Effect of Transendocardial Delivery of Autologous Bone Marrow

ORIGINAL CONTRIBUTION

ONLINE FIRST Effect of Transendocardial Delivery of Autologous Bone Marrow Mononuclear Cells on Functional Capacity, Left Ventricular Function, and Perfusion in Chronic Heart Failure The FOCUS-CCTRN Trial

Emerson C. Perin, MD, PhD, James T. Context Previous studies using autologous bone marrow mononuclear cells (BMCs) in Willerson, MD, Carl J. Pepine, MD, Timothy D. patients with ischemic cardiomyopathy have demonstrated safety and suggested efficacy. Henry, MD, Stephen G. Ellis, MD, David X. M. Objective To determine if administration of BMCs through transendocardial injec- Zhao, MD, Guilherme V. Silva, MD, Dejian Lai, tions improves myocardial perfusion, reduces left ventricular end-systolic volume (LVESV), or enhances maximal oxygen consumption in patients with coronary artery disease or PhD, James D. Thomas, MD, Marvin W. LV dysfunction, and limiting heart failure or angina. Kronenberg, MD, A. Daniel Martin, PhD, PT, Design, Setting, and Patients A phase 2 randomized double-blind, placebo-controlled R. David Anderson, MD, MS, Jay H. Traverse, trial of symptomatic patients (New York Heart Association classification II-III or Canadian MD, Marc S. Penn, MD, PhD, Saif Cardiovascular Society classification II-IV) with a left ventricular ejection fraction of 45% Anwaruddin, MD, Antonis K. Hatzopoulos, or less, a perfusion defect by single-photon emission tomography (SPECT), and coronary artery disease not amenable to revascularization who were receiving maximal medical therapy PhD, Adrian P. Gee, PhD, Doris A. Taylor, at 5 National Heart, Lung, and Blood Institute–sponsored Cardiovascular Cell Therapy Re- PhD, Christopher R. Cogle, MD, Deirdre search Network (CCTRN) sites between April 29, 2009, and April 18, 2011. Smith, RN, Lynette Westbrook, RN, James Intervention Bone marrow aspiration (isolation of BMCs using a standardized au- Chen, RN, Eileen Handberg, PhD, Rachel E. tomated system performed locally) and transendocardial injection of 100 million BMCs or placebo (ratio of 2 for BMC group to 1 for placebo group). Olson, RN, MS, Carrie Geither, RN, Sherry Main Outcome Measures Co-primary end points assessed at 6 months: changes Bowman, RN, Judy Francescon, RN, Sarah in LVESV assessed by echocardiography, maximal oxygen consumption, and revers- Baraniuk, PhD, Linda B. Piller, MD, MPH, ibility on SPECT. Phenotypic and functional analyses of the cell product were per- Lara M. Simpson, PhD, Catalin Loghin, MD, formed by the CCTRN biorepository core laboratory. David Aguilar, MD, Sara Richman, Claudia Results Of 153 patients who provided consent, a total of 92 (82 men; average age: Zierold, PhD, Judy Bettencourt, MPH, Shelly 63 years) were randomized (n=61 in BMC group and n=31 in placebo group). Changes in LVESV index (−0.9 mL/m2 [95% CI, −6.1 to 4.3]; P=.73), maximal oxygen con- L. Sayre, MPH, Rachel W. Vojvodic, MPH, sumption (1.0 [95% CI, −0.42 to 2.34]; P=.17), and reversible defect (−1.2 [95% CI, Sonia I. Skarlatos, PhD, David J. Gordon, MD, −12.50 to 10.12]; P=.84) were not statistically significant. There were no differences PhD, Ray F. Ebert, PhD, Minjung Kwak, PhD, found in any of the secondary outcomes, including percent myocardial defect, total defect size, fixed defect size, regional wall motion, and clinical improvement. Lemuel A. Moye´, MD, PhD, Conclusion Among patients with chronic ischemic heart failure, transendocardial in- Robert D. Simari, MD jection of autologous BMCs compared with placebo did not improve LVESV, maximal for the Cardiovascular Cell Therapy Research oxygen consumption, or reversibility on SPECT. Network (CCTRN) Trial Registration clinicaltrials.gov Identifier: NCT00824005 JAMA. 2012;307(16):1717-1726 ELL THERAPY HAS EMERGED AS Published online March 24, 2012. doi:10.1001/jama.2012.418 www.jama.com an innovative approach for treating patients with ad- Author Affiliations and a list of the CCTRN study group vanced ischemic heart dis- studies have been performed primarily are listed at the end of this article. Cease, including those with refractory an- using autologous stem/progenitor Corresponding Author: Lemuel A. Moye´ , MD, PhD, University of Texas, 1200 Pressler, W-848, Houston, gina and/or heart failure. Early clinical cells.1-13 In patients with ischemic heart TX 77030 ([email protected]).

Downloaded From: http://jama.jamanetwork.com/ by a University of Kansas, Participant User on 05/22/2014 BONE MARROW MONONUCLEAR CELLS FOR CHRONIC HEART FAILURE

disease and heart failure, treatment with administration of 100ϫ106 total BMCs formity of cell preparation. Briefly, autologous bone marrow mononuclear improves measures of LV performance approximately 80 to 100 mL of bone mar- cells (BMCs) has demonstrated safety and perfusion at 6 months compared row was aspirated from the iliac crest and has suggested efficacy.10,14-17 with baseline levels. using standard techniques. The aspirate None of the clinical trials per- Briefly, we enrolled patients aged 18 was processed with a closed, automated formed to date, however, have been years or older with clinically stable coro- cell processing system21 (Sepax, Biosafe powered to evaluate specific efficacy nary artery disease, LV ejection frac- SA). Composition of CD34 and CD133 measures. In addition, autologous cell tion (LVEF) of 45% or less, limiting an- cells was determined by flow cytom- therapy trials have usually enrolled gina (Canadian Cardiovascular Society etry. After the cells passed stipulated lot older patients with acute or chronic left [CCS] class II-IV) and/or congestive release criteria, including viability ventricular (LV) dysfunction without heart failure (New York Heart Associa- (Ͼ70%) and sterility, randomization was evaluating the effect of these param- tion [NYHA] class II-III), a perfusion performed by the data coordinating cen- eters or cell function on clinical out- defect by single-photon emission com- ter. Treatment assignment was masked come. These factors are important puted tomography (SPECT), and no re- to all but 1 designated cell processing because ischemic heart failure has the vascularization options while receiv- team member at each center not involved potential to limit the beneficial influ- ing guideline-based medical therapy.20 in patient care. The target dose was ences of cellular effects.6,18 The study was conducted at the 5 100ϫ106 total BMCs. The BMC final The present study was undertaken by CCTRN centers; the organizational product was suspended in normal saline the National Heart, Lung, and Blood In- structure and oversight of the CCTRN containing 5% human serum albumin stitute–sponsored Cardiovascular Cell have been described.20 The protocol was and adjusted to a concentration of Therapy Research Network (CCTRN).19 reviewed and approved by the local in- 100ϫ106 cells in 3 mL distributed into It builds on work from a pilot study in stitutional review boards at each cen- three 1-mL syringes. The placebo group Brazil,10 which in turn led to the first ter. All patients provided written in- received a cell-free suspension in the phase 1 randomized trial of autologous formed consent. same volume. BMC therapy for heart failure in the Randomization was computer- Within 12 hours of aspiration, BMCs United States (FOCUS-HF) approved by generated and used variable block sizes or placebo were delivered in 15 sepa- the US Food and Drug Administra- of 6 or 9, randomly selected and strati- rate injections (0.2 mL each) to LV en- tion.16 These initial studies showed that fied by center. All randomized patients docardial regions identified as viable transendocardial delivery of BMCs was underwent baseline testing, bone mar- (unipolar voltage Ն6.9 mV) by elec- feasible and appeared safe in patients row harvesting, and automated cell pro- tromechanical mapping as described with chronic heart failure due to multi- cessing that was performed locally.20 Pa- elsewhere.20 A 2-dimensional echocar- vessel coronary artery disease.10,16 Al- tients were randomized in a 2:1 ratio to diogram was performed immediately af- though these preliminary studies also receive either BMCs or a placebo (cell- ter the injection procedure and on the evaluated LV function, perfusion, and free) preparation. The cell-containing or next day before hospital discharge. Se- functional capacity, a definitive assess- cell-free preparation was delivered to vi- rial measurements of creatine kinase, ment of efficacy was not possible due to able myocardial regions identified dur- creatine kinase MB, and troponin also the small number of patients. Thus, the ing electromechanical mapping of the LV were obtained. All patients remained in present trial was designed as a larger endocardial surface (NOGA, Biologics the hospital overnight and were then study to investigate the effects of tran- Delivery Systems, Cordis Corpora- discharged with instructions for guide- sendocardial-delivered BMCs in pa- tion). All caregivers and patients were line-recommended therapy. Patients tients with chronic ischemic heart dis- masked to treatment. At 6 months, all were examined for safety and efficacy ease and LV dysfunction with heart baseline testing was repeated in an iden- at 6 months. All events deemed to be failure and/or angina.20 tical fashion. potential major adverse clinical events Baseline assessments have been de- were assessed by 2 independent cardi- METHODS scribed.20 Demographic and clinical ologists not affiliated with any clinical A phase 2 randomized double-blind, pla- variables were determined by inter- site and masked to treatment assign- cebo-controlled trial, FOCUS-CCTRN view and documented from each pa- ment. Follow-up for safety continues (First Mononuclear Cells injected in the tient’s medical record. Race and eth- (up to 12 months postintervention) United States conducted by the CCTRN) nicity were recorded as self-described with annual telephone calls at 2, 3, 4, was designed to evaluate the safety and by the patients. and 5 years postintervention. efficacy of BMCs in patients with chronic Cell harvesting and processing pro- The CCTRN established a cell bio- ischemic heart disease and LV dysfunc- cedures for all CCTRN protocols have repository core laboratory to advance tion who have no other revasculariza- been reported.21,22 Rigorous automated the understanding of the relationship tion options. The primary objective was methods for local cell processing were between cell product characteristics to determine whether transendocardial implemented to ensure quality and uni- (composition or phenotype and func-

Downloaded From: http://jama.jamanetwork.com/ by a University of Kansas, Participant User on 05/22/2014 BONE MARROW MONONUCLEAR CELLS FOR CHRONIC HEART FAILURE

tion) and clinical outcomes.23 The re- The 3 prespecified primary end Prespecified subgroup analyses for maining processed cells (unused cells points of change (6-month follow-up hypothesis generation examined the ef- in the BMC group and all cells from pa- minus baseline) in LVESV, maximal fects of treatment stratified by age, sex, tients in the placebo group) were oxygen consumption, and defect size race, diabetes, serum BNP levels in pa- shipped to the biorepository core labo- on SPECT were compared between the tients with heart failure, preexisting co- ratory with patient consent. BMC group and the control group. morbidity, number of endothelial Echocardiographic measurements For each co-primary end point, a colony forming cells, and baseline were performed by an echocardiogra- sample size was computed based on es- LVEF. Two-sided significance testing phy core laboratory according to pub- timates of the effect size and the stan- was used; P values of less than .05 were lished guidelines24 and included LV dard deviation of the difference from the deemed statistically significant. end-systolic volume (LVESV), LV end- prior data.20 Type I error was appor- diastolic volume, regional wall mo- tioned at the .05 level to be conducted RESULTS tion, and LVEF. Myocardial contrast at 80% power with 10% of patients an- Screening commenced in March 2009 was used to enhance endocardial defi- ticipated as lost to follow-up. All test- and 153 patients provided consent. Be- nition. These measurements were com- ing was 2-sided. The study was de- tween April 29, 2009, and April 18, puted using the biplane rule methods signed to detect a mean difference 2011, 92 patients were randomized of Simpson as described by Maret et al.25 between the 2 groups of 27 mL for (n=61 in BMC group and n=31 in pla- The LV volume data were normalized LVESV, 5 mL/kg/min for maximal oxy- cebo group). Of the 273 patients for body surface area and indexed data gen consumption, and 10 absolute per- screened, most were excluded due to are presented. centage points for reversible ische- not having evidence of reversibility SPECT or adenosine myocardial per- mia. To ensure adequate power for each (prior to protocol amendment during fusion tests were performed to identify of the 3 end points, the sample size was the final third period of enrollment) or changes in ischemic (reversible) de- computed for each one, and the maxi- having an LVEF greater than 45% fects from rest and after adenosine infu- mum sample size was selected.20 This (FIGURE 1).27 sion over 4 minutes (or if contraindi- produced a sample size of 86 patients, Briefly, this was an older, white male cated, with a regadenoson bolus) using which was administratively increased population (TABLE 1). Most (76%) pa- standardized protocols. To enhance vi- to 92 patients (31 in the placebo group tients had an implantable cardioverter- ability detection on resting images, sub- and 61 in the BMC group). No type I defibrillator. No statistically significant lingual nitroglycerin was administered error adjustment for multiple compari- differences were seen in baseline char- 15 minutes before injection of techne- sons was incorporated because this was acteristics between the BMC and pla- tium Tc 99m sestamibi for the resting im- a phase 2 study. cebo groups, except for greater ranola- age. Changes in fixed perfusion defects All statistical analyses were con- zine use in the BMC group (TABLE 2), by SPECT also were measured. ducted using SAS version 9.2 (SAS In- which was consistent with more pa- Maximal oxygen consumption was stitute Inc).26 Descriptive statistics for tients with CCS class II to IV angina in assessed by using the Naughton tread- baseline characteristics were generated the BMC group. The mean (SD) LVEF mill protocol. Blood flow improve- for demographic variables, medical his- on the qualifying echocardiogram was ment was examined by magnetic reso- tory, physical examination, laboratory 32.4% (9.2%) in the BMC group and nance imaging (MRI) in patients without data, and clinical events. ␹2 Statistics and 30.2% (7.8%) in the placebo group. MRI contraindications. t tests were used to evaluate the differ- All randomized patients had their Clinical improvement by CCS classi- ences between the 2 study groups. Gen- bone marrow processed with Ficoll fication, NYHA classification, and change eral linear modeling techniques as- using the automated Sepax device.21 The in antianginal medications was ex- sessed the effects of treatment on the mean (SD) volume of bone marrow har- plored. Serum brain-type natriuretic pep- continuous primary and secondary out- vested was 93.7 (8.3) mL. The mean tide (BNP) levels were collected in pa- comes of the study. Both unadjusted and (SD) time from aspiration to product tients with congestive heart failure, and baseline covariate-adjusted treatment ef- injection was 8.9 (1.2) hours in the changes in the levels were assessed at 6 fects were computed. Dichotomous sec- BMC group and 8.6 (2.2) hours in the months. Major adverse clinical events ondary end points (ie, clinical improve- placebo group. were assessed and defined as new ment at 6 months, change in CCS Of the 92 patients randomized, 5 pa- myocardial infarction, rehospitaliza- anginal score and NYHA class, de- tients were identified as having a le- tion for percutaneous coronary inter- crease in weekly need for antianginal sion suitable for percutaneous revas- vention in treated coronary artery terri- medication [nitrates]) were analyzed cularization (although no patient had tories, death, and rehospitalization for using ␹2 and Fisher exact tests. The time- such a lesion identified on the qualify- non–myocardial infarction acute coro- to-event end points (ie, major adverse ing angiogram). Per protocol, these 5 nary syndrome or congestive heart clinical events) could not be reliably as- patients underwent revascularization failure. sessed due to the paucity of events. rather than receive the study product.

Downloaded From: http://jama.jamanetwork.com/ by a University of Kansas, Participant User on 05/22/2014 BONE MARROW MONONUCLEAR CELLS FOR CHRONIC HEART FAILURE

A sixth patient experienced a limited of 100ϫ106 nucleated cells, which con- A total of 54 patients in the BMC group retrograde catheter-related dissection of tained an average of 2.6% of CD34 cells and 28 patients in the placebo group had the abdominal aorta that precluded and 1.2% of CD133 cells. Five of these paired LVESV data at baseline and at 6 study product delivery. patients had harvests that contained less months (FIGURE 2). The mean (SD) Electromechanical map–guided injec- than 100 million cells (99.9, 99.6, 99, 80, LVESV index (LVESVI) at baseline was tion of cells or placebo was conducted and 61 million cells). The other patient 57.9 (26.1) mL/m2 in the BMC group per protocol in the remaining 86 pa- experienced recurrent ventricular tachy- and 65.0 (19.8) mL/m2 in the placebo tients. Mean viability of the cell prod- cardia with hypotension after each in- group. At 6 months, the mean (SD) uct (Trypan blue exclusion) was 98.6% jection and received only a small vol- LVESVI was 57.0 (25.5) mL/m2 in the (TABLE 3). In the BMC group, all but 6 ume of cell product (approximately 13 BMC group and 65.0 (23.3) mL/m2 in patients received the targeted total dose million cells). the placebo group. The difference in the change in LVESVI between the BMC group and the placebo group was not Figure 1. Flow Diagram of Patients in FOCUS-CCTRN Trial statistically significant (−0.9 mL/m2 [95% CI, −6.1 to 4.3]; P=.73). 273 Assessed for eligibility A total of 52 patients in the BMC group and 27 patients in the placebo 181 Excluded 116 Did not meet eligibility criteria group had paired maximal oxygen con- 55 No reversible ischemia sumption data at baseline and at 6 32 Other cardiac conditions 29 Left ventricular ejection fraction >45% months (Figure 2). The mean (SD) base- 27 Refused to participate line maximal oxygen consumption was 38 Other reasons 14.6 (3.8) mL/kg/min in the BMC group and 15.3 (4.6) mL/kg/min in the pla- 92 Randomized cebo group. At 6 months, the mean (SD) maximal oxygen consumption was 15.0 61 Randomized to receive transendocardial injection 31 Randomized to receive placebo (4.5) mL/kg/min in the BMC group and of 100 million bone marrow mononuclear cells 2 Did not receive intervention as randomized 4 Did not receive intervention as randomized (revascularizable lesion found at time 14.7 (5.1) mL/kg/min in the placebo 3 Revascularizable lesion found at time of intervention) group. The difference in the change in of intervention 1 Catheter-related dissection of abdominal aorta maximal oxygen consumption between the BMC group and placebo group Primary analysis Primary analysis was not statistically significant (1.0 [95% 54 Assessed for changes in left ventricular end- 28 Assessed for changes in LVESV by CI, −0.42 to 2.34]; P=.17). systolic volume (LVESV) by echocardiography echocardiography 7 Excluded 3 Excluded A total of 52 patients in the BMC 1 Other 1 Other group and 25 patients in the placebo 2 Too ill 0 Transplant group had paired SPECT evaluations at 1 Lost to follow-up 1 LVAD placement 1 Death 1 No show baseline and at 6 months (Figure 2). The 1 LVAD placement mean (SD) percent reversible defect dur- 1 No show 27 Assessed for changes in maximal oxygen consumption ing the baseline period was 25.1% 52 Assessed for changes in maximal oxygen 4 Excluded (27.8%) in the BMC group and 11.8% consumption 0 Other 9 Excluded 1 Transplant (20.4%) in the placebo group. At 6 2 Other 1 LVAD placement months, the mean (SD) percent revers- 3 Too ill 1 No show ible defect was 21.3% (26.6%) in the 1 Lost to follow-up 1 Death 26 Assessed for changes in perfusion defect BMC group and 9.2% (9.1%) in the pla- 1 LVAD placement by SPECT cebo group. The difference in the change 1 No show 5 Excluded 2 Other for percent reversible defect between the 50 Assessed for changes in perfusion defect 1 Transplant BMC group and placebo group was not by single-photon emission computed 1 LVAD placement tomography (SPECT) 1 No show statistically significant (−1.2 [95% CI, 11 Excluded −12.50 to 10.12]; P=.84). 4 Other 3 Too ill There were no significant differ- 1 Lost to follow-up ences in the change between the 2 1 Death groups over time for percent total myo- 1 LVAD placement 1 No show cardial defect (−0.9 [95% CI, −5.0 to 3.3]), total defect size (−1.6 [95% CI, FOCUS-CCTRN indicates First Mononuclear Cells injected in the United States conducted by the Cardiovas- −5.1 to 1.9]), or fixed defect size (−0.7 cular Cell Therapy Research Network. [95% CI, −4.8 to 3.4]).

Downloaded From: http://jama.jamanetwork.com/ by a University of Kansas, Participant User on 05/22/2014 BONE MARROW MONONUCLEAR CELLS FOR CHRONIC HEART FAILURE

The small number of patients with- Findings for stroke volume were simi- CD34 was 0.5% to 6.9% (SD, 1.2%). As- out contraindications for MRI (n=17) lar, with a mean (SD) increase of 2.7 suming that differences of 1.96 for SD precluded performing an informative (12.9) mL in the BMC group and a de- or 2.4% are more likely due to biologi- analysis on the MRI data. There were crease of −5.8 (15.2) mL in the placebo cal variability, the effect of differences no significant differences in the change group; this difference was significant (8.4 in CD34 cell level beyond that ex- between the 2 groups in regional wall [95% CI, 2.1 to 14.8]; P=.01). pected due to natural variability was ex- motion (−0.1 [95% CI, −0.30 to 0.14]; In an exploratory analysis, BMC amined, using a 3% level to be conser- P=.47) and LV end-diastolic volume in- therapy was associated with an improve- vative. Every 3% higher level of CD34 dex (2.5 [95% CI, −4.4 to 9.3]; P=.48). ment in maximal oxygen consumption cells was associated with on average a Forty percent of patients in the BMC for patients with number of endothelial 3.0% greater absolute unit increase in group and 47% of patients in the pla- colony-forming cells greater than the me- LVEF in a multiple variable model that cebo group were NYHA class III at base- dian value of 80 (change: 2.5 [95% CI, included age and treatment as predictor line. The decrease over time in the per- 0.16 to 4.88]). However the interaction variables (3.06 [95% CI, 0.14-5.98]; centage of patients in the BMC group test for this assessment was not signifi- P=.04). An analogous computation for who were NYHA class III was statisti- cant (interaction effect size: 2.61 [95% CD133 cells (range, 0.1%-3.6%; cally significant (40% vs 20%; for differ- CI, −0.30 to 5.51]; P=.08). SD=0.62) revealed that every 3% higher ence: 95% CI, 3% to 37%; P=.02); there A regression analysis showed that level of CD133 cells was associated with was no significant difference in the analo- higher CD34 cell or CD133 cell counts on average a 5.9% greater absolute unit gous change for the placebo group. How- were associated with greater absolute increase in LVEF (5.94% [95% CI, ever, when the between-group analysis unit increase in LVEF. The range of 0.35%-7.57%]; P=.04). was applied, this finding was not statistically significant. Similarly, there were Table 1. Patient Baseline Characteristics no significant differences in the change BMC Group Placebo Group P in CCS class (difference in the percent (n = 61) (n = 31) Value change: 0.18 [95% CI, −0.07 to 0.43]; Mean (SD) P=.49), serum BNP levels (regular BNP: Age, y 63.95 (10.90) 62.32 (8.25) .47 −40.3 pg/mL [95% CI, −120.2 to 39.7]; Height, cm 174.50 (8.79) 177.42 (9.60) .15 Weight, kg 91.53 (22.00) 99.99 (24.23) .10 P=.32 and probrain-type natriuretic pep- Body mass indexa 30.10 (6.14) 31.80 (6.60) .23 tide: 150.2 pg/mL [95% CI, −1215.2 to Blood pressure, mm Hg 1515.6]; P=.82), or decrease in the need Systolic 120.59 (19.69) 122.13 (15.78) .71 for antianginal medication between the Diastolic 70.95 (11.18) 74.77 (10.35) .12 2 groups at 6 months (2 in BMC group Heart rate, beats/min and 0 in placebo group; difference in the Mean (SD) 67.90 (10.45) 72.61 (13.60) .07 percent change: 0.04 [95% CI, −0.01 to Median (range) 65.00 (51.00-100.00) 70.00 (49.00-107.00) 0.09]; P=.28). No. (%) Female sex 8 (13.11) 2 (6.45) .49 Subgroup analyses examined the ef- White race 58 (95.08) 30 (96.77) Ͼ.99 fects of demographics, comorbidities Hispanic ethnicity 3 (4.92) 1 (3.23) Ͼ.99 (age, sex, diabetes mellitus, hyperten- New York Heart Association sion, angina, and hyperlipidemia), and classification cell surface markers (CD34 and CD133) I 6 (9.84) 2 (6.45) II 32 (52.46) 14 (45.16) on end point measures. There were no .59 significant differences. Additional out- III 23 (37.70) 15 (48.39) comes were examined for exploratory IV 0 0 purposes. Canadian Cardiovascular (n = 54) (n = 25) Society classification The baseline LVEF was available in 54 I 13 (24.07) 10 (40.00) patients in the BMC group and in 28 pa- II 24 (44.44) 10 (40.00) .45 tients in the placebo group. The mean III 16 (29.63) 5 (20.00) (SD) baseline LVEF was 34.7% (8.8%) IV 1 (1.85) 0 in the BMC group and 32.3% (8.6%) in Qualifying LVEF by echocardiography, (n = 60) (n = 31) .25 the placebo group. At 6 months, the mean (SD), % 32.43 (9.23) 30.19 (7.76) mean (SD) LVEF change was an in- Aspiration to injection time, h (n = 58) (n = 29) crease of 1.4% (5.2%) in the BMC group Mean (SD) 8.95 (1.18) 8.56 (2.22) .28 and a decrease of −1.3% (5.1%) in the Median (range) 9.01 (6.52-11.40) 8.98 (0.22-11.40) placebo group. This difference was sig- Abbreviations: BMC, bone marrow mononuclear cell; LVEF, left ventricular ejection fraction. aCalculated as weight in kilograms divided by height in meters squared. nificant (2.7 [95% CI, 0.3 to 5.1]; P=.03).

Downloaded From: http://jama.jamanetwork.com/ by a University of Kansas, Participant User on 05/22/2014 BONE MARROW MONONUCLEAR CELLS FOR CHRONIC HEART FAILURE

The patients were divided based on significant effect of therapy was seen for cent reversibility on SPECT) for age. No median age of the population (Յ62 the primary end points (LVESV, maxi- significant differences were seen on the years and Ͼ62 years). No statistically mal oxygen consumption, and per- secondary end points or cell product variables in the subgroup analysis, except for those described below. Table 2. Baseline Patient Medical History, Medication Use at Time of Randomization, and Laboratory Evaluations When LVEF was assessed, patients No. (%) of Patientsa aged 62 years or younger showed a statistically significant effect of therapy. Pa- BMC Group Placebo Group P tients in the BMC group demon- (n = 61) (n = 31) Value strated a mean (SD) increase in LVEF Medical history Diabetes 21 (34.43) 16 (51.61) .12 of 3.1% (5.2%) from baseline to 6 Hypertension 49 (80.33) 24 (77.42) .79 months, whereas patients in the pla- Hyperlipidemia 57 (93.44) 29 (93.55) Ͼ.99 cebo group showed a decrease of −1.6% Angina 21 (34.43) 12 (38.71) .82 (6.6%). The difference in the change be- Former or current smoking 46 (75.41) 20 (64.52) .33 tween groups was significant (4.7% Prior myocardial infarction (n = 57) (n = 31) Ͼ.99 [95% CI, 1.0% to 8.4%]; P=.02). 53 (92.98) 29 (93.55) There were no in-hospital events Prior revascularization 51 (83.61) 26 (83.87) Ͼ.99 Prior coronary artery bypass graft surgery 47 (77.05) 25 (80.65) .79 (other than the dissection noted ear- No. of operations lier). One patient died 29 days after 1 33 (70.21) 21 (84.00) BMC delivery due to pump failure, 2 13 (27.66) 4 (16.00) .39 which was deemed unlikely to be as- 3 1 (2.13) 0 sociated with cell therapy. Another pa- Congestive heart failure tient had a myocardial infarction 61 Yes 36 (59.02) 20 (64.52) .66 days after BMC delivery; the infarc- Prior hospitalization 14 (22.95) 9 (29.03) .61 tion did not occur in the targeted in- Asymptomatic carotid disease 11 (18.03) 3 (9.68) .37 History of stroke or transient ischemic attack 8 (13.11) 1 (3.23) .26 jection area, and the patient was dis- Valvular heart disease 18 (29.51) 8 (25.81) .81 charged from the hospital 4 days later. Peripheral vascular disease 13 (21.31) 3 (9.68) .25 There were no rehospitalizations in History of arrhythmia (n = 56) (n = 28) Ͼ.99 either group for percutaneous coro- 29 (51.79) 14 (50.00) nary intervention prior to the 6-month Cardiac pacemaker 42 (68.85) 23 (74.19) .64 visit. Eight patients (3 in the BMC group Implantable cardioverter-defibrillator 3 (4.92) 2 (6.45) Ͼ.99 and 5 in the placebo group) were re- Dual chamber pacing 17 (18.5) 10 (10.9) .81 hospitalized for congestive heart fail- Medication use at time of randomization ACE inhibitor or ARB 37 (60.66) 22 (70.97) .37 ure, with 1 additional patient in the Aldosterone inhibitor 9 (14.75) 8 (25.81) .26 BMC group rehospitalized for acute Aspirin/P2 Y12 53 (86.89) 29 (93.55) .49 coronary syndrome during this same ␤-Blockers 57 (93.44) 30 (96.77) .66 time frame. One patient in the pla- Warfarin 10 (16.39) 4 (12.90) .77 cebo group underwent heart transplan- Digitalis 4 (6.56) 4 (12.90) .44 tation, and 2 other patients (1 in each Diuretics 41 (67.21) 23 (74.19) .63 group) had LV assist device place- Nitrates 39 (63.93) 18 (58.06) .65 ments before the 6-month visit. Statins 44 (72.13) 21 (67.74) .81 Ranolazine 21 (34.43) 3 (9.68) .01 COMMENT Laboratory evaluations, median (range) The CCTRN was developed by the Na- Hemoglobin, g/dL 14.0 (10.0-16.9) 14.3 (12.4-16.6) .21 tional Heart, Lung, and Blood Insti- High-sensitivity C-reactive protein, mg/L (n = 54) (n = 29) .60 1.4 (0.1-37.0) 1.1 (0-86.4) tute to advance cell therapy for pa- Glomerular filtration rate, mL/min/1.73 m2 (n = 58) (n = 29) .96 tients with cardiovascular diseases by 71.2 (29.6-155.4) 70.1 (30.5-107.3) using a collaborative network ap- Brain-type natriuretic peptide, pg/mL (n = 46) (n = 23) .68 proach to facilitate larger studies with 132.0 (16.0-545.0) 105.0 (26.0-140.0) Probrain natriuretic peptide, pg/mL (n = 15) (n=8) .95 wide applicability. The FOCUS-CCTRN 833.0 (50.0-9793.0) 828.0 (103.0-5778.0) trial is the first, to our knowledge, ad- Abbreviations: ACE, angiotensin-converting enzyme; ARB, angiotensin receptor blocker; BMC, bone marrow mono- equately powered study of cell therapy nuclear cell. SI conversion factors: To convert C-reactive protein to nmol/L, multiply by 9.524; hemoglobin to g/L, multiply by 10.0; in patients with chronic ischemic heart natriuretic peptide to ng/L, multiply by 1.0. disease and LV dysfunction (LVEF aUnless otherwise indicated. Յ45%) to be completed in the United

Downloaded From: http://jama.jamanetwork.com/ by a University of Kansas, Participant User on 05/22/2014 BONE MARROW MONONUCLEAR CELLS FOR CHRONIC HEART FAILURE

States. We found no significant differ- Table 3. Bone Marrow Mononuclear Cell Product Characteristics ences in a priori selected primary end Mean (SD) points between patients treated with BMCs and placebo in this first-in-man BMC Group Placebo Group P study that administered 100 million (n = 61) (n = 31) Value ϫ 6 cells via transendocardial injection. The Total nucleated cells/product, 10 99.03 (5.58) 100.03 (0.18) .32 protocol randomized 92 patients from % Viability/product by Trypan blue exclusion 98.56 (1.11) 98.70 (0.89) .52 % of CD34 cells/producta (n = 57) (n = 30) .67 a cohort of 153 patients who provided 2.71 (1.19) 2.60 (0.93) consent, demonstrating the efficiency % of CD133 cells/producta (n = 57) (n = 30) .59 and expertise of network recruiting as 1.21 (0.62) 1.14 (0.48) well as the perceived need by the com- Colony-forming units-Hill/producta (n = 55) (n = 30) .40 munity with heart failure for therapy 109.41 (206.29) 151.33 (244.20) Endothelial colony-forming cells/producta (n = 49) (n = 28) .60 to address this disease. 131.84 (164.62) 156.44 (240.12) Primary end points used in previous Abbreviation: BMC, bone marrow mononuclear cell. celltherapytrialsofheartfailurehavebeen aFour patients either declined to participate or had insufficient product for the biorepository. arbitrarily chosen due to lack of sufficient historical data in stem/progenitor cell trials.Inthesepreviousstudiesofpatients In the present phase 2 study, explor- tion, a meaningful comparison of the with ischemic heart disease, both LVEF atory analyses revealed that LVEF im- results of the 2 studies is difficult. How- (by echocardiography) and myocardial proved in the BMC group compared ever, both CD133 and CD34 cell popu- ischemia (by SPECT) were used to mea- with the placebo group by 2.7%. This lations have been shown to give rise to sure outcomes of interest and suggested difference is in keeping with results endothelial and vascular progenitor improvement.5,28 However,thesetrialsen- from a previous meta-analysis of BMC cells and to secrete chemokines and cy- rolled patients with mostly preserved therapy in patients with chronic ische- tokines capable of recruiting cells and LVEF (ranging from 48% to 56%). The mic heart disease and in smaller, indi- promoting cell survival.34-37 These find- recentlypublishedFOCUS-HFtrialisone vidual trials that evaluated BMC therapy ings support a model in which CD34 of the first reported studies of autologous in similar patients.1,30-32 The modest im- and CD133 cells might improve myo- BMC therapy in patients with ischemic provement in LVEF in our study is con- cardial oxygenation and LV function in heart failure and low LVEF.14 That phase sistent and may be more meaningful in areas of ischemia and/or hibernating 1 trial also demonstrated a lack of im- light of the larger number of patients myocardium; however, further study is provement in the measures that were se- enrolled. needed. lected for the current study (LVESV, To examine this finding further, we In the FOCUS-HF trial,16 maximal maximal oxygen consumption, and per- assessed LVEF with respect to the bone oxygen consumption improved in the cent reversibility on SPECT); however, marrow characteristics of the patient. younger patients and was correlated at the time the CCTRN-FOCUS was de- Improvement in LVEF correlated with with cell function in an exploratory signed,theresultsofFOCUS-HFwereun- the percentage of CD34 and CD133 analysis. In the present study, maxi- known. cells in BMC samples. This correla- mal oxygen consumption improve- Power calculations for the primary tion was based on a central bio- ment was correlated with endothelial end points selected for the current repository assessment of cell surface cell function, which warrants further in- study assumed ambitious improve- markers present in the cell product. vestigation. Additional analyses of cell ments after BMC injections in maxi- Evaluating inherent variability in the function will be forthcoming from the mal oxygen consumption of 5 mL/kg/ cell product may provide mechanistic CCTRN biorepository and may pro- min, in LVESV of 22 mL, and in insight into the relationship between vide further meaningful correlations reversibility on SPECT of 10% based cell characteristics and both patient with outcome measures. on results from a pilot study from Bra- baseline characteristics and clinical Establishment of the biorepository zil.10 Since then, exercise training in outcomes. core laboratory by the CCTRN marks patients with heart failure and low Losordo et al33 recently found a re- an important step forward in under- LVEF (HF-ACTION [A Controlled duction in angina and increased exer- standing the role of cell function in car- Trial Investigating Outcomes of Exer- cise duration with the delivery of au- diac cell therapy. By providing mecha- cise Training]) resulted in only a 0.6 tologous CD34 cells in patients with nistic insight, cell phenotypic and mL/kg/min improvement in maximal refractory angina who had normal functional studies will help to define oxygen consumption using the same LVEF (Ն55%) overall. Because base- meaningful end points for future stud- protocol used in FOCUS-CCTRN.29 line LVEF in patients in the BMC group ies and will aid in selecting patients Defining end points in this field con- (32.4%) in our study represented pa- most likely to receive maximal ben- tinues to be a major challenge. tients with significant LV dysfunc- efits from autologous therapy.

Downloaded From: http://jama.jamanetwork.com/ by a University of Kansas, Participant User on 05/22/2014 BONE MARROW MONONUCLEAR CELLS FOR CHRONIC HEART FAILURE

Although this study enrolled more SPECT sestamibi often underesti- nificant amount of myocardial viabil- patients than previous heart failure mates myocardial viability and revers- ity was noted on electromechanical trials in patients with ischemic heart ibility in patients with multivessel coro- mapping. For this reason and to facili- disease and low LVEF, the sample nary artery disease compared with tate enrollment because many pa- size was still relatively small. The SPECT thallium or positron emission tients were being excluded due to the sample size chosen required large tomography.38-41 After approximately 20 SPECT reversibility criterion, in No- improvements in selected end points months of enrollment in the present vember 2010 the protocol was amended to show a significant treatment effect. study, investigators noted a discrep- during the final third period of enroll- This choice occurred principally due ancy between the amount of reversibil- ment to include patients with any per- to the paucity of data available for ity present on baseline SPECT and the fusion defects (ie, fixed or reversible). these evaluations. A large percentage subsequent finding of viable myocar- This may have skewed the population of the patients in our study had con- dium by electromechanical mapping as to patients with a lesser degree of myo- traindications for MRI, thus preclud- well as the presence of angina. cardial viability, limiting the areas suit- ing a meaningful evaluation of the Even with minimal reversibility (eg, able for cell injection. In addition, the MRI data. 1%) on baseline SPECT sestamibi, a sig- study size precludes any determina-

Figure 2. Changes in Major Outcomes Over Time by Therapy

Change in left ventricular end systolic volume (LVESV) by echocardiography Change in maximal oxygen consumption

160 35

140 30

120 25 100 20 80 15 LVESV, mL LVESV, 60 10 40

20 5

0 Maximal Oxygen Consumption, mL/kg/min 0 Baseline 6 mo Baseline 6 mo Baseline 6 mo Baseline 6 mo

Bone marrow Placebo BMC Placebo mononuclear cells (n = 28) (n = 52) (n = 27) (BMC) (n = 54)

Change in reversible defect by single-photon emission computed tomography Change in global left ventricular ejection fraction (LVEF)

100 60

90 50 80

70 40 60

50 30

40 % LVEF,

Reversibility, % Reversibility, 20 30

20 10 10

0 0 Baseline 6 mo Baseline 6 mo Baseline 6 mo Baseline 6 mo

BMC Placebo BMC Placebo (n = 52) (n = 25) (n = 54) (n = 28)

Solid circles indicate mean values at baseline and 6 months. Error bars indicate 95% confidence intervals.

Downloaded From: http://jama.jamanetwork.com/ by a University of Kansas, Participant User on 05/22/2014 BONE MARROW MONONUCLEAR CELLS FOR CHRONIC HEART FAILURE

tion of the effect of therapy on the oc- Cogle, Smith, Westbrook, Chen, Handberg, Olson, Funding/Support: Funding for this trial was Geither, Bowman, Francescon, Piller, Zierold, Vojvodic, provided by the NHLBI under cooperative agree- currence of community-wide ac- Moye´. ment 5 U01 HL087318-04. It also was supported cepted clinical outcomes (eg, total Analysis and interpretation of data: Perin, Willerson, in part by NHLBI contracts N01-HB-37164 mortality), which must be addressed in Pepine, Henry, Ellis, Zhao, Silva, Lai, Thomas, and HHSN268201000008C awarded to the Kronenberg, Martin, Penn, Hatzopoulos, Gee, Taylor, Molecular and Cellular Therapeutics Facility, Uni- larger forthcoming studies. Although Cogle, Baraniuk, Piller, Simpson, Loghin, Aguilar, versity of Minnesota, and N01-HB-37163 and there was a difference in LVEF in an ex- Richman, Zierold, Vojvodic, Gordon, Ebert, Kwak, HHSN268201000007C awarded to the Cell Pro- Moye´ , Simari. cessing Facility, Baylor College of Medicine, and ploratory analysis, its clinical signifi- Drafting of the manuscript: Perin, Willerson, Pepine, National Center for Research Resources CTSA grant cance is conditional. Ellis, Silva, Westbrook, Geither, Moye´ , Simari. UL1 TR000064 awarded to the University of Critical revision of the manuscript for important in- Florida. The CCTRN also acknowledges its industry CONCLUSIONS tellectual content: Perin, Willerson, Pepine, Henry, partners: Biosafe, Biologics Delivery Systems Group, Zhao, Silva, Lai, Thomas, Kronenberg, Martin, and Cordis Corporation for their contributions of In the largest study to date of autolo- Anderson, Traverse, Penn, Anwaruddin, Hatzopoulos, equipment and technical support during the con- Gee, Taylor, Cogle, Smith, Westbrook, Chen, duct of the trial. gous BMC therapy in patients with Handberg, Olson, Bowman, Francescon, Baraniuk, Role of the Sponsor: The NHLBI had a role in the de- chronic ischemic heart disease and Piller, Simpson, Loghin, Aguilar, Richman, Zierold, sign and conduct of the study; and had a minimal role Bettencourt, Sayre, Vojvodic, Skarlatos, Gordon, Ebert, in the collection, management, analysis, or interpre- LV dysfunction, we found no effect Kwak, Moye´ , Simari. tation of the data; and preparation, review, or ap- of BMC therapy on prespecified end Statistical analysis: Lai, Thomas, Baraniuk, Vojvodic, proval of the manuscript. points. Further exploratory analysis Moye´. NHLBI Project Office Team: Sonia Skarlatos, PhD, Da- Obtained funding: Willerson, Pepine, Penn, vid Gordon, MD, PhD, Ray Ebert, PhD, Wendy Taddei- showed a significant improvement in Hatzopoulos, Taylor, Handberg, Piller, Simpson. Peters, PhD, Minjung Kwak, PhD, and Beckie Cham- LVEF associated with treatment. Our Administrative, technical, or material support: Perin, berlin. findings provide evidence for further Willerson, Pepine, Zhao, Thomas, Kronenberg, Martin, CCTRN Steering Committee Chair: Robert Simari, MD. Anderson, Traverse, Smith, Westbrook, Chen, FOCUS Trial Investigators and Clinical Teams: Min- studies to determine the relationship Handberg, Olson, Bowman, Francescon, Piller, neapolis Heart Institute Foundation: Timothy Henry, between the composition and func- Simpson, Aguilar, Bettencourt, Sayre, Vojvodic, MD, Jay Traverse, MD, David McKenna, MD, Beth Gordon, Ebert. Jorgenson, RN, and Rachel Olson, RN, MS. Cleve- tion of bone marrow product and Study supervision: Willerson, Zhao, Thomas, land Clinic Foundation: Stephen Ellis, MD, Marc Penn, clinical end points. Understanding Hatzopoulos, Simpson, Loghin, Skarlatos, Kwak, MD, PhD, Saif Anwaruddin, MD, James Harvey, MD, these relationships will improve the Simari. Carrie Geither, RN, Mark Jarosz, RN, Cindy Oblak, and Conflict of Interest Disclosures: The authors have com- Jane Reese Koc, MT. Texas Heart Institute: James design and interpretation of future pleted and submitted the ICMJE Form for Disclosure Willerson, MD, Emerson Perin, MD, PhD, Guilherme studies of cardiac cell therapy. of Potential Conflicts of Interest. All of the authors re- Silva, MD, James Chen, RN, Casey Kappenman, ported receiving research grant funding and support Deirdre Smith, RN, and Lynette Westbrook, RN, MS. Published Online: March 24, 2012. doi:10.1001 for travel expenses to meetings for the Cardiovascu- University of Florida Department of Medicine: Carl /jama.2012.418 lar Cell Therapy Research Network (CCTRN) from the Pepine, MD, Barry Byrne, MD, Christopher Cogle, MD, Author Affiliations: Texas Heart Institute, St Luke’s National Heart, Lung, and Blood Institute (NHLBI). Dr David Anderson, MD, John Wingard, MD, Eileen Episcopal Hospital, Houston (Drs Perin, Willerson, Perin reported serving as a consultant to Amorcyte, Handberg, PhD, Tempa Curry, RN, and Diann Fisk, and Silva, Mss Smith and Westbrook, and Mr Teva, Cytori, and Aldagen. Dr Pepine reported receiv- MT. Vanderbilt University School of Medicine: David Chen); School of Medicine (Drs Pepine, Anderson, ing research grants from Amarin Pharma, Amory- Zhao, MD, Antonis Hatzopoulos, PhD, Allen Naftilan, Cogle, and Handberg) and College of Public Health cyte, Angioblast/Mesoblast, Baxter Healthcare, Bayer MD, Sherry Bowman, RN, Judy Francescon, RN, and and Health Professions (Dr Martin), University of Healthcare, Daiichi-Sankyo, GlaxoSmithKline, Lilly, Karen Prater. Florida, Gainesville; Minneapolis Heart Institute at Merck, the National Institutes of Health, the Na- Data Coordinating Center and Laboratory Teams: Abbott Northwestern Hospital, Minneapolis, Min- tional Heart, Lung, and Blood Institute, Pfizer, sanofi- University of Texas School of Public Health: Lem- nesota (Drs Henry and Traverse and Ms Olson); aventis, Viron Therapeutics; serving as a consultant uel Moye, MD, PhD, Lara Simpson, PhD, Linda School of Medicine, University of Minnesota, Min- to Aastrom, Amarin Pharma, Amorycyte (scientific ad- Piller, MD, MPH, Sarah Baraniuk, PhD, Dejian Lai, neapolis (Drs Henry, Traverse, Taylor, and Zierold); visory board), Angioblast Systems (data and safety PhD, Shreela Sharma, PhD, Judy Bettencourt, MPH, Cleveland Clinic Foundation, Cleveland, Ohio (Drs monitoring board), Lilly and County Community Foun- Shelly Sayre, MPH, Rachel Vojvodic, MPH, Larry Ellis and Thomas and Ms Geither); School of Medi- dation (data and safety monitoring board), and SLACK Cormier, Robert Brown, PhD, Diane Eady, Crystal cine, Vanderbilt University, Nashville, Tennessee Inc; and holding 2 patents with the University of Florida. Maitland, MBA, Courtney Ransom, Maybelle Sison, (Drs Zhao, Kronenberg, and Hatzopoulos and Mss Dr Henry reported serving on steering committees for RN, and Michelle Cohen. Baylor College of Medi- Bowman and Francescon); Schools of Public Health and receiving research grants from Aastrom and Me- cine: Adrian Gee, PhD, Sara Richman, and David (Drs Lai, Baraniuk, Piller, Simpson, and Moye´ and soblast. Dr Martin reported receiving research grants Aguilar, MD. University of Texas Medical School: Mss Bettencourt, Sayre, and Vojvodic) and Medi- from the University of Texas to provide an exercise Catalin Loghin, MD. Cleveland Clinic C5 Research cine (Dr Loghin), University of Texas, Houston; testing core laboratory and received financial sup- Imaging Core: James Thomas, MD, Allen Borowski, Northeast Ohio Medical University, Akron (Dr port for travel expenses. Dr Penn reported receiving Annitta Flinn, and Cathy McDowell (echo core Penn); Penn Heart and Vascular Hospital, University research grants from Athersys Inc; serving as a board laboratory). Vanderbilt University School of Medi- of Pennsylvania, Philadelphia (Dr Anwaruddin); member for Juventas Therapeutics; serving as a con- cine: Marvin Kronenberg, MD, Doreen Judd, and College of Medicine, Baylor University, Houston, sultant to Juventas Therapeutics and Aastrom Biosci- Amy Wright (SPECT core laboratory). University of Texas (Drs Gee and Aguilar and Ms Richman); ences; and holding patents, receiving royalties, and Florida Department of Physical Therapy: Daniel National Heart, Lung, and Blood Institute, owning stock in Juventas Therapeutics. Dr Hatzopou- Martin, PhD, PT, Eileen Handberg, PhD (maximal Bethesda, Maryland (Drs Skarlatos, Gordon, Ebert, los reported receiving payment for lectures from Wash- oxygen consumption core laboratory). Center for and Kwak); and Mayo Clinic, Rochester, Minnesota ington University, University of Minnesota, and Ten- Cardiovascular Repair: Doris Taylor, PhD, Claudia (Dr Simari). nessee State University. Dr Taylor reported receiving Zierold, PhD, and Marjorie Carlson (biorepository Author Contributions: Dr Moye´ had full access to all payment for a lecture at an annual meeting of the core in Minnesota). University of Florida Depart- of the data in the study and takes responsibility for Cell Society. Dr Handberg reported receiving grants ment of Medicine: Christopher R. Cogle, MD, and the integrity of the data and the accuracy of the data from Baxter, BDI, Amrin, Angioblast, Neostem, Elizabeth Wise (biorepository core in Florida). analysis. Harvest Technologies, Daiichi Sankyo, Pfizer, Gilead Additional Contributions: We thank the NHLBI Study concept and design: Perin, Willerson, Pepine, Sciences, and GlaxoSmithKline. Ms Olson reported gene and cell therapies data and safety monitoring Henry, Zhao, Silva, Lai, Thomas, Kronenberg, Traverse, receiving research grants from Mesoblast and board and the protocol review committee for their Penn, Hatzopoulos, Westbrook, Handberg, Olson, Aastrom. Ms Geither reported receiving support review and guidance of the FOCUS trial and Baraniuk, Piller, Simpson, Bettencourt, Sayre, Skarlatos, for travel expenses to meetings from the Cleveland Rebecca A. Bartow, PhD, of the Texas Heart Insti- Gordon, Kwak, Moye´ , Simari. Clinic. Dr Aguilar reported being a consultant to tute at St Luke’s Episcopal Hospital for editorial Acquisition of data: Perin, Willerson, Pepine, Henry, Amylin Pharmaceutical in the area of diabetes and heart assistance. Dr Bartow was not financially compen- Zhao, Silva, Anderson, Traverse, Anwaruddin, Taylor, failure. sated for her contribution.

Downloaded From: http://jama.jamanetwork.com/ by a University of Kansas, Participant User on 05/22/2014 BONE MARROW MONONUCLEAR CELLS FOR CHRONIC HEART FAILURE

REFERENCES nary artery diseases (PROTECT-CAD trial). Eur Heart J. 26. SAS Publishing. SAS 9.2 Language Reference 2007;28(24):2998-3005. Dictionary. Cary, NC: SAS Institute; 2009. 1. Abdel-Latif A, Bolli R, Tleyjeh IM, et al. Adult bone 14. Assmus B, Fischer-Rasokat U, Honold J, et al. Trans- 27. Moher D, Schulz KF, Altman D. The CONSORT marrow-derived cells for cardiac repair. Arch Intern coronary transplantation of functionally competent statement. JAMA. 2001;285(15):1987-1991. Med. 2007;167(10):989-997. BMCs is associated with a decrease in natriuretic pep- 28. van Ramshorst J, Bax JJ, Beeres SL, et al. Intra- 2. Beeres SL, Bax JJ, Dibbets-Schneider P, et al. Sus- tide serum levels and improved survival of patients with myocardial bone marrow cell injection for chronic tained effect of autologous bone marrow mono- chronic postinfarction heart failure. Circ Res. 2007; myocardial ischemia. JAMA. 2009;301(19):1997- nuclear cell injection in patients with refractory an- 100(8):1234-1241. 2004. gina pectoris and chronic myocardial ischemia. Am 15. Beeres SL, Bax JJ, Dibbets-Schneider P, et al. In- 29. O’Connor CM, Whellan DJ, Lee KL, et al. Efficacy Heart J. 2006;152(4):684-686, e11-e16. tramyocardial injection of autologous bone marrow and safety of exercise training in patients with 3. Briguori C, Reimers B, Sarais C, et al. Direct intra- mononuclear cells in patients with chronic myocar- chronic heart failure. JAMA. 2009;301(14):1439- myocardial percutaneous delivery of autologous bone dial infarction and severe left ventricular dysfunction. 1450. marrow in patients with refractory myocardial angina. Am J Cardiol. 2007;100(7):1094-1098. 30. Assmus B, Honold J, Scha¨ chinger V, et al. Trans- Am Heart J. 2006;151(3):674-680. 16. Perin EC, Silva GV, Henry TD, et al. A random- coronary transplantation of progenitor cells after myo- 4. Burt RK, Loh Y, Pearce W, et al. Clinical applica- ized study of transendocardial injection of autolo- cardial infarction. N Engl J Med. 2006;355(12):1222- tions of blood-derived and marrow-derived stem cells gous bone marrow mononuclear cells and cell func- 1232. for nonmalignant diseases. JAMA. 2008;299(8): tion analysis in ischemic heart failure (FOCUS-HF). Am 31. Erbs S, Linke A, Adams V, et al. Transplantation of 925-936. Heart J. 2011;161(6):1078-1087, e3. blood-derived progenitor cells after recanalization of 5. Fuchs S, Kornowski R, Weisz G, et al. Safety and 17. Yao K, Huang R, Qian J, et al. Administration of chronic coronary artery occlusion. Circ Res. 2005; feasibility of transendocardial autologous bone intracoronary bone marrow mononuclear cells on 97(8):756-762. marrow cell transplantation in patients with ad- chronic myocardial infarction improves diastolic 32. Hendrikx M, Hensen K, Clijsters C, et al. Recovery vanced heart disease. Am J Cardiol. 2006;97(6): function. Heart. 2008;94(9):1147-1153. of regional but not global contractile function by the di- 823-829. 18. Kissel CK, Lehmann R, Assmus B, et al. Selective rect intramyocardial autologous bone marrow 6. Heeschen C, Lehmann R, Honold J, et al. Pro- functional exhaustion of hematopoietic progenitor cells transplantation. Circulation. 2006;114(1)(suppl): foundly reduced neovascularization capacity of bone in the bone marrow of patients with postinfarction heart I101-I107. marrow mononuclear cells derived from patients with failure. J Am Coll Cardiol. 2007;49(24):2341- 33. Losordo DW, Henry TD, Davidson C, et al. Intra- chronic ischemic heart disease. Circulation. 2004; 2349. myocardial, autologous CD34ϩ cell therapy for refrac- 109(13):1615-1622. 19. Simari RD, Moye´ LA, Skarlatos SI, et al. Devel- tory angina. Circ Res. 2011;109(4):428-436. 7. Lipinski MJ, Biondi-Zoccai GG, Abbate A, et al. Im- opment of a network to test strategies in cardiovas- 34. Asahara T, Murohara T, Sullivan A, et al. Isolation pact of intracoronary cell therapy on left ventricular cular cell delivery. J Cardiovasc Transl Res. 2010; of putative progenitor endothelial cells for angiogenesis. function in the setting of acute myocardial infarction. 3(1):30-36. Science. 1997;275(5302):964-967. J Am Coll Cardiol. 2007;50(18):1761-1767. 20. Willerson JT, Perin EC, Ellis SG, et al. Intramyo- 35. Ciraci E, Della Bella S, Salvucci O, et al. Adult hu- 8. Losordo DW, Schatz RA, White CJ, et al. Intra- cardial injection of autologous bone marrow mono- man circulating CD34ÃLinÃCD45ÃCD133Ã cells can dif- myocardial transplantation of autologous CD34ϩ stem nuclear cells for patients with chronic ischemic heart ferentiate into hematopoietic and endothelial cells. Blood. cells for intractable angina. Circulation. 2007;115 disease and left ventricular dysfunction (First Mono- 2011;118(8):2105-2115. (25):3165-3172. nuclear Cells injected in the US [FOCUS]). Am Heart J. 36. Mackie AR, Losordo DW. CD34-positive stem cells. 9. Martin-Rendon E, Brunskill SJ, Hyde CJ, et al. Au- 2010;160(2):215-223. Tex Heart Inst J. 2011;38(5):474-485. tologous bone marrow stem cells to treat acute myo- 21. Gee AP, Richman S, Durett A, et al. Multicenter 37. Paprocka M, Krawczenko A, Dus D, et al. CD133 cardial infarction. Eur Heart J. 2008;29(15):1807- cell processing for cardiovascular regenerative medi- positive progenitor endothelial cell lines from human cord 1818. cine applications. Cytotherapy. 2010;12(5):684- blood. Cytometry A. 2011;79(8):594-602. 10. Perin EC, Dohmann HF, Borojevic R, et al. Tran- 691. 38. Bonow RO, Dilsizian V, Cuocolo A, Bacharach SL. sendocardial, autologous bone marrow cell transplan- 22. Traverse JH, Henry TD, Ellis SG, et al. Effect of Identification of viable myocardium in patients with tation for severe, chronic ischemic heart failure. intracoronary delivery of autologous bone marrow chronic coronary artery disease and left ventricular Circulation. 2003;107(18):2294-2302. mononuclear cells 2 to 3 weeks following acute myo- dysfunction. Circulation. 1991;83(1):26-37. 11. Perin EC, Dohmann HF, Borojevic R, et al. Im- cardial infarction on left ventricular function. JAMA. 39. Brunken R, Schwaiger M, Grover-McKay M, et al. proved exercise capacity and ischemia 6 and 12 2011;306(19):2110-2119. Positron emission tomography detects tissue meta- months after transendocardial injection of autolo- 23. Zierold C, Carlson MA, Obodo UC, et al. Devel- bolic activity in myocardial segments with persistent thal- gous bone marrow mononuclear cells for ischemic oping mechanistic insights into cardiovascular lium perfusion defects. J Am Coll Cardiol. 1987; cardiomyopathy. Circulation. 2004;110(11)(suppl 1): cell therapy. Am Heart J. 2011;162(6):973-980. 10(3):557-567. II213-II218. 24. Lang RM, Bierig M, Devereux RB, et al. Recom- 40. Dilsizian V, Arrighi JA, Diodati JG, et al. Myocar- 12. Tse HF, Thambar S, Kwong YL, et al. Safety of mendations for chamber quantification. JAmSoc dial viability in patients with chronic coronary artery catheter-based intramyocardial autologous bone mar- Echocardiogr. 2005;18(12):1440-1463. disease. Circulation. 1994;89(2):578-587. row cells implantation for therapeutic angiogenesis. 25. Maret E, Brudin L, Lindstrom L, et al. Computer- 41. Tamaki N, Yonekura Y, Yamashita K, et al. Rela- Am J Cardiol. 2006;98(1):60-62. assisted determination of left ventricular endocardial tion of left ventricular perfusion and wall motion with 13. Tse HF, Thambar S, Kwong YL, et al. Prospective borders reduces variability in the echocardiographic metabolic activity in persistent defects on thallium-201 randomized trial of direct endomyocardial implanta- assessment of ejection fraction. Cardiovasc Ultrasound. tomography in healed myocardial infarction. Am J tion of bone marrow cells for treatment of severe coro- 2008;6:55. Cardiol. 1988;62(4):202-208.

Downloaded From: http://jama.jamanetwork.com/ by a University of Kansas, Participant User on 05/22/2014