The ARH score, a practical guide to decision-making for retreatment with hepatic arterial infusion chemotherapy in hepatocellular carcinoma patients.

Background: Hepatic arterial infusionchemotherapy (HAIC) is a promising option for large unresectable hepatocellular carcinoma (HCC). Identifying patients who could benefit from continuous HAIC remains a challenge. We aimed to establish an objective model to guide the decision for retreatment with HAIC.

Methods: Between 2015 and 2020, the data of patients with large unresectable HCC without macrovascular invasion or extrahepatic spread undergoing multiple HAIC cycles from 3 different centers were retrieved. We investigated the basic tumor parameters and the effect of HAIC on liver function and tumor response, and their impact on overall survival (OS). A point score (ARH, Assessment for Retreatment with HAIC) was built by using a stepwise Cox regression model in the training cohort (n = 112) and was validated in an independent validation cohort (n = 71).

Results: The high α-fetoprotein before the second cycle of HAIC, an increase in Child-Pugh score, and undesirable radiologic tumor responses remained independent negative prognostic factors and were used to create the ARH score. The prognosis of HCC patients deteriorated significantly with the increase in ARH score. The median OS of patients with ARH score 0-2 points and ≥ 2.5 points were 19.37 months and 11.60 months (P < 0.001). All of these results had been confirmed in the external validation cohort and demonstrated significance across multiple subgroups.

Conclusions: The ARH score makes an excellent prediction of the prognosis of HCC patients who received retreatment of HAIC. Patients with an ARH score ≥ 2.5 prior to the second cycle of HAIC may not profit from further sessions.