Scholarlink

Objective <p>Vessels encapsulating tumor clusters (VETC) and microvascular invasion (MVI) are risk factors for early recurrence after hepatectomy in patients with hepatocellular carcinoma (HCC). This study aimed to assess the performance of multiparameters of dual-phase contrast-enhanced dual-energy CT (DECT) in identifying VETC and MVI statuses noninvasively.</p> Methods <p>This retrospective study enrolled 122 patients with pathologically confirmed HCC (VETC +/−, N = 57/65; MVI +/−, N = 54/68) who underwent contrast-enhanced DECT examinations. Conventional clinical information and radiological features were collected. DECT-derived multiparameters in the arterial phase (AP) and portal venous phase (PVP) were analyzed, including non-normalized effective atomic number (Zeff), electron density (ED), and iodine density (ID), and their corresponding normalized parameters (N1-2 Zeff, N1-2 ED, N1-2 ID) based on two normalization approaches (N1 = tumor-to-aorta ratio; N2 = tumor-to-liver parenchyma). The slope of the energy spectrum curve (λ) was also evaluated. Univariate and multivariate logistic regression analyses were performed to construct nomograms and conventional models. The performance of each parameter, conventional model, and nomogram for assessing VETC and MVI was assessed using the area under the receiver operating characteristic curve (AUC).</p> Results <p>Multiparameters predicted VETC and MVI with AUCs ranging from 0.64 to 0.77 (all <i>p</i> &lt; 0.01). Compared with their non-normalized counterparts, the performances of N1-Zeff<sub>AP</sub> (AUC: 0.77, <i>p</i> = 0.01) and N1-ID<sub>AP</sub> (AUC: 0.70, <i>p</i> = 0.01) in predicting VETC were significantly improved, and N1-ED<sub>PVP</sub> and N1-ID<sub>PVP</sub> showed numerically higher sensitivity (88.2%) and specificity (88.9%) for predicting MVI, respectively. Nomogram<sub>VETC</sub>—integrating N1-Zeff<sub>AP</sub>, AST &gt; 40&#xa0;IU/L, and mosaic appearance—for evaluating VETC, and nomogram<sub>MVI</sub>—integrating N1-ED<sub>PVP</sub>, N1-ID<sub>PVP</sub>, nonsmooth tumor margin, and intratumoral arteries—for evaluating MVI outperformed their corresponding conventional models (AUC: 0.80 vs. 0.70, and 0.83 vs. 0.77, respectively; both <i>p</i> &lt; 0.05).</p> Conclusion <p>DECT-derived multiparameters can be used to noninvasively identify VETC and MVI statuses. Nomograms incorporating these parameters enhance VETC and MVI prediction in HCC.</p> Graphic abstract <p></p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Dual-energy CT-derived multiparameters for noninvasive prediction of vessels encapsulating tumor clusters and microvascular invasion in hepatocellular carcinoma

  • 期刊论文

  • 发表日期：

    2026年5月12日

查看全文
  • Mengsi Li,
  • Huayu You,
  • Ma Luo,
  • Anqi Li,
  • Jinhui Zhou,
  • Lina Zhang,
  • Yuanqiang Xiao,
  • Xiao Yu,
  • Weiwei Deng,
  • Zhongping Zhang,
  • Chuanmiao Xie,
  • Shiting Feng,
  • Jin Wang

摘要

Objective

Vessels encapsulating tumor clusters (VETC) and microvascular invasion (MVI) are risk factors for early recurrence after hepatectomy in patients with hepatocellular carcinoma (HCC). This study aimed to assess the performance of multiparameters of dual-phase contrast-enhanced dual-energy CT (DECT) in identifying VETC and MVI statuses noninvasively.

Methods

This retrospective study enrolled 122 patients with pathologically confirmed HCC (VETC +/−, N = 57/65; MVI +/−, N = 54/68) who underwent contrast-enhanced DECT examinations. Conventional clinical information and radiological features were collected. DECT-derived multiparameters in the arterial phase (AP) and portal venous phase (PVP) were analyzed, including non-normalized effective atomic number (Zeff), electron density (ED), and iodine density (ID), and their corresponding normalized parameters (N1-2 Zeff, N1-2 ED, N1-2 ID) based on two normalization approaches (N1 = tumor-to-aorta ratio; N2 = tumor-to-liver parenchyma). The slope of the energy spectrum curve (λ) was also evaluated. Univariate and multivariate logistic regression analyses were performed to construct nomograms and conventional models. The performance of each parameter, conventional model, and nomogram for assessing VETC and MVI was assessed using the area under the receiver operating characteristic curve (AUC).

Results

Multiparameters predicted VETC and MVI with AUCs ranging from 0.64 to 0.77 (all p < 0.01). Compared with their non-normalized counterparts, the performances of N1-ZeffAP (AUC: 0.77, p = 0.01) and N1-IDAP (AUC: 0.70, p = 0.01) in predicting VETC were significantly improved, and N1-EDPVP and N1-IDPVP showed numerically higher sensitivity (88.2%) and specificity (88.9%) for predicting MVI, respectively. NomogramVETC—integrating N1-ZeffAP, AST > 40 IU/L, and mosaic appearance—for evaluating VETC, and nomogramMVI—integrating N1-EDPVP, N1-IDPVP, nonsmooth tumor margin, and intratumoral arteries—for evaluating MVI outperformed their corresponding conventional models (AUC: 0.80 vs. 0.70, and 0.83 vs. 0.77, respectively; both p < 0.05).

Conclusion

DECT-derived multiparameters can be used to noninvasively identify VETC and MVI statuses. Nomograms incorporating these parameters enhance VETC and MVI prediction in HCC.

Graphic abstract