174,P=0.001)与高TILs显著相关。
  • 超声表现:回声不均匀(OR 7.106,P=0.001)与高TILs显著相关。
  • MRI表现:DCE-MRI肿块样病变类型(OR 0.425,P=0.022)和均匀内部强化(OR 0.144,P=0.001)与高TILs显著相关。

诊断标准与关键数值

  • 以组织学淋巴细胞浸润面积比例20%为阈值区分低TILs和高TILs。
  • 列线图预测高TILs的AUC在训练集为0.884(95%CI 0.846-0.921),在测试集为0.883(95%CI 0.825-0.940)。

临床应用与阅片要点

  • 治疗前可综合超声(形态、边界、回声均匀性)和MRI(病变类型、强化均匀性)特征评估高TILs可能性。
  • 列线图预测高TILs的AUC在训练集为0.884、测试集为0.883,校准曲线显示良好校准能力。
  • 作者认为这些发现可能有助于治疗决策和改善预后评估。

局限性

  • 原文未报告局限性。

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

OBJECTIVES: To explore the potential of non-invasive imaging techniques in predicting tumour-infiltrating lymphocytes (TILs) status in triple-negative (TN) and HER-2 positive breast cancers.

METHODS: Between April 2018 and November 2023, a total of 432 eligible TN and HER-2 positive breast cancers who underwent ultrasound (US) and magnetic resonance imaging (MRI) were retrospectively enrolled in this study. US and MRI variables were collected from the workstation. Fraction of TILs was determined histologically based on the proportion of lymphocytes infiltrated areas. A cutoff of 20% was established to distinguish between low and high levels of TILs. Predictive factors associated with high-TILs were identified using logistic regression analysis and a nomogram was developed. The diagnostic performance of the nomogram was evaluated.

RESULTS: A total of 432 patients with TN and HER2-positive breast cancer were evaluated, among which 228 patients (52.7%) exhibited high levels of TILs. Oval/round shape (odds ratio, [OR], 0.148; P = .001), circumscribed margin (OR, 0.174; P = .001), heterogeneous echotexture (OR, 7.106; P = .001), mass-like lesion type on DCE-MRI (OR, 0.425; P = .022), and homogeneous internal enhancement (OR, 0.144; P = .001) were all significantly linked to high-TILs. The nomogram based on these characteristics yielded an optimal area under the curve (AUC) of 0.884 (95% confidence interval [CI], 0.846-0.921) in the training set and 0.883 (95% CI, 0.825-0.940) in the test set. The calibration curve showed favourable calibration ability.

CONCLUSION: The nomogram includes imaging features from both US and MRI, which are crucial for predicting high levels of TILs in TN and HER2-positive breast cancers.

ADVANCES IN KNOWLEDGE: Pretreatment US and MRI features have the potential to non-invasively predict TILs' levels in TN and HER2-positive breast cancer. These findings may aid in treatment decision-making and improve prognostic assessment.

来源:Br J Radiol,查看原文。版权归原作者所有。

引用

  1. Zhang M, Gong L, Cai M, et al. Prediction of tumour-infiltrating lymphocytes in triple-negative and HER2-positive breast cancer using multi-modality ultrasound and MRI: a single-institution retrospective study. Br J Radiol. 2025. PMID: 40591454. doi:10.1093/bjr/tqaf146

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。