• HCC
• RECIST
• TACE

Abstract
Aims: To retrospectively evaluate agreement between modified RECIST criteria (mRECIST) assessed at Computed Tomography (CT) and pathology in a large series of patients with hepatocellular carcinoma (HCC) who were transplanted after transarterial chemoembolization (TACE).
Materials and Methods: From January 1996 to December 2010, 178 patients (M/F=155/23; mean age 55.8±6.3 years) with HCC underwent TACE followed by liver transplantation (LT). Two blinded independent readers retrospectively reviewed CT examinations, performed in the interval between TACE and LT, to assess tumor response to TACE according to mRECIST. On the explanted livers, percentage of tumor necrosis was pathologically classified as 100%, 90-99%, 50-90%, <50% and 0%.
Results: At latest CT examination, the objective response rate was 77.5% (138/178), with 86 cases (48.3%) of complete response (CR). A good intra- (k=0.75 and 0.86) and inter-observer (k=0.81) agreement was obtained. Out of 302 nodules, sensitivity and specificity of CT in detecting complete necrosis were 87.5% and 68.9%, respectively. On a per-patient basis, agreement between mRECIST and pathology was obtained in 119 patients (66.9%), with 20 cases (11.2%) of underestimation and 39 cases (21.9%) of overestimation of tumor response at CT. The majority of overestimations (26/39, 66.7%) consisted of patients with CR and 90-99% necrosis. CT sensitivity and specificity in differentiating between responders and non-responders were 92.3% and 82.9%, respectively.
Conclusions: CT can overestimate tumor response, by missing minimal residual viable tumor. Nonetheless, mRECIST criteria assessed at CT after TACE are reproducible and reliable, with high accuracy in differentiating responders and non-responders.