By Jeffrey Zimmet, MD, PhD

Associate Professor of Medicine, University of California, San Francisco; Director, Cardiac Catheterization Laboratory, San Francisco VA Medical Center

Dr. Zimmet reports no financial relationships relevant to this field of study.

SYNOPSIS: Investigators tested the mortality benefit of intracoronary bone marrow cells in patients with successfully reperfused acute myocardial infarction. They observed no effect on mortality.

SOURCE: Mathur A, Fernández-Avilés F, Bartunek J, et al. The effect of intracoronary infusion of bone marrow-derived mononuclear cells on all-cause mortality in acute myocardial infarction: The BAMI trial. Eur Heart J 2020; Aug 30;ehaa651. doi: 10.1093/eurheartj/ehaa651. [Online ahead of print].

Using a patient’s own bone marrow mononuclear cells as therapy for ischemic myocardial injury has been a recognized area of study for longer than 20 years. Although small proof-of-concept trials, such as TOPCARE-AMI, made headlines in the early 2000s, no large-scale, Phase III trial had been completed to bring this therapy into the mainstream.

Based in part on the promising results of the Phase II REPAIR AMI study, (N Engl J Med, 2006), the Phase III BAMI trial was designed to demonstrate the safety and efficacy of intracoronary autologous bone marrow mononuclear cells (BM-MNC) in post-MI patients. To this end, patients with ejection fraction < 45% after successful primary PCI for acute MI were enrolled. The treatment group received intracoronary infusion of autologous bone marrow mononuclear cells between two and eight days after the MI event, in addition to standard medical therapy. The control group received medical therapy alone in an open-label design. The trial was powered to detect a 25% reduction in all-cause mortality at a follow-up of two years. It was designed in 2011, and enrollment took place between September 2013 and October 2017. The authors planned to enroll 3,000 patients from at least 17 sites in 11 European countries. The European Commission partly funded the trial, and work was coordinated from the BAMI trial office in London.

Shortfalls in funding and enrollment became evident early. In particular, numbers of acute MI patients with ejection fraction ≤ 45% were markedly less than anticipated. The steering committee reimagined the trial as an observational study. Although 37 sites were enrolled in the study, only 28 contributed patients, and 23 actually delivered the experimental therapy. Ultimately, the authors enrolled 375 patients, of whom 185 received BM-MNC and 190 were randomized to control. Average age was 59 years, more than 80% were men, and more than 90% were white. The mean ejection fraction was 39%, as measured by a core echo laboratory. More than 97% of patients had Killip class I or II MI.

The original power calculations assumed a 12% mortality rate, based on early data from the primary PCI era. Among enrolled patients at two years, there were only six deaths in the BM-MNC group, and seven deaths in the control group, which was not statistically different. Among the secondary endpoints, only rehospitalization due to heart failure demonstrated an advantage among the treated patients, with five BM-MNC patients and 15 control patients recording an event (2.7% vs. 8.1%; hazard ratio, 0.332; 95% confidence interval, 0.12-0.88). The remaining secondary endpoints were not different between the groups, including rehospitalization due to MI, revascularization, ICD implant, and stroke.

The authors concluded the sample sizes and event rates were too low to make any meaningful group comparisons. They estimated future trials would need a prohibitively large number of patients (> 10,000) to demonstrate a treatment effect at the observed average mortality rate of 3.5%.


With two decades and a multitude of clinical trials, what have we to show for these investigations into BM-MNC after acute MI? The authors of earlier trials all have used surrogate endpoints, such as LV function and infarct size, and have shown disparate results. TOPCARE-AMI (Clin Res Cardiol, 2011), BOOST (Eur Heart J, 2009), BONAMI (Eur Heart J, 2017), and REPAIR AMI authors all reported improvements in LVEF, while no significant changes were observed in LEUVEN-AMI (Lancet, 2006), ASTAMI (N Engl J Med, 2006), REGENT (Eur Heart J, 2009), or HEBE (Eur Heart J, 2011). The largest of these, REPAIR AMI, included 202 patients. Those authors reported intracoronary infusion of progenitor cells led to an improvement in LV function as well as a reduction in the composite of death, recurrence of MI, and subsequent revascularization.

The Mathur et al trial was nearly twice as large, and yet did not demonstrate any such benefits. One lesson appears to be that the landscape of acute MI has changed over time. With aggressive use of primary PCI, and current medical therapy, mortality after AMI is significantly lower than it used to be. Even identifying significant numbers of post-MI patients with ejection fraction ≤ 45% turned out to be challenging.

For now, the writing appears to be on the wall: Subsequent trials of BM-MNC in post-MI patients who have undergone successful reperfusion are impractical and are unlikely to be successful. Further trials of cell-based therapies are likely to focus on sicker HFrEF patients who continue to have high rates of clinical endpoints and for whom demonstration of benefit (or lack thereof) is more likely.