Defining Actionable Cell States in Cardiovascular Disease Through Single-Cell and Spatial Technologies

Cardiovascular disease remains the leading cause of death worldwide, yet patients with similar clinical phenotypes often exhibit striking heterogeneity in disease mechanisms, progression rates, and responses to therapy. This variability reflects a fundamental limitation of conventional molecular and histopathological approaches, which rely on bulk tissue measurements that average signals across diverse cellular populations. As a result, critical information about rare cell populations, transitional cell states, and spatially organized microenvironments is obscured. This limitation is particularly consequential in the cardiovascular system, where tissue function emerges from tightly coordinated interactions among highly specialized and structurally diverse cell types, including cardiomyocytes, endothelial cells, fibroblasts, vascular smooth muscle cells, and resident immune populations. Each of these cell types exists not as a single static entity, but as a spectrum of dynamic states shaped by developmental origin, mechanical forces, metabolic demands, and inflammatory or injury signals. Consequently, the clinical phenotype of cardiovascular disease is not solely determined by the presence or absence of specific cell types, but by the distribution and dynamics of cellular states within and across tissues.

As precision cardiovascular medicine advances, a central challenge has emerged: how can we identify the specific cellular states and molecular programs that drive disease, and determine which of these represent actionable therapeutic targets? Addressing this question requires a shift in conceptual framework—from defining disease by anatomy or histopathology and static cell types toward defining disease by molecularly and spatially resolved cellular states embedded within tissue architecture. In this context, “actionable cell states” represent a unifying concept. These states are not static entities, but dynamic configurations of gene regulatory activity, transcriptional output, protein abundance and modification, metabolic activity, and spatial context. Importantly, actionable states are defined not only by molecular features, but by their relationship to clinically relevant outcomes, including disease initiation, progression, regression, and response to therapy. A single cell type may occupy multiple states, and distinct cell types may converge on similar pathological programs, underscoring the need for frameworks that transcend traditional cell type classification. Single-cell and spatial omics technologies have emerged as transformative tools for addressing this complexity. Rather than cataloging cell types in isolation, these approaches resolve the cellular ecosystems underlying cardiovascular biology. Single-cell transcriptomic, epigenomic, proteomic, and multiomic technologies enable the identification of cellular heterogeneity at unprecedented resolution, while spatial technologies preserve the architectural context in which these states arise and interact. Together, they provide a foundation for reconstructing tissue organization as a dynamic, multi-layered system of interacting cellular states.

This series of Circulation Research Reviews recent advances in single-cell and spatial technologies and and highlights their application to cardiovascular biology and disease. Across four complementary manuscripts, the authors describe how emerging methodologies are redefining our understanding of cellular identity, tissue organization, and disease mechanisms. Collectively, these contributions emphasize a central theme: no single technology is sufficient to capture the complexity of cardiovascular disease. Instead, integrative, multimodal approaches are required to define actionable cellular states and translate molecular insights into clinically meaningful frameworks.

Single-cell and spatial technologies across regulatory layers.

A key strength of current single-cell and spatial technologies lies in their ability to interrogate distinct layers of biological regulation. Each modality provides a partial view of cellular state, and integration across these layers is essential for reconstructing a complete picture of cellular function. At the epigenomic level, single-cell assay for transposase-accessible chromatin sequencing (scATAC-seq) identifies regions of open chromatin and infers transcription factor binding potential. This regulatory layer represents the accessibility landscape that constrains and enables gene expression programs, effectively encoding a “primed” state of cellular identity. Epigenomic profiling thus provides insight into lineage commitment, developmental potential, and the latent regulatory capacity of cells prior to transcriptional activation. At the transcriptional level, single-cell RNA sequencing (scRNA-seq) measures gene expression programs that define cellular identity and activation states. scRNA-seq has been instrumental in uncovering previously unrecognized cell populations and disease-associated transcriptional states across the cardiovascular system, including in myocardial infarction, heart failure, and atherosclerosis. Importantly, transcriptional states often reflect both intrinsic regulatory programs and extrinsic environmental cues, making them a central but context-dependent readout of cellular function. At the proteomic level, emerging single-cell and spatial proteomic technologies provide a more direct measure of cellular function. While transcriptomic data reflect gene expression potential, proteins represent the primary effectors of cellular activity, integrating transcriptional regulation with translation, post-translational modification, and degradation. As such, proteomic approaches capture not only expression abundance but also proteostasis—the dynamic balance that determines protein function, localization, and turnover.

In the first contribution, Li D et. al., Single-cell studies advance understanding of the genetic and molecular bases of atherosclerosis¹, demonstrate how integration of scRNA-seq, scATAC-seq, and spatial transcriptomics can be used to dissect the cellular complexity of atherosclerotic plaques. Their work highlights how inflammatory, smooth muscle–derived, and endothelial populations occupy distinct and transitional states within the plaque microenvironment. Importantly, integration with genome-wide association studies (GWAS) links disease-associated genetic variants to specific cell types and regulatory programs, providing a mechanistic bridge between genetic risk and tissue pathology.

In a complementary contribution, Li B et. al., Decoding cardiac development and maturation at single-cell and spatial transcriptomic resolution² illustrate how combined single-cell and spatial profiling can be used to reconstruct developmental trajectories of the heart. Their analysis highlights how distinct embryonic progenitor populations migrate, diversify, and interact within spatially defined niches to generate the mature cardiac architecture. This work underscores the importance of spatial context in interpreting transcriptional states, particularly in systems where tissue structure is tightly coupled to function.

Binek and Angel in “Single Cell Proteomics to Spatial Multiomics in Cardiovascular Disease: Foundations, Technologies, and Biological Insights³” review the remarkable advances in single-cell proteomics and spatial mass spectrometry imaging, emphasizing the growing ability to measure protein, metabolite, lipid and other small molecule abundance and modification states at high spatial resolution. These approaches provide critical insight into signaling pathways and functional outputs that are not always predictable from transcriptomic data alone. They further highlight the application of mass spectrometry–based imaging, which uniquely enables simultaneous profiling of proteins, metabolites, and lipids. This capability provides a multidimensional view of cellular regulation, extending beyond canonical gene expression programs to encompass metabolic and biochemical states. An additional emerging frontier discussed in their work is extracellular matrix (ECM) proteomics at spatial resolution. The ECM is increasingly recognized not as a static scaffold but as a dynamic and information-rich component of tissue biology. High-resolution mapping of ECM composition provides insight into tissue remodeling, fibrosis, and mechanical signaling, all of which are central to cardiovascular pathology.

Ma M et. al., in Decoding cardiovascular disease through spatial proteomics⁴, further expands on these themes by describing rapidly evolving ‘deep’ spatial proteomic platforms. They distinguish between whole-tissue mapping approaches and region-of-interest (ROI)–based strategies, highlighting the trade-offs between spatial coverage and molecular depth. Importantly, they outline the future potential of three-dimensional tissue imaging approaches capable of reconstructing molecular maps of entire organs, including the heart and vasculature. Such approaches may ultimately enable molecule-by-molecule reconstruction of tissue architecture across spatial scales.

Across these contributions, a consistent set of technical and biological challenges emerges. Cardiovascular tissues present unique obstacles due to their structural complexity, mechanical stress environment, and the fragility of key cell types such as cardiomyocytes and vascular smooth muscle cells. Moreover, many disease-relevant states are transient, rare, or spatially localized, making them difficult to capture without integrated multimodal approaches.

Integrating regulatory layers into cellular state space.

A unifying framework emerging from these technologies is the concept of a multidimensional cellular state space. In this framework, each cell is represented not by a single identity label, but by its position within a high-dimensional space defined by epigenomic, transcriptional, proteomic, metabolic, and spatial features. Cellular identity is thus reinterpreted as a region of state space rather than a discrete category. Integrating across modalities enables reconstruction of more complete and mechanistically grounded representations of cellular behavior. Regulatory states defined by chromatin accessibility constrain transcriptional programs, which in turn shape protein abundance and signaling activity, while metabolic and spatial context further modulate cellular function. This hierarchical organization allows identification of stable, transitional, and context-dependent states, each of which may have distinct roles in disease progression or resolution.

Within this framework, actionable cell states are those that are not only molecularly distinct but also linked to clinically relevant outcomes. These include states associated with disease severity, prognosis, and therapeutic responsiveness or resistance. Importantly, therapeutic failure may reflect not only incomplete target engagement, but persistence of pathogenic cellular states that are refractory to intervention despite apparent molecular targeting. This perspective reframes cardiovascular disease as a dynamic system of interacting cellular states rather than a static collection of dysregulated genes or pathways. It also provides a conceptual basis for integrating molecular data with clinical phenotypes, enabling more precise stratification of disease and more targeted therapeutic design.

Clinical implications of single-cell resolution.

Single-cell and spatial technologies also have important implications for translational and clinical research. Beyond identifying candidate therapeutic targets, these approaches enable assessment of whether therapies act on the relevant cellular compartments and whether disease-associated states are effectively reversed. Based on this concept, therapeutic non-response may reflect persistence of pathogenic cellular states, insufficient drug exposure, or target expression. For example, inflammatory signaling inhibitors may fail if immune cells transition into alternative activation states that bypass the targeted pathway. Similarly, fibrotic remodeling may persist if fibroblasts adopt stable pathogenic states resistant to single-pathway modulation. This state-based perspective provides a mechanistic explanation for interpatient variability in therapeutic response and suggests that predictive biomarkers should incorporate cellular state composition rather than rely solely on single-gene or single-protein readouts. It also raises the possibility that effective therapies may need to reprogram cellular states rather than simply inhibit individual molecular targets.

Challenges and future directions.

Despite rapid progress, significant challenges remain in fully realizing the potential of single-cell and spatial technologies in cardiovascular research. Current limitations include incomplete measurement of post-translational modifications, protein complexes, metabolic flux, lipid signaling, and dynamic signaling networks at single-cell resolution. In addition, integrating multimodal datasets across platforms, scales, and cohorts remains computationally and statistically challenging. Future progress will require continued technological innovation, but also standardization of experimental and analytical pipelines, improved reproducibility across platforms, and robust validation in clinically annotated patient cohorts. Equally important is the need to scale these approaches to large, diverse patient populations while maintaining sufficient spatial and molecular resolution. Ultimately, the field must move beyond descriptive atlases toward quantitative, mechanistically grounded models of disease. This transition will depend on computational frameworks capable of integrating multimodal data into interpretable representations of cellular state and tissue organization. Such models must be linked directly to clinically meaningful outcomes, including disease progression, therapeutic response, and survival.

Conclusion.

Taken together, the contributions in this compendium highlight a fundamental shift in cardiovascular biology. The field is moving from the identification of discrete cell types toward the definition of dynamic, multidimensional cellular states that drive disease and repair. Single-cell and spatial technologies provide the foundation for this transition by enabling unprecedented resolution of cellular heterogeneity and tissue organization. As these approaches continue to mature and integrate, they are poised to transform cardiovascular research from descriptive molecular cataloging into a mechanistically grounded and clinically actionable discipline. The ultimate goal is not only to map cellular diversity, but to define actionable cell states that connect molecular mechanisms to patient outcomes, thereby advancing the realization of precision cardiovascular medicine.

HIGHLIGHTS BOX (CIRCULATION RESEARCH STYLE).

What Is Known?

• Single-cell and spatial technologies resolve cardiovascular cellular heterogeneity beyond bulk tissue profiling.

• These approaches interrogate distinct regulatory layers including chromatin accessibility, transcription, protein, metabolism, as independent cells and in the spatial context.

• Integration of these modalities enables reconstruction of disease-associated cellular states.

What Is New?

• These technologies enable definition of actionable cell states that link molecular programs to disease behavior and therapeutic response.

• Treatment non-response may reflect persistence of resistant or non-responsive cell states despite apparent target expression.

• Multiomic integration provides a framework for moving beyond descriptive cell atlases toward mechanistic disease models.

What Are the Clinical Implications?

• Single-cell and spatial profiling can improve identification of therapeutic targets, biomarkers, and disease subtypes.

• Defining actionable cell states may improve prediction of treatment response and patient stratification.

• Translation to the clinic will require quantitative, reproducible, and multimodal integration of cellular data linked to clinical outcomes.