Engraftment failure (EF) is a formidable complication after autologous hematopoietic stem cell transplantation (auto-HSCT). Mesenchymal stem cells (MSCs) and cord blood (CB) have been found to support hematopoiesis. Thus, we designed a multicenter randomized clinical trial to investigate the effects and safety of MSCs alone or combined with CB infusion for patients with EF.


Twenty-two patients were randomly assigned to receive the treatment with MSCs alone (MSCs group, n=11) or MSCs combined with CB (CB group, n=11). MSCs were administered once every 2 weeks (2 doses were a cycle) in both groups, and single-unit CB was administered at the same day with the first application of MSCs in CB group; After one cycle of treatments (within 28 days), the patients who did not response to MSCs would receive the therapeutic schedule in CB group, and those patients with partial response (PR) in MSCs group and those without complete response (CR) in CB group would continue another cycle of MSCs treatment. If patients did not obtain CR after two cycles of treatments (within 56 days), they would receive other treatments including allogeneic HSCT.


After the first treatment cycle, the effect rates were not significant difference in MSCs and CB groups (7/11 vs. 9/11, P=0.635), and the median time of hematopoietic reconstruction was 22 (18-28) and 17 (13-22) days, respectively (P=0.036) in MSCs and CB group. There was statistically significant difference regarding neutrophil engraftment, with 17 (range 9-28) and 8 (range 6-14) days respectively (P=0.030), but no difference regarding platelet engraftment, with 21 (range 18-28) and 18 (range 11-21) days respectively (P=0.092) between MSCs and CB groups. After two cycles of treatments, 17 patients obtained CR, 2 PR and 3 NR. CB chimerisms were not detected by short tandem repeat (STR) at +15 and +30 days after CB infusion. None of the patients experienced any adverse events of grade 3/4 with the Common Terminology Criteria for Adverse Events v3.0 (CTCAE v3.0) and acute GVHD or chronic GVHD during the period of study treatment and follow-up. One patient with PR in MSCs group and 1 NR in CB group received allogeneic HSCT at +249 and +273 days after auto-HSCT because of EF and primary disease relapse, respectively. At a median follow-up time of 345 (range 129–784) days post-transplantation, 16 patients remained alive, 3 died of relapse of primary diseases and 1 died of CMV pneumonia following allo-HSCT. None of patients developed EBV-DNA viremia and EBV-associated diseases in two groups. The 2-year overall survival, disease-free survival and tumor relapse post-transplantation were 75.2% (95% CI, 63.2-87.2%), 79.5% (95% CI, 70.1-88.9%) and 20.5% (95% CI, 11.1-29.9%) respectively.


Our data suggest that ex-vivo-expanded MSCs derived from HLA-mismatched BM alone or combined with unrelated CB are effective to EF after auto-HSCT. CB can facilitate the effect of MSCs to EF. Both two strategies do not result in GVHD or increase the risk of primary diseases relapse in patients with EF.

This trial was registered at www.clinicaltrials.govas#NCT01763099.


Liu:It was supported by 863 Program (No. 2011AA020105) and National Public Health Grand Research Foundation (Grant No. 201202017).: Research Funding; It was supported by National Natural Science Foundation of China (Grant No.81000231, No.81270647) and Science and Technology Program of Guangzhou of China (11A72121174). : Research Funding.

Author notes


Asterisk with author names denotes non-ASH members.