P5385Development and preclinical testing of upscaled engineered heart tissue for use in translational studies

Abstract

Abstract Introduction The lack of efficacy of stem cell therapy for the treatment of heart failure may be related to the poor retention rates offered by existing delivery methods (intra-coronary/ intramyocardial). Tissue engineering strategies improve cell retention in small animal models but data regarding engineered heart tissue (EHT) patches large enough for human studies are lacking. Purpose To upscale EHT to a clinically relevant size and mature the patch in-vitro. Once matured to undergo preclinical testing in a rabbit model of myocardial infarction. Methods We developed an upscaled EHT patch (3cm x 2cm x 1.5mm) able to contain up to 50 million human induced pluripotent stem cell derived cardiomyocytes (hiPSC-CM; Fig A/B). Myocardial infarction model was performed by permanent ligation. Results The patches began to beat spontaneously within 3 days of fabrication and after 28 days of dynamic culture (Late EHTs) showed the development of several mature characteristics when compared to early patches (<14 days from fabrication). For example, late EHTs contained hiPSC-CMs which were more aligned (hiPSC-CM accumulative angle change: early 2702±778 degrees [n=4] vs late 922±186 [n=5], p=0.042); showed better contraction kinetics (early peak contraction amplitude 87.9±5.8a.u. versus late 952±304a.u.; p<0.001) and faster calcium transients (time to peak: early 200.8±8.8ms [n=5] vs late 147.7±10.2ms [n=6], p=0.004; time to 75% decay: early 274±9.7ms vs late 219.9±2.7ms, p=0.0003). We then tested the EHT patch in-vivo using a rabbit model (Fig C). Patches were applied to normal (n=5) or infarcted hearts (n=8). Sham operations used non-cellular fibrin patches (n=5). The mean fraction of troponin positive cells in the graft was 27.8±10.3% at 25.2±1.7 days relative to day 0 [n=5] and KU80 (human specific marker) staining confirmed that this was of human origin. CD31 (Fig D) and KU80 staining revealed that the grafts were well vascularized and that the vasculature was not human in origin (therefore were originating from the host). Ex-vivo optical mapping revealed evidence of electrical coupling between the graft and host at 2 weeks and preliminary experiments indicated that the patch improved left ventricular function when grafted onto infarcted hearts. Telemetry recordings in vivo and arrhythmia provocation protocols (ex vivo) indicated that the patch was not proarrhythmic. Figure 1. A/B) EHT Images; C) 20x troponin T (brown) of rabbit myocardium/EHT (2 weeks after grafting), blue counterstain = haematoxylin, red lines = EHT borders; D) 63x CD31 staining (brown) rabbit/EHT border zone (2 weeks after grafting), blue stain = haematoxylin, red lines = graft/host border zones. Conclusion We successfully upscaled hiPSC-CM derived EHT to a clinically relevant size and demonstrated feasibility and integration using a rabbit model of myocardial infarction. Tissue engineering strategies may be the preferred modality of cell delivery for future cardiac regenerative medicine studies.