Adjuvant lenvatinib plus transarterial chemoembolization is associated with better outcomes than gemcitabine plus cisplatin in combined hepatocellular cholangiocarcinoma

To evaluate and compare the efficacy and safety of lenvatinib combined with transarterial chemoembolization (TACE) and gemcitabine plus cisplatin (GP) as adjuvant treatments for combined hepatocellular-cholangiocarcinoma (cHCC-CCA), and to explore the differential therapeutic benefits stratified by the proportion of hepatocellular carcinoma (HCC) components in tumor tissue. A total of 84 pathologically confirmed cHCC-CCA patients who received adjuvant treatment between January 2019 and December 2025 were retrospectively enrolled from two medical centers (The 960th Hospital of PLA, n = 52; The Affiliated Hospital of Shandong University of Traditional Chinese Medicine, n = 32), and assigned to the Lenvatinib+TACE group ( n = 42) and the GP group ( n = 42) according to the adjuvant treatment regimen. All baseline clinical, pathological and radiological characteristics were balanced between the two groups. The objective response rate (ORR) assessed by RECIST 1.1 criteria was the primary efficacy endpoint, with progression-free survival (PFS), overall survival (OS), tumor recurrence rate and time to recurrence as secondary endpoints. Adverse events (AEs) graded by NCI-CTCAE criteria were used to evaluate treatment safety. Subgroup analyses were performed according to the proportion of HCC components in cHCC-CCA tissue (< 30%, 30%–50%, > 50%). Kaplan-Meier method and Log-rank test were used for survival analysis, and appropriate statistical methods were adopted for intergroup comparison of categorical and continuous variables. The two groups were well balanced in all baseline clinical and tumor-related characteristics (all P > 0.05). In the overall cohort, Lenvatinib+TACE showed significantly superior efficacy to GP, with a markedly higher ORR (64.3% vs. 40.5%, P < 0.001), longer median PFS (8.50 vs. 5.35 months; adjusted HR = 0.44, 95% CI 0.27–0.72, P = 0.001) and median OS (16.10 vs. 9.95 months; adjusted HR = 0.46, 95% CI 0.26–0.81, P = 0.007). No significant differences were observed in tumor recurrence rate (31.0% vs. 28.6%) and median time to recurrence (10.40 months vs. 8.40 months) between the two groups (all *P*>0.05). Stratified analysis revealed that the therapeutic advantage of Lenvatinib+TACE was most pronounced in the mixed component (30%–50% HCC) and predominant HCC component (> 50% HCC) subgroups, with statistically significant differences in ORR, PFS and OS between the two groups (all P < 0.05); while in the predominant cholangiocarcinoma (CC) component subgroup (< 30% HCC), no significant differences in the above efficacy endpoints were found between the two regimens (all *P*>0.05). The overall incidence of AEs was manageable in both groups, with distinct AE profiles. The incidence of severe (Grade 3–4) AEs was slightly higher in the Lenvatinib+TACE group (23.8% vs. 19.0%), and no treatment-related mortality occurred in either group. Hypertension and hand-foot syndrome were the characteristic AEs of Lenvatinib+TACE, while the incidences of elevated liver enzymes, fatigue and diarrhea were comparable between the two groups (all P > 0.05). As an adjuvant treatment for cHCC-CCA, Lenvatinib+TACE has better overall efficacy than the traditional GP regimen and a manageable safety profile. The therapeutic benefit of Lenvatinib+TACE is stratified by the proportion of HCC components in tumor tissue, with the most significant efficacy in patients with mixed or predominant HCC components, which can provide a personalized treatment reference for cHCC-CCA patients.