Assessment of HCC response to Yttrium-90 radioembolization with gadoxetate disodium MRI: correlation with histopathology.
Details
Download: Y90.pdf (8275.26 [Ko])
State: Public
Version: author
License: Not specified
State: Public
Version: author
License: Not specified
Serval ID
serval:BIB_B3BB1D94D489
Type
Article: article from journal or magazin.
Collection
Publications
Institution
Title
Assessment of HCC response to Yttrium-90 radioembolization with gadoxetate disodium MRI: correlation with histopathology.
Journal
European radiology
ISSN
1432-1084 (Electronic)
ISSN-L
0938-7994
Publication state
Published
Issued date
09/2022
Peer-reviewed
Oui
Volume
32
Number
9
Pages
6493-6503
Language
english
Notes
Publication types: Journal Article
Publication Status: ppublish
Publication Status: ppublish
Abstract
Transarterial <sup>90</sup> Y radioembolization (TARE) is increasingly being used for hepatocellular carcinoma (HCC) treatment. However, tumor response assessment after TARE may be challenging. We aimed to assess the diagnostic performance of gadoxetate disodium MRI for predicting complete pathologic necrosis (CPN) of HCC treated with TARE, using histopathology as the reference standard.
This retrospective study included 48 patients (M/F: 36/12, mean age: 62 years) with HCC treated by TARE followed by surgery with gadoxetate disodium MRI within 90 days of surgery. Two radiologists evaluated tumor response using RECIST1.1, mRECIST, EASL, and LI-RADS-TR criteria and evaluated the percentage of necrosis on subtraction during late arterial, portal venous, and hepatobiliary phases (AP/PVP/HBP). Statistical analysis included inter-reader agreement, correlation between radiologic and pathologic percentage of necrosis, and prediction of CPN using logistic regression and ROC analyses.
Histopathology demonstrated 71 HCCs (2.8 ± 1.7 cm, range: 0.5-7.5 cm) including 42 with CPN, 22 with partial necrosis, and 7 without necrosis. EASL and percentage of tumor necrosis on subtraction at the AP/PVP were independent predictors of CPN (p = 0.02-0.03). Percentage of necrosis, mRECIST, EASL, and LI-RADS-TR had fair to good performance for diagnosing CPN (AUCs: 0.78 - 0.83), with a significant difference between subtraction and LI-RADS-TR for reader 2, and in specificity between subtraction and other criteria for both readers (p-range: 0.01-0.04). Radiologic percentage of necrosis was significantly correlated to histopathologic degree of tumor necrosis (r = 0.66 - 0.8, p < 0.001).
Percentage of tumor necrosis on subtraction and EASL criteria were significant independent predictors of CPN in HCC treated with TARE. Image subtraction should be considered for assessing HCC response to TARE when using MRI.
• Percentage of tumor necrosis on image subtraction and EASL criteria are significant independent predictors of complete pathologic necrosis in hepatocellular carcinoma treated with <sup>90</sup> Y radioembolization. • Subtraction, mRECIST, EASL, and LI-RADS-TR have fair to good performance for diagnosing complete pathologic necrosis in hepatocellular carcinoma treated with <sup>90</sup> Y radioembolization.
This retrospective study included 48 patients (M/F: 36/12, mean age: 62 years) with HCC treated by TARE followed by surgery with gadoxetate disodium MRI within 90 days of surgery. Two radiologists evaluated tumor response using RECIST1.1, mRECIST, EASL, and LI-RADS-TR criteria and evaluated the percentage of necrosis on subtraction during late arterial, portal venous, and hepatobiliary phases (AP/PVP/HBP). Statistical analysis included inter-reader agreement, correlation between radiologic and pathologic percentage of necrosis, and prediction of CPN using logistic regression and ROC analyses.
Histopathology demonstrated 71 HCCs (2.8 ± 1.7 cm, range: 0.5-7.5 cm) including 42 with CPN, 22 with partial necrosis, and 7 without necrosis. EASL and percentage of tumor necrosis on subtraction at the AP/PVP were independent predictors of CPN (p = 0.02-0.03). Percentage of necrosis, mRECIST, EASL, and LI-RADS-TR had fair to good performance for diagnosing CPN (AUCs: 0.78 - 0.83), with a significant difference between subtraction and LI-RADS-TR for reader 2, and in specificity between subtraction and other criteria for both readers (p-range: 0.01-0.04). Radiologic percentage of necrosis was significantly correlated to histopathologic degree of tumor necrosis (r = 0.66 - 0.8, p < 0.001).
Percentage of tumor necrosis on subtraction and EASL criteria were significant independent predictors of CPN in HCC treated with TARE. Image subtraction should be considered for assessing HCC response to TARE when using MRI.
• Percentage of tumor necrosis on image subtraction and EASL criteria are significant independent predictors of complete pathologic necrosis in hepatocellular carcinoma treated with <sup>90</sup> Y radioembolization. • Subtraction, mRECIST, EASL, and LI-RADS-TR have fair to good performance for diagnosing complete pathologic necrosis in hepatocellular carcinoma treated with <sup>90</sup> Y radioembolization.
Keywords
Carcinoma, Hepatocellular/diagnostic imaging, Carcinoma, Hepatocellular/pathology, Carcinoma, Hepatocellular/radiotherapy, Gadolinium DTPA, Humans, Liver Neoplasms/diagnostic imaging, Liver Neoplasms/pathology, Liver Neoplasms/radiotherapy, Magnetic Resonance Imaging/methods, Middle Aged, Necrosis, Retrospective Studies, Yttrium Radioisotopes, Embolization, Gadoxetate disodium, Hepatocellular carcinoma, Magnetic resonance imaging, Yttrium radioisotopes
Pubmed
Web of science
Create date
11/04/2022 7:28
Last modification date
06/09/2022 6:12