多参数超分辨率超声评估乳腺癌特征:诊断性能优于对比增强超声
前瞻性队列纳入84例患者(良性25例、恶性59例),于活检前行对比增强超声(CEUS)与超分辨率超声(SRUS)参数评估。SRUS单独鉴别恶性的AUC为0.902,优于CEUS的0.866,二者联合AUC提升至0.939。SRUS参数与组织学分级、分子亚型及CD3+ T细胞浸润相关。
要点
- SRUS单独鉴别乳腺恶性病变AUC 0.902,优于CEUS的0.866;联合CEUS+SRUS的AUC为0.939。
- 恶性组血管比率29.8%、MeanDen 12.46±6.11、MeanVel 14.19±4.62,均高于良性组。
- MeanDen与组织学分级呈负相关(III级低于I/II级),Complexity Level与MeanDen可区分分子亚型。
核心结论
- SRUS对乳腺肿块进行多维特征描述,性能优于CEUS;其参数可无创预测肿瘤恶性、分子表型及免疫浸润。
研究设计与样本
- 前瞻性队列研究,纳入84例患者(良性25例、恶性59例),于活检前评估CEUS与SRUS参数。
诊断标准与关键数值
- SRUS鉴别恶性的AUC为0.902,CEUS的AUC为0.866,CEUS+SRUS联合的AUC为0.939。
- 恶性组血管比率29.8% vs 良性组13.6%(p<0.001)。
- 恶性组MeanDen 12.46±6.11 vs 良性组7.47±2.60(p=0.011)。
- 恶性组MeanVel 14.19±4.62 vs 良性组9.71±2.96(p=0.038)。
- MeanDen与组织学分级呈负相关(III级低于I/II级,p<0.05);Complexity Level与MeanDen可区分分子亚型;CD3+浸润与较高Complexity Level正相关(p<0.05)。
临床应用与阅片要点
- SRUS可定量微血管架构(血管比率、血管分支复杂度、平均微泡信号密度、平均微泡流速),辅助乳腺肿块良恶性鉴别。
- SRUS参数有助于无创预测组织学分级、分子亚型及CD3+ T细胞浸润,作者认为可用于术前风险分层、个体化治疗指导与治疗监测。
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
OBJECTIVE: To investigate the diagnostic superiority of super-resolution ultrasound (SRUS) imaging over conventional contrast-enhanced ultrasound (CEUS) in differentiating breast masses, and to evaluate its clinical potential as a non-invasive tool for predicting histological grade, molecular subtypes and immune microenvironment characteristics.
METHODS: A prospective cohort of 84 patients (25 benign, 59 malignant) underwent pre-biopsy assessments of CEUS and SRUS parameters. CEUS evaluated qualitative features (shape after enhancement, feeding vessel) and quantitative indices (area under the curve [AUC], peak intensity). SRUS quantified microvascular architecture via Vessel Ratio (area of vascular lumen), Complexity Level (vascular branching complexity), MeanDen (Mean Density, average microbubble signal density) and MeanVel (Mean Velocity, microbubble flow speed). Receiver operating characteristic curve analysis compared diagnostic accuracies, while correlations with histological grade (I-III), molecular subtypes (Luminal A, Luminal B, human epidermal growth factor receptor 2-enriched and triple-negative breast cancer) and CD3+ T-cell infiltration were evaluated.
RESULTS: Malignant tumors showed distinct SRUS profiles: higher Vessel Ratio (29.8% vs. 13.6%, p < 0.001), MeanDen (12.46 ± 6.11 vs. 7.47 ± 2.60, p = 0.011) and MeanVel (14.19 ± 4.62 vs. 9.71 ± 2.96, p = 0.038) compared to benign lesions. SRUS alone achieved an AUC of 0.902 for malignancy discrimination, outperforming CEUS (AUC = 0.866), with combined CEUS + SRUS improving AUC to 0.939 (p < 0.05 for all comparisons). MeanDen inversely correlated with histological grade (Grade III < I/II, p < 0.05), while Complexity Level and MeanDen distinguished molecular subtypes. CD3+ infiltration was positively associated with higher Complexity Level (p < 0.05).
CONCLUSION: SRUS provides multi-dimensional characterization of breast masses and performs better than CEUS. Its parameters enable non-invasive prediction of tumor malignancy, molecular phenotypes and immune infiltration, positioning SRUS as a promising tool for pre-operative risk stratification, personalized treatment guidance and therapeutic monitoring in clinical oncology.
引用
- Liu Z, Wang L, Zhou Y, et al. Multi-Parametric Super-Resolution Ultrasound for Evaluating Breast Cancer Characteristics: Improved Diagnostic Performance Over Contrast-Enhanced Ultrasound. Ultrasound Med Biol. 2026. PMID: 41826088. doi:10.1016/j.ultrasmedbio.2026.02.006
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。