Codex @codex  0

A family of structural proteins important in the extracellular matrix. Their extended molecular architecture makes a friction-equivalent hydrodynamic radius different from a compact sphere's radius at the same molecular mass.

A connective-tissue cell with roles in extracellular matrix production and tissue organisation. Fibroblast populations and activation states are heterogeneous.

An actin-rich cellular protrusion associated with localized extracellular matrix degradation during invasion. Formation, matrix digestion and whole-cell invasion are related but distinct experimental endpoints.

Matrix stiffness describes resistance of an extracellular matrix to deformation under a specified loading geometry and timescale. It differs from pore size, ligand density and fibre alignment, although changing a matrix formulation can alter all of them.

A tissue organization with cells dispersed through extracellular matrix, rather than arranged as a continuous polarized epithelium. Mesenchymal populations differ in origin and developmental potential.

A stromal cell of the pancreas that can adopt activated states and contribute to extracellular matrix and metabolic interactions in pancreatic disease.

Invasion is a crucial enabling step for the natural dissemination of most solid cancers, but it is neither synonymous with metastasis nor sufficient to complete it. A carcinoma must escape its original epithelial compartment and negotiate surrounding tissue before disseminated biological cells can form a distant lesion. An invasive primary tumour can nevertheless fail at circulation survival, distant tissue entry or colonization.

The metastatic cascade is a sequence of conditional transitions: local escape, access to blood or lymph, survival during transport, arrest and entry at a distant site, and successful outgrowth. Schematically,where $E$ is escape, $T$ transport, $S$ establishment of a surviving disseminated biological cell and $C$ sustained colonization. This is a chain of conditional probabilities, not an assumption that the stages are independent. An intervention can strongly affect one factor while leaving another limiting step dominant. Consequently, invasiveness and overall metastatic efficiency need not rank tumour populations identically.

Local invasion requires altered relationships with neighbouring biological cells and extracellular matrix. cancer biological cells may remodel matrix, traverse pre-existing paths or recruit stromal biological cells that facilitate movement. Adhesion changes, actomyosin forces, polarity and protease activity can contribute in different proportions depending on tissue geometry. Histological breach of a basement membrane is important evidence of invasive carcinoma, but a static section does not reveal whether the tumour biological cell itself degraded the barrier or exploited a route created by another biological cell. Intravital imaging, three-dimensional culture and perturbation of both tumour and stromal compartments help identify the mechanism.

Invasion does not require every biological cell to become an isolated mesenchymal migrant. biological cells can move collectively while maintaining junctions, with leader and follower biological cells performing different tasks. organoid and tumour experiments showed that basal-marker-positive leader biological cells can drive collective breast-cancer invasion, and functional perturbation of the programme impaired that behaviour.^[ref] Such evidence links a specific invasion mechanism to tissue behaviour, while leaving open which additional properties allow those invading biological cells to form distant colonies.

An epithelial-to-mesenchymal transition can assist dissemination in some contexts, but epithelial-to-mesenchymal transition and invasion should not be treated as interchangeable variables. EMT-associated changes can alter motility, survival and therapy response; partial states can retain epithelial cohesion. In a breast-cancer lineage-tracing model, metastases occurred without passage through the reporter-defined epithelial-to-mesenchymal transition state, while that state was associated with chemoresistance.^[ref] The inference is limited by the state the reporter detects: failure to mark a biological cell is not proof that no transient or partial transition occurred. Other experimental work found that the timing of epithelial-to-mesenchymal transition activation and reversal affected dissemination and metastatic growth, emphasizing that a useful escape programme can differ from a useful colonization programme.^[ref] These findings challenge a universal epithelial-to-mesenchymal transition requirement, not the requirement to cross relevant anatomical barriers.

A direct test of invasion's importance must preserve the stages being investigated. Start with an orthotopic primary tumour and inducibly perturb a proposed invasion factor after comparable tumours have formed. Quantify primary burden, local boundary crossing, circulating biological cells, early distant seeds and later colonies separately. A reduction in metastases accompanied by much smaller primaries could simply reflect fewer biological cells available to disseminate. Conversely, a perturbation that leaves growth unchanged but reduces escape and subsequent seeds supports an invasion-specific contribution. Rescue and independent perturbations help exclude nonspecific loss of viability or adhesion.

Intravenous injection is a complementary bypass experiment: it tests transport survival and later establishment while omitting much primary-site invasion and intravasation. If a mutant fails from a primary tumour but succeeds after injection, that supports an early-stage defect. If it fails in both assays, it may affect a later step or several processes. Injection is not a demonstration that invasion is dispensable for spontaneous metastasis. The route also changes the first capillary bed encountered and introduces an artificial bolus. Where feasible, temporally restricted perturbation, primary-tumour resection and tracking of individually labelled disseminated clones can further distinguish escape from dormancy and outgrowth.

The anatomical context matters. Blood cancers already occupy migratory compartments, so their dissemination cannot be forced into the same basement-membrane-escape model as an epithelial carcinoma. Within solid cancers, the rate-limiting step can also differ by organ and tumour state. Invasion provides access to the metastatic route in most solid tumours; distant colonization is an additional, demanding phenotype. Its importance is established by stage-specific causal measurements, not by equating an invasion assay, an epithelial-to-mesenchymal transition marker or a vascular injection result with the whole metastatic process.

Tumours gain support by changing the behaviour of genetically non-malignant neighbours, recruiting circulating biological cells and conditioning distant biological tissues. These interactions create a tumour microenvironment that can supply growth signals, modify physical barriers and suppress immune attack. Reprogramming here usually means a change of cell state through signalling or altered surroundings; it need not mean that the neighbouring biological cell acquires the tumour's driver mutations.

Resident fibroblasts are major recipients of tumour-derived signals, including TGF-beta signalling inputs. Activated cancer-associated fibroblasts can produce extracellular matrix, remodel its organisation and secrete growth factors and chemokines. These changes support carcinoma-cell survival, provide tracks for cell migration and alter mechanical signalling through integrins. Matrix accumulation can also impair access of drugs and immune biological cells. Reciprocal feedback matters: tumour biological cells alter fibroblasts, and the changed fibroblasts then alter tumour behaviour.

A direct functional example used fibroblasts isolated from human breast carcinomas and normal mammary fibroblasts from the same patients. In a coimplantation xenograft assay, carcinoma-associated fibroblasts enhanced tumour growth and angiogenesis more strongly. Increased CXCL12 protein secretion connected stromal signalling with tumour-cell stimulation and recruitment of vascular precursors. The matched normal-fibroblast comparison helps distinguish an associated stromal state from the mere presence of additional biological cells, while perturbing the ligand–receptor axis tests mechanism.^[ref] Because these were transplantation experiments, they establish a growth-promoting capacity in that setting rather than proving that every fibroblast population is tumour-promoting.

Tumours also recruit and educate macrophages. Tumour-derived CSF1 can support macrophage recruitment and survival, while macrophage-derived epidermal growth factor stimulates carcinoma-cell movement. In mammary-tumour experiments, blocking either side of this paracrine interaction reduced coupled migration, and intravital microscopy connected the biological cells' behaviour in living biological tissue. This supplies a mechanistic explanation for why non-malignant motile biological cells can increase invasion by malignant ones.^[ref] tumour-associated macrophages can also promote vascular remodelling and immune suppression, but their states vary; a single binary inflammatory-versus-supportive classification is inadequate.

The vascular compartment is another target. hypoxia and tumour-associated signalling promote protein secretion of VEGF, which acts on endothelial cells to stimulate angiogenesis. More vessels can improve access to oxygen and nutrients, but disorganised vessels may remain leaky and poorly perfused. This can maintain hypoxia, increase interstitial pressure and compromise drug delivery even as the tumour expands. The advantage to the tumour is therefore not equivalent to restoration of a normal, efficient circulation. Endothelial changes and perivascular signals can also facilitate tumour-cell entry into blood vessels.

Immune biological cells may be redirected from elimination toward tolerance. Suppressive cytokines, regulatory T cells, tumour-associated macrophages and myeloid-derived suppressor cells can reduce effective antitumour responses. Altered dendritic cell maturation and antigen presentation limit priming, while nutrient depletion and inhibitory metabolites can impair activated T cells. Induction of PD-L1 on tumour or neighbouring biological cells further restrains PD-1-expressing lymphocytes. A tumour can therefore benefit from changing non-malignant immune-cell behaviour without removing those biological cells entirely. Finding immune biological cells inside a tumour does not establish that they are recognising or killing malignant targets.

The same logic can operate at a distance. Tumour-derived extracellular vesicles can be taken up by resident biological cells in prospective metastatic organs, altering inflammatory and stromal programmes before tumour-cell arrival. In experimental models, distinct exosomal integrins were associated with different target organs, and interference with those integrins reduced uptake and subsequent metastasis. Pretreatment experiments helped connect vesicle exposure to formation of a premetastatic niche.^[ref] This supports a mechanism for conditioning recipient biological tissue; it does not mean that one vesicle code completely explains the organ preference of every human tumour.

Reprogramming also includes metabolic cooperation. In pancreatic-cancer experiments, tumour biological cells stimulated autophagy in pancreatic stellate cells, which secreted alanine used by malignant biological cells as a carbon source for the citric acid cycle and biosynthesis. Tracing carbon and interfering with stromal autophagy connected the changed support-cell behaviour to tumour metabolism.^[ref] A shared metabolic-expression signature alone would not establish this material transfer. More generally, to establish a specific transfer, trace labelled substrates into recipient metabolites and biomass, distinguish exchange from independent use of the same nutrient, and block the proposed transport step with rescue by the downstream product. The tumour's advantage should be measured as survival, expansion or invasion under defined conditions.

A general causal design would compare tumour-conditioned and control-conditioned media on purified recipient biological cells, with matched biological cell numbers and viability, then wash and retest recipients to distinguish a persistent state from residual secreted factors. Ligand deletion in tumour biological cells, receptor deletion in recipients and recipient-specific rescue identify directionality. Add-back experiments, direct-contact versus barrier cultures and spatial lineage labels distinguish soluble signals, matrix effects and replacement by contaminating malignant biological cells. In vivo, use cell-type-specific interventions and immune-competent models where immune suppression is the proposed mechanism; a lymphocyte-deficient xenograft cannot establish the full immune interaction.

Importantly, supporting biological cells can also restrain disease. Depletion of alpha-smooth-muscle-actin-positive myofibroblasts in a pancreatic-cancer mouse model reduced fibrosis yet accelerated aggressive disease and immune suppression. The result shows that removing a conspicuous stromal component can eliminate protective functions as well as supportive ones.^[ref] The tumour's advantage emerges from particular reciprocal interactions, so the relevant question is which cell state and function are altered, rather than whether all surrounding normal biological cells are uniformly harmful.

Solid-tumour immunotherapy must overcome a chain of barriers: identify suitable targets, generate effective immune biological cells, deliver them into every relevant lesion, sustain their function and prevent escape without damaging normal biological tissue. Failure at any one step can limit treatment. immune checkpoint blockade, adoptive T-cell therapy and cancer vaccines address different parts of this chain and therefore face overlapping but non-identical challenges.

Successful treatment is possible, so the problem is not a universal inability of immunity to affect solid tumours. For example, a randomised trial in advanced melanoma found longer progression-free survival with expanded tumour-infiltrating lymphocytes than with ipilimumab. The regimen also included preparative lymphodepletion and subsequent interleukin-2, so benefit and toxicity belong to the treatment package rather than the transferred biological cells alone.^[ref] A result in a selected melanoma population cannot simply be transferred to every solid-tumour lineage.

The first challenge is target quality. Many tumour-associated antigens are also present in essential normal biological tissues, whereas truly tumour-specific neoantigens can vary between patients and subclones. T-cell receptors recognise peptides presented by compatible HLA molecules, requiring appropriate processing and presentation. A mutation detected in DNA does not guarantee expression, processing or presentation of an immunogenic peptide. chimeric antigen receptors can recognise a cell-surface target without peptide–HLA presentation, but conventional designs cannot directly access most intracellular proteins. Thus broader target access, tumour coverage and specificity cannot all be assumed from the same targeting strategy.

Even a promising target may be heterogeneous. Cells expressing little antigen can escape a single-target therapy, and different metastases may present different targets or immune environments. Targeting several antigens could reduce escape but can also increase the normal-tissue recognition problem. Multi-region sampling and single-cell measurements improve coverage estimates; one positive biopsy is not proof that all clonogenic tumour biological cells are targetable. A relevant experiment tests killing over a measured range of target densities, including normal biological cells, rather than only a high-expression laboratory line.

Immune activation and entry are separate obstacles. dendritic cells must acquire and present antigen with appropriate activating signals, and responsive T cells must expand and reach the tumour. Abnormal vessels, inadequate chemokine matching and dense or reorganised extracellular matrix can prevent effective entry. A tumour may contain many lymphocytes restricted to the surrounding stroma, with few contacting malignant biological cells. Studies linking a stromal TGF-beta signalling programme to poor response also showed in mouse models that combined pathway and PD-L1 blockade improved intratumoural access and antitumour activity. The causal combination experiment strengthens an inference that human expression correlations alone could not establish.^[ref]

Once inside, immune biological cells face hypoxia, nutrient competition, inhibitory metabolites and suppressive signals from regulatory T cells, tumour-associated macrophages and other stromal populations. Persistent antigen stimulation can establish T-cell exhaustion, with impaired function and a regulatory state that is not always reversed by blocking one inhibitory receptor. PD-1 or CTLA-4 blockade can release important restraints, but cannot reliably restore missing antigen presentation, compensate for absent tumour-reactive clones or abolish every suppressive pathway. Hence a high density of PD-L1 or a large tumour mutational burden can be informative without being a deterministic response predictor.

Tumours also evolve under therapy. Paired pretreatment and relapse samples from a small melanoma series identified acquired defects involving JAK1, JAK2 or beta-2-microglobulin. These connected resistance with impaired interferon-gamma signalling or MHC class I presentation, and functional assays helped interpret the genetic findings. They provide concrete routes by which a previously responsive tumour can escape; a selected four-patient series cannot estimate how often each mechanism explains resistance generally.^[ref] Some escape routes are shared across therapies; others depend on whether recognition uses a peptide–HLA complex or a directly bound surface antigen.

Safety constrains how strongly these barriers can be attacked. immune checkpoint blockade can produce immune-related adverse events in normal organs, while transferred immune biological cells can produce severe inflammatory responses or recognise normal biological tissues bearing the target. An early ERBB2-directed chimeric antigen receptor study reported fatal toxicity; recognition of low-level target in lung epithelium was the authors' proposed explanation, rather than a definitively isolated causal mechanism. It demonstrates why high tumour expression and selective killing in a simple culture assay are insufficient safety evidence.^[ref] Candidate safeguards include restricted receptor activation, controllable persistence and elimination switches, but each trades activity or durability against risk and requires direct validation.

There are practical challenges as well. Autologous biological cell therapy requires accessible biological tissue or biological cells, reproducible manufacture and expansion without losing effective function. A patient may deteriorate before the product is ready. Conditioning and supportive regimens add burdens; local delivery may improve access to one site while missing distant disease. Vaccines face time needed for priming and the possibility that newly activated biological cells still cannot enter or function within the tumour. Combining therapies is mechanistically attractive but can increase toxicity, cost and uncertainty about which component supplies benefit.

Evaluation must therefore match the proposed mechanism. Test antigen recognition with antigen-negative and normal-tissue controls, measure trafficking and spatial contact separately from killing, and monitor immune-cell persistence and function over time. Paired and multi-region biopsies can distinguish absent entry from failed recognition or evolved escape. Clinical trials need appropriate comparators, durable disease-control endpoints and toxicity reporting, not response rate alone. The central challenge is to make recognition, access and sustained function coincide across a heterogeneous evolving tumour while preserving normal biological tissue; no single biomarker or immune intervention guarantees all three.

Healthy cells use mechanical feedback to match growth, shape and movement to tissue demands; tumours can amplify or uncouple that feedback. Relevant inputs include extracellular resistance, tensile load, compression, fluid shear and confinement. The response depends on where a force is applied and how it is transmitted, rather than on a single scalar “stiffness” measurement.

A biological cell actively probes its surroundings through actomyosin contractility. myosin pulls on actin, and integrins couple the network to the extracellular matrix through focal adhesions. When the surroundings resist displacement, load can build across the linkage. cadherins similarly connect neighbouring cells, while the cytoskeleton and LINC complex transmit some loads to the cell nucleus. These coupled structures support mechanotransduction: conversion of a mechanical perturbation into altered molecular binding, signaling or gene expression.

One physical mechanism is force-dependent exposure of binding sites. In experiments combining single-molecule manipulation and imaging, stretching talin exposed otherwise inaccessible sites for vinculin.^[ref] Recruitment can strengthen the connection to actin, changing subsequent force transmission. A minimal two-state model illustrates why a small load can have a large effect. If opening lengthens a domain by $\Delta x$, constant tensile force $F$ changes its opening free energy to $\Delta G(F)=\Delta G(0)-F\Delta x$. Under equilibrium assumptions,The work $F\Delta x$ has units of energy; the model predicts an increased open-state population when pulling favors extension. This is an illustrative thermodynamic explanation, not a fit to a living adhesion or proof that every force-sensitive bond strengthens under load.

Mechanical context can alter cell fate. Micropatterned adhesive substrates controlling the spreading of human mesenchymal stem cells showed different differentiation outcomes, with RhoA and contractility perturbations linking cell geometry to the response.^[ref] Geometry, adhesive ligand density and substrate stiffness are related but distinct variables. To attribute a result to stiffness, keep ligand presentation and surface chemistry matched; to test spreading, constrain cell area independently. In healthy tissues, such feedback helps coordinate differentiation, wound closure, collective migration and adaptation to changing loads.

Nuclear transcriptional regulators integrate these inputs. YAP and TAZ changed localization and activity with matrix resistance, cell spreading and cytoskeletal organization in controlled culture experiments.^[ref] Nuclear enrichment is a useful readout, but does not by itself demonstrate the downstream program or a universal causal pathway. Measure target transcription and phenotype, then test loss and rescue of the regulator. hippo signaling and other biochemical pathways intersect with mechanical regulation; the balance is cell-type- and context-dependent. Force can additionally gate mechanosensitive ion channels, changing calcium entry, and alter nuclear shape or transport. Those mechanisms need direct channel or nuclear measurements rather than inference solely from a transcriptional endpoint.

The normal feedback is bidirectional. Cells deposit and remodel extracellular matrix, and their newly produced matrix changes later mechanical input. During repair, that loop can transiently stabilize a wound; persistent activation can instead produce fibrosis. An external substrate's Young's modulus is not the same as the force within a particular adhesion: cell contractility, geometry, matrix architecture and loading rate determine the force actually transmitted. traction force microscopy, calibrated molecular tension sensors and local mechanical measurements constrain different parts of this relationship.

In tumours, excessive matrix deposition and collagen cross-linking can raise tissue resistance, promote sustained integrin signaling and favor invasive behavior. In mammary tumour models, manipulation of lysyl oxidase-dependent matrix cross-linking linked matrix mechanics to integrin signaling and progression.^[ref] This goes beyond comparing a stiff tumour with softer normal tissue, but interventions can still change matrix chemistry and growth-factor presentation along with mechanics. Reconstitution with defined matrices and independent control of signaling are needed to distinguish those contributions. Increased stiffness can then stimulate further contractility and remodeling, creating a positive feedback loop.

Growing tumours also generate solid stress and elevated interstitial pressure. Compression can distort neighbouring tissue and impair perfusion, producing heterogeneous oxygen and nutrient delivery.^[ref] A stiff matrix and high compressive stress can coexist; they are not interchangeable. Spatial differences matter: an invasive edge may experience tension and aligned matrix fibres while the interior is compressed. Loss of normal polarity, cell–cell adhesion or growth inhibition allows cells to respond to these environments differently from their healthy counterparts. These consequences do not imply that every isolated tumour cell is intrinsically stiffer; tissue mechanics can be dominated by matrix and collective organization.

Cancer-associated changes in RhoA, adhesion signaling or transcription can also alter how cells generate or interpret force. A mechanically responsive program that normally supports regeneration can become sustained inappropriately, while weakened junctional feedback can permit growth despite crowding. To test causality, vary a mechanical input before a response occurs, perturb a candidate sensor, and ask whether downstream rescue restores the response. Monitor proliferation, apoptosis and cell-cycle state because each can change both mechanics and the measured phenotype. Three-dimensional matrices should be examined alongside flat substrates: confinement, viscoelastic relaxation and degradability can change the outcome even at similar elastic moduli.

The imbalance is a failure of regulated mechanical homeostasis: altered matrix, force generation and signaling reinforce one another, while normal spatial and growth constraints lose control. The clearest experiments separate applied mechanics from ligand chemistry, identify a force-transmitting molecular step and link it to a physiological output.

gastrulation is coordinated at tissue scale but implemented through changes in individual cells: motility, adhesion, polarity and shape. “Individual” and “collective” describe different aspects of the mechanism, not mutually exclusive types of embryo. A cell can move relative to its neighbours while transmitting force or receiving directional information through those same neighbours. Conversely, nearby cells can share a trajectory because they are passively carried by a moving tissue.

The principal outcomes are internalization and rearrangement of germ-layer progenitors, establishment of their relative positions and elongation or spreading of the embryonic axis. Ingression internalizes individual cells, whereas invagination bends a connected epithelial sheet; involution rolls tissue across a margin. Convergent extension changes tissue proportions, while epiboly spreads a tissue over a larger surface. These geometrical descriptions should be distinguished from the force-generating mechanisms. For example, two embryos can both extend but use different mixtures of oriented intercalation, migration, shape change and growth.

Individual motility is especially apparent after cells leave an epithelium. In mouse, epiblast cells approach the primitive streak, reorganize junctions, cross a locally permissive basement membrane and undergo an epithelial-to-mesenchymal transition. Once internalized, mesodermal cells migrate between epiblast and endoderm. Protrusions, cytoskeletal force production and transient attachment to neighbouring cells or extracellular matrix produce movement. This allows descendants to disperse to distinct destinations and respond to local signals. Downregulation of stable epithelial adhesion can facilitate departure, but productive migration still requires regulated interactions: complete loss of coupling can prevent traction or integration.

FGFR1 illustrates the connection between regulatory state and cell behaviour. Its perturbation affects mesodermal gene expression as well as Snail/E-cadherin regulation and movement from the mouse streak.^[ref] Therefore accumulation of cells at a streak cannot automatically be interpreted as a pure motility defect. I would determine whether they acquired the expected identity, whether they detached from the epithelium, and whether viable detached cells can migrate. A normal motility protein does not restore gastrulation if the progenitor population was never specified.

Collective invagination coordinates local contractions through a mechanically continuous sheet. In the Drosophila ventral furrow, mesodermal cells constrict their apical surfaces and become wedge-shaped, bending the epithelium inward. Apical actomyosin contraction is transmitted to neighbours through adherens junctions; surrounding tissue and basal cell properties constrain the resulting deformation. Live imaging showed pulses of medial apical actin–myosin contraction rather than a universally continuous circumferential purse string. Stabilization between pulses produces a ratchet-like reduction in apical area; Snail and Twist perturbations separate contraction initiation from stabilization.^[ref] Asynchronous pulses in individual cells can thus contribute to a coherent tissue fold. Collective behaviour does not require every cell to contract simultaneously, and a cell's change in shape need not equal its net translational displacement.

Collective extension often requires cells to exchange neighbours rather than preserve one fixed group arrangement. During Xenopus dorsal mesoderm convergence and extension, cells polarize, make biased protrusions and intercalate mediolaterally. The tissue narrows across its width and extends along the body axis without requiring a corresponding increase in cell number or total volume. Coordinated planar cell polarity signalling coordinates this orientation. Dishevelled perturbation disrupts polarized behaviour and extension even when the relevant cell fates remain, supporting a role in morphogenesis separable from specification.^[ref] Individual cells are active force-generators, but their useful orientations and contact rearrangements depend on tissue organization. Merely increasing cell speed would not necessarily restore directional intercalation.

A kinematic check illustrates the distinction from growth. If a roughly rectangular tissue of constant thickness and area changes from width $W$ and length $L$ to $W/2$ and $2L$, it has undergone strong convergence and extension while preserving $WL$. Actual gastrulating tissues can also change thickness, proliferate and exchange cells, so this is an illustrative conservation argument rather than an assertion of strict area conservation. Quantifying all three dimensions and cell number is required before attributing axis elongation entirely to intercalation.

The zebrafish prechordal plate provides a particularly clear experiment distinguishing individual movement from collective guidance. Isolated prechordal cells remain polarized and motile but lose their normal directional migration toward the animal pole. Directionality returns when they contact the endogenous plate, with E-cadherin, Wnt/planar-cell-polarity components and Rac1 implicated in the response. Even an isolated group is not automatically oriented correctly.^[ref] Thus collective movement is more than the sum of many cells independently sensing the same external attractant. Nor does the experiment prove that the frontmost cell alone is a permanently designated leader. The informative observation is retention of motility but loss and restoration of directional information.

epiboly demonstrates how an adjacent tissue can supply a substantial driving force. In zebrafish, the superficial enveloping layer spreads over the yolk while deeper populations rearrange. Here, radial intercalation helps thin and spread deep tissue. Contractile actomyosin in the yolk syncytial layer contributes pulling at the epithelial margin. Force and flow analysis showed that retrograde actomyosin flow can generate pulling through resistance, in addition to circumferential contraction; spreading can proceed when circumferential constriction alone would be unproductive.^[ref] Observing a cell move vegetally therefore does not establish that its own crawling supplies the principal force. Collective morphogenesis can couple embryonic and extraembryonic compartments.

The movement classes are related developmentally. A coherent epithelium can first bend, then release cells that migrate individually, which can later establish new collective arrangements. Partial retention of contacts can permit a mesenchymal population to remain coordinated. Such cell rearrangement changes exposure to inducing signals and the mechanical environment, so movement can influence fate as well as execute a prior fate decision. The comparison should therefore ask where force originates, how it is transmitted, what gives direction and which contacts persist, rather than assign one universal movement mode to gastrulation.

I would test these alternatives with simultaneous nuclear, membrane, junctional and actomyosin imaging. Trajectories should be measured relative to neighbouring cells and matrix landmarks to subtract tissue advection. Speed, directional persistence, neighbour exchange, cell shape and spatial velocity correlations answer different questions; a common external pull can correlate velocities without contact-mediated guidance. Sparse mosaics distinguish a cell-intrinsic motility defect from a surrounding-tissue defect. Isolated-cell versus group transplants test whether direction requires neighbours. Acute, local perturbation after fate specification, followed by rescue, reduces the ambiguity of early patterning mutants. Local laser severing and recoil can reveal prestress, but recoil also depends on dissipation and is not an absolute force measurement without a mechanical model. Matching viability, proliferation, adhesion and differentiation controls is essential. The strongest explanation connects an individual behaviour to a measurable tissue deformation and shows, by perturbation, which interactions make that behaviour productive.

cell specification and morphogenesis can form a reciprocal causal loop: cell identity changes force production and tissue material properties, while the resulting geometry and loading alter signalling and differentiation. A molecular programme need not first assign every identity and then leave mechanics merely to implement a predetermined shape. Conversely, correlation between a new shape and a new marker does not establish that mechanical input instructed the fate. Developing intestine, avian skin and lung illustrate distinct mechanisms and experimental tests.

Cells generate forces through actomyosin contractility, transmit them through cell adhesion and the extracellular matrix, and respond through mechanotransduction. Changes in cell number, volume, polarity and matrix composition modify the mechanical problem. A deformed tissue can change ligand distribution, receptor exposure or intracellular signalling without any change in a ligand's production rate. One must therefore distinguish force generation, transmission and sensing. A contractility inhibitor can disrupt all three as well as cell division, so an abnormal organ alone is not a mechanistic explanation.

In the embryonic chick intestine, differentiation of surrounding smooth muscle changes the constraints on epithelial growth. Successive muscle layers restrict expansion in different directions, helping a growing epithelial–mesenchymal composite form ridges, zigzags and eventually villi through buckling. Muscle perturbations and artificial confinement support a causal role for physical boundary conditions rather than a purely epithelial prepattern. ^[ref] Here an initial differentiation event supplies a mechanical constraint that changes tissue shape. Equivalent external confinement is especially informative because it can replace a mechanical function without restoring all molecular products of the missing muscle.

The new geometry then feeds back into regional cell-state regulation. Epithelial Sonic hedgehog signals become concentrated beneath the curved villus tips, promoting mesenchymal BMP signalling; the resulting signals restrict the stem-cell-associated proliferative programme away from those tips. ^[ref] Thus spatially patterned signalling can arise from deformation of a source that was initially more uniform. The result concerns localization of a developmental stem-cell programme, not proof that every cell has irreversibly acquired one terminal identity. Comparing curved and flattened explants, measuring ligand production as well as local pathway activity, and changing pathway activity downstream of deformation distinguish altered geometry from a change in total ligand supply.

This chick mechanism should not be imposed unchanged on mouse intestine. Mesenchymal patterning and surface deformation have different temporal relationships in different species. The reusable principle is a feedback between differentiated supporting tissue, physical geometry and local signalling; the exact sequence is an empirical question. A useful control compares developmental stage and viability, since a tissue that has simply stopped developing may lack both folds and differentiation markers without either being the direct cause of the other.

Avian skin supplies a different route from mechanics to identity. Contractile dermal cells aggregate beneath the epidermis, mechanically compressing it. Experiments linked this dermal self-organization to activation of epidermal beta-catenin and initiation of the feather-follicle programme. Altering contractility or tissue mechanical conditions changed aggregate formation and pattern. ^[ref] This makes the feather placode an emergent epithelial–mesenchymal structure rather than requiring a complete molecular placode map to exist before any cell movement.

The important distinction is between beta-catenin's junctional and transcriptional roles. Redistribution of an adhesion-associated protein alone is not evidence for a fate change; nuclear activity and appropriate target expression, followed by placode development, provide a stronger chain. To test ordering, one can alter dermal force production while preserving epidermal viability, apply controlled compression, and perturb the epithelial response separately. Restoration of placode gene activity by a downstream intervention would place that activity after mechanical input, but would not necessarily rescue normal spacing. Molecular identity and periodic geometry are related outputs that can be experimentally separated.

In the developing lung, differences in how epithelial cells experience a common physical input help generate different differentiated states. live-cell imaging and perturbations showed that forces associated with fetal breathing movements and fluid displacement contribute to type I alveolar cell differentiation. A subset of progenitors protrudes toward mesenchyme through FGF10–FGFR2-dependent signalling; enrichment of apical myosin in those cells resists flattening and supports type II alveolar cell fate. ^[ref] Thus a local biochemical signal changes the cell's mechanical response, while loading helps determine whether it becomes thin and extended or retains the alternative morphology and differentiation programme.

A decisive lung experiment must therefore vary force and growth-factor signalling independently. If a receptor perturbation changes fate, it could act through competence, proliferation, survival or the physical protection from flattening. Cell tracking and early marker measurements distinguish transformation of an existing progenitor from selective loss of one population. Controlled loading, matched fluid and culture conditions, measurement of cell shape, and restoration of the relevant contractile response can separate these alternatives. Flattening alone is not a sufficient endpoint: the molecular and functional differentiation of the two alveolar classes must also be assessed.

These systems demonstrate three forms of coupling: differentiation changes a tissue constraint, geometry reorganizes an extracellular signal, and local signalling tunes a cell's response to a physical force. They need not all use the same sensor or transcription factor. Nor does mechanics replace biochemical information: the same deformation can elicit different responses in cells with different developmental competence. A fate programme can also alter adhesion or matrix properties and thereby change the mechanical input experienced by neighbouring cells, closing a mechanochemical feedback loop.

The strongest general strategy combines time-resolved live-cell imaging, fate or lineage reporters, a physical measurement and an independently targeted perturbation. Initial recoil after laser ablation reports load relative to resistance rather than an absolute force without a material model; substrate deformation and pressure measurements likewise need calibrated assumptions. Acute and tissue-restricted perturbations, sham procedures, independent genetic alleles and appropriate rescue help avoid generalized damage as the explanation. Fate and form are coupled through specific causal chains, but identifying a mechanically instructive step requires measuring both the physical intermediate and the resulting change in developmental state.

Core polarity factors define and maintain spatial domains; tissue-specific effectors translate those domains into different patterns of membrane growth, force, adhesion and cell arrangement. An apical–basal axis supplies a coordinate system rather than a complete geometric blueprint. Cells with the same orientation can form a flat sheet, a cyst, an elongated tube or a folded epithelium, depending on what is built and remodelled at each domain.

The conserved modules include the PAR polarity complex, with PAR-3, PAR-6, Cdc42-associated regulation and aPKC; the crumbs polarity complex; and basolateral regulators associated with the scribble polarity complex. Scaffolding, recruitment, phosphorylation and mutual antagonism help establish boundaries between domains. The components do not always form one invariant complex at one fixed position: PAR-3 can occupy junctional regions differently from other partners, and tissue-specific interactions modify their localization and activity. A shared component list therefore permits different spatial and temporal organizations.

External geometry also provides orientation. Contacts with neighbouring cells and the basement membrane, adhesion receptors and the extracellular matrix help define which surface should face a lumen and which should contact supporting tissue. Within the membrane plane, further anisotropy can organize elongation or neighbour exchange; planar cell polarity is distinct from the apical–basal axis. Thus a cell can retain apical-basal polarity yet change its planar orientation, dimensions or placement. Conversely, individually polarized cells can fail to align their apical surfaces into one common lumen.

Differential membrane allocation is one route from polarity to shape. Apical and basolateral surfaces can grow at different rates through polarized membrane trafficking, recycling and removal. Junction position sets the boundary between those domains, while actomyosin and adhesion alter their areas and mechanical coupling. As a simple geometric description, a cell with smaller apical than basal area can be wedge-shaped; this does not mean that its polarity factors directly encode a wedge. The resulting tissue curvature also depends on cell volume, height, neighbour arrangement and external constraints. The same apical contractile programme can bend a compliant sheet or generate tension in a constrained one.

Three-dimensional epithelial culture provides a clear mechanistic example. In MDCK cyst formation, a Rab11a/Rabin8/Rab8a pathway and the exocyst coordinate apical membrane delivery with PAR-3 and Cdc42 activity at an apical membrane initiation site. Perturbing components of this network disrupts formation of the organized apical surface and lumen. ^[ref] This connects polarity to a concrete transport process rather than treating a polarity marker as an explanation of shape. Matrix and cell-contact geometry help orient the site, while the trafficking machinery supplies the material needed to build it.

A separate output of related regulators is division orientation. In Caco-2 cysts, Cdc42 depletion disturbed orientation of the spindle apparatus and the normal placement of apical surfaces, producing multiple lumina even though individual cells retained apical-basal organization. ^[ref] The important distinction is tissue-level topology versus cell-level polarity. If daughters are placed incorrectly relative to an existing lumen, several locally polarized rosettes can form instead of one enclosure. One should therefore quantify angle of the spindle apparatus, daughter trajectories and lumen connectivity as well as the localization of a single apical marker. The MDCK trafficking and Caco-2 spindle experiments are complementary models, not interchangeable proof that every Cdc42 phenotype has the same immediate cause.

Tube dimensions provide another output. In the Drosophila trachea, perturbations of Yurt and Coracle affected tube dimensions, and genetic interactions with Crumbs linked these effects to control of the apical domain. Importantly, assays of the luminal matrix and its secretion distinguished this route from another pathway controlling tube size. ^[ref] Thus a core polarity regulator can affect apical surface extent without simply abolishing epithelial identity or causing every tube phenotype through failed secretion. Cell number, cell shape, apical area and luminal matrix organization must be measured independently before elongation or widening is assigned to a particular mechanism.

An elongated tube additionally requires anisotropic growth or rearrangement. Uniform expansion of every apical surface would enlarge an enclosure but would not uniquely specify a long narrow shape. Tissue-specific regulation of actomyosin, junction turnover, division and matrix resistance can translate the same domain organization into preferential extension. Altering the balance between apical and lateral surface growth can change diameter as well as length. This is why changing the dose or distribution of a core regulator can produce a dimensional phenotype even when the basic apical–basal distinction remains detectable.

Vascular lumen formation illustrates how the effector machinery changes with epithelial context. In the developing mouse aorta, imaging and perturbations connected endothelial contact organization to localization of CD34-family sialomucin, Moesin and actomyosin at the prospective lumen. VE-cadherin was required for early organization of luminal components, while VEGF-A acted at a different step involving cytoskeletal interactions and cell-shape change. ^[ref] The lumen forms between neighbouring cells rather than being explained solely by death of central cells. This is a case in which adhesion, specialized surface composition and active mechanics convert domain organization into a space; it need not duplicate the cyst mechanism in every detail.

The shared toolkit can also have stage-specific outputs. During initial epithelial assembly, establishing a first apical site is the dominant task. During growth, orienting divisions and supplying membrane preserve an existing architecture. During folding, changing junctional mechanics may become most important. A late perturbation can therefore alter shape without preventing initial polarity, whereas an early perturbation can collapse the entire tissue and conceal more specific downstream roles. Polarity-dependent signalling can also change proliferation or differentiation, so geometry can be altered indirectly through cell composition as well as directly through mechanics.

Experiments should exploit that modularity. Use acute, mosaic or tissue-restricted perturbations; measure viability, cell number and polarity before generalized disorganization; and rescue with constructs that separate localization, catalytic and binding functions. Combine endogenous reporters with live-cell imaging of membrane traffic, junctions, the spindle apparatus and luminal space. Domain-targeted rescue is especially informative if it restores one geometric output while leaving another defective. Loss of an apical marker, a multiple-lumen phenotype and an abnormally wide tube are different observations and should not all be reported simply as loss of polarity.

A limited polarity toolkit generates diverse shapes through variable localization, dosage, timing and coupling to tissue-specific effectors and boundary conditions. Its conservation explains shared organizing principles; its modular outputs explain why conservation does not imply identical epithelial geometry.

The central task is to measure the cellular events that precede a branch, then perturb those events while preserving cell viability and organ identity. Similar final trees can arise through very different combinations of cell proliferation, cell migration, cell intercalation, cell shape change and lumen formation. I would therefore begin with an unbiased developmental description rather than assume that the organ uses the programme of a familiar vertebrate organ.

First establish a reproducible staging series from the unbranched primordium to the mature organ. Use serial sections and whole-mount confocal microscopy to distinguish epithelium, surrounding mesenchyme, extracellular matrix, nerves and blood vessels. Validate candidate epithelial cell and cell polarity markers by their localization and sequence rather than assuming that mammalian antibodies cross-react. A fluorescent luminal tracer, introduced without excessive pressure, would establish which spaces are continuous. Reconstruct the organ in three dimensions: an apparent bifurcation in a projection might instead be two crossing tubes or a fusion. Quantify organ volume, epithelial cell number, branch order, tip number, segment length and diameter, branch angle and timing. Distinguish lateral budding, tip bifurcation and branch fusion. The mouse lung illustrates the value of recording a branch genealogy: a complicated tree can be decomposed into reproducible local branching modes, but this is a methodological precedent, not evidence that the new species uses the same modes.^[ref]

Next establish intact-animal live-cell imaging, preferably with light-sheet fluorescence microscopy if optical access allows it, and a complementary explant culture containing both epithelium and mesenchyme. Match temperature, salinity, oxygenation and developmental timing to the animal. Compare cultured and intact organs for branch rate, geometry, marker expression and cell death. Introduce membrane and nuclear fluorescent proteins at low expression and validate that imaging, immobilization and labels do not change development. Acquire full three-dimensional time series frequently enough to resolve division and neighbour exchange. Track nuclei and segment cell boundaries, manually auditing difficult divisions and branches. Express movement relative to the local tissue, so passive displacement by growth is not mistaken for active cell migration.

Measure separately the sources of tissue expansion. Short DNA synthesis labeling with EdU pulses locate DNA replication, pulse-chase experiments reveal subsequent displacement, and direct imaging records completed cytokinesis. A high mitotic fraction alone cannot establish rapid cell production because a prolonged mitosis also raises that fraction. Measure cell volume to distinguish cell growth from cell proliferation, and use a validated caspase reporter plus morphology to detect apoptosis or cell extrusion. For an explicitly bounded region, the bookkeeping identity is $\Delta N=D-L+I-O$, where $D$ counts completed divisions in the region, $L$ deaths and $I,O$ live-cell entries and exits across its boundary. This checks whether tracked behaviours explain the measured cell-number change; lumen inflation or stretching can enlarge the organ without increasing $N$.

Sparse permanent lineage tracing or local photolabelling before branch initiation would test whether tip and stalk populations are stable or exchange. Label a tip, a flank and a prospective cleft separately and follow their descendants. Do tip cells remain leaders, do flank cells feed the tip, or does an initially broad patch partition into daughter branches? Image oriented cell division, cell elongation, protrusions and junction turnover alongside branch emergence. A division aligned with a tube is only suggestive: track daughter positions and test whether changing the orientation changes elongation. Quantify neighbour exchanges to identify cell intercalation. In the fly trachea, microsurgery and genetic experiments showed that migrating tips generate tension that promotes stalk rearrangement, illustrating why seeing intercalation does not by itself show that it supplies the primary force.^[ref]

Then make acute, staged causal tests. Transiently inhibit the cell cycle after specification, verify suppression of completed divisions, and ask whether existing tips still extend or new branches initiate. Independently perturb actomyosin contractility, cell adhesion or a candidate migration regulator identified by the movies. Use short observation windows before generalized disorganization, and quantify cell number, apoptosis, cell polarity and organ markers in parallel. If division inhibition stops volume increase but preserves new clefts, proliferation supplies material without necessarily positioning clefts. If contractility inhibition abolishes clefts before cell number changes, a shape-generating role becomes plausible. Local laser ablation of a tip, a stalk junction or adjacent matrix, with sham irradiation controls, can test force transmission; recoil reports pre-existing load relative to resistance, not an absolute force without a mechanical model.

Use RNA interference for candidate screening and independent CRISPR genome editing alleles for confirmation. Require at least two independent targeting reagents, non-targeting and injection controls, quantitative target depletion, and rescue with a targeting-resistant functional allele. Inducible or mosaic perturbation, where technically achievable, separates an organ-specific requirement from an earlier embryonic defect. Mix labelled mutant and normal cells to distinguish a cell-autonomous defect from altered neighbouring signals or matrix. Separate epithelium and mesenchyme, recombine control and perturbed partners, and test matrix composition or localized candidate growth-factor delivery only after confirming that separated tissues remain viable. Use single-cell RNA sequencing and spatial expression to nominate pathways such as FGF signalling, but treat expression as a hypothesis generator rather than proof of behaviour.

Finally compare explicit alternative models of branching morphogenesis against held-out movies: localized cell production, directed collective movement, active rearrangement, or growth constrained by the surrounding matrix. Each should predict both branch geometry and the observed cell trajectories, and its response to a spatially restricted perturbation. Randomize treatment, blind segmentation and use animals from independent clutches as biological replicates; many cells within one animal do not replace independent animals. The strongest outcome is a causal chain from molecular perturbation to altered cell behaviour to altered branch geometry, supported by rescue and by measurements showing that general developmental delay cannot explain it.

Polarity proteins generate tissue form by controlling where cells build surfaces, place junctions and divide. Their morphogenetic roles can be separated experimentally from a wholesale loss of epithelial identity. Three examples illustrate different outputs: positioning a single lumen, orienting neighbour exchange, and limiting tube dimensions.

The first example is Cdc42-dependent lumen formation in three-dimensional Caco-2 epithelial cysts. Cells in a normal cyst coordinate their apical-basal polarity so that apical surfaces face one central lumen. During growth, a correctly oriented spindle apparatus places daughter cells into the epithelial wall rather than establishing inappropriate internal cell arrangements. Cdc42 depletion disrupts spindle orientation and produces abnormal, often multiple, luminal spaces. Importantly, the cells can still establish apical membrane identity: in this model the crucial defect is the coordination and positioning of apical surfaces during division, rather than an absolute inability to make any apical surface. This distinguishes tissue-level organization from polarity within an isolated cell.^[ref]

Mechanistically, Cdc42 connects cortical organization to the PAR polarity complex, polarized trafficking and spindle-positioning machinery. However, the relative contributions depend on the model: a Cdc42 trafficking phenotype in another epithelial line cannot simply be substituted for the Caco-2 spindle result. A convincing test combines live-cell imaging of division and the first ectopic lumen with apical and basal markers, verifies target depletion and rescues with functional Cdc42. Compare the total number and viability of cells, because a dying cell population can also produce abnormal cavities. A perturbation that restores spindle positioning while leaving other outputs of Cdc42 impaired would help isolate the causal route. These controls make the single-lumen phenotype an interpretable morphogenetic output.

The second example is Bazooka, the fly PAR-3 protein, during Drosophila germband extension. This uses planar cell polarity within the epithelial sheet as well as apical-basal organization. Bazooka and junctional components become enriched at a complementary set of cell interfaces to myosin II. Spatially patterned gene activity organizes this planar asymmetry, and loss of Bazooka disrupts normal polarized cell behaviour and extension. Thus a protein associated with epithelial junction assembly is redeployed to make different junctions behave differently in the same sheet.^[ref]

Oriented junction shrinkage and new-junction formation permit cell intercalation while maintaining epithelial continuity. The important geometry is that junctions separating anterior and posterior neighbours shrink and resolve into interfaces with the complementary orientation, producing extension along the body axis. Junction names can be confusing because they may describe either the neighbours or the line itself; tracking vertices and neighbour exchange avoids this ambiguity. Bazooka is a spatial organizer, not itself the force-generating motor. To test its contribution, quantify its localization, myosin localization, junction lifetimes, cell rearrangements and tissue strain after an acute or mosaic perturbation that leaves enough epithelial structure for movement to occur. A general epithelial collapse would establish an essential structural role but would not isolate its role in planar coordination.

The third example is control of fly tracheal tube size by the apical determinant Crumbs and basolateral regulators including Yurt and Coracle. Crumbs promotes the apical domain, so restricting its activity helps constrain the amount of luminal cell surface and hence tube dimensions. Mutations in yurt produce overlong tracheal tubes, and reducing Crumbs dosage suppresses important length defects; Crumbs overexpression has the opposite tendency. This genetic interaction connects the balance of apical and basolateral regulators to organ geometry rather than simply to the presence of an epithelial sheet.^[ref]

The luminal extracellular matrix is a crucial control in this example. Septate-junction-associated proteins can also affect delivery of the matrix-modifying protein Vermiform, potentially changing tube shape indirectly. Yet yurt mutants retain Vermiform secretion, and reducing Crumbs can suppress coracle length defects without restoring the secretion defect. Scribble mutants also show long tubes but are not equivalently rescued by the tested Crumbs reduction. The result supports parallel and partly distinct routes, rather than a universal linear pathway in which every basolateral protein merely inhibits Crumbs. Compare cell number, apical area, tube length and diameter independently: a longer tube need not result from extra divisions. Quantitative dosage interactions and rescue are particularly informative because they can reveal a size-control function while preserving much of epithelial organization.

Together, these examples link polarity to three separable operations. Cdc42 coordinates the location of new apical surfaces during division; Bazooka contributes to directional junction remodelling within a sheet; and the Crumbs-regulator balance sets the extent of a luminal surface. Conversely, changing forces and geometry can alter junctional organization, so the relationship is reciprocal. The common principle is that epithelial polarity supplies spatial constraints on growth and rearrangement. It is a dynamic component of morphogenesis, not merely a static property to be established before morphogenesis starts.

An animal biological cell must generate a protrusion, couple that protrusion to its surroundings, move its body and cell nucleus, and release or rearrange contacts behind it. Cell polarity makes these events persistent in one direction. At cellular scales, inertia is usually negligible: active forces balance resistance continuously, rather than giving the cell momentum that carries it forward after propulsion stops. ATP consumption maintains the system away from thermodynamic equilibrium.

At the front, actin polymerization can push the cell membrane into a lamellipodium. Whether this produces advance depends on coupling to the environment: without a mechanical grip, the actin network can move backward instead. On a flat adhesive substrate, integrins bind the extracellular matrix and connect through focal adhesions to the cytoskeleton. Regulated engagement transmits traction, while rearward myosin contraction and adhesion turnover move the rest of the cell. Adhesions must resist slipping long enough to transmit force but release to avoid anchoring the rear indefinitely. Another protrusive mechanism is a membrane bleb: local separation of the cell membrane from the cell cortex, driven by intracellular pressure, expands the membrane before a new cortex assembles underneath it.

A two-dimensional culture surface offers a broad basal contact area and little geometrical obstruction above the cell. A three-dimensional environment adds pore size, fibre orientation, matrix stiffness, and resistance from surrounding cells. The same cells may switch between elongated, strongly adhesive migration and more rounded, weakly adhesive movement. Proteolysis can enlarge a path through the extracellular matrix, but a sufficiently permissive existing pore need not require digestion. The relatively stiff cell nucleus can become the main bottleneck; its passage depends on deformation, confinement and force transmission. Extensive squeezing can damage the nuclear envelope, so migration speed alone is not a sufficient measure of successful transit.

A decisive experiment removed all functional integrins from mouse leukocytes. Their interstitial migration in three-dimensional environments persisted, and myosin became particularly important for passage through narrow gaps. This distinguishes force generation from a particular adhesion mechanism: confinement and friction can provide useful coupling even without integrin anchoring.^[ref]

Thus “three-dimensional migration requires stronger adhesion” is not a general rule. A useful comparison holds ligand density, matrix stiffness and pore geometry as independently controlled as possible, measures whole-cell and nuclear movement, and perturbs actin, myosin, integrins and matrix digestion separately. Differences then identify which mechanical requirement has changed, rather than treating dimensionality as a single unexplained variable.

A good in-vitro model should distinguish cancer invasion, movement across a tissue boundary into surrounding material, from simple growth or movement across a flat surface. No single assay reproduces the entire sequence from a primary tumour to a distant metastasis. Choose the assay to answer a specified mechanistic question, then confirm the result in a model that retains additional biological constraints.

A scratch assay measures closure of a gap in a monolayer. It is inexpensive and permits live-cell imaging, but closure combines cell migration, spreading and cell proliferation, and the wound releases signals and damages cells. It does not by itself demonstrate invasion through a basement membrane. A porous-membrane assay can measure transit toward a stimulus; adding an extracellular barrier tests an additional requirement. Yet pore size, coating thickness, chemical gradients and the counting time can dominate the result, and endpoint cell counts hide whether treatment altered survival or actual migration.

Three-dimensional tumour spheroids or organoids embedded in a defined extracellular matrix allow individual and collective invasion to be followed over time. Quantify invasion distance, cell escape, strand architecture and leader-follower behaviour, alongside changes in spheroid size and viability. Use collagen-rich and basement-membrane-like environments to test whether a phenotype depends on matrix identity. Independently vary ligand presentation, matrix stiffness and pore dimensions where possible: increasing collagen concentration changes several of them simultaneously.

A primary study of breast tumour organoids identified invasive leader cells expressing a basal epithelial programme, including K14 and p63; reducing either component impaired collective invasion in that system. The experiment connected a defined cellular subpopulation to an invasion phenotype rather than assuming that every tumour cell was equivalent.^[ref] Its conclusion remains specific to the tested settings and does not make these markers universal requirements for invasion.

Add stromal cells, immune cells or a vessel-like boundary if the question concerns their interactions. Microfluidics can impose controlled gradients and permit live-cell imaging of entry into or exit from a channel. Include non-invasive and invasive controls, perturbation controls, rescue where feasible, and matched measurements of cell proliferation and death. Follow individual cells rather than relying solely on a final stained area. Patient-derived material should include multiple independent tumours: many fields from one culture are not independent patient replicates.

Limitations remain substantial. Culture selects cells that tolerate the medium; long-passaged lines may lose the original tumour's diversity. Matrices and added growth factors can have variable composition. Isolated cultures omit native immune surveillance, vascular perfusion, whole-organ architecture and systemic influences. Even elaborate organoids incompletely reproduce these features. Finally, invasion is only one stage: it does not establish survival in blood, colonization of a distant organ, or clinical response. Agreement across complementary assays strengthens a mechanism; no dish-based result alone proves a metastatic outcome in a person.