In Heart Failure, Cellular Therapy Is Active

CHICAGO — For patients with advanced heart failure, an injection of autologous mononuclear cells into damaged myocardia missed the primary endpoint in the sham-controlled, randomized, blinded CardiAMP-HF trial, but there was a strong signal of benefit, supporting further assessment.

The study had a hierarchical win-ratio design, and the intervention met two of its three endpoints and suggested clinical activity on other measures, providing the motivation for another study, which has already received US Food and Drug Administration (FDA)–approval and will start this summer, according to Amish N. Raval, MD, director of Clinical Cardiovascular Research at the Stem Cell & Regenerative Medicine Center, University of Wisconsin–Madison.

The sham-controlled, randomized trial builds on a 1src-patient roll-over feasibility study that was published in 2src21. That study and other work provided the basis for the CardiAMP-HF trial, which Raval presented at the American College of Cardiology (ACC) Scientific Session 2src25.

Cell Potency Screening Might Be Needed

In the feasibility trial, mononuclear cells from bone marrow aspirate that has high numbers of specific cell types, such as CD34+ positive cells, produced greater improvement in heart failure symptoms. As a result, candidates for the CardiAMP-HF trial were screened for cell potency.

On a win ratio, protection against events in the first of three hierarchical endpoints — a composite of all-cause mortality, cardiac transplant, and implantation of a left ventricular assist device — was numerically greater with mononuclear injection than with the sham treatment.

Mononuclear injection also had a numerical advantage over sham treatment for the second hierarchical endpoint, which was a composite of myocardial infarction, stroke, and heart failure hospitalization. However, because mononuclear injection was numerically inferior to the sham treatment for the third hierarchical endpoint — change from baseline in the 6-minute walk test — the study’s primary endpoint, which required all three endpoints to show an advantage, was missed.

Despite a substantially greater number of wins for the first and second hierarchical endpoints, the final tally of wins, due to the inferior effect on walk test, was similar in the two groups, with no statistical advantage for the cell-therapy group (win ratio, 1.src1; P=.src954).

The data safety monitoring committee, which became aware toward the end of the trial that the primary endpoint would not be reached, terminated the study early, but Raval provided several sets of data to suggest that this treatment is active.

Cell Therapy Outperformed Sham for All Hard Endpoints

Events were lower in the cell-therapy group than in the sham group for every hard endpoint measured, including all-cause death, heart transplantation, myocardial infarction, and a combination of major adverse cardiac events. Of these, the total rate of death and adverse events was lower in the cell-therapy group than in the sham group (2src.3% vs 31.7%; P=.17).

Also, when the Minnesota Living With Heart Failure Questionnaire — a prespecified secondary outcome — replaced the walk test as the third hierarchical endpoint, the overall win ratio climbed to 1.39 in favor of cell therapy, bringing the numerical advantage much closer to statistical significance (P=.14src).

In a post hoc analysis of the subgroup of patients who had a baseline level of N-terminal pro B-type natriuretic peptide (NT-proBNP)> 5srcsrc pg/mL, cell therapy had a statistically significant advantage over sham therapy (P=.src2).

In retrospect, Raval said he questions the use of the walk test as an endpoint when it was also a screening criterion for trial entry. The change in walk test over the course of the study was unexpectedly variable, he said, suggesting that this might have been a study confounder.

There were no device-related or procedure-related events associated with death or major adverse events. Three pericardial perfusions were observed in the cell-therapy group, but all were successfully drained without major sequelae, Raval reported.

A small sample of bone marrow was taken from the hip of trial candidates and then screened for favorable characteristics, including cell potency, in accordance with the study protocol.

Once candidates were selected for the trial, a larger amount of bone marrow was harvested and then prepared with BioCardia’s CardiAMP cell-therapy system for injection at the point of care. This system has been designated a breakthrough therapy by the FDA.

Patients were randomized in a 3:2 ratio at 18 study sites in the United States and Canada. Complete 2-year follow-up data were available for 72 of the 74 patients (97.3%) in the cell-therapy group and for 39 of the 41 patients (95.1%) in the sham group. In the cell-therapy group, 8-1src injections were made into the myocardia. During the procedure, sham patients, who did not receive any injections, wore headphones, and underwent a scripted protocol to maintain the blinding.

Candidates for the trial were required to have a left ventricular ejection fraction of at least 2src%, but no> 4src%, and to be in New York Heart Association class II or III heart failure, despite receiving optimized guideline-directed medical therapy.

Not All Patients With Heart Failure Are Candidates

Of the 3src3 candidates screened, 115 (38%) were enrolled. Of the 188 who were deemed ineligible for the study, 72 (38%) were excluded on the basis of inadequate cell potency, and 3src (16%) met the exclusion criteria, such as having inadequate renal function — an estimated glomerular filtration rate <3src mL/min per 1.73 m² — an arrhythmia, or a low hematocrit level.

Despite the negative primary endpoint, “the results over 2 years suggest that autologous mononuclear cells processed and delivered intramyocardially may be beneficial,” Raval said. This is the justification for the planned Cardia-AMP-HF II trial.

Yet in response to several questions posed by Mary Walsh, MD, medical director of the St. Vincent Cardiovascular Research Institute in Indianapolis, who was the ACC-invited discussant, Raval acknowledged that autologous cell therapy is likely to be viable for only a subset of patients with severe heart failure.

In addition to the 38% of patients who were excluded from the CardiAMP-HF trial because their bone marrow aspirate characteristics were predictive of an inadequate response, post hoc evidence that elevated levels of NT-proBNP might be needed to increase the likelihood of benefit suggests that numerous criteria might need to be met by candidates.

Yet Walsh, like Raval, agreed that the data support efforts to further develop this therapeutic approach for patients who qualify.

For patients with advanced functional heart failure impairments despite optimized guideline-directed medical therapy, “we have little left to offer,” said Walsh. More studies are needed to confirm a clinically meaningful benefit, but she is intrigued by the “elegant” design of the CaridAMP-HF trial and is hopeful that the next series of studies will provide more detailed information about the patients most likely to benefit and confirm a viable new therapeutic option.

“We really need new therapies,” Walsh said. Given positive reports from the earlier studies, patients are aware of this resea