Three-dimensional spheroid models for cardiovascular biology and pathology

Abstract

Scaffold-free three-dimensional (3D) cellular spheroid cultures better replicate the in vivo cellular microenvironments of complex tissues than traditional two-dimensional (2D) cell cultures, as they promote more intricate cell-cell and cell-extracellular matrix (ECM) interactions. In the context of cardiovascular research, 3D spheroids have emerged as valuable models for studying angiogenesis, modeling the cardiac microenvironment, and advancing drug development and cardiac tissue repair. Given that cardiovascular disease remains the leading cause of morbidity worldwide, exploring 3D spheroids as in vitro models in cardiovascular research holds potential for advancing the field. Despite their promise, the experimental potential of 3D spheroids in cardiovascular disease and biology has yet to be realized. Therefore, this review discusses the advantages and limitations of 3D spheroid models for studying angiogenesis and cardiovascular pathobiology, their applications in cardiac drug development and tissue repair, and how these models can advance cardiovascular research.

Graphical abstract

3D spheroids incorporate various cell types to model in vivo cardiovascular structures and pathologies.

1. Introduction

For decades, two-dimensional (2D) cell culture models have provided critical insights into molecular mechanisms regulating cell function in vivo, advancing our understanding of normal cellular processes and diseases progression while contributing to therapeutic development.¹ However, 2D systems have inherent limitations in replicating the complex microenvironment of living tissues.1, 2, 3, 4, 5 Specifically, they lack essential cell-cell and cell-extracellular matrix (ECM) interactions that regulate essential biological functions, including cell morphology, polarity, differentiation, migration, proliferation, survival, and responses to mechanical and chemical stimuli.^2,6 Consequently, data obtained from 2D cultures often differ from outcomes observed in animal studies and clinical trials.^7,8 Overall, these limitations restrict the study of disease mechanisms and translational relevance, particularly in cardiovascular disease (CVD) research (Table 1).

Table 1.

An overview comparing 2D and 3D cell culture across various cellular aspects.

	2D Cell Culture	3D Spheroid Culture	References
Cell Morphology	Do not mimic in vivo tissue structures causing changes in cell morphology	Replicate in vivo-like tissue structure and cell morphology due to their 3D organization	2,8,33,111
Cell Organization	Cellular monolayer	Three distinct cell layers (Fig. 1): a central necrotic core, an inner quiescent layer, and an outer proliferating layer	7,8,39,98,99,105
Cell Behavior	Active cell division, dynamic changes in gene expression, and loss of cell polarity	Preserved cell division and polarity with gene expression and cellular biochemistry closely mimicking in vivo cell behaviors	8,10,28,32,76,81,91,92,105,109,111
Cell Phenotype & Differentiation	Loss of diverse phenotypes by induced or prevented differentiation	Diverse phenotypes as seen in vivo	2,10,33,39,81,111
ECM Microenvironment	Loss of cell-matrix interactions crucial for differentiation, proliferation, and cell signaling	In vivo-like 3D microenvironment; more complex and physiologically relevant cell-ECM dynamics	4,10,32,75, 76, 77, 78,140
Cell-Cell Interactions	Cell-cell interactions only between neighboring cells on the horizontal plane	Multidirectional cell-cell interactions among heterogeneous cell populations allowing for in vivo-like modeling of cell communication	4,9,10,33,39,76,98,105
Cell Culture Quality	2D cell culture methods provide consistent reproducibility, simple protocols, and low cost	Reproducibility may be inconsistent; however, various established culture methods are available, and costs remain low for most spheroid systems	2,8,38,78
Predictive Accuracy	Uniform access to nutrients and therapeutic treatments in 2D systems often leads to overestimation of drug sensitivity and effectiveness	Nutrient and oxygen gradients in 3D systems influence cell metabolism and drug susceptibility, yielding more in vivo-like drug responses	2,13,14,36, 37, 38,44,137,144
Disease Fidelity	Lack of tissue architecture and limited cell-cell and cell-ECM interactions restrict pathophysiological signaling in 2D cultures	3D models better mimic tissue structure and preserve pathological cell-cell and cell-ECM interactions, consistent with in vivo disease models	8,32,36,37,44,90,137, 138, 139,144

To address these challenges, there has been increased attention on developing reproducible and physiologically relevant three-dimensional (3D) cell culture systems, such as spheroids, that better replicate in vivo conditions. Studies have shown that 3D cell culture systems more accurately recapitulate the in vivo microenvironment, particularly with respect to cellular and ECM organization, signaling activity, drug sensitivity, and the preservation of native cell shape and phenotype.9, 10, 11, 12 3D organization of cells creates gradients of oxygen, nutrients, and waste, resulting in heterogenous cell populations and responses that more closely reflect in vivo conditions. These gradients influence cellular metabolism and drug susceptibility, providing a more predictive platform for evaluating drug efficacy and therapeutic resistance, particularly in cancer research compared to 2D culture methods.^13,14 Moreover, the 3D architecture enables cells to migrate into the ECM and establish essential cell-cell and cell-ECM interactions in all directions. In contrast, 2D monolayer cultures allow contact only along the rigid surface of the underlying substratum and between horizontal neighboring cells.¹¹ Additionally, the increased complexity of interactions in 3D models significantly impacts key cellular processes, including proliferation, differentiation, and survival.^4,15 Given these advantages (Table 1), 3D cell culture has become an invaluable research tool, offering in vitro systems that better represent the in vivo microenvironment and preserve cellular function, thereby expanding research possibilities.16, 17, 18

Given that CVD remains the leading cause of death worldwide, innovative experimental approaches and model systems are essential for advancing our understanding of its pathogenesis and for improving therapeutic strategies. 3D models, such as spheroids, are increasingly used to study CVD, including its underlying pathology and vascular remodeling.19, 20, 21 This review focuses on the application of 3D spheroid models in cardiovascular research, with an emphasis on their potential in drug development, tissue repair, and mechanosensation studies.

2. Overview of 3D spheroid models

3D cell culture techniques vary in complexity, including scaffold-based, scaffold-free, and resource-intensive organoid models.22, 23, 24 Scaffold-based methods involve culturing cells on or within synthetic biomaterials, such as hydrogels or polymeric scaffolds, which provide mechanical support and spatial cues. In contrast, scaffold-free methods enable cells to self-assemble into tissue-like structure without the support of synthetic materials, employing techniques such as hanging drop, low-attachment plates, microfluidic systems, and bioreactors.24, 25, 26 Among scaffold-free strategies, 3D spheroid culture is a suspension-based method that facilitates cell aggregation into spherical structures,²⁷ mimicking solid tumors in organization and ECM composition, making it a cost-effective model for cancer research.²⁸ Note that a wide range of techniques for generating spheroids is summarized in Table 2, including hanging drop, low-attachment plates, liquid overlay, spinner flasks, pellet culture, acoustic levitation, magnetic levitation, microfluidics, and bioprinting. Regardless of the method selected, it is essential to optimize the protocol according to the desired size and cell type to ensure high viability and uniformity. Among the more accessible techniques, hanging drop, low-attachment plates, and liquid overlay are commonly used due to their simplicity, low cost, and minimal equipment requirements. These methods rely on gravitational and hydrophobic forces to facilitate cell aggregation and spheroid formation.29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44 Other relatively simple methods include spinner flasks and pellet culture, which involve spinning or rotating the cells to facilitate spheroid formation.^13,27,45, 46, 47, 48, 49, 50, 51, 52 These models are often used in regenerative medicine, where they support stem cell differentiation. More recent techniques use specialized equipment, such as acoustic and magnetic levitation, to rapidly generate spheroids and enable their transfer into various experimental systems.^13,53, 54, 55, 56, 57, 58, 59, 60, 61 Although acoustic and magnetic levitation have not yet been applied to cardiac spheroid formation, their high-throughput capacity and consistent self-organization suggest strong potential for future cardiovascular research. The most advanced techniques, including microfluidics and bioprinting, offer precise control over spheroid structure and microenvironment, making them powerful tools for cardiovascular applications.^25,37,62, 63, 64, 65, 66, 67, 68, 69, 70, 71 In this review, we focus on the hanging drop, low-attachment, and liquid overlay methods, alongside hydrogel scaffolds and microfluidic systems to enhance the cellular microenvironment.

Table 2.

Summary of scaffold-free spheroid generation techniques and their applications. Various spheroid generation techniques are summarized to evaluate their uniformity, throughput capability, cell viability, specialized equipment, and their applications. The term "large spheroid culture” in the application column refers to spheroids with diameters of 400 ​μm or greater. “Long-term culture” refers to spheroids maintained in culture for more than 2 weeks.

	Uniformity	Throughput	Cell Viability	Specialized Equipment	Applications	Ref
Hanging Drop	Produces relatively uniform spheroids based on droplet size and cell number, though variability may arise from manual handling and gradual disaggregation	Simple and easily scalable, but labor-intensive	Good for ≤2 weeks in culture >92 ​% live cells	No	Affordable & simple culture method Short-term culture Easy imaging Drug screening Well-established for tumor modeling Implemented for cardiovascular modeling	29, 30, 31, 32, 33
Low-attachment Plates	Forms single, uniformly sized and shaped spheroids with consistent circularity, though compactness may be reduced	Compatible with multi-well formats and suitable for large-scale experiments and spheroid generation	Some brands support long-term viability, but most are used for short-term culture High viability is maintained for up to 7 days but declines by 21 days in many commercially available brands	No	Affordable & simple culture method Mainly short-term, but can be long-term Easy maintenance & imaging Drug screening Well-established for tumor modeling Implemented for cardiovascular modeling	29,34, 35, 36, 37, 38, 39
Liquid Overlay	Forms moderately reproducible spheroids in both mono- and co-culture, though formation requires a longer time Agarose concentration, suspension volume, and cell number must be optimized based on the cell type	Compatible with multi-well formats and suitable for large scale experiments and spheroid generation, but require an additional agarose-coating step	Stable viability for 14 days Tend to form larger necrotic core	Yes Agarose-coated plates	Affordable and simple culture method Mainly short-term culture Co-culture ability Drug Screening Well-established for tumor modeling Implemented for cardiovascular modeling	40, 41, 42, 43, 44
Spinner Flasks	Forms moderately reproducible spheroids with irregular borders and variable size	Suitable for producing large amounts of spheroids in a single flask	Good nutrient & waste diffusion supports high viability, making it suitable for long-term culture Loss of cells over time due to spheroid disintegration; some studies report minimal viability and disaggregation by day 9	No	Large spheroid culture Long-term culture if carefully optimized Tissue Regeneration – chondrogenesis and MSCs Established for tumor modeling Implemented for cardiovascular modeling	45, 46, 47, 48, 49, 50
Pellet Culture	Produces highly reproducible spheroid size and volume with consistent seeding density, though dependent on manual precision Forms compact spheroids after a 5-min centrifugation and 24-hr incubation	Requires many tubes, limiting scalability	Shown stable viability for 7 days Mesenchymal stem cells (MSCs) shown to be viable for 28 days	No	Large spheroid culture Short-term or long-term spheroid culture based on cell type Tissue Regeneration – chondrogenesis and MSCs Established for tumor modeling	13,27,51,52
Acoustic Levitation	Facilitates immediate, homogeneous self-organization of spheroids with reproducible size and shape Provides greater compaction and pore development compared to other techniques	Forms a cell layer within seconds and circular spheroids in approximately 6–12 ​h Thousands can be cultured in parallel easily	≥90 ​% cell viability after 24 ​h ≥80 ​% cell viability after 6 days in culture	Yes- Waveform generators, signal transducers, acoustic reflector, and acoustofluidic chip	Good for initial spheroid formation Cell/spheroid self-organization studies Co-culture ability Drug Screening Regenerative Medicine – increased therapeutic efficacy of MSCs	53, 54, 55, 56
Magnetic Levitation	Shape did not vary with seeding density Increased cell number leads to greater spheroid size variability. Larger aggregates form a more cylindrical shape due to a nonuniform magnetic field	Enables rapid spheroid formation within 24 ​h and can be scaled for mass production	≥80 ​% viability after 24 ​h High proliferative capacity with minimal or no necrotic core	Yes Metal nanoparticles, magnetic plate, and magnetic culture plate lid	Good for initial spheroid formation Co-culture ability Drug Screening Regenerative Medicine – MSCs and osteoblasts Established for tumor modeling	13,57, 58, 59, 60, 61
Micro-fluidics	Size of individual spheroids is nearly identical, controlled by the size of the microwell	Spheroid assembly is more advanced compared to other methods Enables precise control over each spheroid size Can produce thousands of spheroids per hour through automation	83–95 ​% viability after 3 or more days Facilitates long term culture through vascularization and nutrient/exchange	Yes Microfluidic device, oil/water, and flow controller Can be integrated with application-specific enhancements, such as sensors	Long-term culture Co-culture ability Often used in concert with other spheroid generation techniques Drug screening Simulates fluid shear effects Can incorporate sensors for real-time data collection Well-established for tumor modeling Implemented for cardiovascular modeling	25,37,62, 63, 64, 65, 66, 67
Bioprinting	Highly controlled spheroid size and organization	Printing techniques vary: extrusion-based printing deposits bioink layer by layer and is less suited for mass production, while drop-on-demand printing offers high-throughput capabilities	81–98 ​% viability after 24 ​h Little cell death after 42 days	Yes Bioink and bioprinter	Long-term culture Co-culture ability Cell/spheroid self-organization studies Drug Screening Tissue regeneration Well-established for tumor modeling Implemented for cardiovascular modeling Metabolism models	68, 69, 70, 71

Spheroid models support monoculture or co-culture, enabling the replication of diverse in vivo microenvironments by incorporating various cell types (Fig. 1).^4,27,72,73 This capability allows them to mimic native cell-cell interactions and tissue organization. In the absent of exogenous scaffolds or substrates, cells within spheroids secrete their own ECM proteins and rely on cell-cell and cell-ECM interactions to sustain functions such as proliferation.⁷⁴ Compared to other 3D culture systems, spheroids feature three distinct layers: a central necrotic core, an inner quiescent layer, and an outer proliferating layer (Fig. 1), making them particularly useful for studying drug responses, cell proliferation, and apoptosis.¹⁰ Spheroids can be cultured independently or embedded in 2D or 3D matrices to modulate their microenvironment and to better replicate diverse physiological and pathological conditions. This adaptability has been leveraged to investigate the role of ECM stiffening in tumor progression, proliferation, and invasion.75, 76, 77, 78 Beyond cancer research, spheroid models can provide a platform for studying the impact of mechanical properties on cellular functions, suggesting that potential application in stiffness-associated cardiovascular biology and pathology.²⁷

Fig. 1.

3D spheroids incorporate various cell types to model in vivo cardiovascular structures and pathologies. Spheroids modeling healthy arteries and angiogenesis show organizational significance, with the VSMC core mimicking the tunica media of the vessel, while ECs form a distinct layer adjacent to the VSMCs, resembling the endothelium. Co-culture spheroids exhibit enhanced angiogenic ability due to EC and MSC interactions, leading to the expression of pro-angiogenic factors. In cardiac tissue, a diverse cell population contributes to its distinct function, and co-culturing these cells in a 3D structure help maintain phenotype and function via cell-cell interaction. 3D spheroids also provide valuable models for in vivo biological processes contributing to cardiovascular pathologies, such as atherosclerotic plaques. Incorporating macrophages and inducing their differentiation into foam cells in co-culture with healthy vessel cells creates a structurally accurate model, better representing drug efficacy. The spheroid structure, including the outer proliferating layer, central necrotic core, and inner quiescent layer, remains consistent across different cell types. The model's adaptability to various cardiovascular conditions and its ability to mimic the in vivo environment make it a useful tool for cardiovascular research. Created in BioRender. Krug, A. (2025) https://BioRender.com/h2hky9q.

3D spheroids are also well-suited for high-throughput applications, offering significant promise for personalized medicine and early disease diagnosis, particularly when integrated with Machine Learning (ML) approaches.⁷⁹ Building on this potential, spheroids generated from patient-derived cells, such as induced pluripotent stem cells (iPSCs), may serve as predictive models for early-stage CVD, capturing patient-specific variations and pathological cues. These models may exhibit distinct gene expression profiles and reveal early pathophysiological changes, such as alterations in cell-cell and cell-ECM interactions, which occur before clinical symptoms emerge. Furthermore, they provide a platform for evaluating pharmacological interventions by tracking changes in spheroid size and morphology. Although ML techniques have been applied to analyze spheroid architecture and behavior, their use in cardiovascular diagnostics remains limited, highlighting a promising direction for future personalized CVD research.^32,80 As their versatility becomes increasingly recognized, 3D spheroid models have expanded into a broad range of applications, including drug screening, tissue engineering, personalized medicine, and cardiovascular research (Fig. 1).^25,81,82

3. Spheroids for vascular biology and pathology

3.1. Blood vessel spheroids

Blood vessels function as delivery systems, transporting oxygen and nutrients to tissues while removing metabolic waste.⁸³ These conduits consist of a single layer of endothelial cells (ECs) lining the vascular lumen and overlaying vascular smooth muscle cells (VSMCs), arranged in layers varying by vessel type and function (e.g., arteries or veins). The dynamic interactions between ECs and VSMCs are essential for maintaining vascular stability and function, with their disruption contributing to vascular pathology.^84,85 Therefore, studying EC and VSMC biology is pivotal for understanding the mechanisms underlying vascular diseases. To better recapitulate how vascular cells assemble, mature, and maintain homeostasis within blood vessels, ECs are often co-cultured with VSMCs. In addition, the ECM and support cells—including pericytes, macrophages, fibroblasts, and dendritic cells—are critical to vascular development and regulation, making their integration into co-culture systems essential for creating more physiologically relevant vascular models.⁸⁶

To explore in vitro vascular cell behavior, Korff et al. co-cultured equal numbers of human umbilical vein endothelial cells (HUVECs) and human umbilical artery smooth muscle cells (HUASMCs) in suspension within non-adhesive 96-well round-bottom plates. Over four days, these cells spontaneously aggregated, forming a HUASMC core with an HUVEC surface layer, resembling an inside-out vessel structure.³⁹ Co-culture successfully induced a mature yet quiescent HUVEC phenotype, characterized by increased inter-endothelial junctional complexes, suppressed secretion of platelet-derived growth factor beta (PDGF-β), and increased cell survival,³⁹ critical for driving VSMC migration and proliferation observed in neointima formation.87, 88, 89, 90 Conversely, HUVEC spheroids in monoculture exhibited fewer inter-endothelial junctional complexes, abundant PDGF-β secretion, and increased apoptosis, demonstrating their inability to replicate the in vivo cellular microenvironment.³⁹ Furthermore, HUVECs became refractory to vascular endothelial growth factor (VEGF)-induced stimulation when HUASMCs and HUVECs were co-cultured at the same ratio, thereby limiting sprouting angiogenesis. However, angiogenesis was restored when spheroids were co-stimulated with VEGF and angiopoietin-2 (Ang-2).³⁹ This study demonstrates the importance of EC-SMC interactions in maintaining vessel homeostasis, which influences biological processes such as vascular remodeling. This co-culture model more accurately reflects blood vessel physiology than monoculture systems, enabling the examination of paracrine signaling between ECs and SMCs. Therefore, it provides valuable insight into molecular pathways driving normal vascular processes and the development of pathological conditions such as atherosclerosis.

To investigate the mechanisms of atherosclerosis, a vascular pathology characterized by arterial stiffening, one study aimed to closely mimic the cellular characteristics of fibroatheroma (Fig. 1), a stage in atherosclerotic plaque development where plaques are prone to rupture and occlude vessels, to gain deeper insight into the molecular and cellular interactions within the vessel that contribute to poor patient outcomes.⁹⁰ The plaque formation process leads to vessel remodeling, involving inflammatory cell infiltration and VSMC phenotypic switching, both of which can be effectively studied using this model. Other aspects of atherosclerosis have also been proposed and modeled using 3D spheroids.¹⁹ This includes the spheroid culture of foam cells, which are macrophages that have undergone differentiation with increased lipid uptake and are the primary components of atherosclerotic plaques. Nguyen et al. developed a 3D foam cell spheroid system (Fig. 1) to evaluate the efficacy of various drugs in limiting lipid accumulation and inflammation, two key processes driving atherosclerosis progression.⁹¹ This system proved useful for testing multiple therapeutic agents, including fluocinolone acetonide, dexamethasone, and nanoliposomal delivery of STAT.^91,92

3.2. Angiogenesis spheroids

Angiogenesis is a complex physiological process in which new blood vessels form from pre-existing vasculature.^93,94 This process occurs in two forms: sprouting angiogenesis and intussusceptive angiogenesis.⁹³ Sprouting angiogenesis is driven by endothelial tip cells that migrate through the ECM and proliferate, often in response to hypoxia in poorly perfused tissue. Intussusceptive angiogenesis, or “splitting angiogenesis,” occurs when the endothelium of adjacent capillaries splits and rearranges, eventually forming a new pillar or capillary between them.⁹³

Initially, 3D spheroids were used to model tumor angiogenesis and explore its association with prognosis.95, 96, 97 One of the earlier examples was implantation of Lewis lung carcinoma spheroids onto mice with dorsal skinfold chambers by Filho et al.⁹⁶ This approach was soon followed by simplified in vitro models such as multicellular tumor spheroids co-cultured with HUVECs to induce angiogenesis.⁹⁸ The success of the tumor spheroid models in studying angiogenesis led to their broader application in vascular biology. Concurrently, investigations into the mechanisms and conditions enabling ECs to form new blood vessels were conducted using 3D spheroid monoculture (Fig. 1).⁹⁹

Recent studies have focused on advancing the traditional mono-culture spheroid angiogenesis model by developing co-culture spheroid systems. While ECs drive angiogenesis, pericytes and SMCs play a crucial role in vessel stabilization and maturation¹⁰⁰. The interaction among these vascular cell types is fundamental to maintaining vascular function and integrity.¹⁰¹ To better model spheroid-based angiogenesis in vitro, incorporating ECs, SMCs and pericytes is essential.102, 103, 104 One such approach was developed by Shah et al. who engineered a 3D angiogenesis model using co-culture spheroids (Fig. 1) composed of endothelial colony forming cells (ECFCs), circulating vascular cell precursors, and mesenchymal stem cells (MSCs), which are thought to differentiate into pericytes.^105,106 Specifically, ECFCs and MSCs were co-cultured to 70–80 ​% confluency before generating spheroids using the hanging drop method. After two days, the spheroids were embedded in neutralized type I collagen gel to simulate the in vivo extracellular microenvironment. A real-time cell recorder within a cell culture incubator monitored sprout formation in each spheroid at multiple time points. Co-cultured spheroids composed of both ECFCS and MSCs exhibited greater sprout number and length, with continuous growth for up to 12 ​h, compared to mono-culture spheroids formed from only ECFCs or MSCs embedded in collagen gels.¹⁰⁵ This study highlights the enhanced ability of co-culture spheroids to recapitulate the multistep angiogenesis process, including ECFC-MSC interactions, sprouting, tube formation, and vessel maturation.¹⁰⁵

Co-culture spheroid models incorporating ECs and SMCs provide a valuable in vitro system for studying angiogenesis, as their interactions promote VEGF expression and other pro-angiogenic factors, while SMCs adherence to ECs support vessel maturation.^107,108 Zuo et al. created EC-SMC spheroids to study cell migration, culturing HUVECs and VSMCs in round-bottom 96-well plate for 24 ​h (Fig. 1).¹⁰⁹ The formed spheroids were then encapsulated in fibrinogen hydrogels to study migration. Although initially focused on cell migration, the study revealed upregulated angiogenesis-related genes, with EC-SMC spheroids exhibiting increased expression of SDF-1, HIF-1 and angiopoietin-1 compared HUVECs or VSMCs mono-culture spheroids. These results suggest that co-culture spheroids promote angiogenesis and are an advantageous model system for studying its mechanisms.¹⁰⁹

Despite their potential, EC-SMC-pericyte co-culture spheroids remain underexplored for in vitro angiogenesis studies, offering a promising avenue for mechanistic investigations. However, these models oversimplify physiological and pathological angiogenesis by limiting cellular interactions to only those between cell types included in the spheroid. Integrating additional cell types, such as immune cells and fibroblasts, could better replicate the dynamic cell-cell interactions in vivo. Further refinement of these co-culture models will allow for a more precise investigation of the mechanisms regulating angiogenesis.

4. Spheroids for cardiac biology and pathology

The heart is at the center of the cardiovascular system and consists of multiple layers, including the innermost endocardium, myocardium, and outer epicardium.¹¹⁰ The myocardium serves as the contractile muscle layer, while each layer has distinct morphological features and functions. These layers are embedded in an ECM and a complex microenvironment with functional connections to adjacent layers. Consequently, co-culture spheroids may better recapitulate physiological conditions in vitro. In line with this, the vascularized cardiac spheroid (VCS) model was created by incorporating human iPSC-derived cardiomyocytes (CMs), cardiac fibroblasts (CFs), and human coronary artery ECs,¹¹¹ enabling the study of multicellular interactions within the cardiac microenvironment. Designed to improve in vitro cardiac models for studying cardiovascular function, CVD, drug responses, and tissue regeneration, the VCS was formed within 96-well hanging drop plates after three days in culture. The CM:EC:CF ratio (2:1:1) was selected to approximate in vivo cellular proportions in the human heart. The hanging drop method for spheroid culturing creates a 3D cellular aggregate mimicking a physiologically relevant ECM and microvascular network.¹¹¹ The VCS model provides a valuable platform for exploring complex physiological processes and can be adapted for specific cardiac disease models.

Recently, Kore et al. developed a cardiac spheroid model using cardiomyocytes (Fig. 1) to investigate molecular mechanisms and potential treatments for cardiomyocyte cytoprotection under hypoxic conditions.¹¹² This model incorporated mesenchymal stromal cell-derived exosomes, which reduced apoptosis and promoted cardiomyocyte survival following lipopolysaccharide-induced myocardial injury, an inflammatory and apoptotic stimulus. However, despite its valuable findings, the model's mono-culture design limits its ability to fully replicate complex cellular interactions and structure organization of the human heart. Exosomes from various cell types, such as circulating endothelial progenitor cells, play roles in biological processes like angiogenesis¹¹³ and may benefit from a co-culture spheroid model to better simulate in vivo cellular interactions. Despite these limitations, cardiac spheroid models^111,112 demonstrate promise as in vitro platforms for developing more advanced heart models, including those for human heart disease. Incorporating 3D bioprinting to organize multiple cardiac cell types, hydrogel and polymer scaffolds to replicate the ECM microenvironment when embedding the spheroids, and microfluidic chips to facilitate vascularization within spheroids could further enhance their physiological relevance and functionality.^71,114, 115, 116

Heart development and function depend on intricate interactions between CMs and CFs.¹¹⁷ Co-culture spheroids of these cell types model their dynamic relationship and the processes underlying cardiac development, but the system remains incomplete. To refine this model, Beauchamp et al. examined the properties of co-cultured CMs and CFs (Fig. 1), comparing their tissue-like characteristics to those of mono-culture spheroids.³³ Using the hanging drop method, human iPSC-derived cardiomyocytes (hiPSC-CMs) and CFs were co-cultured into 3D spheroids and analyzed using confocal microscopy, action potential recording, and quantitative PCR. Electrophysiological analysis revealed that 3D co-culture cardiac spheroids exhibited spontaneous action potential activity. While both 2D and 3D cultured hiPSC-CMs displayed spontaneous action potential activity, the spheroid model promoted a more mature electrical phenotype by sustaining action potential activity and synchronized beating patterns throughout the spheroid. In contrast, 2D-cultured hiPSC-CMs formed isolated beating clusters, resulting in patchy electrical activity compared to the stable contractile behavior observed in 3D spheroids. To further assess the effects of long-term culture, 2D culture of CFs, 3D mono-culture of CFs, and 3D co-culture spheroids of hiPSC-CMs and CFs were maintained for up to one month. In both CF mono-culture and hiPSC-CM/CF co-culture spheroids, CFs closely resembled in vivo ventricular cardiac cells in size, morphology, actin organization, and phenotype.³³ While 2D CF culture exhibited prominent actin stress fibers and smooth muscle actin expression, indicative of substantial myofibroblast activation, CF mono-culture spheroids maintained a resting CF phenotype with minimal myofibroblast activation throughout the culture period.³³ Both mono-culture and co-culture spheroids effectively preserved physiologically relevant CF phenotypes, establishing spheroid culture as valuable approach for studying cardiac biology. The success of co-culture spheroids in replicating in vivo-like CMs and CFs highlights their value in modeling phenotypically accurate CM-CF interactions. These findings highlight the advantages of cardiac spheroids in advancing cardiac research. Continued innovations in media supplements, ECM-like materials, microfluidic systems, and nanoscale coatings will further refine cardiac spheroid culture models, enhancing their physiological relevance and translational potential.

5. Spheroids for mechanobiology

Mechanical forces, such as tissue stiffness and compliance, along with chemical signals, play a central role in regulating molecular pathways and cellular function in vivo. This interaction forms the basis of mechanobiology, a rapidly expanding field that has been extensively reviewed.118, 119, 120 Aberrantly increased tissue stiffness contributes to the progression of CVD and cancer by modulating cell behavior and phenotype.121, 122, 123, 124, 125 For instance, in cardiac fibrosis, increased tissue stiffness promotes the trans-differentiation of quiescent cardiac fibroblasts into myofibroblasts, leading to excessive extracellular matrix (ECM) deposition and impaired ventricular function.^122,126,127 Similarly, in atherosclerosis and vascular pathologies, arterial stiffness plays a major role in regulating VSMC and EC behavior. It induces VSMC proliferation, migration, vascular remodeling, and exacerbates pro-inflammatory responses,128, 129, 130, 131 while also driving EC dysfunction, a key hallmark of atherosclerosis.^132,133 Given the central role of stiffness in cardiac and vascular pathologies, investigating mechanical signaling is essential in CVD research models. Although 3D spheroid models have been increasingly applied to cancer and cardiovascular research, their application in mechanotransduction remains limited.

In cancer study, McKenzie et al. used investigated the effect of ECM stiffness on ovarian cancer by generating spheroids from SKOV3 cells, derived from human ovarian adenocarcinoma, in agarose-coated microwells and seeding them onto fibronectin-coated hydrogels of varying stiffness.⁷⁶ Spheroid behavior was stiffness-dependent: on soft hydrogels (∼3 ​kPa), spheroids remained spherical with minimal spreading after 24 ​h, whereas on stiff hydrogels (25−125 ​kPa), they exhibited increased spreading and loss of sphericity. Moreover, SKOV3 spheroids on stiff hydrogels showed increased nuclear YAP localization and expression, consistent with observations in 2D cultures, while those on soft hydrogels displayed reduced nuclear YAP levels.⁷⁶ These findings highlight how matrix stiffness regulates spheroid morphology and intracellular signaling, underscoring the need for further investigations using other cell types to deepen our understanding of mechanotransduction in 3D spheroids.

In cardiovascular applications, Chin et al. investigated the effects of substrate stiffness on H9C2 cardiac-derived myoblast spheroids encapsulated in gelatin methacryloyl hydrogels with a stiffness gradient (3.68–17.52 ​kPa) and cultured for 10 days.¹³⁴ Morphologically, while nuclei in H9C2 spheroids exposed to stiff environments appeared more compact and exhibited larger cytoplasmic volumes than those in softer hydrogels, no significant changes were observed with increasing stiffness. Mechanistically, YAP and MRTF-A, mechanosensors typically localized to the nucleus in response to increased stiffness,^135,136 were instead more cytoplasmic, while higher hydrogel stiffness was associated with increased lamin-A and vinculin expression. Although this study supports the use of spheroids in mechanotransduction research, further studies with diverse cardiovascular cell types in pathological contexts are needed to better understand stiffness-dependent morphological and mechanistic responses in 3D.

Taken together, manipulating ECM stiffness within spheroid cultures offers a promising strategy for modeling stiffness-driven processes in cardiac fibrosis, atherosclerosis, and other CVDs. The ability of spheroids to replicate native 3D cell–ECM interactions (Table 1) may help elucidate the roles of stiffness-modulated biomolecules in promoting pathological cell behavior and disease progression.

6. Spheroids for drug testing and tissue regeneration

Challenges in using 3D spheroid models to quantify drug efficacy include limited drug penetration, difficulties in assessing viability or growth, immunostaining clarity, and high-throughput screening limitations.¹³⁷ To address these challenges, Christoffersson et al. developed a high-throughput 3D cardiac spheroid-based assay to evaluate drug effects on cardiac cell outgrowth (Fig. 2 and Table 3).^36,37 Cardiac spheroids were generated from pluripotent stem cell-derived CMs, seeded into low-attachment plates for spheroid formation.³⁷ They were then captured (20 spheroids per channel) in a laminin-coated microfluidic device for high-content screening, with drug treatment, immunostaining, and imaging performed in the same fluidic channel to enable rapid, consistent readouts using noninvasive light and fluorescence microscopy. This system is particularly valuable, as discrete spheroids establish gradients of oxygen, nutrients, and metabolic wastes, maintaining a viable microenvironment.^138,139

Fig. 2.

Cardiac spheroid methodology for drug testing and tissue regeneration. Hanging drop and low-attachment plates are simple methods for generating cardiac spheroids with minimal steps. These techniques have been further enhanced by incorporating specialized microfluidic devices, silicon nanowires, and carbon electrodes, enabling high-throughput drug testing and improved functionality. Created in BioRender. Krug, A. (2025) https://BioRender.com/qsqeqcf.

Table 3.

Comparison of cardiac spheroid models for drug screening and therapeutic applications. Cardiac spheroid models for evaluating drug and therapeutic potential were assessed using key parameters relevant to cardiovascular research and drug discovery: disease fidelity, drug screening accuracy, reproducibility, and throughput. Each parameter was qualitatively scored as Low, Medium, or High reflecting the relative performance of each model. These rankings represent comparative evaluations across the model systems and aim to guide model selection based on specific research goals. CM: cardiomyocyte, VSMC: vascular smooth muscle cell, HUVEC: human umbilical vein endothelial cell, CF: cardiac fibroblasts, EC: endothelial cell, hiPSC-CM: human induced stem cell-cardiomyocyte.

Spheroid models have been adapted to simulate vascular pathologies such as atherosclerosis and neointimal hyperplasia, providing a cost-effective alternative to animal models while preserving a 3D pathological microenvironment.^32,90,140 Vaidyanathan et al. developed a VSMC-based spheroid model to mimic neointimal formation driven by VSMC accumulation and proliferation, a hallmark of restenosis and atherosclerosis (Fig. 2 and Table 3).^32,90 In this study, human VSMC spheroids, generated using the hanging drop method, were treated with drugs that regulate VSMC morphology, proliferation, and neointimal formation. A machine learning approach assessed the effects of focal adhesion kinase (FAK) and small GTPase (Rac, Rho, and Cdc42) inhibitors on spheroid morphology and formation, revealing heterogeneous, dose-dependent changes. While further evaluation of VSMC proliferation is needed, this model integrates computational approach for early-stage drug screening and mechanistic studies of CVD progression.

Di Cio et al. demonstrated the 3D spheroid model by developing a vascularized cardiac spheroids-on-a-chip system for drug testing.³⁸ The system utilized a microfluidic device to mimic systemic drug delivery and facilitate microvasculature development (Fig. 2 and Table 3). It incorporated co-culture spheroids composed of CMs/ECs/CFs and HUVEC/placenta-derived pericyte/fibrinogen gels for vasculature formation. The microfluidic devices were created using PDMS through photo and soft lithography. Co-culture cardiac spheroids containing CMs, ECs, and CFs were created using ultra-low attachment 96-well plates. HUVECs or HUVEC/placenta-derived pericytes, mixed with thrombin and fibrinogen, were first injected into the microfluidic device's gel chamber to initiate vasculature formation, while side channels were filled with EC growth medium 2 and VEGF to support vasculature growth. After four days, the cardiac spheroids were embedded into the central culture well along with the fibrinogen and thrombin solution to facilitate vascular invasion of the spheroid. Cultured for ten days, the HUVEC spheroids displayed interconnected vasculature, enhanced nutrient delivery, and perfusability. The spheroids remained functionally active, with beating observed. Although this cardiac spheroid model displays enhanced functionality and complex vascularization, the role of co-culture in this model is not fully understood. The model itself differs from in vivo conditions, primarily because the inclusion of pericytes with HUVECs did not improve vascular integrity as it does in the body. Instead, the spheroids are believed to generate NG2+ cells that take on the functional role of pericytes. Overall, the use of cardiac spheroids for drug testing and tissue regeneration is still evolving and has yet to reach its full potential in cardiovascular research. Although these models are not yet fully understood, studies like this highlight their promise as biologically relevant platforms and warrant further investigation.

Human iPSC-CM technology has been explored for CVD treatment through regenerative medicine and tissue enhancement. However, a major limitation in myocardial regeneration is the arrhythmogenic risk and limited functional improvement following transplantation of unspecific hiPSC-CM.141, 142, 143 To address this, Richards et al. developed a method integrating nanowires and electrical stimulation to enhance hiPSC cardiac spheroid development, improving their suitability for cell transplantation (Fig. 2 and Table 3).⁴⁴ Electrically conductive silicon nanowires promoted structural and functional maturation in hiPSC cardiac spheroids, while the combination of electrical stimulation and nanowires enhanced microtissue development, improved contractile properties, strengthened cell-cell junctions, and reduced arrhythmogenic risk by decreasing the spontaneous beat rate of hiPSC cardiac spheroids.^44,144 This model effectively mitigates arrhythmogenic risks while enhancing functional outcomes, supporting cardiac repair and advancing cell therapy approaches. However, its complexity and long preparation time may limit its practicality for studying molecular pathways in cardiac spheroids, despite its use for in vitro regeneration and in vivo transplantation. Another challenge in using hiPSC-CM for tissue enhancement is their low engraftment ratio. Given the lengthy generation process and the high number of hiPSC-CMs required for cardiac repair, maximizing retention is crucial. Studies suggest that implanting hiPSC-CMs as spheroids with polymeric hydrogels improves retention compared to non-aggregated cells.145, 146, 147 However, whether this enhancement leads to improved cardiac tissue regeneration remains uncertain and requires long-term investigation.

7. Spheroid-organoid model systems for drug perfusion

Spheroid systems incorporating co-culture and microfluidics are valuable platforms for drug testing but can be further enhanced through the integration of spheroid-organoid hybrids or organoids. Organoids are self-assembled 3D structures derived from stem or tumor cells that more closely recapitulate tissue architecture and function by accommodating multiple cell types and their spatial organization.^12,148 Many organoids begin from simple cell aggregates or spheroids, making hybrid systems a natural extension.^149,150 Directed differentiation using growth factors and matrix gels allows stem cells to self-organize into complex, tissue-like structures. However, organoids typically have limited perfusability due to their larger size. As with spheroids, microfluidic systems and bioprinting have been applied to enhance drug delivery and nutrient exchange in organoid culture.151, 152, 153 These advances have enabled the development of cardiac organoids for drug testing. For example, Zhang et al. created an endothelialized-myocardium-on-a-chip model using organoids to evaluate drug toxicity.¹⁵² Upon treatment with the cardiotoxic agent doxorubicin, the model demonstrated a dose-dependent decline in cardiomyocyte beating rate and von Willebrand factor secretion by endothelial cells, supporting the use of organoids to measure functional responses in cardiac tissue. Despite these advantages, organoids remain less reproducible and are more labor- and cost-intensive than spheroids.¹⁵⁴ In contrast, spheroids offer a high-throughput, biologically relevant alternative to 2D culture for assessing drug toxicity and efficacy. When combined with microfluidics, spheroid-on-a-chip systems enhance perfusability, support prolonged viability, and enable the formation of structurally robust microvasculature. These features make them more practical and reproducible models for drug perfusion studies. Once drug safety and efficacy are established using spheroid systems, organoid integration can provide added functional insight into adverse tissue responses prior to animal testing.

8. Translational bottlenecks and ethical considerations regarding spheroid models

Despite the growing use of spheroid models in cardiovascular research, it is important to acknowledge both translational bottlenecks and ethical considerations to support their effective clinical application. A major limitation lies in their incomplete replication of complex in vivo physiology. Although spheroids provide a more physiologically relevant alternative to 2D cultures (Table 1), they often lack essential features such as key cell types and their interactions, vascularization, immune responses, ECM components, and the full complexity of native tissue architecture.^12,155 Furthermore, their long-term culture is constrained by limited oxygen and nutrient diffusion, which can result in a necrotic core, reduced cell viability, and altered metabolic activity. These conditions may shift gene expression profiles in ways that do not accurately represent the pathological states being studied.¹⁵⁶ As a result, these limitations reduced the utility of spheroids for modeling long-term disease processes, such as fibrotic progression, or for evaluating therapeutic responses to chronic drug exposure.¹³⁷

Another persistent challenge in spheroid modeling is the inter-laboratory variability in spheroid generation and maintenance, driven by the use of diverse culture techniques (Table 2). These techniques frequently require extensive optimization, with outcomes highly dependent on the cell types and experimental goals. Since reproducibility is critical for translational research, establishing standardized protocol is essential to enhance clinical applicability.¹⁵⁶ Building on this, a major translational bottleneck lies in the scalability and integration of spheroid models with high-throughput generation and screening platforms. Although recent advances, such as microfluidic chambers, have improved the consistency and scale of spheroid production, several challenges persist. These challenges include variability in spheroid size, as well as oxygen and nutrient gradients that affect drug perfusion, ultimately limiting their utility in large-scale drug screening and pharmacokinetic modeling.^12,155 Furthermore, interpreting data from 3D culture often requires specialized imaging and analytical tools, creating practical barriers to broader adoption in drug development pipelines (Table 3).^32,37,38

A potential ethical consideration in advancing spheroid models is developing individualized spheroid systems using patient-derived cells. These personalized spheroid models enable clinically relevant drug testing with an accurate replication of physiological microenvironments, providing a mechanism for patient-tailored treatments. However, obtaining patient cells via biopsies requires informed consent and careful clinical risk assessment. When implementing patient-derived spheroids into a therapeutic regimen, it is imperative that the sensitive data obtained from these models stay securely protected. While spheroids remain promising for disease modeling and therapy development, their use demands strict adherence to ethical standards to ensure patient safety and trust.

9. Conclusion

3D spheroid models are powerful platforms for advancing our understanding of cardiovascular biology, disease mechanisms, tissue regeneration, and drug development. Compared to traditional 2D culture systems, these models provide a more physiologically relevant environment, enabling the study of complex cell-cell and cell-ECM interactions, as well as mechanotransduction within CVD contexts. While challenges remain such as limited drug penetration, variability in assay reproducibility, and the technical demands of model generation, 3D spheroid systems have demonstrated significant potential for drug screening, tissue enhancement, and disease modeling. Ongoing innovations in spheroid culture methods, including the integration of microfluidic systems, conductive scaffolds, and machine learning approaches, continue to enhance the versatility and precision of these models. With these advancements, 3D spheroid systems are increasingly positioned to contribute to the development of personalized medicine approaches and next generation therapeutic strategies for CVD.

CRediT authorship contribution statement

Alanna Krug: Writing – review & editing, Writing – original draft, Visualization, Investigation, Conceptualization. Gabrielle Inserra: Writing – review & editing, Writing – original draft, Investigation, Conceptualization. Rhonda Drewes: Writing – original draft. Amanda Krajnik: Writing – original draft, Conceptualization. Joseph A. Brazzo: Writing – original draft. Thomas Mousso: Writing – original draft. Su Chin Heo: Writing – review & editing, Conceptualization. Yongho Bae: Writing – review & editing, Writing – original draft, Supervision, Resources, Project administration, Investigation, Funding acquisition, Conceptualization.

Ethical approval

This study does not contain any studies with human or animal subjects performed by any of the authors.

Funding

This work was supported by an NIH/NHLBI grant (R01HL163168) to Y. Bae and S. Heo.

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.