Computed tomography-based radiomics for predicting lymphovascular invasion in rectal cancer. Issue 146 (January 2022)

• Record Type: Journal Article
• Authors: Li, Mou; Jin, Yumei; Rui, Jun; Zhang, Yongchang; Zhao, Yali; Huang, Chencui; Liu, Shengmei; Song, Bin
• Is Part Of: European journal of radiology. Issue 146(2022)
• Publication Date: 2022-01
• Subjects: Radiomics -- Computed tomography -- Rectal cancer -- Lymphovascular invasion; LVI lymphovascular invasion -- RC rectal cancer -- CT computed tomography -- ROC receiver operating characteristic -- AUC the area under the receiver operating characteristic curve -- CI confidence interval -- MRI magnetic resonance imaging -- HE hematoxylin-eosin -- cT CT-reported T stage -- cN CT-reported N stage -- VOI volume of interest -- ICC intra (inter)-observer correlation coefficient -- CA19-9 carbohydrate antigen 19-9 -- OR odds ratio -- CEA carcinoembryonic antigen -- CA125 carbohydrate antigen 125 -- SEN sensitivity -- SPE specificity; Medical radiology -- Periodicals; Radiology -- Periodicals; Radiologie médicale -- Périodiques; Medical radiology; Periodicals; 616.075705
• DOI: 10.1016/j.ejrad.2021.110065
• Languages: English
• ISSNs: 0720-048X

Abstract

Highlights: Lymphovascular invasion is an unfavorable prognostic factor in rectal cancer. External validation allows more insight to generalizability of the model. Radiomics of peritumoral regions can offer biological information about tumors. Abstract: Purpose: To develop and externally validate a computed tomography (CT)-based radiomics model for predicting lymphovascular invasion (LVI) before treatment in patients with rectal cancer (RC). Method: This retrospective study enrolled 351 patients with RC from three hospitals between March 2018 and March 2021. These patients were assigned to one of the following three groups: training set (n = 239, from hospital 1), internal validation set (n = 60, from hospital 1), and external validation set (n = 52, from hospitals 2 and 3). Large amounts of radiomics features were extracted from the intratumoral and peritumoral regions in the portal venous phase contrast-enhanced CT images. The score of radiomics features (Rad-score) was calculated by performing logistic regression analysis following the L1-based method. A combined model (Rad-score + clinical factors) was developed in the training cohort and validated internally and externally. The models were compared using the area under the receiver operating characteristic curve (AUC). Results: Of the 351 patients, 106 (30.2%) had an LVI + tumor. Rad-score (comprised of 22 features) was significantly higher in the LVI + group than in the LVI- group (0.60 ± 0.17 vs. 0.42 ± 0.19, P = 0.001). The combined model obtained good predictive performance in the training cohort (AUC = 0.813 [95% CI: 0.758–0.861]), with robust results in internal and external validations (AUC = 0.843 [95% CI: 0.726–0.924] and 0.807 [95% CI: 0.674–0.903]). Conclusions: The proposed combined model demonstrated the potential to predict LVI preoperatively in patients with RC.