1. LETTERS
Intracoronary Bone Marrow Mononuclear Cells to “guard against both premature declarations of victory
After Myocardial Infarction and premature abandonment of a promising therapeutic
strategy.”5
To the Editor: The LateTIME trial findings1 suggest a lack Hung Q. Ly, MD, MSc, FRCPC
of benefit from intracoronary delivery of bone marrow mono- Author Affiliation: Interventional Cardiology Division, Montreal Heart Institute,
nuclear cells (BMCs) 2 to 3 weeks after primary percuta- Universite de Montreal, Montreal, Quebec, Canada (Qh.ly@montreal.ca).
´ ´
neous coronary intervention (PCI) in patients with myo- Conflict of Interest Disclosures: The author has completed and submitted the ICMJE
Form for Disclosure of Potential Conflicts of Interest and reported receiving grants
cardial infarction (MI) and depressed left ventricular (LV) from the Fonds de Recherche en Sante du Quebec, the TheCell Network, the Ca-
´ ´ ´
function. However, further clarification is needed to better nadian Stem Cell Network, the Heart and Stroke Foundation of Canada, the Mon-
treal Heart Institute Foundation, and the DesGroseillers-Berard-Universite de Mon-
´ ´
appreciate the study conclusions. treal Chair in Interventional Cardiology; and serving as a consultant to Cook Medical.
´
While timing can affect efficacy of cardiac cell-based
1. Traverse JH, Henry TD, Ellis SG, et al; Cardiovascular Cell Therapy Research
therapies, patient selection remains of paramount impor- Network (CCTRN). Effect of intracoronary delivery of autologous bone marrow
tance. It is unclear why patients with lesions in coronary mononuclear cells 2 to 3 weeks following acute myocardial infarction on left ven-
tricular function: the LateTIME randomized trial. JAMA. 2011;306(19):2110-
arteries other than the left anterior descending (4 right 2119.
coronary arteries and 1 left circumflex coronary artery, all 2. Ly HQ, Hoshino K, Pomerantseva I, et al. In vivo myocardial distribution of mul-
tipotent progenitor cells following intracoronary delivery in a swine model of myo-
in the intervention group) were included. Because the cardial infarction. Eur Heart J. 2009;30(23):2861-2868.
mean qualifying LV ejection fraction (LVEF) was 35% to 3. Meyer GP, Wollert KC, Lotz J, et al. Intracoronary bone marrow cell transfer
36%, could a mixed cardiomyopathy have explained the after myocardial infarction: 5-year follow-up from the randomized-controlled BOOST
trial. Eur Heart J. 2009;30(24):2978-2984.
initial LV dysfunction and thus partly biased the efficacy 4. Eitel I, Kubusch K, Strohm O, et al. Prognostic value and determinants of a hy-
findings in cell-treated patients? Moreover, was consider- pointense infarct core in T2-weighted cardiac magnetic resonance in acute rep-
erfused ST-elevation-myocardial infarction. Circ Cardiovasc Imaging. 2011;
ation given to including only patients with anterior wall 4(4):354-362.
MIs associated with severe but persistent LV dysfunction 5. Rosenzweig A. Cardiac cell therapy—mixed results from mixed cells. N Engl J
Med. 2006;355(12):1274-1277.
(ie, LVEF Յ35% at baseline and not just at screening),
who would be more likely to benefit from BMC injection To the Editor: Dr Traverse and colleagues observed no dif-
relative to a population with only moderate LV impair- ferences at 6 months in global or regional LV function,
ment? While more likely to induce recruitment delays, infarct size, or infarct volumes in patients with LV dys-
such a strategy might have resulted in greater benefit in function as a consequence of an MI who received intra-
the intervention group. coronary infusion of autologous BMCs compared with
As mentioned by Dr Traverse and colleagues, other cell controls.1 The authors attributed the lack of efficacy when
populations (bone marrow–derived or not) could ulti- compared with similar trials that found an improvement in
mately prove more beneficial in such a clinical setting. How- LV function in part to the late time of infusion (median:
ever, alternative routes of delivery, such as intramyocar- 17.4 days vs 7 days in prior studies) when BMC homing
dial or transcoronary, might prove to be more efficacious. signals from ischemic cardiomyocytes have diminished.
Although intracoronary injection remains a widely used We believe the authors did not address a more likely cause
method, it is associated with poor rates of cell retention.2 for lack of efficacy in their trial: infusion of an insufficient
In addition, the Bone Marrow Transfer to Enhance ST- number of potent cells.
Elevation Infarct Regeneration (BOOST) investigators, in
their long-term follow-up analysis,3 reported that mag-
GUIDELINES FOR LETTERS. Letters discussing a recent JAMA article should
netic resonance imaging (MRI) characterization of myocar- be submitted within 4 weeks of the article’s publication in print. Letters received
dial damage suggested subgroups that may have a greater after 4 weeks will rarely be considered. Letters should not exceed 400 words of
text and 5 references and may have no more than 3 authors. Letters reporting
clinical response to cardiac cell-based therapies. Were dif- original research should not exceed 600 words of text and 6 references and may
ferences observed in intramyocardial hemorrhage, trans- have no more than 5 authors. They may include up to 2 tables or figures but on-
mural infarction, or microvascular obstruction between line supplementary material is not allowed. All letters should include a word count.
Letters must not duplicate other material published or submitted for publication.
groups in this study? Such MRI characteristics reflect an al- Letters not meeting these specifications are generally not considered. Letters will
tered course in myocardial salvage and LV remodeling4 and be published at the discretion of the editors and are subject to abridgement and
editing. Further instructions can be found at http://jama.com/instructions. A signed
therefore might modulate the effects of cardiac cell-based statement for authorship criteria and responsibility, financial disclosure, copyright
therapies. transfer, and acknowledgment and the ICMJE Form for Disclosure of Potential Con-
flicts of Interest are required before publication. Letters should be submitted via
It will be challenging for any adjunctive therapeutic strat- the JAMA online submission and review system at http://manuscripts.jama.com
egy to show an incremental benefit in morbidity or mortal- (note: do not include “www” before the URL). For technical assistance, please
ity above that of the well-established clinical success of pri- contact jama-letters@jama-archives.org.
mary PCI. Nevertheless, the LateTIME study will contribute Letters Section Editor: Jody W. Zylke, MD, Senior Editor.
1022 JAMA, March 14, 2012—Vol 307, No. 10 ©2012 American Medical Association. All rights reserved.
Downloaded from jama.ama-assn.org at American Medical Association on March 14, 2012

2. LETTERS
The potency of intracoronary-infused BMCs appears to ment, but only 5 patients had non–anterior MIs. Their in-
depend on the number of CD34 cells (a subset of BMCs) clusion did not appear to influence outcomes because the
infused and the mobility of CD34 cells measured in vitro results were similar when analyzed without these patients.
in a stromal-derived factor 1 (SDF-1) gradient. A meta- We have no evidence suggesting our patients had other forms
analysis of 811 patients with MI demonstrated the impor- of cardiomyopathy contributing to the reduced LVEF on
tance of the number of cells infused because the greatest im- qualifying echocardiogram (mean [SD], 36.4% [6.5%])
provement in LV function and volumes were observed among within 48 hours of their PCI. When measured at the time
patients receiving more than 109 BMCs.2 In addition to quan- of the intracoronary infusion several weeks later, LVEF had
tity, Britten et al3 observed that the ex vivo migratory ca- increased (mean [SD], 44.3% [8.4%]) likely due to resolu-
pacity of BMCs (the migratory subset being CD34 cells co- tion of myocardial stunning. Still, it remained below the base-
expressing CXCR-4 receptors) in an SDF-1 gradient was the line median of 48.9% seen in the Reinfusion of Enriched Pro-
strongest independent predictor of LV infarct size reduc- genitor Cells and Infarct Remodeling in Acute Myocardial
tion after intracoronary infusion of BMCs. Seeger et al4 later Infarction (REPAIR-AMI) study in which the greatest ben-
showed preparing BMCs in heparin and saline, and not au- efit of BMCs on LVEF recovery was observed.1 We agree that
tologous serum without heparin, significantly reduced SDF-1 patients with persistent LV dysfunction after MI (LVEF
mobility. Our study of intracoronary infused purified CD34 Ͻ35%) constitute an important target for future studies that
cells demonstrated that a minimum of 10ϫ106 CD34 cells should include new cell types and delivery methods such
was required to improve perfusion and again showed a sig- as intramyocardial or transcoronary delivery. We are cur-
nificant positive correlation between the number of mobile rently performing more detailed MRI analyses to address the
(measured in an SDF-1 gradient) CD34 cells infused and extent of infarct transmurality and microvascular obstruc-
improvement in perfusion and decrease in infarct size.5 tion that may also influence outcomes.
In the LateTIME trial, treated patients received a median Drs Quyyumi and Pecora speculate that insufficient cell
CD34 cell dose of 3.8 ± 1.5ϫ106 cells, well below 10ϫ 106 numbers may have caused lack of efficacy, note the role of
CD34 cells. Additionally, in vitro SDF-1 mobility of the in- heparin, and state that “in vitro SDF-1 mobility of the in-
fused cells was not measured and may have been adversely fused cells was not measured.” The latter statement is in-
affected by the absence of autologous serum in the infused correct because SDF-1 mobility, as well as nitric oxide pro-
product. Future studies must account for the quantity and duction and other measures, are part of our biorepository
mobility of infused (potent) cells before conclusions re- studies2 to develop mechanistic insights into cardiovascu-
garding efficacy are made. lar cell therapy. These studies are ongoing and beyond the
Arshed Quyyumi, MD, FRCP scope of the clinical results presented in our article.
Andrew L. Pecora, MD Our delivered dose exceeded the greater than 100 million
BMC threshold identified by meta-analysis to be efficacious,3
Author Affiliations: Division of Cardiology, Emory University, Atlanta, Georgia (Dr
Quyyumi); and John Theurer Cancer Center, Hackensack University Medical Cen-
and the number of CD34 cells infused was similar to the
ter, Hackensack, New Jersey (Dr Pecora) (apecora@humed.com). REPAIR-AMI trial, which had a positive outcome.1 Different
Conflict of Interest Disclosures: The authors have completed and submitted the cell isolation procedures generate different BMC numbers (re-
ICMJE Form for Disclosure of Potential Conflicts of Interest. Dr Quyyumi reported
being a consultant to Amorcyte. Dr Pecora reported being chief medical officer of covery), composition, and function (colony assays and in-
NeoStem (manufacturer and equity owner of Amorcyte). vivo angiogenesis), but the specific details responsible for these
1. Traverse JH, Henry TD, Ellis SG, et al; Cardiovascular Cell Therapy Research differences are unclear. Possibilities include, but are not lim-
Network (CCTRN). Effect of intracoronary delivery of autologous bone marrow ited to, heparin concentration, density gradient medium, cen-
mononuclear cells 2 to 3 weeks following acute myocardial infarction on left ven-
tricular function: the LateTIME randomized trial. JAMA. 2011;306(19):2110- trifuge speed and time, wash steps (number, diluents, and cen-
2119. trifuge settings), overnight storage, and red blood cell content.
2. Martin-Rendon E, Brunskill SJ, Hyde CJ, Stanworth SJ, Mathur A, Watt SM.
Autologous bone marrow stem cells to treat acute myocardial infarction: a sys- With so many variables, head-to-head comparisons are re-
tematic review. Eur Heart J. 2008;29(15):1807-1818. quired to determine specific contributions. Our final cell sus-
3. Britten MB, Abolmaali ND, Assmus B, et al. Infarct remodeling after intracoro-
nary progenitor cell treatment in patients with acute myocardial infarction (TOPCARE-
pension volume contained very little heparin (after 3 washes,
AMI): mechanistic insights from serial contrast-enhanced magnetic resonance the 43 mL of cell suspension contained 0.49 units of heparin
imaging. Circulation. 2003;108(18):2212-2218. total) and was not stored overnight. We are not aware of re-
4. Seeger FH, Tonn T, Krzossok N, Zeiher AM, Dimmeler S. Cell isolation proce-
dures matter: a comparison of different isolation protocols of bone marrow mono- ports directly addressing heparin’s effect on BMC migration.
nuclear cells used for cell therapy in patients with acute myocardial infarction. Eur The study by Seeger et al4 tested 2 cell separation protocols
Heart J. 2007;28(6):766-772.
5. Quyyumi AA, Waller EK, Murrow J, et al. CD34(ϩ) cell infusion after ST el- with many variables and did not directly measure the effects
evation myocardial infarction is associated with improved perfusion and is dose of heparin. In addition, heparin expanded the number of he-
dependent. Am Heart J. 2011;161(1):98-105.
matopoietic progenitor cells in an umbilical cord blood study,
In Reply: We support Dr Ly’s comments on the impor- illustrating the complexity of this issue.5
tance of patient selection in these trials. We elected to in- Therefore, we believe the timing of BMC delivery after
clude patients with large infarcts from lesions in non–left MI remains a critical factor in the results of the LateTIME
anterior descending coronary arteries to facilitate enroll- trial. The results of the TIME trial,6 available later this year,
©2012 American Medical Association. All rights reserved. JAMA, March 14, 2012—Vol 307, No. 10 1023
Downloaded from jama.ama-assn.org at American Medical Association on March 14, 2012

3. LETTERS
will provide insights into the results of delivery on day 3 low separate analyses of clopidogrel escalating doses by dia-
compared with day 7 after MI. betes status in carriers and noncarriers.
Jay H. Traverse, MD Patients with chronic kidney disease frequently present with
Timothy D. Henry, MD increased platelet reactivity, and worse renal function dimin-
Carl J. Pepine, MD ishes the platelet inhibitory effect of clopidogrel.3 In the PLATO
Author Affiliations: Minneapolis Heart Institute Foundation at Abbott Northwest- trial, ticagrelor compared with clopidogrel resulted in a greater
ern Hospital, Minneapolis, Minnesota (Drs Traverse and Henry) (Jay.Traverse absolute risk reduction in patients with chronic kidney dis-
@allina.com); and University of Florida College of Medicine, Gainesville (Dr Pep-
ine), for the Cardiovascular Cell Therapy Research Network.
ease than in those with normal renal function.4 It would be
Conflict of Interest Disclosures: The authors have completed and submitted the interesting to know if impaired renal function differentially
ICMJE Form for Disclosure of Potential Conflicts of Interest. Drs Traverse, Henry, affected the improvement in platelet inhibition with increas-
and Pepine reported receiving grants, travel support, and research support to their
institutions from the National Heart, Lung, and Blood Institute. Dr Pepine also re- ing doses of clopidogrel between carriers and noncarriers.
ported receiving research support to his institution from BDS Cordis Corp- We agree with the authors’ suggestion of alternate anti-
(NOGA) and serving on the scientific advisory board for Amorcyte.
platelet agent use in cases of low or nonresponsiveness to
1. Schachinger V, Erbs S, Elsasser A, et al; REPAIR-AMI Investigators. Intracoro-
¨ ¨ clopidogrel in carriers of CYP2C19*2 loss-of-function al-
nary bone marrow-derived progenitor cells in acute myocardial infarction. N Engl
J Med. 2006;355(12):1210-1221. lele. We found that prasugrel was superior to a double dose
2. Zierold C, Carlson MA, Obodo UC, et al. Developing mechanistic insights into of clopidogrel, particularly in carriers of the CYP2C19*2 loss-
cardiovascular cell therapy: Cardiovascular Cell Therapy Research Network Bio-
repository Core Laboratory rationale. Am Heart J. 2011;162(6):973-980. of-function allele.5
3. Martin-Rendon E, Brunskill SJ, Hyde CJ, Stanworth SJ, Mathur A, Watt SM.
Autologous bone marrow stem cells to treat acute myocardial infarction: a sys-
Dimitrios Alexopoulos, MD, FACC, FESC
tematic review. Eur Heart J. 2008;29(15):1807-1818. Ioanna Xanthopoulou, MD
4. Seeger FH, Tonn T, Krzossok N, Zeiher AM, Dimmeler S. Cell isolation proce-
dures matter: a comparison of different isolation protocols of bone marrow mono- Author Affiliations: Department of Cardiology, Patras University Hospital, Rio, Pa-
nuclear cells used for cell therapy in patients with acute myocardial infarction. Eur tras, Greece (dalex@med.upatras.gr).
Heart J. 2007;28(6):766-772. Conflict of Interest Disclosures: The authors have completed and submitted the
5. Okamoto T, Takagi M, Soma T, et al. Effect of heparin addition on expansion ICMJE Form for Disclosure of Potential Conflicts of Interest. Dr Alexopoulos re-
of cord blood hematopoietic progenitor cells in three-dimensional coculture with ported receiving speaker fees from sanofi-aventis, Pharmaserve, Eli Lilly, Astra-
stromal cells in nonwoven fabrics. J Artif Organs. 2004;7(4):194-202. Zeneca, and Boehringer Ingelheim. Dr Xanthopoulou reported no disclosures.
6. Traverse JH, Henry TD, Vaughan DE, et al; Cardiovascular Cell Therapy Re-
search Network (CCTRN). Rationale and design for TIME: a phase II, randomized, 1. Mega JL, Hochholzer W, Frelinger AL III, et al. Dosing clopidogrel based on
double-blind, placebo-controlled pilot trial evaluating the safety and effect of tim- CYP2C19 genotype and the effect on platelet reactivity in patients with stable car-
ing of administration of bone marrow mononuclear cells after acute myocardial diovascular disease. JAMA. 2011;306(20):2221-2228.
infarction [published correction appears in Am Heart J. 2009;158(6):1045]. Am 2. Wallentin L, James S, Storey RF, et al; PLATO investigators. Effect of CYP2C19
Heart J. 2009;158(3):356-363. and ABCB1 single nucleotide polymorphisms on outcomes of treatment with ti-
cagrelor versus clopidogrel for acute coronary syndromes: a genetic substudy of
the PLATO trial. Lancet. 2010;376(9749):1320-1328.
Clopidogrel Dosing Based on Genotype 3. Htun P, Fateh-Moghadam S, Bischofs C, et al. Low responsiveness to clopido-
grel increases risk among CKD patients undergoing coronary intervention. J Am
To the Editor: Dr Mega and colleagues1 presented a study Soc Nephrol. 2011;22(4):627-633.
4. James S, Budaj A, Aylward P, et al. Ticagrelor versus clopidogrel in acute coro-
showing that increasing the clopidogrel dose resulted in plate- nary syndromes in relation to renal function: results from the Platelet Inhibition
let inhibitory effects in patients heterozygous for the and Patient Outcomes (PLATO) trial. Circulation. 2010;122(11):1056-1067.
5. Alexopoulos D, Dimitropoulos G, Davlouros P, et al. Prasugrel overcomes high
CYP2C19*2 loss-of-function allele, but not in homozy- on-clopidogrel platelet reactivity post-stenting more effectively than high-dose (150-
gotes. We consider a few points that require clarification. mg) clopidogrel: the importance of CYP2C19*2 genotyping. JACC Cardiovasc Interv.
2011;4(4):403-410.
Platelet reactivity was assessed at least 1 month after myo-
cardial infarction (MI), percutaneous coronary interven- In Reply: Drs Alexopoulos and Xanthopoulou ask about the
tion (PCI), or both. However, it is well recognized that most timing of the intervention in ELEVATE-TIMI 56 relative to
events, including stent thrombosis, occur early, within 30 the patients’ clinical course. The aim of ELEVATE-TIMI 56
days of PCI. In the Platelet Inhibition and Patient Out- was to evaluate whether higher maintenance doses of clopi-
comes (PLATO) genetic substudy,2 the event rate at 30 days dogrel could improve the pharmacological response in the set-
was higher in clopidogrel-treated patients with CYP2C19 ting of loss-of-function CYP2C19 genotypes. Thus, by de-
alleles than those without CYP2C19 alleles, leading to ear- sign, patients were in the maintenance phase of P2Y12 adenosine
lier separation of event rates between the ticagrelor and clopi- diphosphate (ADP) receptor blockade therapy and were en-
dogrel groups in patients with loss-of-function alleles. In con- rolled between 4 weeks and 6 months following PCI or acute
trast, beyond 30 days, no difference in event rates between MI. Clinical circumstances such as PCI or acute MI can affect
treatment groups for patients with any loss-of-function al- platelet reactivity. Therefore, the protocol specified that en-
lele was observed. Therefore, genotyping and appropriate rollment should occur at a time in which platelet reactivity
clopidogrel dose escalation outside the setting of MI and PCI should have stabilized to minimize the risk of these clinical
may not provide incremental clinical utility. circumstances influencing results with the different doses of
Diabetes mellitus is widely accepted as a major adverse clopidogrel, which were administered in various temporal se-
clinical factor affecting platelet reactivity. In the present study, quences. The findings in ELEVATE-TIMI 56, which evalu-
35.4% (36.4% in noncarriers and 32.6% in carriers) of pa- ated maintenance doses of clopidogrel, are consistent with the
tients had diabetes, a sample probably large enough to al- findings of the Clopidogrel and Response Variability Investi-
1024 JAMA, March 14, 2012—Vol 307, No. 10 ©2012 American Medical Association. All rights reserved.
Downloaded from jama.ama-assn.org at American Medical Association on March 14, 2012