Tissue-Intrinsic Shape Mechanics in Growing Pre-Migratory Tumor Spheroids

One of the hallmarks of pre-migratory tumors is the progressive loss of compact morphology. To investigate how tumors may intrinsically regulate their shape during growth, we employ a three-dimensional (3D) vertex model of multicellular aggregates that incorporates key structural features of tumor spheroids, including its surface, a proliferative rim, and a necrotic core. Focusing exclusively on tumor-intrinsic mechanical interactions, we examine how their collective effects guide morphological evolution en route to metastasis. We show that spheroids acquire lobulated morphologies through an interplay between differential tensions at the spheroid surface and the living-necrotic interface (LNI), together with differential growth within the proliferative rim. In addition, spheroid shapes can be substantially modulated by tissue rheological properties emerging from active, cell-scale forces. Our cell- and tissue-scale simulations of tumor morphologies are enabled by a computational framework that overcomes a major limitation of 3D vertex models - the lack of cell-division - by introducing a graph-based polyhedral-division algorithm within the Graph Vertex Model (GVM).

💡 Research Summary

The authors address a fundamental yet under‑explored aspect of early‑stage solid tumours: how intrinsic mechanical interactions shape a growing spheroid before cells acquire migratory capabilities. To study this, they extend a three‑dimensional vertex model by embedding it in a Graph Vertex Model (GVM) framework that treats the tissue topology as a graph. The key technical advance is a graph‑based polyhedral division algorithm that enables realistic cell division in fully 3D aggregates. In practice, a randomly oriented cleavage plane is introduced for a selected cell; all intersected faces and edges are identified, a local sub‑graph is extracted via pattern matching, and a set of graph transformations simultaneously deletes old relationships and creates new ones, thereby generating two daughter cells while preserving topological consistency. This method works for both interior and boundary cells and thus removes the long‑standing limitation of 3D vertex models that could not handle proliferating populations.

Mechanically, the model’s energy functional consists of a surface term ΣΓ_mn A_mn, where Γ_mn denotes effective interfacial tension (cortical tension plus adhesion) and A_mn the area of the face between cells m and n, and a bulk incompressibility term κ_V (V_m – V_0)^2 that keeps cell volumes near a preferred value V_0. Three distinct tension values are used: a bulk tension Γ_0, a surface tension Γ (for faces exposed to the extracellular environment), and a living‑necrotic interface (LNI) tension Γ_LNI (for faces separating viable rim from necrotic core). Lengths are nondimensionalized by V_0^{1/3} and tensions by Γ_0; time is scaled by τ = η/Γ_0, where η is a friction coefficient.

Nutrient transport is modeled by a steady‑state reaction‑diffusion equation with diffusion coefficient D and consumption rate k. By discretizing along the logical layer index ℓ (ℓ = 1 for the outermost layer, ℓ = 2 for the next, etc.), the authors obtain a geometric decay c_ℓ = c_0 / 2^{ℓ‑1}. Cells divide only when c_ℓ exceeds a threshold c_TH, which defines a proliferative rim of thickness λ (the number of layers that satisfy c_ℓ > c_TH). Within the rim, cells grow at a rate k_0 (chosen small enough to approximate quasistatic growth) and divide once their preferred volume doubles, using the aforementioned polyhedral division rule. The cleavage plane is oriented roughly perpendicular to the spheroid surface but otherwise random, mimicking realistic mitotic orientation.

Simulations start from a disordered aggregate of 50 cells and run until 2000 cells are reached. For each combination of parameters (Γ_LNI, λ, Γ, and an active noise timescale τ_relax), 50 independent runs are performed to obtain statistically robust averages. Morphology is quantified by two dimensionless shape metrics: reduced volume v = 6√π V / A^{3/2} (v = 1 for a perfect sphere, decreasing with surface irregularity) and reduced perimeter S = P_cs / √(4π A_cs) for 2‑D cross‑sections (S = 1 for a circular section).

The results reveal a clear mechanical triad governing shape evolution:

1. Interfacial tension at the LNI (Γ_LNI). High Γ_LNI creates a stiff, well‑defined internal boundary that suppresses surface undulations; spheroids remain compact (high v, low S) regardless of rim thickness. Low Γ_LNI weakens the coupling between the necrotic core and the outer surface, allowing internal growth stresses to deform the surface.

2. Proliferative rim thickness (λ). When λ is small (only one or two layers proliferate), the LNI lies close to the surface and its stabilizing effect dominates. As λ increases, the rim thickens, decoupling the LNI from the surface; surface undulations become more pronounced, leading to lobulated shapes (decreased v, increased S). If λ becomes very large, the necrotic core essentially disappears and the spheroid again approaches a smoother shape.

3. Active mechanical noise (τ_relax). By introducing stochastic fluctuations in interfacial tension, the authors model tissue rheology as a viscoelastic material with a relaxation timescale. Larger noise amplitudes accelerate the growth of small perturbations, producing highly irregular, branched morphologies even when Γ_LNI is moderate.

Systematic parameter sweeps show non‑monotonic dependencies: for low Γ_LNI, both v and S exhibit extrema as λ varies, indicating an optimal rim thickness that maximizes lobulation. For high Γ_LNI, morphology changes monotonically with λ. The interplay of the three factors defines a “tissue‑intrinsic triad” that predicts whether a growing pre‑migratory spheroid will stay smooth or develop the lobulated morphology associated with early invasion and poorer patient prognosis.

In the discussion, the authors emphasize that these purely mechanical determinants operate independently of extracellular matrix cues, highlighting the importance of intrinsic tissue rheology in tumor evolution. They suggest that therapeutic strategies aimed at modulating effective surface tensions (e.g., through drugs that alter cortical contractility or cell‑cell adhesion) or reducing mechanical noise (e.g., by stabilizing cytoskeletal dynamics) could potentially impede the morphological transition toward invasive phenotypes.

Overall, the paper delivers a novel computational platform that overcomes a major limitation of 3D vertex models, provides mechanistic insight into shape regulation of growing tumor spheroids, and identifies key physical parameters that could be targeted to influence cancer progression.