| CT细胞外体积分数对上皮性卵巢癌分型的预测价值 |
| 投稿时间：2025-07-11 修订日期：2025-07-17 点此下载全文 |
| 引用本文：王勇,陈鹏,林运智,付秀虹.CT细胞外体积分数对上皮性卵巢癌分型的预测价值[J].医学研究杂志,2025,54(12):106-111 |
| DOI：
10.11969/j.issn.1673-548X.2025.12.018 |
| 摘要点击次数: 19 |
| 全文下载次数: 16 |
|
| 基金项目:河南省医学科技攻关计划联合共建项目(LHGJ20240763)；河南省中央引导地方科技发展资金项目(Z20221343023)；漯河市妇科肿瘤多模态影像学评估创新型科技团队 |
|
| 中文摘要:目的 探讨基于CT生成的细胞外体积分数(extracellular volume fraction, ECV)结合临床及影像特征构建列线图模型预测术前上皮性卵巢癌(epithelial ovarian cancer, EOC)分型的价值。方法 回顾性分析2022年6月～2024年8月经术后病理确诊为EOC并在术前接受全腹部增强CT扫描的124例患者。收集患者的临床资料包括年龄、绝经状态,以及肿瘤标志物CA125、HE4等,评估影像学特征包括病灶最大径、形态、边界等,并基于平扫和平衡期CT值及红细胞比容计算肿瘤ECV值。比较Ⅰ型与Ⅱ型EOC的组间差异,采用多因素Logistic回归分析筛选独立预测因子,构建预测模型,并使用R语言绘制列线图。通过受试者工作特征(receiver operating characteristic, ROC)曲线、校准曲线和决策曲线分析评估模型的效能。结果 Ⅱ型EOC患者的CA125、HE4水平高于Ⅰ型(P<0.05),病灶最大径更小(P<0.05)、形态更不规则(P<0.05)、边界更不清(P<0.05),腹膜转移更多见(P<0.05)、ECV值更高(P <0.05)。多因素Logistic回归分析结果显示,肿瘤CA125、边界及ECV值是预测Ⅱ型EOC的独立预测因子(P<0.05)。由三者构建的联合模型预测效能最佳,曲线下面积(area under the curve, AUC)为0.921(95% CI:0.855～0.988),敏感度为84.2%,特异性为91.7%。基于此模型构建的列线图具有良好的校准度和临床实用性。结论 基于CT生成的ECV结合临床标志物CA125及影像特征构建的列线图模型能在术前有效、无创地预测EOC分型,为个体化治疗决策提供参考依据。 |
| 中文关键词:上皮性卵巢癌 细胞外体积分数 体层摄影术 X线计算机 列线图 |
| |
| Predictive Value of CT-based Extracellular Volume Fraction for the Classification of Epithelial Ovarian Cancer. |
|
|
| Abstract:Objective To explore the value of preoperative nomogram combining CT-based extracellular volume fraction (ECV) with clinical and imaging features for predicting preoperative classification of epithelial ovarian cancer (EOC). Methods A retrospective study included 124 pathologically confirmed EOC patients who underwent preoperative contrast-enhanced CT scan of the entire abdomen from June 2022 to August 2024. The clinical data of the patients were collected, including age, menopausal status, and tumor markers such as CA125 and HE4. The imaging features including the maximum diameter, shape, and boundary of the lesion were evaluated. The ECV value of tumor were was calculated based on the CT values of plain scan and equilibrium phase and the hematocrit. The differences between typeⅠand type Ⅱ EOC were compared. Multivariate Logistic regression analysis was used to screen the independent predictors to build a prediction model, and the nomogram was drawn using R language. The performance of the model was evaluated via receiver operating characteristic (ROC) curve analysis, calibration curve analysis and decision curve analysis. Results Compared with type Ⅰ EOC, type ⅡEOC patients exhibited significantly higher serum levels of CA125 and HE4 (P<0.05), smaller lesion maximum diameters (P<0.05), more irregular morphology (P<0.05), more ill-defined boundaries (P<0.05), a higher frequency of peritoneal metastasis (P<0.05), and significantly elevated ECV values (P<0.05). The results of multivariate Logistic regression analysis showed that CA125, boundary and ECV value of tumors were the independent predictors for type ⅡEOC (P<0.05). The combined model incorporating these three predictors demonstrated optimal performance, achieving an area under the curve (AUC) of 0.921 (95% CI:0.855-0.988), with a sensitivity of 84.2% and specificity of 91.7%. The nomogram constructed based on this model showed good calibration and clinical utility. Conclusion The nomogram model constructed based on CT-derived ECV combined with the clinical marker CA125 and imaging features provides a reliable non-invasive tool for preoperative classification of EOC, providing a reference basis for individualized treatment planning. |
| keywords:Epithelial ovarian cancer Extracellular volume fraction Tomography, X-Ray computed Nomogram |
| 查看全文 查看/发表评论 下载PDF阅读器 |
|
|
|
