Chemoembolization for hepatocellular carcinoma: 1-month response determined with apparent diffusion coefficient is an independent predictor of outcome.

To evaluate the predictive utility of apparent diffusion coefficient (ADC) changes at diffusion-weighted (DW) magnetic resonance (MR) imaging 1 month after transcatheter arterial chemoembolization (TACE) for hepatocellular carcinoma (HCC) compared with the Response Evaluation Criteria in Solid Tumors (RECIST) version 1.1, European Association for the Study of the Liver (EASL) criteria, and modified RECIST (mRECIST). Institutional review board approval and informed consent were obtained for this prospective study. Thirty-eight patients with inoperable HCC underwent 1.5-T MR imaging, including DW imaging, before and 1 month after TACE. Responses 1 month after TACE were assessed with the ADC change relative to baseline (ADC ratio), RECIST, EASL criteria, and mRECIST. Eight patients underwent transplantation 4 months after TACE, allowing the association between response and histopathologic necrosis ratio to be determined. In 30 patients, Kaplan-Meier and log-rank tests were used to correlate the response variables with progression-free and overall survival. Median progression-free survival (PFS) was 5 months, and overall survival was 17.5 months. Survival analyses showed significant effects of pretreatment α-fetoprotein level (P = .03) and ADC ratio (P < .0001) on PFS and substantial effects of mRECIST, RECIST, and EASL criteria (.05 < P < .1). ADC ratio was a significant predictor of 5-month PFS (P < .001), with an accuracy of 93.3% compared with 53.3% with mRECIST and EASL criteria and 66.7% with RECIST. No response variables correlated with overall survival. Only the ADC ratio was significantly associated with histopathologic tumor necrosis (P = .03). The ADC ratio 1 month after TACE was an independent predictor of PFS, which showed stronger association with tumor response than did RECIST, EASL criteria, or mRECIST.