Transarterial chemoembolization (TACE) is used as a locoregional treatment for liver-limited hepatocellular carcinoma (HCC) and can induce immunologic changes within the tumor environment. This provided the rationale for investigating whether combining TACE with immune checkpoint inhibition could be safely delivered and potentially enhance antitumor activity. A multicenter phase I study has now reported safety and antitumor outcomes with nivolumab administered at different time points around drug-eluting bead TACE (deb-TACE).
The study, titled “A multicenter pilot study of nivolumab with drug-eluting bead transarterial chemoembolization in patients with liver-limited hepatocellular carcinoma,” was published in ESMO Gastrointestinal Oncology in September 2026.
Authors: J.J. Harding, O.M. Fitzpatrick, J.F. Chou, H. Yarmohammadi, K.A. Reiss, R.K. Do, P. Wong, D.N. Khalil, I. El Dika, C. Ferrer, O. Heffernan, A. Yaqubie, J.D. Giardina, M. D’Angelica, W.R. Jarnagin, G. Nadolski, A. Covey, J.P. Erinjeri, K. Brown, M.C. Soulen, B. Tan, M. Capanu and G.K. Abou-Alfa.
Study Design
The investigator-initiated, multicenter phase I study, CA209-731/NCT03143270, included patients with unresectable, liver-limited HCC who were not candidates for curative resection or liver transplantation. Eligible patients had Child-Pugh A liver function, measurable disease according to RECIST v1.1 and an ECOG performance status of 0 or 1.
The primary objective was to assess the safety and tolerability of nivolumab combined with deb-TACE. Secondary objectives included best objective response rate (ORR), 6- and 12-month progression-free survival (PFS), and overall survival (OS), while exploratory analyses assessed circulating immune-cell populations and serum cytokines.
Nineteen patients were enrolled across three cohorts. Nivolumab was administered intravenously at 240 mg every 2 weeks for up to 1 year, with the timing relative to deb-TACE differing between cohorts.
In cohort 1, patients underwent deb-TACE on day 0 and started nivolumab 14 days later. In cohort 2, nivolumab was started 4 weeks before deb-TACE, withheld on the day of the procedure and resumed 2 weeks afterward. In cohort 3, nivolumab was also started 4 weeks before deb-TACE but continued without interruption through the procedure. Cohort 3 was subsequently expanded for additional safety assessment.
Patients were enrolled between June 2017 and July 2021. The median age was 67 years, and 15 of the 19 patients were male. Six patients had BCLC B1 disease, eight had BCLC B2 disease and five had BCLC C disease, with the latter classification based on segmental portal vein involvement. No patients had main portal vein invasion or extrahepatic disease.
Safety
No dose-limiting toxicity was observed in cohorts 1 or 2. One dose-limiting toxicity occurred in cohort 3, consisting of grade 3 transaminitis after three doses of nivolumab and one deb-TACE procedure. Imaging demonstrated bilomas and possible biliary tree injury at the site of embolization. The transaminitis resolved with supportive measures, and treatment was restarted without recurrence.
The most common treatment-related adverse events were fatigue in 53% of patients, transaminase elevation in 42% and fever in 37%. No immune-mediated hepatitis, treatment-related liver failure or treatment-related deaths were reported.
The median number of nivolumab doses was nine, ranging from three to 28. Two patients had nivolumab interruptions. Three immune-related adverse events were reported: one patient developed grade 2 hypothyroidism and grade 2 adrenal insufficiency, while another developed grade 3 transaminitis.
Antitumor Activity
Among the 19 patients, the ORR was 21% (95% CI, 6%-44%), with four patients achieving a partial response. Stable disease was observed in 11 patients (58%), progressive disease in three patients (16%) and one patient was not assessable for response. Median PFS was 5.9 months (95% CI, 3.6-23 months), while median OS was 23 months (95% CI, 19-62 months). Estimated OS rates were 68% at 12 months and 46% at 24 months. Estimated PFS rates were 23% at 12 months and 8.5% at 24 months.
At the July 30, 2025 data cutoff, four patients remained alive, with a median follow-up of 72.7 months among surviving patients. Fifteen patients had died from the disease.
Exploratory Immune Findings
Exploratory immune profiling showed transient increases in CD8-positive T-cell proliferation after deb-TACE, particularly in the cohorts in which nivolumab was initiated before embolization. Cytokine changes were modest and transient, with no sustained differences between cohorts. Minimal changes were observed in TIM-3, LAG-3 and CTLA-4 expression, while regulatory T-cell frequencies remained stable. The investigators noted that these immune findings were modest and variable, and the study was not powered to determine the optimal sequencing of nivolumab and deb-TACE.
Interpretation and Limitations
The combination of nivolumab and deb-TACE was feasible and generally well tolerated across the different schedules evaluated. However, the antitumor activity observed in the study was comparable with historical outcomes reported for deb-TACE or TACE alone. The authors noted that historical studies of TACE have reported ORRs of approximately 20%-30%, median PFS of 4-8 months and median OS of 20-30 months. In this study, the ORR was 21%, median PFS was 5.9 months and median OS was 23 months, providing no clear evidence of enhanced activity from adding single-agent nivolumab.
Interpretation is limited by the small phase I sample size and heterogeneous patient distribution across cohorts. Some patients had received prior locoregional treatment, which may have affected the tumor immune microenvironment. The translational analyses were also limited by lower cellular viability, the possibility that sampling missed short-lived cytokine changes and the possibility that peripheral blood analyses may not fully capture intratumoral immune dynamics.
Takeaway
Nivolumab combined with deb-TACE demonstrated acceptable tolerability across the treatment schedules evaluated in patients with unresectable, liver-limited HCC. However, antitumor activity did not exceed historical benchmarks for deb-TACE alone. The authors concluded that the findings do not demonstrate clear synergy between deb-TACE and single-agent checkpoint inhibition and support further investigation of rational multi-agent approaches together with improved patient selection.
The full article is available in ESMO Gastrointestinal Oncology.
Find more similar articles on OncoDaily.
