State Key Laboratory of Oncology in South China, Guangzhou, China.
Collaborative Innovation Center for Cancer Medicine, Guangzhou, China.
Am Surg. 2023 May;89(5):1468-1478. doi: 10.1177/00031348211063549. Epub 2021 Dec 11.
There is no satisfactory indicator for monitoring recurrence after resection of hepatocellular carcinoma (HCC). This retrospective study aimed to design and validate an HCC monitor recurrence (HMR) model for patients without metastasis after hepatectomy.
A training cohort was recruited from 1179 patients with HCC without metastasis after hepatectomy between February 2012 and December 2015. An HMR model was developed using an AdaBoost classifier algorithm. The factors included patient age, TNM staging, tumor size, and pre/postoperative dynamic variations of alpha-fetoprotein (AFP). The diagnostic efficacy of the model was evaluated based on the area under the receiver operating characteristic curves (AUCs). The model was validated using a cohort of 695 patients.
In preoperative patients with positive or negative AFP, the AUC of the validation cohort in the HMR model was .8877, which indicated better diagnostic efficacy than that of serum AFP (AUC, .7348). The HMR model predicted recurrence earlier than computed tomography/magnetic resonance imaging did by 191.58 ± 165 days. In addition, the HMR model can predict the prognosis of patients with HCC after resection.
The HMR model established in this study is more accurate than serum AFP for monitoring recurrence after hepatectomy for HCC and can be used for real-time monitoring of the postoperative status in patients with HCC without metastasis.
目前尚无满意的指标用于监测肝癌（HCC）切除后复发。本回顾性研究旨在设计并验证一种适用于肝癌切除术后无转移患者的 HCC 复发监测（HMR）模型。
本研究纳入了 2012 年 2 月至 2015 年 12 月期间 1179 例肝癌切除术后无转移患者的资料。采用 AdaBoost 分类器算法建立 HMR 模型，模型中包含的因素有患者年龄、TNM 分期、肿瘤大小以及甲胎蛋白（AFP）的术前术后动态变化。根据受试者工作特征曲线（ROC）下面积（AUC）评估模型的诊断效能。使用另一组 695 例患者的数据对模型进行验证。
在术前 AFP 阳性或阴性的患者中，验证队列中 HMR 模型的 AUC 为 0.8877，其诊断效能优于血清 AFP（AUC 为 0.7348）。HMR 模型比计算机断层扫描/磁共振成像早 191.58±165 天预测复发。此外，HMR 模型可预测 HCC 患者术后的预后。
与血清 AFP 相比，本研究中建立的 HMR 模型更能准确地监测 HCC 患者肝切除术后的复发情况，可用于实时监测无转移 HCC 患者的术后状态。
