Cardiovascular disease remains a leading global health burden, yet many preclinical systems still struggle to predict how candidate drugs will perform in the human heart. Rodent models are limited by species differences, while conventional two-dimensional cell cultures cannot fully reproduce the architecture, mechanics, electrical activity, metabolism, and cellular communication of human cardiac tissue. In a review published in Research, researchers from Shanghai University and collaborating institutions highlight how 3D cardiac constructs are emerging as more clinical relevant platforms for cardiovascular drug discovery and safety assessment.

3D cardiac constructs are typically generated from human pluripotent stem cells and assembled through scaffold-based systems, suspension culture, microfluidic devices, or self-organizing protocols. Compared with two-dimensional cultures, they better capture cardiac contraction, electrophysiology, metabolic activity, and multicellular interactions. The review compares major model types, including engineered heart tissues, heart-on-chip systems, scaffold-free cardiac microtissues, and cardiac organoids. Each serves a different purpose: engineered tissues are strong tools for contractility and tissue-level functional testing; heart-on-chip systems allow controlled flow and dynamic stimulation; microtissues support scalable screening; and cardiac organoids are especially useful for modeling early heart development and genetic disease.

For disease modeling, patient-derived induced pluripotent stem cells allow researchers to preserve disease-associated genetic variants and build models of dilated cardiomyopathy, hypertrophic cardiomyopathy, arrhythmogenic cardiomyopathy, and other inherited disorders. Non-genetic conditions can also be modeled by applying metabolic stress, inflammatory factors, environmental toxins, or hypoxic stimulation. These approaches enable 3D cardiac systems to reproduce oxidative stress, inflammation, vascular dysfunction, contractile abnormalities, and other clinically relevant phenotypes, creating more informative settings for drug response test.

The review emphasizes that the value of a 3D cardiac model depends not only on whether it resembles heart tissue, but also on whether it provides stable and quantitative functional readouts. Key measurements include electrophysiology, action potentials, calcium signaling, contractility, myocardial metabolism and genetic profiling. Technologies such as multielectrode arrays, patch-clamp recording, optical mapping, calcium imaging, contractility analysis, extracellular flux analysis, and multi-omics profiling can help determine whether a drug affects rhythm, contraction, energy use, or cardiotoxicity. However, many current assays were originally designed for two-dimensional systems, and the field still needs standardized functional platforms tailored to intact 3D tissues.

Biomaterials and artificial intelligence are accelerating the transition of 3D cardiac constructs from experimental models to drug discovery platforms. Conductive materials can improve electrical coupling between cardiomyocytes; hydrogels provide matrix-like support; and 3D bioprinting using biocompatible and supportive materials allows spatial organization of multiple cell types. AI is increasingly being used for image-based phenotyping, cell-type annotation, and model design. Because 3D cardiac systems can be costly and time-consuming to generate, AI-assisted workflows may help narrow candidate compounds before functional validation and accelerate drug discovery.
Despite this progress, important barriers remain. Many stem cell-derived cardiomyocytes still resemble fetal or neonatal cells rather than adult myocardium, affecting sarcomere organization, calcium handling, metabolism, electrophysiology, and drug responses. Limited vascularization restricts construct size, long-term culture, and cell survival in deeper regions. Reproducibility also remains a challenge because results can vary with cell source, differentiation efficiency, culture batch, cell-type ratio, and assay workflow. The review argues that 3D cardiac constructs should complement, rather than replace, existing animal and in vitro models. Their broader adoption will depend on improved maturation, vascularization, immune-cell integration, standardization, scalable manufacturing, cost control, and clearer regulatory pathways.

The complete study is accessible via DOI:10.34133/research.1165