AUTHOR=Xu Jiachen , Yin Yu , Yang Jun , Chen Li , Li Zhi , Shen Jian , Wang Wansheng , Ni Caifang 

TITLE=Modified quantitative and volumetric response evaluation criteria for patients with hepatocellular carcinoma after transarterial chemoembolization

JOURNAL=Frontiers in Oncology

VOLUME=Volume 13 - 2023

YEAR=2023

URL=https://www.frontiersin.org/journals/oncology/articles/10.3389/fonc.2023.957722

DOI=10.3389/fonc.2023.957722

ISSN=2234-943X

ABSTRACT=Objective: To investigate the cut-off value of quantitative and volumetric response evaluation criteria for patients with hepatocellular carcinoma (HCC) after transarterial chemoembolization (TACE) and compare the performance of the modified criteria to one-dimensional criteria in survival prediction.
Methods: A retrospective single-center study was performed for treatment-naïve patients with HCC who underwent initial TACE between June 2015 and June 2019. Treatment response assessment was performed after the first observation by contrast CT or MRI, with the measurement of diameters by mRECIST (modified Response Evaluation Criteria in Solid Tumors) and volumes by qEASL (quantitative European Association for Study of the Liver). Overall survival (OS) was the primary endpoint of this study. The new cut-off value for volumetric response evaluation criteria was created using restricted cubic splines. The performance of modified qEASL (mqEASL, with the new cut-off value) and mRECIST on survival prediction was compared by Cox regression models in internal and external validation.
Results: A total of 129 patients (mean age, 60 years ± 11 [standard deviation]; 111 men) were included and divided into training (n=90) and validation (n=39) cohorts. The cut-off value for the viable volume reduction was set at 57.0%. The mqEASL enabled separation of nonresponders and responders in terms of median OS (p<0.001), 11.2 months (95% CI, 8.5-17.2 months) vs 31.5 months (95% CI, 25.5-44.0 months). Two multivariate models were developed with independent prognostic factors (tumor response, metastasis, portal vein tumor thrombus, and subsequent treatment) to predict OS. Model 2 (for mqEASL) had a greater Harrel's C index, higher time-dependent AUROC and more precise calibration on 6-month survival rates than Model 1 (for mRECIST).
Conclusions: With the modified cut-off value, the quantitative and volumetric response of HCC patients to TACE becomes a precise predictor of overall survival. Further studies are needed to verify this modification before applied in clinical practice.