ORIGINAL ARTICLE 18F-FDG PET/CT predicts microvascular invasion and early recurrence after liver resection for hepatocellular carcinoma A prospective observational study Chetana Lim 1 , Chady Salloum 1 , Julia Chalaye 2 , Eylon Lahat 1 , Charlotte E. Costentin 3 , Michael Osseis 1 , Emmanuel Itti 2 , Cyrille Feray 2 & Daniel Azoulay 1 1 Department of Hepatobiliary and Pancreatic Surgery and Liver Transplantation, 2 Department of Nuclear Medicine, and 3 Department of Hepatology, Henri Mondor Hospital, Créteil APHP, France Abstract Background: This study assessed the prognostic value of 18F-fluorodeoxyglucose (18F-FDG) positron emission tomography (PET)/computed tomography (CT) in the prediction of MVI and early recurrence following resection. Method: This prospective study (ClinicalTrials.gov ID: NCT02145013) included 78 consecutive HCC patients who underwent 18F-FDG PET/CT before curative-intent resection from 2014 to 2017. Prog- nostic factors available before surgery for predicting MVI and early recurrence (2 years) were identified by univariate and multivariate analyses. Results: The 18F-FDG PET/CT result was positive in 30 (38%) patients. MVI was present in 33% (26/78) of specimens. Early recurrence occurred in 19% (14/74) of surviving patients. PET/CT positivity was the sole independent predictor of MVI (odds ratio [OR] = 3.6, 95% confidence interval [CI] = 1.1 – 11.2; p = 0.03), with a specificity and sensitivity for predicting MVI of 73% and 62%, respectively. Analysis of variables available before surgery showed that PET/CT positivity (hazard ratio [HR] = 5.8, 95% CI = 1.6 – 20.4; p = 0.006) and the male sex (HR = 6.6; 95% CI = 1.8 – 24.2; p = 0.005) were independent predictors of early recurrence. Conclusion: 18F-FDG PET/CT predicts MVI and early recurrence after surgery for HCC and could be used to select patients for neoadjuvant treatment. Received 21 May 2018; accepted 10 October 2018 Correspondence Daniel Azoulay, Department of HPB and Liver Transplantation, Henri Mondor Hospital, 51 avenue de Lattre de Tassigny, 94010 Créteil, France. E-mail: daniel.azoulay@aphp.fr Introduction Surgery is the mainstay for the management of hepatocellular carcinoma (HCC). Before proceeding to liver resection, 3 interconnected issues should be assessed, namely, i) the technical resectability of the tumor, ii) the operability in patients, and iii) the oncological utility of surgery. Thus far, the last aspect can be addressed with limited accuracy by clinical models including all or some of the following: cross- sectional imaging findings, the alpha-fetoprotein (AFP) level, liver disease, and liver function. This limited accuracy in predicting biological aggressiveness is underlined by i) the high tumor recurrence rate (the main factor of survival), which reaches up to 70% at 5 years, despite initial curative- intent liver resection and ii) the good results obtained in a significant subset of patients not fulfilling even the strictest criteria for resection. 1 Indeed, microvascular invasion (MVI) 1 – 5 is the main acknowledged surrogate of biological aggressiveness, but it can only be assessed a posteriori by analyzing the resected specimen, i.e., after exposing the pa- tient to surgical mortality, morbidity, and, from an onco- logical perspective, a potentially futile surgery. Preresection HPB 2018, -,1 – 9 © 2018 International Hepato-Pancreato-Biliary Association Inc. Published by Elsevier Ltd. All rights reserved. https://doi.org/10.1016/j.hpb.2018.10.007 HPB Please cite this article in press as: Lim C, et al., 18F-FDG PET/CT predicts microvascular invasion and early recurrence after liver resection for hepatocellular carcinoma, HPB (2018), https://doi.org/10.1016/j.hpb.2018.10.007