Premium
Determination of hepatocellular carcinoma grade by needle biopsy is unreliable for liver transplant candidate selection
Author(s) -
Court Colin M.,
HarlanderLocke Michael P.,
Markovic Daniela,
French Samuel W.,
Naini Bita V.,
Lu David S.,
Raman Steven S.,
Kaldas Fady M.,
Zarrinpar Ali,
Farmer Douglas G.,
Finn Richard S.,
Sadeghi Saeed,
Tomlinson James S.,
Busuttil Ronald W.,
Agopian Vatche G.
Publication year - 2017
Publication title -
liver transplantation
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.814
H-Index - 150
eISSN - 1527-6473
pISSN - 1527-6465
DOI - 10.1002/lt.24811
Subject(s) - medicine , hepatocellular carcinoma , milan criteria , concordance , grading (engineering) , liver transplantation , biopsy , predictive value , gastroenterology , transplantation , oncology , pathology , civil engineering , engineering
The objective of this article is to evaluate the utility of preoperative needle biopsy (PNB) grading of hepatocellular carcinoma (HCC) as a biomarker for liver transplantation (LT) candidate selection. Given the prognostic significance of HCC tumor grade, PNB grading has been proposed as a biomarker for LT candidate selection. Clinicopathologic characteristics of HCC LT recipients (1989‐2014) with a PNB were analyzed, and the concordance of PNB grade to explant grade and vascular invasion was assessed to determine whether incorporation of PNB grade to accepted transplant criteria improved candidate selection. Of 965 patients undergoing LT for HCC, 234 (24%) underwent PNB at a median of 280 days prior to transplant. Grade by PNB had poor concordance to final explant pathology (κ = 0.22; P = 0.003), and low sensitivity (29%) and positive predictive value (35%) in identifying poorly differentiated tumors. Vascular invasion was predicted by explant pathologic grade ( r s = 0.24; P < 0.001) but not PNB grade ( r s = −0.05; P = 0.50). Increasing explant pathology grade ( P = 0.02), but not PNB grade ( P = 0.65), discriminated post‐LT HCC recurrence risk. The incorporation of PNB grade to the established radiologic Milan criteria (MC) did not result in improved prognostication of post‐LT recurrence (net reclassification index [NRI] = 0%), whereas grade by explant pathology resulted in significantly improved reclassification of risk (NRI = 19%). Preoperative determination of HCC grade by PNB has low concordance with explant pathologic grade and low sensitivity and positive predictive value in identifying poorly differentiated tumors. PNB grade did not accurately discriminate post‐LT HCC recurrence and had no utility in improving prognostication compared with the MC alone. Incorporation of PNB to guide transplant candidate selection appears unjustified. Liver Transplantation 23 1123–1132 2017 AASLD.