The Continuous Spectrum of Epithelial Mesenchymal Transitions

31 concepts19 papers1 proposal

Historically, the epithelial-to-mesenchymal transition (EMT) was viewed as a binary toggle: cells are either tightly bound in an epithelial sheet or fully detached and mobile as mesenchymal cells. However, a growing body of research demonstrates that EMT operates as a continuous spectrum where cells frequently occupy stable or transient hybrid epithelial/mesenchymal (E/M) states. As reviewed by Yutong Sha et al. in Intermediate cell states in epithelial-to-mesenchymal transition, treating EMT as a strict binary switch oversimplifies the true biological continuum, masking the critical roles these intermediate cell states play in development and disease.

Evidence of this spectrum spans multiple clinical contexts. For instance, Shaofeng Yan et al. observed a highly heterogeneous spectrum of protein expression rather than a clean cadherin switch in patient samples in Epithelial–Mesenchymal Expression Phenotype of Primary Melanoma and Matched Metastases and Relationship with Overall Survival. Similarly, profiling circulating tumor cells (CTCs) reveals a fluid EMT spectrum rather than discrete endpoints, as demonstrated across various malignancies by Srujana S. Yadavalli et al. in Data-Driven Discovery of Extravasation Pathway in Circulating Tumor Cells and Arvind Iyer et al. in Integrative Analysis and Machine Learning Based Characterization of Single Circulating Tumor Cells. Even in specialized microenvironments like the cerebrospinal fluid, Haoyu Ruan et al. identified subsets of CTCs trapped in a partial EMT state in Circulating tumor cell characterization of lung cancer brain metastases in the cerebrospinal fluid through single‐cell transcriptome analysis, retaining epithelial features while gaining invasive capabilities.

The Dynamical Architecture of Hybrid States

Why can a cell remain stable in the middle of a transition? The answer lies in the topology of gene regulatory networks (GRNs). Rather than acting as simple on/off switches, these networks contain multi-stable feedback loops. Tian Hong et al. demonstrated in An Ovol2-Zeb1 Mutual Inhibitory Circuit Governs Bidirectional and Multi-step Transition between Epithelial and Mesenchymal States that a mutual inhibitory circuit between the transcription factors Ovol2 and Zeb1 expands the standard EMT landscape into a four-state system, providing a concrete molecular mechanism that stabilizes intermediate hybrid phenotypes.

From a systems and control theory perspective, these intermediate states are not merely random fluctuations. Indika Rajapakse et al. suggested in What Can Systems Theory of Networks Offer to Biology? that during differentiation, cells pass through intermediate states characterized by high entropy and high controllability—critical windows where the cell is highly receptive to external guidance. To map these paths across space and time, Peijie Zhou developed a multiscale dynamical model in Spatial transition tensor of single cells, showing that complex cellular transitions require cells to pass through high-entropy intermediate hubs rather than moving directly between stable endpoints.

Furthermore, these transitions exhibit complex temporal dynamics like hysteresis. Federico Bocci et al. noted in Single-Cell RNA-Seq Analysis Reveals the Acquisition of Cancer Stem Cell Traits and Increase of Cell–Cell Signaling during EMT Progression that cells may not fully revert to their baseline state after a stimulus is removed, indicating that intermediate states possess distinct thermodynamic and network energy profiles. By analyzing RNA splicing dynamics, Manuel Barcenas et al. framed the EMT transition as a saddle-node bifurcation in Tipping points in epithelial-mesenchymal lineages from single-cell transcriptomics data, where an unstable intermediate region acts as a critical "tipping point" that drives final cell fate commitment. This aligns with broader dynamical modeling frameworks, such as the neural network Hidden Markov Models proposed by Jens Kohlmorgen et al. in Analysis of Drifting Dynamics with Neural Network Hidden Markov Models, which demonstrate that drifting transitional phases in complex biological systems are best understood as mixtures of primary stable modes rather than isolated discrete states.

Epigenetic Regulation and Structural Biomarkers

To maintain these intermediate states, cells actively remodel their chromatin structure. The transition is not merely driven by the presence or absence of transcription factors, but by precise epigenetic control. Vin Yee Chung et al. demonstrated in The role of GRHL2 and epigenetic remodeling in epithelial–mesenchymal plasticity in ovarian cancer cells that the transcription factor GRHL2 regulates DNA methylation and histone modifications specifically within intermediate states, proving that these phases are actively maintained regulatory regimes rather than passive, accidental steps. In highly plastic populations like cancer stem cells, identifying these specific epigenetic drivers remains challenging, though machine learning frameworks like the one developed by Yusuke Suita et al. in Machine learning on multiple epigenetic features reveals H3K27Ac as a driver of gene expression prediction across patients with glioblastoma highlight specific histone modifications (such as H3K27Ac) as critical predictors of these state changes.

These internal epigenetic shifts manifest physically in the nucleus. Using high-resolution texture descriptors to analyze three-dimensional chromatin morphology, Priyanka Rana et al. showed in Estimation of three-dimensional chromatin morphology for nuclear classification and characterisation that intermediate cellular states can be explicitly classified by their unique spatial chromatin condensation patterns, providing a structural biomarker for transitional cell populations that evade traditional binary classification.

The Functional Advantages of Being Intermediate

Why do cells—especially cancer cells—benefit from staying in the middle of the spectrum?

First, hybrid states provide a functional compromise. As Tian Hong et al. noted in An Ovol2-Zeb1 Mutual Inhibitory Circuit Governs Bidirectional and Multi-step Transition between Epithelial and Mesenchymal States, retaining a balance of epithelial adhesion and mesenchymal mobility allows cells to undergo collective cell migration, a highly efficient strategy for tumor invasion.

Second, intermediate states are tightly linked to developmental immaturity and stemness. Tasha Thong et al. discovered in Hybrid Stem Cell States: Insights Into the Relationship Between Mammary Development and Breast Cancer Using Single-Cell Transcriptomics that normal human mammary cells naturally exist in hybrid E/M states that mirror the transcriptomic profiles of embryonic stem cells. In a malignant context, the authors of Intermediate cell states in epithelial-to-mesenchymal transition and Federico Bocci et al. in Single-Cell RNA-Seq Analysis Reveals the Acquisition of Cancer Stem Cell Traits and Increase of Cell–Cell Signaling during EMT Progression demonstrated that these hybrid states exhibit enhanced cancer stem cell (CSC) traits, which directly contribute to drug resistance, tumor initiation, and metastasis.

Third, intermediate states facilitate smoother, more efficient phenotypic transitions. Computational modeling by Yutong Sha et al. in Inference and multiscale model of epithelial-to-mesenchymal transition via single-cell transcriptomic data revealed that intermediate cell states reduce stochastic noise and optimize transition efficiency, acting as stepping stones that prevent the cell from getting stuck or failing to adapt.

Finally, these intermediate states actively shape their surroundings. Using single-cell and spatial transcriptomics, Teng Wang et al. showed in Single-cell and spatial transcriptomic analysis reveals tumor cell heterogeneity and underlying molecular program in colorectal cancer that specific intermediate transition programs in colorectal cancer cells directly drive the activation of surrounding stromal fibroblasts. This demonstrates that the EMT spectrum is not just an internal cellular program, but a dynamic engine for remodeling the entire tumor microenvironment.

Key Foundational Work

What Can Systems Theory of Networks Offer to Biology? (2012) — contributed 1 concept elements to this direction.
Analysis of Drifting Dynamics with Neural Network Hidden Markov Models (1997) — contributed 1 concept elements to this direction.
History & citationVersion 1 · 2026-10-04

Current · Permanent link

View frozen version 1

Cite this direction

Latest version (a living page):

@misc{lacuna_emt_as_a_spectrum_08b026e0,
  title = {EMT as a spectrum},
  author = {{Lacuna}},
  howpublished = {Lacuna research direction},
  year = {2026},
  url = {https://lacuna.tiptreesystems.com/direction/emt-as-a-spectrum/rdr_08b026e03aa74b66a721904d96410d48},
  note = {Latest version of a living research direction; cite a version for a fixed text. Accessed 2026-11-20},
}

This exact version (version 1, frozen text):

@misc{lacuna_emt_as_a_spectrum_08b026e0_v1,
  title = {EMT as a spectrum},
  author = {{Lacuna}},
  howpublished = {Lacuna research direction},
  year = {2026},
  url = {https://lacuna.tiptreesystems.com/direction/emt-as-a-spectrum/rdr_08b026e03aa74b66a721904d96410d48/v/1},
  note = {Version 1, frozen 2026-10-04. Accessed 2026-11-20},
}
About

These directions overlap with this one through shared papers.

See something wrong? Report a mistake