A pooled analysis of mesenchymal stem cell-based therapy for liver disease

Abstract

Background: Liver disease is a major cause of death and disability. Mesenchymal stem cells (MSCs) show promise for the treatment of liver disease. However, whether MSC-based therapy is more effective than conventional treatment is unclear, as are the optimal MSC source, the administration frequency, and the most effective MSC delivery route. We therefore undertook a systematic review and meta-analysis of the therapeutic efficacy of MSCs against liver disease and the related factors.

Methods: We systematically searched Medline (PubMed), Cochrane Library, EMBASE, ClinicalTrials.gov, and SinoMed CBM to identify studies published up to June 2017 involving liver disease patients receiving MSC-based therapy and which reported estimates of liver function during the follow-up period.

Results: Thirty-nine studies were selected from 672 publications. According to a meta-analysis of 23 controlled studies, compared with conventional treatment MSC therapy significantly improves liver function in patients with liver disease in terms of the model of end-stage liver disease score, albumin, alanine aminotransferase, and total bilirubin levels, and prothrombin time, up to 6 months after administration. However, it has no beneficial effects in terms of prothrombin activity, international normalized ratio, or cholinesterase level. Considerable heterogeneity was identified at most time points. Subgroup analyses showed that a single MSC injection was more effective than multiple injections, MSC administration was more effective via the hepatic artery than the peripheral vein, and MSCs derived from bone marrow were more effective than those derived from the umbilical cord.

Conclusions: MSC-based therapy is relatively safe and improves liver function during the first 6 months after administration. A single injection administration via the hepatic artery and MSCs derived from bone marrow are optimal in terms of improving liver function. However the significant heterogeneity among studies and discontinuous results of the subgroup meta-analysis should be addressed; moreover the long-term efficacy of MSC therapy warrants further investigation.

Keywords: Cell therapy; Liver disease; Mesenchymal stem cell; Pooled analysis.

Fig. 1

Flowchart of study selection. MSC mesenchymal stem cell, nRCT, nonrandomized controlled trial, RCT randomized controlled trial

Fig. 2

Changes in liver function from baseline to week 48. A total of 983 patients received mesenchymal stem cell (MSC)-based therapy, and 557 patients underwent conventional supportive treatment. After treatment, the model for end-stage liver disease (MELD) score, and the alanine aminotransferase (ALT), total bilirubin (TBiL), prothrombin time (PT), and international normalized ratio (INR) levels showed decreasing trends, while the albumin (ALB), prothrombin activity (PTA), and cholinesterase (CHE) levels showed increasing trends. Data are means ± SEM

Fig. 3

Forest plot of liver function. Compared with the control group (C), the model for end-stage liver disease (MELD) score in the MSC group (M) was nonsignificantly lower at baseline, significantly lower at weeks 2, 4, 8, 12, and 24, and nonsignificantly lower at weeks 36 and 48. The albumin (ALB) level in the MSC group was significantly higher at baseline and at weeks 2, 4, 8, 12, 24, and 36, and nonsignificantly higher at week 48. The alanine aminotransferase (ALT) level in the MSC group was nonsignificantly lower at baseline and at weeks 8, 36, and 48, and significantly lower at weeks 2, 4, 12, and 24. The total bilirubin (TBiL) level in the MSC group was non-significantly higher at baseline, significantly lower at weeks 2, 4, 8, 12, and 24, and non-significantly lower at weeks 36 and 48. The prothrombin time (PT) in the MSC group was significantly higher at baseline, non-significantly lower at weeks 2, 4, and 8, significantly lower at weeks 12 and 24, and non-significantly lower at weeks 36 and 48. The prothrombin activity (PTA), international normalized ratio (INR), and cholinesterase (CHE) values did not differ significantly between the MSC and control groups at any time point. CI confidence interval. SMD standardized mean difference

Fig. 4

Meta-analysis of patient subgroups. Red indicates significant improvement in the mesenchymal stem cell (MSC) group compared with the control group; blue indicates no significant improvement. Subgroup analyses at other time points and of other parameters could not be conducted because of an insufficient number of studies. ALB albumin, ALT alanine aminotransferase, B-MSC bone marrow-derived mesenchymal stem cell, CI confidence interval, MELD model for end-stage liver disease, nRCT nonrandomized controlled trial, RCT randomized controlled trial, TBiL total bilirubin, UC-MSC umbilical cord-derived mesenchymal stem cell