AUTHOR=Li Jia , Ma Yunhui , Yang Chunyu , Qiu Ganbin , Chen Jingmu , Tan Xiaoliang , Zhao Yue 

TITLE=Radiomics analysis of R2* maps to predict early recurrence of single hepatocellular carcinoma after hepatectomy

JOURNAL=Frontiers in Oncology

VOLUME=Volume 14 - 2024

YEAR=2024

URL=https://www.frontiersin.org/journals/oncology/articles/10.3389/fonc.2024.1277698

DOI=10.3389/fonc.2024.1277698

ISSN=2234-943X

ABSTRACT=Objectives: To determine the function of radiomics analysis based on R2* maps for predicting early recurrence (ER) in single hepatocellular carcinoma (HCC) following partial hepatectomy.Methods: This retrospective study involved 202 patients with surgically verified single HCC who underwent preoperative magnetic resonance imaging between 2018 and 2021 at one of two institutions. The patients from institution I (n = 126) and institution II (n = 76) were assigned to training and validation sets, respectively. A logistic regression with the least absolute shrinkage and selection operator (LASSO) regularization was used to select features to build a radiomic score (Rad-score). Correlations of clinicopathological or texture features and Rad-score with ER were calculated using uni-and multi-variable tests. We established a final combined model including the optimal Rad-score and clinical-pathologic risk factors. We formulated and validated a predictive nomogram for predicting ER in HCC. The discrimination, calibration, and clinical utility of the nomogram were assessed.Results: Multivariate logistic regression showed that the Rad-score, microvascular invasion (MVI), and α fetoprotein (AFP) level > 400 ng/mL were significant independent predictors of ER in HCC. Using these significant factors, a predictive nomogram was built. The areas under the receiver's operator characteristic curve of the nomogram and precision-recall curve were 0.901 and 0.753, respectively, and the F1 was 0.831 in the training set. In the validation set, the values were 0.827, 0.659, and 0.808.The nomogram that integrates the radiomic score, MVI, and AFP demonstrates high predictive efficacy for estimating the risk of ER in HCC. It facilitates personalized risk classification and therapeutic decision-making for HCC patients.