基于钆塞酸二钠MRI的影像组学列线图预测肝细胞癌微血管侵犯及患者术后辅助TACE获益的多中心研究
回顾性纳入三家医院260例肝细胞癌患者,基于钆塞酸二钠MRI提取影像组学特征并构建列线图预测微血管侵犯,训练组AUC 0.982、标准化外部验证组0.969、非标准化外部验证组0.981;m-score分层显示高危组术后辅助TACE显著降低早期复发(P=0.006),低危组无显著获益(P=0.270)。
要点
- 列线图整合影像组学标签、最大径>5.1 cm、瘤周低信号、包膜不完整及形态不规则预测微血管侵犯。
- 训练组AUC 0.982,标准化外部验证组0.969,非标准化外部验证组0.981。
- 基于列线图的m-score可分层,高危组术后辅助TACE显著降低早期复发,低危组无显著获益。
核心结论
- 基于钆塞酸二钠MRI的影像组学列线图结合影像组学标签与临床影像特征,可实现术前无创微血管侵犯风险预测及术后辅助TACE获益评估,可能帮助临床医生实施更合适的干预。
研究设计与样本
- 回顾性多中心研究,共纳入三家医院260例符合条件患者,训练队列140例、标准化外部验证队列65例、非标准化外部验证队列55例。
关键影像表现/征象
- 列线图纳入的影像学预测因子包括最大径>5.1 cm、瘤周低信号、包膜不完整及形态不规则。
- 影像组学特征与影像学特征均从肝切除术前钆塞酸二钠MRI图像中提取。
诊断标准与关键数值
- 列线图在训练队列AUC=0.982,标准化外部验证队列AUC=0.969,非标准化外部验证队列AUC=0.981。
- 基于列线图构建m-score用于患者分层并评估术后辅助TACE获益。
- 高危组术后辅助TACE显著降低早期复发(P=0.006),低危组无显著效果(P=0.270)。
临床应用与阅片要点
- 列线图可用于术前无创评估肝细胞癌微血管侵犯风险。
- m-score可分层识别可能从术后辅助TACE获益的患者,高危组显著降低早期复发,低危组无显著获益。
- 决策曲线分析证实该列线图具有临床 usefulness。
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 evaluate the performance of a radiomics nomogram developed based on gadolinium-ethoxybenzyl-diethylenetriamine penta-acetic acid (Gd-EOB-DTPA) MRI for preoperative prediction of microvascular invasion (MVI) of hepatocellular carcinoma (HCC), and to identify patients who may benefit from the postoperative adjuvant transarterial chemoembolization (PA-TACE).
METHODS: A total of 260 eligible patients were retrospectively enrolled from three hospitals (140, 65, and 55 in training, standardized external, and non-standardized external validation cohort). Radiomics features and image characteristics were extracted from Gd-EOB-DTPA MRI image before hepatectomy for each lesion. In the training cohort, a radiomics nomogram which incorporated the radiomics signature and radiological predictors was developed. The performance of the radiomics nomogram was assessed with respect to discrimination calibration, and clinical usefulness with external validation. A score (m-score) was constructed to stratify the patients and explored whether it could accurately predict patient who benefit from PA-TACE.
RESULTS: A radiomics nomogram integrated with the radiomics signature, max-D(iameter) > 5.1 cm, peritumoral low intensity (PTLI), incomplete capsule, and irregular morphology had favorable discrimination in the training cohort (AUC = 0.982), the standardized external validation cohort (AUC = 0.969), and the non-standardized external validation cohort (AUC = 0.981). Decision curve analysis confirmed the clinical usefulness of the novel radiomics nomogram. The log-rank test revealed that PA-TACE significantly decreased the early recurrence in the high-risk group (p = 0.006) with no significant effect in the low-risk group (p = 0.270).
CONCLUSIONS: The novel radiomics nomogram combining the radiomics signature and clinical radiological features achieved preoperative non-invasive MVI risk prediction and patient benefit assessment after PA-TACE, which may help clinicians implement more appropriate interventions.
CLINICAL RELEVANCE STATEMENT: Our radiomics nomogram could represent a novel biomarker to identify patients who may benefit from the postoperative adjuvant transarterial chemoembolization, which may help clinicians to implement more appropriate interventions and perform individualized precision therapies.
KEY POINTS: • The novel radiomics nomogram developed based on Gd-EOB-DTPA MRI achieved preoperative non-invasive MVI risk prediction. • An m-score based on the radiomics nomogram could stratify HCC patients and further identify individuals who may benefit from the PA-TACE. • The radiomics nomogram could help clinicians to implement more appropriate interventions and perform individualized precision therapies.
引用
- Zhang, Zhang, Li, et al. Radiomics nomogram for the prediction of microvascular invasion of HCC and patients' benefit from postoperative adjuvant TACE: a multi-center study. Eur Radiol. 2023. PMID: 37368104. doi:10.1007/s00330-023-09824-5
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。