Durable scar size reduction due to allogeneic mesenchymal stem cell therapy regulates whole-chamber remodeling

Abstract

Background: Intramyocardial injection of mesenchymal stem cells (MSCs) in chronic ischemic cardiomyopathy is associated with reverse remodeling in experimental models and humans. Here, we tested the hypothesis that allogeneic MSC therapy drives ventricular remodeling by producing durable and progressive scar size reduction in ischemic cardiomyopathy.

Methods and results: Gottingen swine (n=12) underwent left anterior descending coronary artery myocardial infarction (MI), and 3 months post-MI animals received either intramyocardial allogeneic MSC injection (200 mol/L cells; n=6) or left ventricle (LV) catheterization without injection (n=6). Swine were followed with serial cardiac magnetic resonance imaging for 9 months to assess structural and functional changes of the LV. Intramyocardial injection was performed using an integrated imaging platform combining electroanatomical mapping unipolar voltage and 3-dimensional cardiac magnetic resonance imaging angiography-derived anatomy to accurately target infarct border zone injections. MSC-treated animals had a 19.62 ± 2.86% reduction in scar size at 3 months postinjection, which progressed to 28.09 ± 2.31% from 3 to 6 months postinjection (P<0.0001). MSC-treated animals had unchanged end-diastolic volume (EDV; P=0.08) and end-systolic volume (ESV; P=0.28) from preinjection to 6 months postinjection, whereas controls had progressive dilatation in both EDV (P=0.0002) and ESV (P=0.0002). In addition, MSC-treated animals had improved LV sphericity index. Percentage change in infarct size correlated with percentage change in EDV (r=0.68; P=0.01) and ESV (r=0.77; P=0.001). Ejection fraction increased from 29.69 ± 1.68% to 35.85 ± 2.74% at 3 months post-MSC injection and progressed to 39.02 ± 2.42% 6 months postinjection (P=0.0001), whereas controls had a persistently depressed ejection fraction during follow-up (P=0.33).

Conclusion: Intramyocardial injection of allogeneic MSCs leads to a sustained and progressive reduction in infarct size, which in turn drives reverse remodeling and increases in ejection fraction. These findings support ongoing biological activity of cell therapy for substantial periods and suggest optimal end points for future clinical trials.

Keywords: cardiac magnetic resonance imaging; ischemic cardiomyopathy; stem cell therapy.

Figure 1.

Merging CMR angiography and electroanatomical mapping to guide transendocardial mesenchymal stem cell injection. A, CMR angiography–derived 3‐dimensional reconstruction of the heart and great vessels. B, Segmented CMR LV reconstruction with corresponding (C) EAM unipolar violated LV map. D, Merged CMR‐EAM image showing EAM unipolar voltage projected onto a detailed CMR angiography reconstruction of the LV. Bull's‐eye 17‐segment plot of corresponding (E) EAM unipolar voltage–derived infarct (red indicates scar with depressed voltage) and (F) delayed‐enhancement CMR‐derived scar (darker areas indicate infarct). G, EAM unipolar voltages for scar tissue compared with normal myocardium. H, EAM creates larger LV volumes compared with CMR, which (I) correlated with the degree of LV dysfunction. CMR indicates cardiac magnetic resonance imaging; LIMA, left internal mammary artery; RVOT, right ventricle outflow tract; LV, left ventricle; Ao, aorta; EAM, electroanatomic map; EF, ejection fraction; EDV, end‐diastolic volume; LA, left atrium.

Figure 2.

Durable and progressive scar size reduction due to intramyocardial injection of allogeneic bone marrow MSCs. A, Delayed‐enhancement CMR scar size images show durable reduction in scar size with MSC therapy. B, Control animals have stable infarct size in ischemic cardiomyopathy. C, Infarct size was reduced with MSC therapy compared with control when measured by absolute scar volume. D, Scar size reduction with MSC therapy was durable and progressive from 3 to 6 months postinjection as measured by infarct size as a percentage of LV mass. (*P<0.0001,** P<0.001 vs 3 months, ***P<0.05, †P<0.01 MSC vs placebo, ††P<0.05 at 3 vs 6 months; between‐group ANOVA interaction P value is plotted in each graph). MSCs indicates mesenchymal stem cells; MI, myocardial infarction; LV, left ventricle; CMR, cardiac magnetic resonance imaging; ANOVA, analysis of variance.

Figure 3.

Allogeneic MSC therapy reverses remodeling in ischemic cardiomyopathy. Example of a swine heart (A) 6 months postinjection of allogeneic MSCs compared with a (B) no‐injection control (note: both hearts had comparable EDV, ESV, and infarct size 3 months post‐MI). The control heart demonstrated progressive LV dilatation, whereas the MSC‐treated heart had significantly smaller LV chamber volumes accompanied by reduced infarct size. Graphs showing percentage change in (C) EDV and (D) ESV. No injection control animals had progressive dilatation of both EDV and ESV from 3 months (preinjection point) to 9 months, whereas the MSC‐treated group had stable LV volumes. (*P<0.001, **P<0.05 at 3 vs 6 months and 6 vs 9 months, †P<0.01 at 6 vs 9 months; between‐group ANOVA interaction P value is plotted in each graph). MSC indicates mesenchymal stem cells; MI, myocardial infarction; EDV, end‐diastolic volume; ESV, end‐systolic volume; LV, left ventricle; ANOVA, analysis of variance.

Figure 4.

Allogeneic MSC therapy improves LV sphericity index. A, Formula for calculating the LV sphericity index (SI). B and C, CMR 4‐chamber diastolic SI images showing the LV chamber changing from the spherical shape of heart failure toward a more normal elliptical configuration after MSC therapy. D and E, No‐injection control animal's heart has worsening sphericiy (increased SI indicates a more spherical chamber, whereas a smaller SI models a more elliptical shape). Graphs of (F) diastolic SI and (G) systolic SI, showing improved sphericity after allogeneic MSC therapy. (*P=0.11, **P<0.01, †P<0.05 at 3 vs 9 months; between‐group ANOVA interaction P value is plotted in each graph). RV indicates right ventricle; LV, left ventricle; EDV, end‐diastolic volume; ESV, end‐systolic volume; MI, myocardial infarction; MSC, mesenchymal stem cells; CMR, cardiac magnetic resonance imaging; ANOVA, analysis of variance.

Figure 5.

Infarct size reduction determines degree of left ventricle remodeling with allogenic MSC therapy. Example of (A) MSC‐treated heart showing a smaller LV chamber and scar (white tissue) compared with a (B) no‐injection control with a larger chamber and infarct. Correlation between the percentage change in infarct size from 3 months post‐MI (preinjection) to 9 months post‐MI (6 months postinjection) with the percentage change in (C) EDV and (D) ESV suggests that infarct size reduction with MSC therapy drives reverse remodeling in ischemic cardiomyopathy. MSC indicates mesenchymal stem cells; LV, left ventricle; MI, myocardial infarction; EDV, end‐diastolic volume; ESV, end‐systolic volume.

Figure 6.

Progressive improvement in LV function with allogeneic MSC therapy. A, MSC therapy showed improved LVEF at 3 months postinjection, which progressively increased at 6 months postinjection. (*P<0.0001, **P<0.01 at 3 vs 6 months, †P<0.05 at 6 vs 9 months; between‐group ANOVA interaction P value plotted in the graph). LV indicates left ventricle; MSC, mesenchymal stem cells; LVEF, left ventricular ejection fraction; MI, myocardial infarction; EF, ejection fraction.