文献速递肿瘤 PET/CT

18F-FDG PET/CT预测Luminal A型乳腺癌腋窝淋巴结转移的诊断效能

Diagnostic performance of 18F-FDG PET/CT in the prediction of axillary lymph node metastasis in Luminal A breast cancer.
PubMed 文献 · Eur J Nucl Med Mol Imaging · 2026-06-16

该回顾性研究纳入279例初治Luminal A型乳腺癌女性患者,评估18F-FDG PET/CT定量参数对腋窝淋巴结(ALN)转移的术前预测价值。结果显示腋窝SUVmax(AUC 0.881)显著优于视觉评估(AUC 0.851,p=0.031),联合模型诊断效能最高(AUC 0.889),并能正确重分类29%的视觉评估假阴性患者。

要点

  • 腋窝SUVmax以0.93为截断值,预测ALN转移灵敏度83.6%、特异度79.5%。
  • 联合模型AUC 0.889,显著优于单独腋窝SUVmax(p=0.022)。
  • 定量数据使29%视觉假阴性转移患者被正确重分类。

核心结论

  • 在Luminal A型乳腺癌中,客观定量18F-FDG PET/CT参数及其多参数联合模型显著超越传统视觉评估的诊断局限性。
  • 联合模型在手术决策中提供较高净临床获益,有望作为高危患者管理和低危人群降级手术的个性化决策支持工具。

研究设计与样本

  • 回顾性评估279例经组织病理学确诊的初治Luminal A型乳腺癌女性患者,于2012年1月至2025年12月期间行18F-FDG PET/CT分期及后续ALN取样。
  • 由两名核医学专家共识分析视觉评估与定量参数(原发灶SUVmax、SUVmean、MTV、TLG,腋窝SUVmax及腋窝/原发灶SUVmax比值SUVr)。
  • 采用ROC分析和DeLong检验评估诊断效能,通过NRI和IDI评估模型增量价值,DCA确定临床效用。

主要发现

  • 152例(54.5%)患者确诊ALN转移,转移组所有定量参数均显著更高(p < 0.001)。
  • 专家视觉分析准确率84.6%,腋窝SUVmax的AUC为0.881(95% CI: 0.841-0.920),显著优于视觉评估(AUC 0.851,p = 0.031)。
  • 腋窝SUVmax以0.93为截断值,灵敏度83.6%、特异度79.5%。
  • 联合模型整合视觉与定量数据,诊断效能最高(AUC 0.889),显著优于单独腋窝SUVmax(p = 0.022)。
  • 定量数据使29%(NRI)视觉分析初报假阴性的转移患者被正确重分类,整体判别能力提高41.4%(IDI)。

关键数值与推荐

  • 腋窝SUVmax预测ALN转移的AUC为0.881(95% CI: 0.841-0.920),截断值0.93时灵敏度83.6%、特异度79.5%。
  • 视觉评估AUC为0.851,腋窝SUVmax显著优于视觉评估(p = 0.031)。
  • 联合模型AUC为0.889,显著优于单独腋窝SUVmax(p = 0.022)。
  • 联合模型较视觉分析NRI为29%,IDI为41.4%。
  • 联合模型在20-90%风险阈值范围内净临床获益高于标准手术方法。

临床怎么用

  • 对Luminal A型乳腺癌患者术前可结合腋窝SUVmax等定量参数辅助预测ALN转移,弥补视觉评估的不足。
  • 联合视觉与定量数据的模型有助于识别视觉评估假阴性的转移患者,支持高危患者管理及低危人群降级手术决策。

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)。

原文

PURPOSE: This study aims to assess the diagnostic performance of 18F-FDG PET/CT parameters for the preoperative prediction of axillary lymph node (ALN) metastasis in Luminal A-type breast cancer patients, and to investigate the net clinical benefit of combining visual analysis with quantitative metrics. MATERIALS AND METHODS: We retrospectively evaluated 279 treatment-naive female patients histopathologically diagnosed with Luminal A type breast cancer who underwent 18F-FDG PET/CT for staging and subsequent ALN sampling between January 2012 and December 2025. Both visual assessment and quantitative parameters-including maximum standardized uptake value (SUVmax), mean standardized uptake value (SUVmean), metabolic tumor volume (MTV), and total lesion glycolysis (TLG) of the primary tumor, as well as axillary SUVmax and the axilla-to-primary SUVmax ratio (SUVr)-were analyzed by two nuclear medicine specialists in consensus. Diagnostic performance was measured using ROC analysis and DeLong's test. The incremental value of the models was evaluated through Net Reclassification Improvement (NRI) and Integrated Discrimination Improvement (IDI), while clinical utility was determined via Decision Curve Analysis (DCA). RESULTS: ALN metastasis was confirmed in 152 (54.5%) patients. All quantitative parameters were significantly higher in the metastatic group (p < 0.001). While expert-based visual analysis showed 84.6% accuracy, axillary SUVmax (AUC: 0.881, 95% CI: 0.841-0.920) significantly outperformed visual assessment (AUC: 0.851, p = 0.031). At a cutoff of 0.93, axillary SUVmax yielded 83.6% sensitivity and 79.5% specificity. The Combined Model, integrating visual and quantitative data, achieved the highest diagnostic power (AUC: 0.889) and was statistically superior to axillary SUVmax alone (p = 0.022). The incorporation of quantitative data correctly reclassified 29% (NRI) of metastatic patients initially reported as false negatives in visual analysis and improved overall discrimination by 41.4% (IDI). DCA revealed that the combined model offered a higher net clinical benefit than standard surgical approaches within a 20-90% risk threshold range. CONCLUSION: In Luminal A-type breast cancer, objective quantitative 18F-FDG PET/CT parameters and their derived multiparametric models significantly exceed the diagnostic limitations of conventional visual assessment. By providing high net clinical benefit in the surgical decision-making process, the proposed combined model has the potential to serve as a valuable personalized decision support tool for the management of high-risk patients and the de-escalation of invasive surgery in low-risk cohorts.

来源:Eur J Nucl Med Mol Imaging,查看原文。版权归原作者所有。

引用

  1. Şeker K, Genç HÇ, Koç T, et al. Diagnostic performance of 18F-FDG PET/CT in the prediction of axillary lymph node metastasis in Luminal A breast cancer. Eur J Nucl Med Mol Imaging. 2026. PMID: 42297930. doi:10.1007/s00259-026-07986-0

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。