Granulocyte-colony stimulating factor in acute-on-chronic liver failure: Systematic review and meta-analysis of randomized controlled trials

Background and aims: A dysfunctional immune response is key to the pathogenesis of acute-on-chronic liver failure (ACLF). It has been suggested that treatment with granulocyte colony-stimulating factor (G-CSF) increases survival in patients with ACLF by improving immune cell dysfunction and promoting liver regeneration. The aim of the study is to evaluate the survival benefit associated with G-CSF administration compared with standard medical therapy (SMT) in ACLF.

Methods: Systematic review and meta-analysis of randomized controlled trials. The primary outcome was survival at 60-90 days. We searched Ovid Medline, EMBASE, and Cochrane Central Register of Controlled Trials from inception to August 2021. Manual searches of reference lists in relevant articles and conference proceedings were also included. The revised Cochrane risk-of-bias tool was used for quality and risk of bias assessment. Two independent investigators extracted the data, and disagreements were solved by a third collaborator.

Results: The initial search identified 142 studies. Four randomized controlled trials were selected for quantitative analysis including 310 patients (154 G-CSF and 156 SMT). Significant heterogeneity was observed (I²=74%, Chi²=11.57, p=0.009). G-CSF administration did not improve survival in patients with ACLF (random-effects model, risk ratio=0.64 [95% CI 0.39, 1.07]). However, when considering only the results from the studies performed in Asia, a significant decrease on mortality was observed (risk ratio=0.53 [95% CI 0.35, 0.81]). Severity scores (MELD and Child) and CD34+ peripheral cells mobilization did not significantly improve with G-CSF.

Conclusion: In a systematic review and meta-analysis, G-CSF administration did not significantly improve overall survival compared to SMT in patients with ACLF. The beneficial effects observed in Asian studies, as opposed to the European region, suggest that specific populations may benefit from further research aiming to identify certain subgroups with favourable outcomes when using G-CSF.