Department of Interventional Therapy, The Second Affiliated Hospital of Guangzhou University of Chinese Medicine, No. 111 Dade Road, Guangzhou, 510080, Guangdong, People's Republic of China.
Department of Ultrasound, The Sixth Medical Center of Chinese PLA General Hospital, Beijing, 100048, China.
J Cancer Res Clin Oncol. 2024 Oct 8;150(10):448. doi: 10.1007/s00432-024-05941-w.
Surgical resection (SR) following transarterial chemoembolization (TACE) is a promising treatment for unresectable hepatocellular carcinoma (uHCC). However, biomarkers for the prediction of postoperative recurrence are needed.
To develop and validate a model combining deep learning (DL) and clinical data for early recurrence (ER) in uHCC patients after TACE.
A total of 511 patients who received SR following TACE were assigned to derivation (n = 413) and validation (n = 98) cohorts. Deep learning features were taken from the largest tumor area in liver MRI. A nomogram using DL signatures and clinical data was made to forecast early recurrence risk in uHCC patients. Model performance was evaluated using area under the curve (AUC).
A total of 2278 subsequences and 31,346 slices multiparametric MRI including contrast-enhanced T1WI, T2WI and DWI were input in the DL model simultaneously. Multivariable analysis identified three independent predictors for the development of the nomogram: tumor number (hazard ratio [HR]:3.42, 95% confidence interval [CI]: 2.75-4.31, P = 0.003), microvascular invasion (HR: 9.21, 6.24-32.14; P < 0.001), and DL scores (HR: 17.46, 95% CI: 12.94-23.57, P < 0.001). The AUC of the nomogram was 0.872 and 0.862 in two cohorts, significantly outperforming single-subsequence-based DL mode and clinical model (all, P < 0.001). The nomogram provided two risk strata for cumulative overall survival in two cohorts, showing significant statistical results (P < 0.001).
The DL-based nomogram is essential to identify patients with uHCC suitable for treatment with SR following TACE and may potentially benefit personalized decision-making.
经动脉化疗栓塞（TACE）后手术切除（SR）是治疗不可切除肝细胞癌（uHCC）的一种有前途的治疗方法。然而，需要预测术后复发的生物标志物。
开发和验证一种结合深度学习（DL）和临床数据的模型，用于预测 TACE 后 uHCC 患者的早期复发（ER）。
共纳入 511 例接受 TACE 后 SR 的患者，分为推导（n=413）和验证（n=98）队列。深度学习特征取自肝脏 MRI 中最大肿瘤区域。使用 DL 签名和临床数据制作列线图，以预测 uHCC 患者的早期复发风险。使用曲线下面积（AUC）评估模型性能。
共输入 2278 个子序列和 31346 个切片多参数 MRI，包括对比增强 T1WI、T2WI 和 DWI。多变量分析确定了列线图的三个独立预测因素：肿瘤数量（风险比[HR]：3.42，95%置信区间[CI]：2.75-4.31，P=0.003）、微血管侵犯（HR：9.21，6.24-32.14；P<0.001）和 DL 评分（HR：17.46，95%CI：12.94-23.57，P<0.001）。列线图在两个队列中的 AUC 分别为 0.872 和 0.862，明显优于基于单序列的 DL 模式和临床模型（均 P<0.001）。列线图在两个队列中为累积总生存提供了两个风险分层，具有显著的统计学结果（P<0.001）。
基于 DL 的列线图对于识别适合 TACE 后 SR 治疗的 uHCC 患者至关重要，并且可能有助于个性化决策。
