This study sought to identify factors that are predictive of a therapeutic response to hepatic arterial infusion chemotherapy (HAIC) by focusing on the number of prior transcatheter arterial chemoembolization (TACE) sessions. To determine the parameters predicting a good response to HAIC, we retrospectively analyzed 170 patients with hepatocellular carcinoma (HCC) who received HAIC regimens comprising low-dose cisplatin combined with 5-fluorouracil (LFP) or cisplatin (CDDP) for the first time. In both the LFP and CDDP regimens, the response rates were significantly lower in patients with three or more prior TACE sessions than in those with two or fewer prior TACE sessions (LFP 57% versus 28%; p=0.01, CDDP 27% versus 6%; p=0.01). Multivariable logistic regression analysis revealed that the number of prior TACE sessions (≥ 3) was significantly associated with non-responder status (odds ratio 4.17, 95% Confidence Interval (CI) 1.76-9.86) in addition to the HAIC regimen. Multivariable analysis using the Cox proportional hazards model revealed that a larger number of prior TACE sessions (≥ 3) was a significant risk factor for survival (hazard ratio 1.60, 95% CI 1.12-2.29) in addition to Child-Pugh class, serum alpha-fetoprotein concentration, and maximum diameter of HCC. HCC patients who receive fewer prior TACE sessions (≤ 2) were found to be better responders to HAIC.
Keywords: hepatic arterial infusion chemotherapy; hepatocellular carcinoma; refractory; transcatheter arterial chemoembolization.