Résumé
•Response or non-response as defined by the Choi criteria exhibited the highest predictive value for overall survival compared with the EASL, RECIST 1.1 and mRECIST criteria.•Patients with progression at the first follow-up examination are never downstaged.•Neither tumor necrosis nor tumor hypervascularization were independent predictors of tumor response.
The first aim was to compare Response Evaluation Criteria in Solid Tumor (RECIST) 1.1, modified Response Evaluation Criteria in Solid Tumor (mRECIST), Choi and European Association for the Study of the Liver (EASL) evaluations to assess the response to sorafenib for hepatocellular carcinoma (HCC).
The second aim was to describe the evolution of HCC and to identify whether some imaging features are predictive of the absence of response.
This retrospective study included 60 patients with advanced HCC treated with sorafenib. Patients must have undergone a scan prior to treatment to identify the number of lesions, size, enhancement and endoportal invasions, and repeat scans thereafter. Computed tomography (CT) scans were analyzed using RECIST 1.1, mRECIST, Choi and EASL criteria. Overall survival was analyzed.
The median overall survival was 10.5 months. On the first CT reevaluation, the sorafenib response rates were 20%, 5%, 7% and 3% according to Choi, EASL, mRECIST and RECIST 1.1. The responders based on Choi exhibited significantly better overall survival compared with non-responders (20.4 months; hazard ratio (HR) 0.042, 95% confidence interval (CI): 0.186–0.94, p=0.035). A modification of imaging findings was observed in 48.3% of patients, and necrosis was present in 44.1% of patients.
This study found a significant difference between Choi versus RECIST 1.1, mRECIST and EASL when evaluating the response to sorafenib in HCC patients.