Autologous atrial appendage micrografts transplanted during coronary artery bypass surgery: design of the AAMS2 randomized, double-blinded, and placebo-controlled trial

Abstract

Background: The AAMS open-label clinical study demonstrated the safety and feasibility of epicardial transplantation of autologous right atrial appendage micrografts (AAMs) during coronary artery bypass grafting (CABG) surgery. The study also provided the first indications of therapeutic efficacy of the AAMs, as delivered within an extracellular matrix patch, to reduce ischemic scar and increase viable ventricular wall thickness. To further evaluate the initial beneficial effects observed in the AAMS study, we designed the randomized, double-blinded, and placebo-controlled AAMS2 trial. Focusing on patients with ischemic heart disease (IHD) and myocardial scar, the AAMS2 trial aims to generate state-of-the-art structural and functional imaging data of the myocardium treated with an AAMs-patch during CABG.

Methods: The AAMS2 trial recruits IHD patients who are set to undergo non-urgent CABG and present with an ischemic myocardial scar in preoperative cardiac magnetic resonance imaging (CMRI) with late gadolinium enhancement. Patients are randomized (1:1) to receive a collagen-based matrix patch (Hemopatch®), with or without AAMs, epicardially onto the scar border. The primary endpoint, assessed by CMRI preoperatively and at 6 months post-operative follow-up, focuses on the left ventricle scar mass. The secondary endpoints center on the change in scar mass by the AAMs-patch site and evaluation of therapy safety and feasibility as well as its effects on myocardial structure and function by echocardiography. Change in blood N-terminal-pro-BNP levels in the timeframe is the co-primary endpoint.

Discussion: Data from the AAMS2 trial provides the first randomized, blinded, and placebo-controlled evaluation of efficacy on epicardial AAMs transplantation for ischemic myocardial scar. This data will pave the road towards rational design of larger AAMs therapeutic efficacy-addressing trial(s).

Trial registration: ClinicalTrials.gov, NCT05632432, registered 30 November 2022, https://clinicaltrials.gov/study/NCT05632432 .

Keywords: AAMs-patch; Atrial appendage; Cardiac surgery; Cell therapy; Epitranscriptomics; Heart failure; Ischemic heart disease; Micrografts; Tissue-engineering.

Fig. 1

The results of the clinical AAMs-patch open-label pilot study. The preclinical results of an epicardial AAMs-patch transplantation are presented in Fig. 4. It prompted an open-label study, focusing on ischemic HF patients with a myocardial scar in preoperative CMRI, which supported the feasibility and safety of AAMs-patch transplantation as an intraoperative therapy adjuvant to CABG in a clinical setting [20]. This study revealed the median time for the AAMs-patch setup to be 33 min, while only minutes were needed for the attachment onto the epicardium. The 6-month follow-up found good tolerability of the AAMs-patch. Moreover, in the 6-month follow-up CMRI records, the AAMs-patch was intractable; however, the transplant area showed significant increase in the live ventricular wall thickness as compared to the baseline [20]. An indicative trend for reduction in scar mass by the site was also found. The results support a larger randomized, double-blinded, and placebo-controlled clinical trial described herein. AAMs, atrial appendage micrografts; AAMs-patch, collagen-based matrix patch encasing atrial appendage micrografts; CABG, coronary artery bypass grafting; CMRI, cardiac magnetic resonance imaging with late gadolinium enhancement; LVEF, left ventricular ejection fraction. The CMRI results are reprinted as minutely modified from Frontiers in Cardiovascular Medicine, 2021 Nummi A, Mulari S, Stewart JA, et al. Epicardial Transplantation of Autologous Cardiac Micrografts During Coronary Artery Bypass Surgery (2021) [20] with permission from publisher under the Creative Commons Attribution License (CC BY 4.0)

Fig. 2

Outline of the AAMS2 trial. The main inclusion criteria (“ECHO”), screening failure criteria (“CMRI”), and the primary (“CMRI”) and the co-primary (“NT-proBNP”) endpoint of the trial are shown in red. In addition to transthoracic echocardiography (“ECHO”), which is performed: (i) at recruitment, (ii) preoperatively, and postoperatively at (iii) hospital discharge and (iv) at 3-month follow-up, transesophageal echocardiography is also done by the perfusion anesthesiologist at the beginning of CABG surgery to assess RAA anatomy and any presence of sludge. Clinical metadata (“OTHER”) include the recording of Framingham cardiovascular risk factors, electrocardiogram, medication with changes, MACCE, quality-of-life numeration with the SF-36 questionnaire, and a survey of dyspnea and angina pectoris symptoms with NYHA and CCS classifications. The RNA-stabilized study blood samples, focusing on adenosine-based epitranscriptomic profile characterization, are collected as previously published [23]. Abbreviations: AAMs, atrial appendage micrografts; AAMs-patch, collagen-based matrix patch encasing atrial appendage micrografts; CABG, coronary artery bypass grafting; CCS, Canadian Cardiovascular Society (angina pectoris grading classification); CMRI, cardiac magnetic resonance imaging with late gadolinium enhancement; LVEF, left ventricular ejection fraction; NT-proBNP, N-terminal pro-B-type natriuretic peptide; NYHA, New York Heart Association (dyspnea grading classification); MACCE, major adverse cardiac and cerebrovascular events; RAA, right atrial appendage; SF-36, 36-item short form survey (standardized questionnaire for assessment of overall quality of life); 6MWT, 6-min walking test

Fig. 3

Power analysis for the AAMS2 trial. The power analysis is based on the unblinded and non-randomized data from the open-label safety and feasibility study of the perioperative AAMs-patch method (n = 6, AAMs-patch-treated; n = 5 controls without patch) [20]. As calculated by the POWER Procedure Wilcoxon-Mann–Whitney Test with fixed scenario elements O’Brien-Castelloe approximation method and two-sided statistical evaluation, the change (Δ) in infarction scar area (%), and mass (g), as evaluated from the 5SD CMRI imaging data preoperatively and 6 months postoperatively at the AAMs-patch transplantation site, the total sample size of 50 (two groups, group size 25, distribution 1:1), yields a power greater than 80% at an α level of 0.05. Analysis was carried out with SAS 9.4 TS Level 1M4 software (SAS Institute Inc., Cary, NC, USA). The indicative effect size on fibrosis is shown in Fig. 1 [20]. A trend for a decrease in the change of NT-proBNP levels was found in those patients receiving an AAMs-patch [20]. A Evaluation using change in infarction area percentage (%). B Evaluation using change in infarction area mass (g). C Evaluation using change in NT-pro-BNP circulatory concentrations (plasma sample analysis preoperative vs. 6 months postoperatively). The red line represents 80% power at a total sample size of 50. A detailed consideration of the nature of the power analysis is provided at the end of the Discussion. α, level of type I error (α-error level representing the proportional level for false-positive result assessment); CMRI, cardiac magnetic resonance imaging with late gadolinium enhancement; NT-proBNP, N-terminal pro-B-type natriuretic peptide; 5SD, 5-standard deviation

Fig. 4

Preclinical AAMs-patch therapy effects in a myocardial ischemia model. A In our preclinical mouse study with LAD-ligation-induced MI and ischemic HFrEF. B Even transplantation of only an epicardial patch without the AAMs attenuated scarring and persistently salvaged myocardial function, which was further enhanced when AAMs were included (AAMs-patch) [16]. C The site-specific untargeted proteomics revealed the molecular-level blueprints and AAMs’ putative mechanisms of action. This analysis revealed that below the AAMs-patch transplantation, in comparison with the patch-only transplantation, more than 200 proteins were expressed differentially in the injured heart, which were associated with upregulated cell viability, protein synthesis, muscle formation, angiogenesis, and glycolysis (a metabolic shift associated with an enhanced regenerative ability [–68]), while the attenuation of inflammation, oxidative stress, and cell death were noted in tandem [16]. The interventricular septal areas also showed significant associations for cell viability. AAMs, atrial appendage micrografts; LAD, left anterior descending (coronary artery); MI, myocardial infarction. Subplot (B) is reprinted as minutely modified from the Journal of Heart and Lung Transplantation, 39/7, Xie Y, et al. Epicardial transplantation of atrial appendage micrograft patch salvages myocardium after infarction, 707–718 (2020), with permission from the publisher (Elsevier) under the Creative Commons Attribution–NonCommercial–NoDerivs (CC BY-NC-ND 4.0 DEED) license

Fig. 5

Possible mechanism of epicardial AAMs–patch transplantation. Upper panel. Studies with species that regenerate their hearts endogenously well have recently shown the instrumentality of cardiac tissue macrophages (CTMs) in the process. They are of early embryonic origin, while the bone marrow, monocyte-derived macrophages command the early healing phases from the abrupt inflammation to rapid fibrosis. Classically, the phenotypes governing the phases have been denoted as proinflammatory (M1) and anti-inflammatory (M2). They operate while the CTMs activate upon necrosis to traverse the subendothelial space to form a niche with the injury-activated epicardium [70, 71]. Once the fibrosis is established, the CTMs and epicardium activate regeneration, a process experimentally linked to the metabolic rewiring of the cardiomyocytes by the scar border [72], which has been modelled to be dictated by the CTMs [73]. Several factors have been described for this sequence (see Discussion) [64]. Lower panel. In humans, like in most adult mammals, myocardial healing after its necrosis is virtually fully halted at fibrosis. Studies on CTM biology have shown them to deplete by age, especially from the subepicardial compartment [69], which might be one reason for the halted cardiac regeneration in adult mammals, although it is seen in neonate humans and mice pups. Zebrafish hearts also start to fail to regenerate after repeated injuries, which might reflect CTM depletion [74]. The preclinical data (Fig. 4) [16], and the first indication of increased live myocardium, with a trend for a reduced scar mass by the AAMs-patch site clinically (Fig. 1) [20], support a model of AAMs-patch transplantation to provide a reparative niche for the scarred and aged adult human myocardium. Here, the inevitable necrosis of some AAMs can be argued to activate the adjacent epicardium alike the CTMs within them. This design awards the molecular and cellular pathophysiological rationale to assess the AAMs-patch therapy effects on the scar in this trial. AAMs, atrial appendage micrografts; ANGPTL4, Angiopoietin-like 4; IL-6, interleukin 6; CLCF1, Cardiotrophin-like cytokine factor 1; Csf1a, (Macrophage) colony stimulating factor 1a; CTM, cardiac tissue macrophage; CXCL12, CXC motif chemokine 12, a.k.a. stromal cell-derived factor 1 (SDF-1); FSTL1, Follistatin-related protein 1; OSM, oncostatin M; RAA, right atrial appendage; RA, retinoic acid; TXA2, Thromboxane A2