ORIGINAL ARTICLE

Cell Therapy & Organ Transplantation

http://dx.doi.org/10.3346/jkms.2014.29.1.23 • J Korean Med Sci 2014; 29: 23-31

A Randomized, Open-Label, Multicenter Trial for the Safety and

Efficacy of Adult Mesenchymal Stem Cells after Acute Myocardial
Infarction
Jun-Won Lee,1 Seung-Hwan Lee,1 Recent studies suggest that the intracoronary administration of bone marrow (BM)-derived
Young-Jin Youn,1 Min-Soo Ahn,1 mesenchymal stem cells (MSCs) may improve left ventricular function in patients with
Jang-Young Kim,1 Byung-Su Yoo,1 acute myocardial infarction (AMI). However, there is still argumentative for the safety and
Junghan Yoon,1 Woocheol Kwon,2 efficacy of MSCs in the AMI setting. We thus performed a randomized pilot study to
In-Soo Hong,2 Kyounghoon Lee,3 investigate the safety and efficacy of MSCs in patients with AMI. Eighty patients with AMI
Jun Kwan,4 Keum Soo Park,4 after successful reperfusion therapy were randomly assigned and received an intracoronary
Donghoon Choi,5 Yang Soo Jang,5 administration of autologous BM-derived MSCs into the infarct related artery at 1 month.
and Mun K. Hong6 During follow-up period, 58 patients completed the trial. The primary endpoint was
1
Division of Cardiology, and 2Department of
changes in left ventricular ejection fraction (LVEF) by single-photon emission computed
Radiology, Yonsei University Wonju College of tomography (SPECT) at 6 month. We also evaluated treatment-related adverse events. The
Medicine, Wonju; 3Department of Cardiology, absolute improvement in the LVEF by SPECT at 6 month was greater in the BM-derived
Gachon Medical School, Gil Medical Center, MSCs group than in the control group (5.9% ± 8.5% vs 1.6% ± 7.0%; P = 0.037). There
Incheon; 4Division of Cardiology, Department of
Internal Medicine, Inha University Hospital, Incheon;
was no treatment-related toxicity during intracoronary administration of MSCs. No
5
Yonsei Cardiovascular Center and Cardiovascular significant adverse cardiovascular events occurred during follow-up. In conclusion, the
Research Institute, Yonsei University College of intracoronary infusion of human BM-derived MSCs at 1 month is tolerable and safe with
Medicine, Seoul, Korea; 6Division of Cardiology, modest improvement in LVEF at 6-month follow-up by SPECT. (ClinicalTrials.gov
Department of Medicine, St. Luke’s Roosevelt
Hospital, Columbia University College of Physicians
registration number: NCT01392105)
and Surgeons, New York, NY, USA
Keywords: Mesenchymal Stem Cells; Myocardial Infarction; Ventricular Dysfunction, Left
Received: 9 July 2013
Accepted: 6 November 2013

Address for Correspondence: INTRODUCTION

Seung-Hwan Lee, MD
Division of Cardiology, Department of Internal Medicine, Yonsei
University Wonju College of Medicine, 20 Ilsan-ro, Remarkable advances of early reperfusion therapy in acute myocardial infarction (AMI)
Wonju 220-701, Korea
Tel: +82.33-741-0907, Fax: +82.33-741-1219 have contributed to a reduction of early mortality as well as complications of post-AMI
E-mail: carshlee@yonsei.ac.kr
(1-3). Nevertheless, delayed treatment leads to subsequent loss of cardiomyocyte and
This study was funded by FCB-Pharmicell Company Limited heart failure, which is a major cause of long term morbidity and mortality. In this re-
(Seongnam, Korea).
spect, stem cell therapy has emerged as a novel alternative option for repairing the da­
maged myocardium (4).
The type and time of administration of stem cells are important issues. First, bone
marrow (BM)-derived mesenchymal stem cells (MSCs) are considered to be an attrac-
tive candidate because of high replicability, paracrine effect, ability to preserve poten-
cy, and no adverse reactions to allogeneic transplants (5, 6). However, the practical use
of MSCs is limited because of time-consuming processes, expensive cost, need for
strict control of infection and so on. Second, most studies were performed around 1
week after AMI with autologous bone marrow-derived progenitor cells (BMCs) (7-12).
Although STAR-Heart study showed beneficial effects of BMCs in patients with chronic
heart failure (13), there is little evidence of best time to treat AMI with stem cells (14).
Assmus et al. demonstrated that the contamination of isolated BMCs with red blood
cells reduced the function of BMCs and the recovery of left ventricular ejection fraction
(LVEF) (15). However, purified MSCs can be expanded from BM and have no concern
about contamination. Therefore, we hypothesized that treatment with purified BM-de-
rived MSCs would be effective in patients with AMI despite of delayed administration.

© 2014 The Korean Academy of Medical Sciences. pISSN 1011-8934

This is an Open Access article distributed under the terms of the Creative Commons Attribution Non-Commercial License (http://creativecommons.org/licenses/by-nc/3.0)
which permits unrestricted non-commercial use, distribution, and reproduction in any medium, provided the original work is properly cited. eISSN 1598-6357
 Lee J-W, et al. • SEED-MSC Trial

We designed a randomized, multicenter, pilot study to deter- ratio to the MSCs group or control group. Control group receiv­
mine whether intracoronary infusion of autologous BM-derived ed optimal medical therapy alone.
MSCs at 1 month is safe and effective in patients with AMI.
Preparation of autologous MSCs
MATERIALS AND METHODS Twenty to twenty-five milliliters (mean ± SD: 23.1 ± 11.5 mL) of
BM aspirates were obtained under local anesthesia from the
Study design and population posterior iliac crest in the MSCs group on 3.8 ± 1.5 days after
From March 2007 to September 2010, total 80 patients were en- admission. All manufacturing and product testing procedures
rolled from three tertiary hospitals in Korea. Patients were eligi- for the generation of clinical-grade autologous MSCs were car-
ble if 1) they were aged 18-70 yr; 2) they had ischemic chest ried out under good manufacturing practice (FCB-Pharmicell
pain for > 30 min; 3) they were admitted to hospital < 24 hr af- Company Limited, Seongnam, Korea). Mononuclear cells were
ter the onset of chest pain; 4) electrocardiography (ECG) showed separated from the BM by density gradient centrifugation (HIS-
ST segment elevation > 1 mm in two consecutive leads in the TOPAQUE-1077; Sigma-Aldrich, St. Louis, MO, USA) and wash­
limb leads or > 2 mm in the precordial leads; and 5) they could ed with phosphate-buffered saline (PBS). Cells were resuspend-
be enrolled in the study < 72 hr after successful revasculariza- ed in Dulbecco’s modified Eagle’s medium-low glucose (DMEM;
tion (defined as residual stenosis < 30% of the infarct-related Gibco, Grand Island, NY, USA) containing 10% fetal bovine se-
artery [IRA]). rum (Gibco), 100 U/mL penicillin/100 μg/mL and streptomy-
We excluded patients with cardiogenic shock, life-threaten- cin (Gibco). They were plated at 2-3 × 105 cells/cm2 into 75 cm2
ing arrhythmia, advanced renal or hepatic dysfunction, history flasks. Cultures were maintained at 37°C in a humidified atmo-
of previous coronary artery bypass graft, history of hematologic sphere containing 5% CO2. After 5-7 days, non-adherent cells
disease and malignancy, major bleeding requiring blood trans- were removed by replacing the medium; adherent cells were
fusion, stroke or transient ischemic attack in the previous 6 mon­ cultured for another 2-3 days. When the cultures were near con-
ths, use of corticosteroids or antibiotics during the previous fluence (70%-80%), adherent cells were detached by using tryp-
month, major surgical procedure in the previous 3 months, car- sin containing ethylene diamine tetra-acetic acid (EDTA; Gib-
diopulmonary resuscitation for > 10 min within the previous 2 co) and replated at 4-5 × 103 cells/cm2 in 175 cm2 flasks. Cells
weeks, positive skin test for penicillin, positive result for viral were serially subcultured up to passage 4 or passage 5 for infu-
markers (human immunodeficiency virus [HIV], hepatitis B vi- sion (mean ± SD: 4.4 ± 0.5 passages).
rus [HBV], hepatitis C virus [HCV] and Venereal Disease Re- On the day of administration, MSCs were harvested using
search Laboratory [VDRL] test), pregnant woman and possible trypsin and EDTA, washed twice with PBS and once with saline
candidate for pregnancy. solution, and resuspended to a final concentration of 1 × 106
cells/kg. The criteria for the release of MSCs for clinical use in-
Primary care and randomization cluded viability > 80%, absence of microbial contamination
All patients were required to have successful revascularization (bacteria, fungus, virus, and mycoplasma) if undertaken 3-4
of an IRA on coronary angiography at the time of randomiza- days before administration, and expression of CD73 and CD105
tion. All patients received aspirin (300 mg loading dose, then by > 90% of cells and absence of CD14, CD34, and CD45 by
100 mg daily) and clopidogrel (600 mg loading dose, then 75 < 3% of cells as assessed by flow cytometry (data not shown).
mg daily) with optimal medical therapy according to the Amer- Also, the in vitro osteogenic and cardiomyogenic differentiation
ican College of Cardiology (ACC)/American Heart Association potential of MSCs in passage 0 or 1 was tested before release as
(AHA) guidelines for treatment of ST-segment elevation myo- a potency test. Alkaline phosphatase staining was used to dem-
cardial infarction (STEMI) (16-18), including aspirin, clopido- onstrate the osteogenic differentiation. Immunostaining with
grel, beta blocker, angiotensin-converting enzyme (ACE) inhib- α-sarcomeric actin and troponin T was used to demonstrate the
itor (or angiotensin-receptor blocker) and statin unless these cardiomyogenic differentiation. Qualitative analysis showed
drugs were contraindicated. The use of aspiration thrombecto- well differentiation potential of all MSCs.
my or a glycoprotein IIb/IIIa inhibitor during percutaneous
coronary intervention (PCI) was left to the investigator’s discre- Cell injection
tion. If primary PCI was not available, a thrombolytic agent was Injection of MSCs has been described elsewhere (11). The final
used to reperfuse the occluded artery. We performed rescue preparation of MSCs (7.2 ± 0.90 × 107 cells) contained into ster-
PCI when ST-segment resolution was < 50% at follow-up elec- ilized syringe was gently transferred and mixed to infusion sy-
trocardiography 90 min after thrombolytic therapy. Patients ringe to minimize cell aggregation and then infused into the
who were successfully reperfused with thrombolytic agents un- IRA via the central lumen of an over-the-wire balloon catheter
derwent elective PCI. Patients were randomly allocated in a 1:1 (Maverick®, Boston Scientific, Natick, MA, USA). To allow the

24  
http://jkms.org http://dx.doi.org/10.3346/jkms.2014.29.1.23
Lee J-W, et al. • SEED-MSC Trial

maximum contact time of MSCs with the microcirculation of System, Tel Aviv, Israel) equipped with a low-energy, high-reso-
the IRA, the balloon was inflated inside the stent at a low pres- lution collimator. Sixty-four images were obtained over a 180°
sure to transiently interrupt antegrade blood flow during infu- orbit using 90° between the heads. Acquisitions were attenua-
sions. The entire cell injection was done during three transient tion-corrected and gated for 16 frames/cardiac cycle. Total ac-
occlusions, each lasting 2 to 3 min. Between occlusions, the quisition time was ~20 min. Vendor-specific, computer-en-
coronary artery was reperfused for 3 min. After cell injection, hanced edge detection methods were used to assess the LV
repeated coronary angiography was undertaken to identify an- epicardial and endocardial margins during the entire cardiac
tegrade flow and the absence of other possible complications. cycle. The computer calculated resting global LVEF from the
Measurements of cardiac enzymes and electrocardiography gated SPECT images using an automated algorithm (20). The
were repeated to assess periprocedural myocardial infarction analysis of SPECT images was performed by blinded indepen-
(MI). The mean duration of cultured MSCs from BM aspiration dent investigators at each participating center.
to intracoronary injection was 25.0 ± 2.4 days. Regional and global LV function were measured by two-di-
mensional echocardiography according to the recommenda-
Follow-up visit and endpoints tions of the American Society of Echocardiography (21). LVEF
Study visits were scheduled at 1, 2, and 6 months after hospital was measured from the end-diastolic and end-systolic volumes
admission for the clinical and functional evaluation. Coronary calculated by the Simpson method from two orthogonal apical
angiography, electrocardiogram-gated single-photon emission views. LV regional wall motion analyses were based on grading
computed tomography (SPECT) and echocardiography were the contractility of individual segments. The left ventricle was
done at baseline and 6 months. Twenty-four hour ambulatory divided into three levels (basal, mid, apical) and 16 segments.
ECG (Holter) monitoring was done at baseline, 1 month and 6 The basal and mid-levels were subdivided into six segments,
months. and the apical level subdivided into four segments. Numerical
The primary endpoint of the study was absolute changes in scoring was adopted on the basis of the contractility of the indi-
global LVEF from baseline to 6 months after the MSCs adminis- vidual segments. In this scoring system, higher scores indicated
tration measured by SPECT. Secondary endpoints were chang- more severe abnormality in the motion of the wall: 1) normoki-
es in left ventricular end-diastolic volume (LVEDV), left ventric- nesis, 2) hypokinesis, 3) akinesis, 4) dyskinesis, and 5) aneurysm.
ular end-systolic volume (LVESV), regional wall motion score The WMSI was derived by dividing the sum of wall motion score
index (WMSI) and major adverse cardiac events (MACE). MACE by the number of visualized segments; a normal WMSI was 1.
was defined as the composites of any cause of death, myocardi- Off-line assessment of all echocardiographic images was per-
al infarction, revascularization of the target vessel, re-hospital- formed by one blinded independent investigator.
ization for heart failure, and life-threatening arrhythmia. MI
was defined following the consensus statement of the Joint Eu- Sample size and statistical analyses
ropean Society of Cardiology (ESC)/American College of Cardi- Sample size calculation is based on the result of BOOST trial (7).
ology (ACC)/American Heart Association (AHA)/World Heart Study hypothesis is to demonstrate the superiority of MSCs treat-
Federation (WHF) Task Force for the Redefinition of Myocardial ment compared with control group. Type I and II error is set to
Infarction for clinical trials on coronary intervention (19). Hence, 0.05 and 0.20 (statistical power 80%). The changes of LVEF and
periprocedural MI was defined as the levels of cardiac biomark- standard deviation are 6.7% ± 6.5% in BMCs group and 0.7% ±
ers (troponin or creatine kinase-MB [CK-MB]) > 3 times the 8.1% in control group. Based on the assumption of 6% differen­
99th percentile of upper limit of normal (ULN) in patients with ces of LVEF and 1:1 allocation ratio with 27% drop-out rate, to-
normal baseline levels, and as a subsequent elevation > 3 times tal 80 patients (40 patients in each group) are necessary.
in CK-MB or troponin in patients with raised baseline levels. Continuous variables are presented as mean ± standard de-
Target-vessel revascularization (TVR) included bypass surgery viation. Categorical data are presented as frequencies and per-
or repeat PCI of the target vessel(s). centages. Comparisons of continuous variables at baseline with
those at follow-up were done with the paired t-test. Compari-
Assessment of left ventricular (LV) function son of non-parametric data between groups was undertaken
SPECT was used for the non-invasive measurement of LVEF. A using the Wilcoxon rank sum test and the Mann-Whitney test.
single dose of technetium-99m (99mTc)-sestamibi (Cardiolite® Statistical significance was set at P < 0.05. Data were analyzed
kit for the preparation of Technetium-99m Sestamibi for Injec- using SPSS for Windows ver. 15 (SPSS, Chicago, IL, USA).
tion; Dupont Merck Pharmaceutical Company, Billerica, MA,
USA) was administered intravenously at rest, and data acquisi- Ethics statement
tion started 30-60 min later. SPECT data were acquired with a This study (SEED-MSC) was a randomized, open-label, multi-
dual-headed gamma camera (Infinia H3000WT; GE Medical center phase-II/III clinical trial, which was approved by the Ko-

http://dx.doi.org/10.3346/jkms.2014.29.1.23 http://jkms.org  25
 Lee J-W, et al. • SEED-MSC Trial

rean Food and Drug Administration (KFDA) and registered with significant differences. The changes of LVEDV and LVESV also
clinicaltrials.gov, number NCT01392105. The institutional re- did not significantly different at 6-month follow-up in either
view board of each participating center approved the treatment group. There were no significant differences with regard to WMSI
protocol before the initiation of enrollment. All patients provid- and change in WMSI.
ed written informed consent for inclusion in the trial.
LV function as revealed by echocardiography
RESULTS Baseline LVEF was similar in the MSCs group and the control
group (48.1% ± 8.0% and 51.0% ± 9.2%, respectively, P = 0.215).
Study participants and baseline characteristics Echocardiographic evaluation revealed a significant increase in
Eighty patients were screened and 69 patients (86.3%) were in- LVEF from baseline to 6 months in the MSCs group but not in
cluded and randomly assigned to the MSCs group (n = 33) or the control group (1.9% ± 2.7% and 0.5% ± 1.8%, P < 0.001). Vol-
control group (n = 36). After enrollment, 11 patients were ex- umetric analyses of LV end-diastole and end-systole at baseline
cluded for the reasons listed in Fig. 1. The main cause of exclu- and 6 months and the changes at 6 months showed no signifi-
sion during follow-up was poor image quality. Two patients cant differences between groups.
were excluded because of long-term medication of prohibited
drug (corticosteroid). Subgroup analyses according to time interval
Table 1 illustrated that the two groups of patients were well We analyzed the subgroup population treated < 6 hr from symp-
matched. There were no differences with respect to cardiovas- tom onset to first balloon inflation. Twenty-one patients of the
cular risk factors and medical treatments. The Killip classifica- stem cell group and 20 patients of the control group were ana-
tion and angiographic characteristics including IRA were also lyzed. The improvement in LVEF was more significant in the
similar between two groups. MSCs group than in the control group according to SPECT (8.3%
Primary PCI was carried out in most cases, except 7 patients ± 8.3% and 1.3% ± 7.5%, P = 0.007) and echocardiography (2.0%
treated with thrombolytic agents (Table 2). Rescue PCI was done ± 2.8% and -0.3% ± 1.5%, P = 0.003).
in 1 out of 7 patients because of reperfusion failure. There were
no significant differences in procedural characteristics and time Safety and clinical outcomes
intervals from chest pain onset to treatment (Table 3). All procedures related to the BM aspiration and stem cell trans-
plantation were well tolerated. There were no serious inflam-
Quantitative analyses of LV function by SPECT matory reactions or bleeding complications at the site of iliac
Baseline LVEF was similar between the two groups (49.0% ± 11.7% puncture after BM aspiration. Patients had no or mild angina
in the MSCs group, and 52.3% ± 9.3% in the control group, P = during balloon inflation for infusion of MSCs. There were no
0.247) (Table 4). The absolute change in global LVEF from base- serious procedural complications related to intracoronary ad-
line to 6 months was significantly improved in the MSCs group ministration of MSCs, such as ventricular arrhythmias, throm-
than the control group. (5.9% ± 8.5% vs 1.6% ± 7.0%, P = 0.037) bus formation or dissection. Periprocedural MI was occurred in
(Fig. 2). Baseline and 6 months LVEDV and LVESV showed no 2 patients. The peak levels of CK-MB and troponin I were 2.38

Patients with ST segment elevation myocardial infarction

Exclusion (n = 7) 80 patients (1:1 ratio) Exclusion (n = 4)

Screening and Randomization
Protocol violation 2 Loss to follow-up 2
Study refusal 2 Study refusal 1
Opinion of investigator 3 Opinion of investigator 1
Stem cell group Control group
n = 33 n = 36

Follow-up after 1, 2, and 6 months

Exclusion (n = 3) Exclusion (n = 8)

Protocol violation 1 Protocol violation 2

Poor data image 1 Poor data image 5
Administration of prohibited drug 1 Final analysis Final analysis Administration of prohibited drug 1
n = 30 n = 28

Fig. 1. Study design.

26  
http://jkms.org http://dx.doi.org/10.3346/jkms.2014.29.1.23
Lee J-W, et al. • SEED-MSC Trial

Table 1. Baseline characteristics between the MSCs group and control group Table 2. Procedural characteristics between the MSCs group and control group
MSCs group Control group MSCs group Control group
Characteristics P value Procedures P value
(n = 30) (n = 28) (n = 30) (n = 28)
Age (yr) 53.9 ± 10.5 54.2 ± 7.7 0.920 Primary PCI 26 (86.7) 25 (89.3) 0.621
Male 27 (90.0) 25 (89.3) 0.929 PCI strategy for multivessel disease 0.716
Body mass index (kg/m2) 25.7 ± 2.6 24.9 ± 2.8 0.259 Culprit only 6 (42.9) 6 (50.0)
Risk factors Complete revascularization 8 (57.1) 6 (50.0)
Hypertension 14 (46.7) 12 (42.9) 0.798 Ad-hoc 7 (87.5) 3 (50.0)
Diabetes mellitus 5 (16.7) 8 (28.6) 0.352 Staged 1 (12.5) 3 (50.0)
Cerebrovascular accident 1 (3.3) 1 (3.6) 0.960 Stent type 0.296
Smoking 19 (63.3) 20 (71.4) 0.583 Drug eluting stent 30 (100) 27 (96.4)
Killip classification 0.191 1st generation 9 (30.0) 10 (37.0)
Killip I 28 (93.3) 23 (82.1) 2nd generation 21 (70.0) 17 (63.0)
Killip II 2 (6.7) 5 (17.9) Bare metal stent 0 (0) 1 (3.6)
Coronary artery disease 0.760 TIMI flow grade
1 vessel 16 (53.3) 16 (57.1) Before PCI 0.405
2 vessel 11 (36.7) 8 (28.6) 0 19 (63.3) 19 (67.9)
3 vessel 3 (10.0) 4 (14.3) 1 2 (6.7) 3 (10.7)
Location 0.694 2 5 (16.7) 1 (3.6)
Left anterior descending artery 22 (73.3) 18 (64.3) 3 4 (13.3) 5 (17.9)
Left circumflex artery 2 (6.7) 2 (7.1) After PCI 0.513
Right coronary artery 6 (20.0) 8 (28.6) 0 and 1 0 (0) 0 (0)
2 1 (3.3) 2 (7.1)
Medications
3 29 (96.7) 26 (92.9)
Aspirin 30 (100) 28 (100) > 0.999
Clopidogrel 30 (100) 28 (100) > 0.999 Infarct related artery
Cilostazol 7 (23.3) 5 (17.9) 0.748 Stent number 1.4 ± 0.6 1.6 ± 0.7 0.212
Beta blocker 28 (93.3) 26 (92.9) 0.943 Stent diameter (mm) 3.08 ± 0.50 3.17 ± 0.39 0.451
ACEi or ARB 27 (90) 26 (92.9) 0.698 Stent length (mm) 31.9 ± 12.9 41.8 ± 19.3 0.028
Statin 27 (90) 25 (89.3) 0.929 Non-infarct related artery
Statin type 0.144 Stent number 1.0 ± 0.0 1.7 ± 0.8 0.102
Atorvastatin 16 (59.3) 20 (80.0) Stent diameter (mm) 3.25 ± 0.38 3.03 ± 0.49 0.356
Rosuvastatin 3 (11.1) 2 (8.0) Stent length (mm) 23.0 ± 6.0 30.5 ± 21.0 0.431
Pitavastatin 7 (25.9) 1 (4.0) Thrombolytic agent before PCI 4 (13.3) 3 (10.7) 0.760
Others 1 (3.7) 2 (8.0) Aspiration thrombectomy 13 (43.3) 14 (50.0) 0.793
Atorvastatin equivalent dose (mg) 0.846 Glycoprotein IIb/IIIa inhibitor 3 (10) 2 (7.1) 0.698
10 13 (48.1) 11 (44.0)
Values are expressed as mean ± SD or number of patients (%). MSC, mesenchymal
20 11 (40.7) 12 (48.0)
stem cell; PCI, percutaneous coronary intervention; TIMI, thrombolysis In Myocardial
40 3 (11.1) 2 (8.0)
Infarction.
Vital signs
Initial systolic BP (mmHg) 135.9 ± 31.3 141.4 ± 28.6 0.490
Initial diastolic BP (mmHg) 85.2 ± 20.8 87.0 ± 18.8 0.732 Table 3. Time intervals (minutes) from symptom to treatment
Initial pulse rate (beat per minute) 81.0 ± 14.3 79.3 ± 15.2 0.646
Intervals No. MSCs group No. Control group P value
Symptom to door time (hr) 0.562
≤2 14 (46.7) 11 (39.3) Symptom to door 30 229.5 ± 260.3 28 188.1 ± 136.1 0.457
2-6 11 (36.7) 14 (50.0) Door to balloon 26 63.7 ± 24.0 25 64.8 ± 27.9 0.875
>6 5 (16.7) 3 (10.7) Symptom to balloon 26 279.4 ± 253.7 25 261.7 ± 133.1 0.758
Symptom to balloon time (hr) 0.775 Door to needle 4 27.0 ± 2.9 3 32.0 ± 2.6 0.069
≤2 5 (19.2) 3 (12.0) Symptom to needle 4 350.8 ± 325.4 3 115.3 ± 35.5 0.277
2-6 16 (61.5) 17 (68.0)
>6 5 (19.2) 5 (20.0) MSC, mesenchymal stem cell.
Symptom to initial SPECT (day) 5.1 ± 2.3 4.8 ± 2.0 0.578
Symptom to follow-up SPECT (day) 185.3 ± 7.5 182.4 ± 9.2 0.184
Symptom to initial Echo (day) 1.7 ± 0.8 1.7 ± 0.9 0.815 There were no deaths, MI, TVR, stent thrombosis, life-threat-
Symptom to follow-up Echo (day) 182.6 ± 6.0 179.5 ± 7.4 0.084 ening arrhythmia or stroke in both groups during 6-month fol-
Values are expressed as mean ± SD or number of patients (%). MSC, mesenchymal low-up. No significant arrhythmic events were recorded on 24
stem cell; ACEi, Angiotensin-converting enzyme inhibitor; ARB, angiotensin receptor hr ambulatory ECG (Holter) monitoring. Paroxysmal non-sus-
blocker; BP, blood pressure; SPECT, single-photon emission computed tomography;
tained atrial fibrillation was found in 2 patients of the MSCs
Echo, echocardiography.
group (1 patient after PCI and 1 patient at 1 month follow-up)
ng/mL (reference range < 5 ng/mL) and 2.848 ng/mL (refer- and in 1 patient of the control group after PCI.
ence range < 0.078 ng/mL) in one patient and 38.36 ng/mL and
5.304 ng/mL in the other. However, they had no symptoms and DISCUSSION
spontaneously recovered without additional treatment during
6-month follow-up. In our study, the main finding is that the intracoronary admin-

http://dx.doi.org/10.3346/jkms.2014.29.1.23 http://jkms.org  27
 Lee J-W, et al. • SEED-MSC Trial

Table 4. Ejection fraction and left ventricular volume as determined by SPECT and 90 MSCs group
echocardiography
MSCs group Control group 80
Measurements P value
(n = 30) (n = 28)
SPECT 70
Global LVEF (%)
Baseline 49.0 ± 11.7 52.3 ± 9.3 0.247
6 months 55.0 ± 11.8 53.9 ± 10.2 0.704 60
LVEDV (mL)
Baseline 115.4 ± 41.8 106.2 ± 31.4 0.349
50

LVEF (%)
6 months 106.4 ± 43.3 100.0 ± 33.9 0.537
LVESV (mL)
Baseline 62.2 ± 36.2 52.6 ± 22.0 0.232 40
6 months 51.6 ± 33.5 48.5 ± 24.1 0.696
Echocardiography
30
Global LVEF (%)
Baseline 48.1 ± 8.0 51.0 ± 9.2 0.215
6 months 50.0 ± 8.4 50.4 ± 9.4 0.862 20
LVEDV (mL)
Baseline 84.7 ± 18.5 78.5 ± 15.9 0.178
6 months 88.9 ± 27.5 81.7 ± 23.4 0.294 10
Mean LVEF change 5.9% ± 8.5%
LVESV (mL)
Baseline 44.5 ± 14.2 39.2 ± 12.6 0.134 0
6 months 45.7 ± 19.6 41.4 ± 16.5 0.368 Baseline 6 months
WMSI
Baseline 1.53 ± 0.27 1.53 ± 0.37 0.944 90 Control group
6 months 1.46 ± 0.39 1.42 ± 0.28 0.658
Changes at 6 months
SPECT 80
LVEF (%) 5.9 ± 8.5 1.6 ± 7.0 0.037
LVEDV (mL) -9.0 ± 22.2 -6.2 ± 22.1 0.626 70
LVESV (mL) -10.6 ± 15.7 -4.1 ± 15.9 0.120
Echocardiography
LVEF (%) 1.9 ± 2.7 -0.5 ± 1.8 < 0.001 60
LVEDV (mL) 4.2 ± 23.3 3.3 ± 24.4 0.880
LVESV (mL) 1.2 ± 12.3 2.2 ± 13.3 0.764
50
LVEF (%)

WMSI -0.07 ± 0.30 -0.11 ± 0.25 0.404

Values are expressed as mean ± SD. MSC, mesenchymal stem cell; SPECT, single-
40
photon emission computed tomography; LVEF, left ventricular ejection fraction; LVEDV,
left ventricular end-diastolic volume; LVESV, left ventricular end-systolic volume; WMSI,
wall motion score index. 30

istration of autologous purified BM-derived MSCs at 1 month 20

after STEMI is tolerable without serious complications and pro-
vides modest improvement in LVEF at 6-month follow-up by 10
Mean LVEF change 1.6% ± 7.0%
SPECT.
0
The stem cell therapy for AMI increased LVEF by 2.99% (95%
Baseline 6 months
confidence interval [CI], 1.26%-4.72%, P = 0.0007) in meta-anal-
ysis (4). In chronic ischemic heart failure, STAR-Heart study Fig. 2. Impact of MSCs treatment on LVEF by SPECT. MSCs, mesenchymal stem cells;
LVEF, left ventricular ejection fraction; SPECT, single-photon emission computed to-
demonstrated that intracoronary BM cell therapy improved
mography.
ventricular performance, quality-of-life and even survival (13).
Our results met the primary endpoint of ≥ 4.3% improvement Strauer et al. (22, 23). Moreover, ischemic pre-conditioning in-
in LVEF compared with the control group. However, it is uncer- duced by transient balloon occlusion seems to be important to
tain whether the relatively small increase in systolic function is recruit MSCs into the infarcted myocardium (24, 25).
meaningful in real-life situations (and not just a statistical dif- Numerous in vitro and in vivo studies have shown the pleio-
ference). tropic effects of MSCs such as antifibrotic, immunomodulatory,
In terms of safety, intracoronary administration of MSCs antiapototic and proangiogenic features as well as the impact
showed no serious adverse events, although periprocedural MI of inflammation/cytokine expression on the different aspects
was developed in 2 patients. It seems to be a safe method to de- of homing, including chemokine-chemokine receptor interac-
liver stem cells via intracoronary route, since its introduction by tions, adhesion on endothelial cells, transendothelial migration,

28  
http://jkms.org http://dx.doi.org/10.3346/jkms.2014.29.1.23
Lee J-W, et al. • SEED-MSC Trial

SPECT could minimize inter-observer variability. Hovland et al.

and invasion through the extracellular matrix (26). Stromal cell- provided good evidence that SPECT showed the improvement
derived factor (SDF)-1α is a major chemotactic paracrine factor of inter-observer agreement compared with echocardiography
for homing stem cells. SDF-1α-modified MSCs enhanced the despite the possibility of an overestimation (32). Third, many
tolerance of engrafted MSCs to hypoxic injury in vitro and im- patients were excluded from this study. Total patients of exclu-
proved their viability in a rat model of infarcted hearts, thus sion were 22/80 (27.5%). Ten of 40 (25%) in treatment group
helping preserve the contractile function and attenuate LV re- and 12/40 (30%) in control group were excluded. It was larger
modeling (27). Moreover, MSC-conditioned medium directly than we had anticipated. After randomization, 11 of 69 (15.9%)
inhibited the function of cardiac fibroblasts and resulted in a patients were excluded and the main cause was poor image
decrease of myocardial fibrosis with the consequent improve- quality. More patients in control group (5/36, 13.9%) were ex-
ment of cardiac function by secreting antifibrotic factors such cluded than those in treatment group (1/33, 3.0%). However,
as adrenomedullin (28). there may be little possibility for selection bias, because radiol-
One of the main concerns is the time limitation for using au- ogists were blinded to treatment information. The selection of
tologous MSCs in acute setting. It is impossible to use autolo- inappropriate images was left at the discretion of radiologists.
gous MSCs immediately because it takes time to harvest and The final decision was confirmed by the agreement of investi-
culture cells for over 3 weeks. However, the optimal time for gators in each participating center. Fourth, we did not use di-
stem-cell therapy was not precisely identified. The temporal verse assessment tools such as the 6-min walking distance, ex-
window of opportunity to maximize efficacy seems to be be- ercise tolerance, pulmonary function test and quality of life. Fi-
tween the acute inflammatory response and scar formation. nally, we did not evaluate the inter- and intra-observer variabil-
Several experimental studies and clinical subgroup analyses ity in radiographic measurements. Sixth, there was no death in
provided important clues that stem cell therapy might be effec- either group. Our study group showed a relatively preserved LV
tive within the first month after AMI, but not in very acute phase systolic function and low risk profiles. Long-term follow-up is
(24 hr after AMI) (29). Further clinical randomized trials need needed to define the safety and beneficial effect of MSCs.
to confirm the optimal time to treatment. In conclusion, this pilot study was designed to identify the
The use of statin may ameliorate the microenvironment of safety and practical efficacy of intracoronary purified autolo-
injured myocardium and protect implanted MSCs. Animals gous BM-derived MSCs in patients with STEMI. Intracoronary
treated with atorvastatin showed the improvement of myocar- administration of autologous BM-derived MSCs at 1 month is
dial perfusion and contractility compared with untreated ani- tolerable and safe with modest improvement in LVEF at 6-month
mals by Yang et al. (30). The combined treatment of atorvastatin follow-up.
with MSCs reduced myocardial apoptosis, oxidative stress and
expression of the inflammatory cytokines (30). Nearly 90% of DISCLOSURE
our patients were prescribed statin at discharge. This synergic
effect of statin might contribute to the functional recovery of FCB-Pharmicell Company Limited (Seongnam, Korea) sup-
damaged myocardium in our study. ported this study but played no role in the study design, data
The total ischemic time from symptom to treatment is the collection, data analysis and interpretation, manuscript writing,
most important factor related to adverse outcomes. Mortality and decision to proceed with publication. The authors indicate
reduction was greatest in the first 2-3 hr after the onset of AMI, no potential conflicts of interest.
as a consequence of myocardial salvage (31). However, there
was no mortality benefit by opening the occluded artery after 6 REFERENCES
hr (31). Our subgroup analysis supports that BM-derived MSCs
may help damaged myocardium to recover, if treated within 6 1. Gibson CM, Pride YB, Frederick PD, Pollack CV Jr, Canto JG, Tiefenb-
runn AJ, Weaver WD, Lambrew CT, French WJ, Peterson ED, et al. Trends
hr after AMI.
in reperfusion strategies, door-to-needle and door-to-balloon times, and
Our study has several limitations. First, our study enrolled
in-hospital mortality among patients with ST-segment elevation myo-
relatively small number of patients. This limitation may attenu-
cardial infarction enrolled in the National Registry of Myocardial Infarc-
ate the efficacy of intracoronary purified MSCs. Further large-
tion from 1990 to 2006. Am Heart J 2008; 156: 1035-44.
scale randomized trials are needed. Second, there may be a 2. McManus DD, Gore J, Yarzebski J, Spencer F, Lessard D, Goldberg RJ.
technical problem for the assessment of LVEF. Although CMR Recent trends in the incidence, treatment, and outcomes of patients with
was considered to be the “gold standard” for evaluation of LV STEMI and NSTEMI. Am J Med 2011; 124: 40-7.
function, it was impossible to use CMR at all institutions for the 3. Roger VL, Go AS, Lloyd-Jones DM, Adams RJ, Berry JD, Brown TM, Car­
first time. We therefore measured LVEF with SPECT because nethon MR, Dai S, de Simone G, Ford ES, et al. Heart disease and stroke
we concluded that software-based automated analysis using statistics: 2011 update: a report from the American Heart Association.

http://dx.doi.org/10.3346/jkms.2014.29.1.23 http://jkms.org  29
 Lee J-W, et al. • SEED-MSC Trial

Circulation 2011; 123: e18-209. 17. Antman EM, Hand M, Armstrong PW, Bates ER, Green LA, Halasya-
4. Martin-Rendon E, Brunskill SJ, Hyde CJ, Stanworth SJ, Mathur A, Watt mani LK, Hochman JS, Krumholz HM, Lamas GA, Mullany CJ, et al.
SM. Autologous bone marrow stem cells to treat acute myocardial infarc- 2007 focused update of the ACC/AHA 2004 guidelines for the manage-
tion: a systematic review. Eur Heart J 2008; 29: 1807-18. ment of patients with ST-elevation myocardial infarction: a report of the
5. Parekkadan B, Milwid JM. Mesenchymal stem cells as therapeutics. Annu American College of Cardiology/American Heart Association Task Force
Rev Biomed Eng 2010; 12: 87-117. on Practice Guidelines. J Am Coll Cardiol 2008; 51: 210-47.
6. Hare JM, Traverse JH, Henry TD, Dib N, Strumpf RK, Schulman SP, Ger­ 18. Kushner FG, Hand M, Smith SC Jr, King SB 3rd, Anderson JL, Antman
stenblith G, DeMaria AN, Denktas AE, Gammon RS, et al. A random- EM, Bailey SR, Bates ER, Blankenship JC, Casey DE Jr, et al. 2009 focused
ized, double-blind, placebo-controlled, dose-escalation study of intrave- updates: ACC/AHA guidelines for the management of patients with ST-
nous adult human mesenchymal stem cells (prochymal) after acute myo­ elevation myocardial infarction (updating the 2004 guideline and 2007
cardial infarction. J Am Coll Cardiol 2009; 54: 2277-86. focused update) and ACC/AHA/SCAI guidelines on percutaneous coro-
7. Wollert KC, Meyer GP, Lotz J, Ringes-Lichtenberg S, Lippolt P, Breiden- nary intervention (updating the 2005 guideline and 2007 focused up-
bach C, Fichtner S, Korte T, Hornig B, Messinger D, et al. Intracoronary date) a report of the American College of Cardiology Foundation/Amer-
autologous bone-marrow cell transfer after myocardial infarction: the ican Heart Association Task Force on Practice Guidelines. J Am Coll Car-
BOOST randomised controlled clinical trial. Lancet 2004; 364: 141-8. diol 2009; 54: 2205-41.
8. Lunde K, Solheim S, Aakhus S, Arnesen H, Abdelnoor M, Egeland T, 19. Thygesen K, Alpert JS, White HD; Joint ESC/ACCF/AHA/WHF Task
Endresen K, Ilebekk A, Mangschau A, Fjeld JG, et al. Intracoronary in- Force for the Redefinition of Myocardial Infarction. Universal definition
jection of mononuclear bone marrow cells in acute myocardial infarc- of myocardial infarction. J Am Coll Cardiol 2007; 50: 2173-95.
tion. N Engl J Med 2006; 355: 1199-209. 20. Germano G, Kiat H, Kavanagh PB, Moriel M, Mazzanti M, Su HT, Van
9. Schächinger V, Erbs S, Elsässer A, Haberbosch W, Hambrecht R, Höls- Train KF, Berman DS. Automatic quantification of ejection fraction from
chermann H, Yu J, Corti R, Mathey DG, Hamm CW, et al. Intracoronary gated myocardial perfusion SPECT. J Nucl Med 1995; 36: 2138-47.
bone marrow-derived progenitor cells in acute myocardial infarction. N 21. Lang RM, Bierig M, Devereux RB, Flachskampf FA, Foster E, Pellikka
Engl J Med 2006; 355: 1210-21. PA, Picard MH, Roman MJ, Seward J, Shanewise JS, et al. Recommenda-
10. Meluzín J, Mayer J, Groch L, Janousek S, Hornácek I, Hlinomaz O, Kala P, tions for chamber quantification: a report from the American Society of
Panovský R, Prásek J, Kamínek M, et al. Autologous transplantation of Echocardiography’s Guidelines and Standards Committee and the Cham-
mononuclear bone marrow cells in patients with acute myocardial in- ber Quantification Writing Group, developed in conjunction with the
farction: the effect of the dose of transplanted cells on myocardial func- European Association of Echocardiography, a branch of the European
tion. Am Heart J 2006; 152: 975.e9-15. Society of Cardiology. J Am Soc Echocardiogr 2005; 18: 1440-63.
11. Janssens S, Dubois C, Bogaert J, Theunissen K, Deroose C, Desmet W, 22. Strauer BE, Brehm M, Zeus T, Köstering M, Hernandez A, Sorg RV, Kö-
Kalantzi M, Herbots L, Sinnaeve P, Dens J, et al. Autologous bone mar- gler G, Wernet P. Repair of infarcted myocardium by autologous intra-
row-derived stem-cell transfer in patients with ST-segment elevation myo- coronary mononuclear bone marrow cell transplantation in humans.
cardial infarction: double-blind, randomised controlled trial. Lancet Circulation 2002; 106: 1913-8.
2006; 367: 113-21. 23. Strauer BE, Brehm M, Zeus T, Gattermann N, Hernandez A, Sorg RV,
12. Ge J, Li Y, Qian J, Shi J, Wang Q, Niu Y, Fan B, Liu X, Zhang S, Sun A, et Kögler G, Wernet P. Intracoronary, human autologous stem cell trans-
al. Efficacy of emergent transcatheter transplantation of stem cells for plantation for myocardial regeneration following myocardial infarction.
treatment of acute myocardial infarction (TCT-STAMI). Heart 2006; 92: Dtsch Med Wochenschr 2001; 126: 932-8.
1764-7. 24. Lu G, Haider HK, Jiang S, Ashraf M. Sca-1+ stem cell survival and en-
13. Strauer BE, Yousef M, Schannwell CM. The acute and long-term effects graftment in the infarcted heart: dual role for preconditioning-induced
of intracoronary Stem cell Transplantation in 191 patients with chronic connexin-43. Circulation 2009; 119: 2587-96.
heARt failure: the STAR-heart study. Eur J Heart Fail 2010; 12: 721-9. 25. Kamota T, Li TS, Morikage N, Murakami M, Ohshima M, Kubo M, Ko-
14. Pereira MJ, Carvalho IF, Karp JM, Ferreira LS. Sensing the cardiac envi- bayashi T, Mikamo A, Ikeda Y, Matsuzaki M, et al. Ischemic pre-condi-
ronment: exploiting cues for regeneration. J Cardiovasc Transl Res 2011; tioning enhances the mobilization and recruitment of bone marrow stem
4: 616-30. cells to protect against ischemia/reperfusion injury in the late phase. J
15. Assmus B, Tonn T, Seeger FH, Yoon CH, Leistner D, Klotsche J, Schäch- Am Coll Cardiol 2009; 53: 1814-22.
inger V, Seifried E, Zeiher AM, Dimmeler S. Red blood cell contamina- 26. Van Linthout S, Stamm Ch, Schultheiss HP, Tschöpe C. Mesenchymal
tion of the final cell product impairs the efficacy of autologous bone mar- stem cells and inflammatory cardiomyopathy: cardiac homing and be-
row mononuclear cell therapy. J Am Coll Cardiol 2010; 55: 1385-94. yond. Cardiol Res Pract 2011; 2011: 757154.
16. Antman EM, Anbe DT, Armstrong PW, Bates ER, Green LA, Hand M, 27. Tang J, Wang J, Guo L, Kong X, Yang J, Zheng F, Zhang L, Huang Y. Mes-
Hochman JS, Krumholz HM, Kushner FG, Lamas GA, et al. ACC/AHA enchymal stem cells modified with stromal cell-derived factor 1 alpha im­
guidelines for the management of patients with ST-elevation myocardial prove cardiac remodeling via paracrine activation of hepatocyte growth
infarction: executive summary: a report of the American College of Car- factor in a rat model of myocardial infarction. Mol Cells 2010; 29: 9-19.
diology/American Heart Association Task Force on Practice Guidelines 28. Li L, Zhang S, Zhang Y, Yu B, Xu Y, Guan Z. Paracrine action mediate
(Writing Committee to revise the 1999 guidelines for the management of the antifibrotic effect of transplanted mesenchymal stem cells in a rat
patients with acute myocardial infarction). J Am Coll Cardiol 2004; 44: model of global heart failure. Mol Biol Rep 2009; 36: 725-31.
671-719. 29. Ter Horst KW. Stem cell therapy for myocardial infarction: are we miss-

30  
http://jkms.org http://dx.doi.org/10.3346/jkms.2014.29.1.23
Lee J-W, et al. • SEED-MSC Trial

ing time? Cardiology 2010; 117: 1-10. tation of primary percutaneous coronary intervention for acute myocar-
30. Yang YJ, Qian HY, Huang J, Geng YJ, Gao RL, Dou KF, Yang GS, Li JJ, Shen dial infarction: is the slope of the curve the shape of the future? JAMA
R, He ZX, et al. Atorvastatin treatment improves survival and effects of 2005; 293: 979-86.
implanted mesenchymal stem cells in post-infarct swine hearts. Eur Heart 32. Hovland A, Staub UH, Bjørnstad H, Prytz J, Sexton J, Støylen A, Vik-Mo
J 2008; 29: 1578-90. H. Gated SPECT offers improved interobserver agreement compared with
31. Gersh BJ, Stone GW, White HD, Holmes DR Jr. Pharmacological facili- echocardiography. Clin Nucl Med 2010; 35: 927-30.

http://dx.doi.org/10.3346/jkms.2014.29.1.23 http://jkms.org  31