To demonstrate the safety of transendocardial stem cell injection (TESI) with autologous MSCs and BMCs in patients with ICM.• To assess prespecified outcomes of efficacy.

Background While human mesenchymal stem cells (hMSCs) have been tested in ischemic cardiomyopathy, few studies exist in chronic non-ischemic dilated cardiomyopathy (NIDCM). Objectives The POSEIDON-DCM trial is a randomized comparison of safety and efficacy of autologous (auto) vs. allogeneic (allo) bone marrow-derived hMSCs in NIDCM. Methods Thirty-seven patients were randomized to either allo- or auto-hMSCs in a 1:1 ratio. Patients were recruited between December 2011 and July 2015 at the University of Miami Hospital. Patients (age: 55.8 ± 11.2; 32% female) received hMSCs (100 million) by transendocardial stem cell injection (TESI) in ten left ventricular sites by NOGA Catheter. Treated patients were evaluated at baseline, 30 days, 3-, 6-, and 12-months for safety: serious adverse events (SAE), and efficacy endpoints: Ejection Fraction (EF), Minnesota Living with Heart Failure Questionnaire (MLHFQ), Six Minute Walk Test (6MWT), MACE, and immune-biomarkers. This trial is registered with ClinicalTrials.gov, #NCT01392625. Results There were no 30-day treatment-emergent (TE)-SAEs. 12-month SAE incidence was 28.2% (95% CI: 12.8, 55.1) in allo, and 63.5% (95% CI: 40.8, 85.7; p=0.1004) in auto. One allo-group patient developed an elevated donor specific cPRA. EF increased in allo by 8.0 units (95% Cl: 2.8, 13.2; p=0.004), and in auto: 5.4 units (95% Cl: −1.4, 12.1; p=0.116, allo vs. auto p=0.4887). 6MWT increased for allo: 37.0 meters (95% Cl: 2.0 to 72.0; p=0.04), but not auto: 7.3 meters (95% Cl: −47.8, 33.3; p=0.71, auto vs. allo p=0.0168). MLHFQ score decreased in allo (p=0.0022), and auto (p=0.463; p=0.172). The MACE rate was lower in allo vs. auto (p=0.0186). Tumor necrosis factor alpha (TNF-α) decreased (p=0.0001 for each), to a greater extent in allo vs. auto at six-months (p=0.05). Conclusion These findings demonstrate safety and support greater, clinically meaningful efficacy of allo-hMSC vs. auto-hMSC in NIDCM patients. Pivotal trials of allo-hMSCs are warranted based on these results.

Rationale While accumulating data support the efficacy of intramyocardial cell-based therapy to improve LV function in patients with chronic ischemic cardiomyopathy undergoing CABG, the underlying mechanism and impact of cell injection site remain controversial.Mesenchymal stem cells (MSCs) improve LV structure and function through several effects including: reducing fibrosis, neoangiogenesis and neomyogenesis. Objective To test the hypothesis that the impact on cardiac structure and function following intramyocardial injections of autologous MSCs results from a concordance of pro-recovery phenotypic effects. Methods and Results Six patients were injected with autologous MSCs into akinetic/hypokinetic myocardial territories not receiving bypass graft for clinical reasons. MRI was used to measure scar, perfusion, wall thickness and contractility at baseline, 3, 6 and 18 months and to compare structural and functional recovery in regions that received MSC injections alone, revascularization alone, or neither. A composite score of MRI variables was used to assess concordance of antifibrotic effects, perfusion, and contraction at different regions. After 18 months, subjects receiving MSCs exhibited increased LVEF (+9.4±1.7%, p=0.0002) and decreased scar mass (-47.5±8.1%; p<0.0001) compared to baseline. MSC-injected segments had concordant reduction in scar size, perfusion and contractile improvement (concordant score: 2.93±0.07), whereas revascularized (0.5±0.21) and non-treated segments (-0.07±0.34) demonstrated non-concordant changes (p<0.0001 vs. injected segments). Conclusions Intramyocardial injection of autologous MSCs into akinetic yet non-revascularized segments produces comprehensive regional functional restitution, which in turn drives improvement in global LV function. These findings, although inconclusive due to lack of placebo group, have important therapeutic and mechanistic hypothesis-generating implications.