Quantification of Functional MR Predicts Early Response in Post-doxorubicin Drug-Eluting Beads Chemoembolization for Hepatocellular Carcinoma

• Published in: Digestive diseases and sciences 2020-08, Vol.65 (8), p.2433-2441
• Main Authors: Ou, Hsin-You, Cheng, Yu-Fan, Chuang, Yi-Hsuan, Hsu, Hsien-Wen, Chen, Chao-Long, Lazo, Marirose Zingapan, Weng, Ching-Chun, Yu, Chun-Yen, Tsang, Leo Leung-Chit, Huang, Tung-Liang, Tong, Yu-Shun
• Format: Article
• Language: English
• Subjects: Antimitotic agents; Antineoplastic agents; Biochemistry; Biomarkers; Chemoembolization; Disease; Gastroenterology; Hepatology; Hepatoma; Liver cancer; Magnetic resonance imaging; Medical centers; Medicine; Medicine & Public Health; Nuclear magnetic resonance spectroscopy; Oncology; Original Article; Spectrum analysis; Stent (Surgery); Tomography; Transplant Surgery; Tumors; Veins & arteries
• ISSN: 0163-2116; 1573-2568
• DOI: 10.1007/s10620-019-05951-6

Background The purpose of this study was to assess the value of functional MRI (fMRI) of post-doxorubicin drug-eluting beads transcatheter arterial chemoembolization (DEB-TACE) for hepatocellular carcinoma (HCC) as an early imaging biomarker of response to therapy. Methods This prospective analysis included 21 consecutive patients undergoing fMRI before and after DEB-TACE at a single medical center from January 2013 to December 2014. Functional MRI, including relative changes in apparent diffusion coefficient (ADC) and choline levels measured at hydrogen-1 magnetic resonance spectroscopy (MRS) of treated lesions, was recorded at baseline before DEB-TACE, and at 1, 2, and 4 weeks after DEB-TACE therapy. Assessment of tumor response was based on dynamic contrast-enhanced computer tomography imaging response according to modified response evaluation criteria in solid tumors. Results At post-therapy, 76% ( n  = 16) of patients demonstrated objective tumor response, 10% ( n  = 2) had stable disease, and 3 (14%) had progressive disease. Stable disease and progressive disease were designated as non-response. At week 2, the mean change in ADC value of responsive tumors was 0.35 ± 0.24 mm 2 /s, which was greater than that of non-response tumors (mean 0.01 ± 0.13 × 10 −3  mm 2 /s) ( P  = 0.006). Significant differences were found in mean choline/water ratio between responsive (7.8 ± 4.9 × 10 −3 ) and non-responsive (17.2 ± 4.9 × 10 −3 ) tumors ( P  = 0.005). Composite scores of choline/water ratio and relative change of ADC showed significantly better diagnostic accuracy in non-responsive tumors than responsive tumors (area under the curve = 1.0; P