We found seven published clinical

We found seven published clinical trials studying the effects of MSC transplantation in patients with liver cirrhosis (listed in Table 2). A pilot phase 1 study published in 2007 included four patients with decompensated liver cirrhosis (Mohamadnejad et al., 2007). Cells were derived from iliac crest puncture and expanded in culture during 73days with 2.5 passages on average. Thirty-two million autologous MSCs were administrated through a single injection into the cubital vein of the arm. No severe side-effects were observed until the end of the follow-up at 12months after the transplantation. Three out of four patients improved their Model for End-Stage Liver Disease (MELD) scores at 6months after transplantation and two of them continued to improve up to 12months. In another pilot phase 1 study, 30 to 50million autologous iliac crest-derived MSCs were injected in the peripheral or portal vein of eight patients with end-stage liver disease (Kharaziha et al., 2009). The treatment was well tolerated and the average MELD score among patients improved from 18 to 11 after 24weeks. In a phase 2 study, Amer et al. randomized 40 patients with end-stage liver failure due to chronic hepatitis C into two groups of 20 patients: the first group received autologous bone marrow-derived MSCs previously transdifferentiated in hepatocyte-like tranylcypromine manufacturer in vitro, the second group received standard supportive treatment (Amer et al., 2011). The patients receiving MSCs had significant improvement in Child–Pugh and MELD scores appearing after 2weeks and maintained for 6months when compared to 20 patients who received traditional supportive treatment. Scores measured by fatigue scale and performance status were also improved. Patients receiving MSCs through intrahepatic route had pronounced stronger improvement of MELD and fatigue scores compared to MSCs infused through intrasplenic route. In a phase 2 trial, Peng et al. transplanted autologous MSCs from iliac bone aspirates in 53 patients with liver failure caused by hepatitis B virus infection (Peng et al., 2011). Levels of albumin, bilirubin, prothrombin time and MELD score were improved at two to three weeks after MSC transplantation, compared with 105 control-patients. Long-term follow-up revealed similar results between groups in terms of hepatocellular carcinoma incidence and mortality. In another phase 2 trial, El-Ansary et al. transplanted autologous iliac bone-derived MSCs in patients with hepatitis C-related cirrhosis and MELD score >12 (El-Ansary et al., 2012). Fifteen patients received 1million MSCs per kg-weight intravenously and were compared to 10 patients with conventional supportive treatment. Follow-up of MSC-transplanted patients at 3 and 6months showed a significant improvement in serum albumin, prothrombin, and total bilirubin levels, resulting in an improved MELD score. In this study, there was no difference in curative potential between MSCs and MSC-derived hepatocyte-like cells. More recently, in phase 2 trial Zhang et al. randomized (2:1) 46 patients with chronic hepatitis B receiving either three injections with 0.5million/kg allogeneic umbilical cord-derived MSCs (n=31) or saline solution (n=15) (Zhang et al., 2012). Patients receiving MSC infusion had improved MELD score, ascites and fibrosis marker levels up to 48weeks after the treatment. The major concerns of this study are the lack of evidence that the patients have been given appropriate conventional anti-viral therapy and the absence of explicit data about the course of illness before inclusion, the virus genotype, or the length of abstinence from alcohol. A common problem in these studies was the lack of data regarding the MSC\'s fate once injected into the body. Gholamrezanezhad et al. addressed this issue and injected 111In-oxine-labeled human bone marrow-derived MSCs in a peripheral vein of cirrhotic patients (Gholamrezanezhad et al., 2011). They observed an initial accumulation of MSCs in the lungs with a peak 20min after injection, followed by a decrease of the lung signal and a gradual increase in the liver and spleen signal that was detected up to 10days after injection (111In half-life is 67h), indicating that MSCs do not only entrap in the lungs, but also migrate to the liver following peripheral intravenous injection. Terai et al. transplanted an average of 70million/kg unsorted bone marrow mononuclear cells from bone marrow aspirates (including other stem cell types as hematopoietic stem cells, MSCs, multipotent adult progenitor cells (Jiang et al., 2002)) in cirrhotic patients (Terai et al., 2006). This therapy allowed an improvement in serum albumin levels, Child–Pugh scores, alpha-fetoprotein levels and liver proliferating cell nuclear antigen up to 24weeks after transplantation. Similar observations reported improvements with the infusion of unsorted bone marrow cells in cirrhotic patients (Amin et al., 2006; Lyra et al., 2007, 2010; Kim et al., 2010) and no improvement in patients with alcoholic steatohepatitis (Spahr et al., 2013). The pilot studies mentioned above suggested that autologous MSC infusion allows mild biological improvements in patients with liver diseases, but clear and significant clinical benefit was not yet reported. To our knowledge, none of these studies provided histologic evidence of improvement with MSC treatment. Of note, intraportal infusion seemed to be more efficient than peripheral route (Amer et al., 2011), and differentiation toward hepatocytes prior to infusion seemed not to increase MSC curative potential (El-Ansary et al., 2012). Overall, evidences provided by most of these clinical studies are quite poor. The cells are often poorly characterized; and improvements are claimed where there are insufficiently powered experimental/control groups, or lack of randomization to make this claim.