Department of Radiology, First Hospital of China Medical University, 155 Nanjing St, Shenyang, 110001, China.
Institute of Cancer Research, First Hospital of China Medical University, Shenyang, China.
Abdom Radiol (NY). 2024 Dec;49(12):4239-4248. doi: 10.1007/s00261-024-04290-z. Epub 2024 Jul 29.
Histopathological growth patterns (HGPs) of colorectal liver metastases (CRLMs) have prognostic value. However, the differentiation of HGPs relies on postoperative pathology. This study aimed to develop a magnetic resonance imaging (MRI)-based radiomic model to predict HGP pre-operatively, following the latest guidelines.
This retrospective study included 93 chemotherapy-naïve patients with CRLMs who underwent contrast-enhanced liver MRI and a partial hepatectomy between 2014 and 2022. Radiomic features were extracted from the tumor zone (R), a 2-mm outer ring (R), a 2-mm inner ring (R), and a combined ring (R) on late arterial phase MRI images. Analysis of variance method (ANOVA) and least absolute shrinkage and selection operator (LASSO) algorithms were used for feature selection. Logistic regression with five-fold cross-validation was used for model construction. Receiver operating characteristic curves, calibrated curves, and decision curve analyses were used to assess model performance. DeLong tests were used to compare different models.
Twenty-nine desmoplastic and sixty-four non-desmoplastic CRLMs were included. The radiomic models achieved area under the curve (AUC) values of 0.736, 0.906, 0.804, and 0.794 for R, R, R, and R, respectively, in the training cohorts. The AUC values were 0.713, 0.876, 0.785, and 0.777 for R, R, R, and R, respectively, in the validation cohort. R exhibited the best performance.
The MRI-based radiomic models could predict HGPs in CRLMs pre-operatively.
结直肠癌肝转移（CRLM）的组织病理学生长模式（HGPs）具有预后价值。然而，HGPs 的区分依赖于术后病理。本研究旨在根据最新指南，建立一种基于磁共振成像（MRI）的放射组学模型，以预测术前 HGP。
本回顾性研究纳入了 93 例接受化疗的 CRLM 患者，这些患者于 2014 年至 2022 年间接受了增强型肝 MRI 检查和部分肝切除术。在 MRI 动脉晚期图像上，从肿瘤区（R）、2mm 外环（R）、2mm 内环（R）和联合环（R）中提取放射组学特征。采用方差分析（ANOVA）和最小绝对收缩和选择算子（LASSO）算法进行特征选择。使用 5 折交叉验证的逻辑回归进行模型构建。使用受试者工作特征曲线、校准曲线和决策曲线分析来评估模型性能。使用 DeLong 检验比较不同模型。
共纳入 29 例促结缔组织增生型和 64 例非促结缔组织增生型 CRLM。在训练队列中，R、R、R 和 R 的放射组学模型的曲线下面积（AUC）值分别为 0.736、0.906、0.804 和 0.794。在验证队列中，R、R、R 和 R 的 AUC 值分别为 0.713、0.876、0.785 和 0.777。R 表现最佳。
基于 MRI 的放射组学模型可以术前预测 CRLM 的 HGPs。
