Magnetic Resonance Imaging Predictors of Hepatocellular Carcinoma Progression and Dropout in Patients in Liver Transplantation Waiting List. Issue 11 (18th October 2022)

• Record Type: Journal Article
• Authors: Baghdadi, Azarakhsh; Luu, Harry T.; Shaghaghi, Mohammadreza; Ghadimi, Maryam; Simsek, Cem; Xu, Ziyi; Hazhirkarzar, Bita; Motaghi, Mina; Hammami, Muhammad; Clark, Jeanne M.; Gurakar, Ahmet; Kamel, Ihab R.; Kim, Amy K.
• Is Part Of: Transplantation direct. Volume 8:Issue 11(2022)
• Page Start: e1365
• Publication Date: 2022-10-18
• Subjects: Transplantation of organs, tissues, etc -- Periodicals; Transplantation -- Periodicals; 362.19795
• DOI: 10.1097/TXD.0000000000001365
• Languages: English
• ISSNs: 2373-8731

Abstract

Abstract : Background: With the rising incidence of hepatocellular carcinoma (HCC), more patients are now eligible for liver transplantation. Consequently, HCC progression and dropout from the waiting list are also anticipated to rise. We developed a predictive model based on radiographic features and alpha-fetoprotein to identify high-risk patients. Methods: This is a case-cohort retrospective study of 76 patients with HCC who were listed for liver transplantation with subsequent liver transplantation or delisting due to HCC progression. We analyzed imaging-based predictive variables including tumor margin (well- versus ill-defined), capsule bulging lesions, volumetric analysis and distance to portal vein, tumor numbers, and tumor diameter. Volumetric analysis of the index lesions was used to quantify index tumor total volume and volumetric enhancement, whereas logistic regression and receiver operating characteristic curve (ROC) analyses were used to predict the main outcome of disease progression. Results: In univariate analyses, the following baseline variables were significantly associated with disease progression: size and number of lesions, sum of lesion diameters, lesions bulging the capsule, and total and venous-enhancing (viable) tumor volumes. Based on multivariable analyses, a risk model including lesion numbers and diameter, capsule bulging, tumor margin (infiltrative versus well-defined), and alpha-fetoprotein was developed to predict HCC progression and dropout. The model has an area under the ROC of 82%, which was significantly higher than Milan criteria that has an area under the ROC of 67%. Conclusions: Our model has a high predictive test for patient dropout due to HCC progression. This model can identify high-risk patients who may benefit from more aggressive HCC treatment early after diagnosis to prevent dropout due to such disease progression.