LAUSR

Purpose We aimed to compare different radiological criteria in evaluating the early tumor response of patients with unresectable hepatocellular carcinoma (uHCC) treated with an anti-programmed cell death protein 1 (PD-1) antibody plus bevacizumab. Method From October 2018 to January 2020, 58 patients [49 (84.5%) men, age = 55.2 ± 10.6 years] receiving both anti-PD-1 antibody and bevacizumab were retrospectively included. Pre- and the first posttreatment contrast-enhanced computed tomography (CE-CT) scans were performed in all patients. The Response Evaluation Criteria in Solid Tumors, version 1.1 (RECIST 1.1), modified RECIST (mRECIST), Choi, and the revised Choi (rChoi) criteria were applied to evaluate tumor response. The endpoint event was defined as overall survival (OS). Results Six (10.3%), 9 (15.5%), 30 (51.7%), and 12 (20.7%) patients were diagnosed as responders by RECIST 1.1, mRECIST, Choi, and rChoi, respectively. The RECIST 1.1 and mRECIST criteria failed to correlate the evaluation categories with OS (p = 0.130 and 0.253, respectively), while both Choi and rChoi significantly correlated with OS (p = 0.002 and 0.006, respectively). Among the four criteria, only those patients identified as responders by Choi (p = 0.0005) and rChoi (p = 0.005) showed significantly better OS than the non-responders. The cumulative 1- and 2-year OS rates by Choi were 93.3% and 79.8% in responders and 69.3% and 30.3% in non-responders, respectively; these rates were 100.0% and 100.0% in responders and 74.9% and 43.1% in non-responders by rChoi, respectively. Conclusions The evaluation of early tumor response using Choi and rChoi instead of RECIST 1.1 and mRECIST significantly correlated with the OS of patients with uHCC treated with an anti-PD-1 antibody plus bevacizumab. Moreover, patients identified as responders by Choi and rChoi showed significantly better OS than the non-responders.