BackgroundCellular immunotherapy has become a new and promising treatment for patients with liver tumor. However, as most immune cells are delivered by intravenous injection, the effect is limited and is likely to produce systemic toxicity. Here, the objective was to investigate the efficacy and safety of cellular immunotherapy by local infusion, which seems to be a promising approach and has not been well-studied.MethodsThe PubMed, Web of Science, Embase, and Cochrane Library databases were searched to obtain literature. The overall response rate (ORR), overall survival (OS) rates, and adverse events were investigated to evaluate the effectiveness and safety of locoregional therapy. The methodological quality of the articles was assessed using the methodological index for non-randomized studies (MINORS) score. The meta-analysis was performed using Stata 15.0.ResultsThe eligible 17 studies involved a total of 318 patients. The random-effects model demonstrated that the ORR of local cell infusion therapy was 48% (95% confidence interval [CI]: 26%–70%). The pooled OS rate was 94% (95% CI: 83%–100%) at 6 months, 87% (95% CI: 74%–96%) at 12 months, and 42% (95% CI: 16%–70%) at 24 months. Subgroup analyses suggested that minimally invasive treatment and absence of metastasis were significantly associated with better ORR. Fourteen studies reported a variety of adverse events related to cell therapy by local perfusion. The most common complications after regional infusion of immune cells were myelosuppression (66%), fever (50%), gastrointestinal toxicity (22%), hepatic dysfunction (15%), and pleural effusion and/or ascites (14%).ConclusionsImmune cell therapy through local perfusion is effective for patients with liver cancer, with manageable toxicity. It demonstrates better prognosis when combined with minimally invasive therapy. Considering the potential limitations, more randomized controlled trials are needed to provide solid evidence for our findings.