Initial HAIC versus TACE with lenvatinib and PD-1 inhibitors for hepatocellular carcinoma with portal vein tumor thrombus: a retrospective cohort study

The choice between hepatic arterial infusion chemotherapy (HAIC) and transarterial chemoembolization (TACE) remains uncertain for hepatocellular carcinoma (HCC) with portal vein tumor thrombus (PVTT) treated with lenvatinib and a programmed cell death protein 1 (PD-1) inhibitor. This single-center retrospective cohort included 243 patients with HCC and PVTT without extrahepatic metastasis treated between June 2020 and May 2024. Patients received initial HAIC ( n = 120) or TACE ( n = 123), each combined with lenvatinib and a PD-1 inhibitor. Outcomes were compared after propensity score matching, with patients retained in their initial group after arterial switching. Overall survival (OS) and progression-free survival (PFS) were the co-primary outcomes. Follow-up ended on October 31, 2025. Matching yielded 99 pairs. Median OS was 19.6 months with HAIC and 11.9 months with TACE (HR, 0.617; 95% CI, 0.449–0.848; Holm-adjusted P = 0.006). Median PFS was 7.7 and 5.7 months, respectively (HR, 0.723; 95% CI, 0.541–0.966; Holm-adjusted P = 0.028), but the pair-stratified log-rank test was not significant ( P = 0.128). Matched objective response rates were 54.5% and 39.4% ( P = 0.028), and disease control rates were 97.0% and 87.9% ( P = 0.035). Grade 3–4 adverse events occurred in 30.8% and 30.9% of the original HAIC and TACE cohorts. No treatment-related deaths were observed. Initial HAIC was associated with longer OS and higher response rates than initial TACE in patients receiving lenvatinib and a PD-1 inhibitor. HAIC was associated with significantly longer PFS in the primary matched Cox analysis, but the pair-stratified log-rank test did not reach statistical significance. Chinese Clinical Trial Registry, ChiCTR2500096470; retrospectively registered on January 24, 2025.