A large, multicenter, sham-controlled trial in heart failure showed no benefit at all from stem cell delivery on the primary outcome of recurrent nonfatal decompensated HF events, but the results were still promising, according to the DREAM-HF study’s principal investigator.
Dr. Emerson C. Perin
When added to guideline-directed medical therapy in patients with HF, a single dose of mesenchymal progenitor cells (MPC) significantly reduced major adverse cardiovascular events (MACE) – a composite of cardiac death, nonfatal MI, and nonfatal stroke – and all-cause death in New York Heart Association (NYHA) class II (but not class III patients), reported Emerson C. Perin, MD, PhD, at the American Heart Association scientific sessions.
The problem is that none of these outcomes were included in the primary endpoint, which was recurrent events because of nonfatal decompensated heart failure. On this endpoint, the hazard ratio for events by the end of follow-up was nonsignificantly but slightly increased among those randomized to MPCs rather than sham control (HR, 1.2; P = .406).
“We learned a lot in this trial,” said Dr. Perin, who is medical director of the Texas Heart Institute in Houston, acknowledging that the expectation of benefit on the primary endpoint now appears to have been misplaced, but the positive result on other outcomes opens a new research direction.
With a negative result on the primary endpoint, a benefit on secondary endpoints is considered hypothesis generating. But Dr. Perin defended his sense of overall optimism about the results, because all of the endpoints on which benefit was demonstrated were prespecified. The positive findings “are not from a post hoc analyses,” he emphasized.
DREAM-HF
In the trial, 537 patients with chronic ischemic or nonischemic heart failure with NYHA class II or III symptoms and a left ventricular ejection fraction of 40% or lower were randomized at 51 sites in the United States and Canada. Patients were required to have elevated N-terminal of the prohormone brain natriuretic peptide levels, at least one prior hospitalization for heart failure, and have been on positive inotropic therapy for more than 1 month (but less than 9 months).
The intracardiac administration of MPCs, which are derived from adult human bone marrow, were delivered by injection guided with the NOGA left ventricular electromechanical mapping system. Multiple transendocardial injections were delivered, all in a single session.
There were no differences in baseline characteristics between those receiving MPCs and those who underwent a sham procedure. In both groups, more than half of patients had a previous MI and a coronary revascularization. Nearly 85% had an implanted defibrillator. Roughly two-thirds were in NYHA class III HF and the remaining were in class II.
Over the follow-up, the lines on a graph documenting nonfatal decompensated heart failure events were largely superimposed for the MPC-treated and sham-treated patients, with no significant differences seen over time.
However, the differences on the secondary events were sizable. For the composite outcome of nonfatal MI and nonfatal stroke over a mean follow-up of about 30 months, the rate of events was less than half as great in those randomized to MPCs (4.6% vs. 13.0%). This translated into about 65% reduction in risk (HR, 0.346; P = .001) overall, and the reduction was about the same in class II or III patients.
For a composite endpoint of MACE, events in the group treated with MPCs were about one-third lower than in the sham procedure group (20.3% vs. 30.1%), a difference that also reached significance (HR, 0.667; P = .021).
For this MACE endpoint, response was evaluated by systemic inflammation. For those with a high-sensitivity C-reactive protein (hsCRP) level of less than 2 mg/L, the risk reduction was small and not significant (HR, 0.843; P = .519). Conversely, there was a large risk reduction in those with hsCRP of at least 2 mg/L that did reach statistical significance (HR, 0.551; P = .012).
Inflammation was also found to be a discriminator for time to cardiac death among the patients with NYHA class II HF. Again, there was no benefit among those with hsCRP below 2 mg/L (HR, 1.355; P = .672), but an 80% risk reduction for those with hsCRP of at least 2 mg/L (HR, 0.204; P = .005).
In class II patients with hsCRP at least 2 mg/L, there was also a 60% reduction in all-cause death (HR, 0.401; P = .027). Neither the reduction in cardiac death nor all-cause death was observed in class III HF patients whether or not they had elevated hsCRP.
These signals of benefit provide a direction for a new set of studies, but Dr. Perin said that safety analyses from the DREAM-HF trial are reassuring for further clinical development.
In addition to the fact that “treatment-emergent adverse events and serious adverse events were similar in the MPC-treated and control patients,” Dr. Perin said that MPC administration was not associated with any clinically meaningful immune responses.
MPCs were first injected into a human 15 years ago, according to Dr. Perin. While a phase 2 trial published several years ago did show an association of MPC administration with a reduction in HF-associated events as well as a reduction in adverse ventricular remodeling, the ischemic benefits observed in this trial, particularly in those with elevated hsCRP, provide a new direction for future trials.
“This turns the page in heart failure research. We now have a new mechanism to consider,” Dr. Perin said.
Not so fast, expert says
This might be a reasonable conclusion, but the AHA-invited discussant, Larry Allen, MD, believes there is essentially no clinical message from this trial. He reiterated multiple times that this trial was neutral with no trend for benefit on the primary outcome.
“There was benefit on the secondary outcomes of nonfatal MI or stroke, but these are not the outcomes we follow in heart failure patients,” he said, noting that benefit from regenerative therapy on ischemic events has not been a major focus of the trials that preceded DREAM-HF.
Despite these intriguing results, “patients should understand that stem cells remain experimental,” he said. For the patient, it is “more important to double down on the importance of guideline directed medical therapy,” which is still being administered at levels that are suboptimal, according to Dr. Allen, medical director of advanced heart failure at the University of Colorado at Denver, Aurora.
“Keep up the investment” in the promise of stem cell therapy, he said, but he cautioned that some of the secondary benefits observed in DREAM-HF, such as the greater response in patients with elevated hsCRP, appear to be new observations that will require a great deal more study to validate.
Dr. Perin has a financial relationship with Mesoblast, which provided funding for the DREAM-HF trial. Dr. Allen reported no relevant conflicts of interest.