Background: A previous study by our group using autologous monocyte-derived DC pulsed ex vivo with HepG2 cell lysate showed some clinical benefit with evidence of antigen-specific T-cell responses in patients with advanced HCC. The current trial reports the activity of this vaccine in combination with TACE in patients with HCC. All patients also received low-dose cyclophosphamide to deplete regulatory T cells and thereby enhance vaccination. Methods: Patients with intermediate stage HCC (performance status 0-2, Child Pugh A/B7) were randomised 1:1 to TACE plus low-dose cyclophosphamide (Group 1) or TACE plus low-dose cyclophosphamide plus dendritic cell vaccination (Group 2). Cyclophosphamide was administered on Day 1 and 29 followed by TACE on Day 31 (+/- DC infusion), with further cyclophosphamide on Days 60, 90 and 120 (+/- additional DC infusions on Days 62, 92 and 122). The primary endpoint was progression free survival (PFS) by RECIST v1.1. Secondary endpoints included radiological response by RECIST v1.1, PFS and radiological response according to modified (m) RECIST, overall survival (OS), immune response and toxicity. Target recruitment was 48 evaluable patients (24 patients in each arm) to detect a 20% increase in PFS rate at 1 year (30% vs 50%) with a relaxed one-sided statistical significance level of 20% and 80% power using a logrank test. Results: Between March 2016 and October 2019, 55 patients from 3 UK centres were randomised of whom 48 were evaluable (24 each arm). Median PFS by RECIST criteria was significantly longer in Group 2 compared to Group 1 (18.6 vs 10.4 months: hazard ratio (HR) 0.43, 80% CI -∞-0.59; one-sided p = 0.02). Median PFS using mRECIST criteria showed a similar magnitude of benefit (18.6 vs 10.8 months: HR 0.48, 95% CI 0.22-1.02). Median OS was 25.7 months in Group 2 vs 21.5 months in Group 1 (HR 0.61, 95% CI 0.27-1.38). Group 2 showed a higher overall response rate (complete and partial response) by RECIST (54% vs 29%) and mRECIST (75% vs 54%) and a higher disease control rate (complete and partial response and stable disease) by RECIST (92% vs 67%) and mRECIST (88% vs 67%). Treatment with DC infusions was well tolerated; the most common adverse events were chills (30%), fatigue (22%) and nausea (22%), all of which were low grade. Immune response analyses are currently ongoing. Conclusions: The addition of tumour lysate pulsed DC infusions to treatment with TACE plus low-dose cyclophosphamide significantly increased PFS in patients with HCC. To the best of our knowledge, this is the first randomised study to demonstrate efficacy using DC in HCC. Further investigation of the role of DC infusions in the treatment of HCC are warranted but will need to take into account the current evolving immunotherapy landscape. Clinical trial information: 11889464.