Flat Biology Is Lying to You: The Case for Leaving 2D Behind

Team Dynamic Matrices | 2026-08-04

Plastic-bound monolayers still anchor most cell biology, not because they are accurate but because they are cheap and familiar. This post examines what a flattened cell actually gets wrong about gene expression, drug response, and disease mechanism, and argues that 2D culture should be treated as the exception that needs justifying, not the default.

A cell growing on a polystyrene dish is not resting in a neutral, simplified version of its natural state. It is responding to an environment that does not exist anywhere in the body: infinitely stiff, uniformly flat, bathed in oxygen and nutrients with no gradient to navigate, and missing most of the three-dimensional contacts that tell it what kind of cell to be. Researchers have known this for decades and used 2D culture anyway, because it is cheap, fast, and compatible with every imaging and screening pipeline built since the 1950s. What has changed is the evidence for how much that convenience costs. The data now show that flat culture does not simplify biology, it distorts it, often in the exact direction that produces false confidence in a drug candidate.

The Flattening Problem Is Not Cosmetic

When a cell adheres to plastic, it spreads, polarizes incorrectly, and loses the cell-cell and cell-matrix contacts that organize normal tissue architecture. This is not a minor perturbation. Comparative gene expression studies show that 3D growth changes expression of genes governing integrin signaling, cell shape, and cell-cell contact relative to the same cells in monolayer, and cytochrome P450 xenobiotic metabolism genes are measurably deregulated in spheroid models relative to 2D. That last point deserves more attention than it gets: if a cell's drug-metabolizing machinery is different in 2D than in 3D, every pharmacokinetic inference drawn from a flat monolayer screen is built on altered biochemistry, not a simplified version of the real thing.

The mechanical environment compounds the problem. Tissue culture plastic has an elastic modulus in the gigapascal range, several orders of magnitude stiffer than most soft tissue. Since integrin-mediated mechanotransduction directly shapes differentiation, proliferation, and drug sensitivity, culturing a cell on plastic is not withholding a variable, it is actively supplying the wrong one. A cell on a dish is not experiencing "no mechanical signal." It is experiencing the mechanical signal of concrete.

Where the Lie Shows Up in the Data

The clearest evidence comes from oncology, where the stakes of getting drug response wrong are direct and expensive. Head-to-head comparisons show that 3D tumor models exhibit resistance to paclitaxel, docetaxel, and HER-targeted therapies that flat cultures of the identical cell line do not reproduce, and in EGFR inhibition studies, oncoprotein addiction and drug efficacy in 3D models tracked with actual patient response, while 2D cultures showed only a highly attenuated version of the same signal. In practice, this means a compound can look effective in a 2D screen and fail in the clinic for reasons that were visible in a spheroid model the whole time. The historical reliance on monolayer screening is now widely implicated as a contributor to the gap between promising preclinical data and disappointing trial outcomes, particularly for cytotoxic and targeted oncology agents where drug penetration through a compacted, gradient-bound tissue matters as much as the target-binding chemistry itself.

Diffusion is the other variable 2D cannot represent. In a monolayer, every cell has unrestricted access to media, drug, oxygen, and nutrients. In a tumor, in an organ, in essentially any real tissue, cells exist along a gradient, and the cells furthest from a vessel experience a different concentration, a different oxygen tension, and often a different phenotype than the cells at the edge. A 2D assay cannot produce that gradient because there is no depth for it to form across. Screening a compound's efficacy without a diffusion barrier is screening a question the drug will never actually face.

Why the Default Persists Anyway

None of this is news to the field. So why does 2D remain the default screening format in so many labs? Partly infrastructure: decades of assay development, automation, and imaging systems are built around flat-bottom plates. Partly cost and throughput: monolayer culture is faster to set up and easier to standardize, at least on the surface. And partly inertia: a method's familiarity gets mistaken for its validity, especially when the downstream failure (a drug that dies in phase 2 for reasons the 2D screen never flagged) is temporally and organizationally distant from the bench where the screen was run. The economic case for 2D culture is real, but it is a case about convenience, not about biological accuracy, and those two justifications get conflated more often than they should.

The honest framing is not "3D culture is better." It is that 2D culture makes specific, consequential claims about mechanics, gradients, and cell shape that are almost never true of the tissue being modeled, and every time a lab defaults to a flat monolayer, it is implicitly asserting that those variables do not matter for the question being asked. Sometimes that is a defensible bet. Often it is simply the bet nobody examined.

The Case for Treating 3D as the Default

If a monolayer cannot reproduce the mechanotransduction, the metabolic profile, or the diffusion gradient of the tissue in question, then 2D data is not a conservative simplification, it is a different experiment wearing the same cell line's name. The burden of proof should sit with the flat model, not the dimensional one. That does not mean every assay needs a full organoid or organ-on-chip system on day one. It means the choice to run cells flat should be a deliberate methodological decision, made with the same rigor as a control, rather than a default inherited from what the last person in the lab happened to have on hand. The tools to move past that default (defined 3D matrices that are as easy to pipette as a 2D coating) already exist. What is missing is not capability, it is the willingness to stop treating flat as neutral.

Sources: Comparative Assay of 2D and 3D Cell Culture Models, PubMed; Genome-Wide Gene Expression Analysis in Cancer Cells Reveals 3D Growth Effects, PMC; The relevance of 3D cell cultures in evaluating breast cancer drug sensitivity, PMC; Organotypic 3D cancer cell cultures mirror drug responses in vivo, PMC; Three-Dimensional Cell Cultures in Drug Discovery and Development, PMC; Role of three-dimensional cell culture in therapeutics and diagnostics, PMC