Department of Radiology and Research Institute of Radiology, Asan Medical Center, University of Ulsan College of Medicine, 88, Olympic-ro 43-gil, Songpa-Gu, Seoul, 05505, South Korea.
Abdom Radiol (NY). 2021 Aug;46(8):3729-3737. doi: 10.1007/s00261-020-02845-4. Epub 2020 Nov 3.
To evaluate associations between pathology and CT assessments made according to the mRECIST in HCC treated by conventional TACE (cTACE), and to identify predictors of complete tumor necrosis.
From March 2016 to July 2018, 83 patients with a total of 100 masses were retrospectively included. Patients underwent sequential cTACE and portal vein embolization, and later hepatic surgery. Evaluation of treatment response and measurement of baseline lipiodol accumulation as mean HU was performed on CT at the time point closest to the time of operation (mean, 54.5 days after cTACE). Significant predictors associated with complete necrosis were identified by multivariate analysis. The optimal cut-off HU value of lipiodol accumulation for prediction of complete necrosis was determined using a ROC analysis.
According to mRECIST, complete response (CR, n = 70) and partial response (n = 30) were classified. 34.3% (24/70) masses classified as CR according to mRECIST were found to have viable lesions on pathology. On multivariate analysis, mean HU of lipiodol accumulation was the only significant predictor of complete necrosis (p = .003, odds ratio 1.746, 95% CI 1.201-2.539). On ROC analysis, 460 HU as a cut-off value was significantly associated with complete necrosis (67.4% sensitivity, 75.0% specificity).
A threshold value for lipiodol accumulation > 460 HU was highly sensitive and specific for complete necrosis, even in complete response according to mRECIST. Therefore, if lipiodol accumulation is insufficient in post-TACE CT, recurrence should be monitored more sensitively.
评估常规 TACE(cTACE)治疗 HCC 时根据 mRECIST 进行的病理和 CT 评估之间的相关性,并确定完全肿瘤坏死的预测因素。
回顾性纳入 2016 年 3 月至 2018 年 7 月期间共 83 例 100 个病灶的患者。患者接受序贯 cTACE 和门静脉栓塞术,随后进行肝切除术。在最接近手术时间的 CT 上评估治疗反应并测量基线碘化油积聚的平均值 HU(cTACE 后平均 54.5 天)。通过多变量分析确定与完全坏死相关的显著预测因素。使用 ROC 分析确定预测完全坏死的碘化油积聚最佳截断值 HU 值。
根据 mRECIST,将完全缓解(CR,n=70)和部分缓解(n=30)进行分类。根据 mRECIST 分类为 CR 的 70 个肿块中有 34.3%(24/70)在病理上发现有活性病变。多变量分析表明,碘化油积聚的平均 HU 是完全坏死的唯一显著预测因素(p=0.003,优势比 1.746,95%CI 1.201-2.539)。ROC 分析显示,460 HU 作为截断值与完全坏死显著相关(67.4%的敏感性,75.0%的特异性)。
在 mRECIST 完全缓解的情况下,碘化油积聚值>460 HU 是完全坏死的高度敏感和特异性指标。因此,如果 TACE 后 CT 显示碘化油积聚不足,应更敏感地监测复发情况。